rev‐Gelatin: A Gelatin with Reverse Thermo‐Responsive Behavior Inspired by Candy and Ice Cubes Phase Dynamics

ABSTRACT

Conventional gelatin's gel‐to‐sol transition upon heating restricts its utility in biomedical applications that benefit from a gel state at physiological temperatures such as Pluronic F127 and poly(NIPAAm). Herein, we present “rev‐Gelatin”, a gelatin engineered with reverse thermo‐responsive properties that undergoes a sol‐to‐gel transition as temperature rises from ambient to body temperature. Inspired by the phase dynamics of common materials like candy and ice cubes, whose surfaces soften or partially melt under warming, facilitating inter‐object adhesion‐ rev‐Gelatin leverages this concept to achieve fluidity at room temperature for easy injectability. At ambient temperature, rev‐Gelatin exists as a microgel solution with sufficient fluidity in the sol state. However, upon exposure to elevated temperatures approaching physiological temperature, rev‐Gelatin microgels coalesce through surface melting, forming a stable gel. This sol‐to‐gel transition is especially advantageous for hemostatic applications. Upon contact with blood, the temperature elevation induces rapid gelation of rev‐Gelatin, effectively creating a barrier that reduces bleeding time and blood loss. Additionally, rev‐Gelatin shows promise as a submucosal injection agent for gastrointestinal surgeries, making it a new class of thermo‐sensitive biomaterials.

rev‐Gelatin, inspired by the surface melting behavior of candy and ice cubes, exhibits reverse thermo‐responsive properties. Opposite to the typical thermal behavior of conventional gelatin, it transitions from sol‐to‐gel as temperatures rise from room to body temperature. Its unique ability to remain fluid at room temperature and solidify at physiological levels offers significant potential for hemostatic applications and gastrointestinal surgeries.

1. Introduction

Gelatin, a widely used biomaterial, has a longstanding history that dates back centuries. It is derived from collagen‐the primary structural protein found in animal connective tissues‐through the hydrolysis processes that typically employ bovine hides, porcine skins, and fish scales as raw materials [1, 2, 3]. The transformation from collagen to gelatin involves breaking down the tightly wound collagen fibers into smaller, more manageable polypeptide chains through processes like thermal treatment and enzymatic hydrolysis. Due to its biocompatibility and biodegradability [4, 5, 6, 7], gelatin has found widespread applications across biomedical fields, from drug delivery to tissue engineering. Its low production cost, combined with ease of processing, elevates gelatin as a biomaterial well‐suited to meet the stringent demands of contemporary biomedical engineering [8, 9, 10, 11].

Gelatin exhibits unique thermo‐responsive behavior, maintaining a gel state below approximately 30°C and transitioning to a sol state as the temperatures increases [12, 13, 14]. For biomedical applications, however, an inverse phase behavior is often preferred: a sol at ambient temperatures that transitions to a gel under physiological conditions. Biomaterials exhibiting lower critical solution temperature (LCST) behavior, such as poly(N‐isopropyl acrylamide) (pNIPAAm) [15, 16, 17], Pluronic F127 [18, 19, 20, 21], and analogous triblock copolymers with modified central blocks [22, 23, 24, 25], are particularly valuable, as they undergo sol‐to‐gel transition at specific temperature thresholds. For instance, pNIPAAm transitions at around 32°C [16, 17], while the transition temperature of Pluronic F127 can be adjusted between 25°C and 37°C by altering its concentration [18, 19, 20, 21]. These phase transitions facilitate the encapsulation of drugs and their controlled release in vivo, thereby enhancing therapeutic efficacy. In contrast, gelatin's inherent gel‐to‐sol transition with increasing temperature constrains its utility as an injectable biomaterial, as it struggles to retain a stable gel state under physiological conditions. Addressing this limitation is thus critical for broadening the scope of gelatin's applications in advanced biomedical engineering.

To circumvent the limitations of gelatin's intrinsic temperature‐responsive behavior, researchers have investigated modifications that mitigate temperature sensitivity and introduce alternative stimuli, such as light. For instance, methacrylated gelatin (GelMA) disrupts the helical structure of gelatin [26, 27, 28], allowing it to remain in a sol state at room temperature for easy injection. When exposed to UV light, these methacryloyl groups crosslink to form a stable hydrogel, making GelMA an attractive candidate for applications in tissue engineering and regenerative medicine. However, the GelMA has notable limitations. The limited penetration of UV light restricts its utility for in vivo applications, particularly for complex or large tissue structures, and the use of photoinitiators generates residual radicals that pose cytotoxic risks. To address these challenges, catechol, a phenolic moiety, has been incorporated to mitigate toxicity without relying on radical crosslinking mechanisms [29, 30, 31, 32]. Despite these advancements, achieving a reverse directional phase behavior in gelatin, where it transitions from sol‐to‐gel as temperature increases, remains an ongoing challenge.

Interestingly, it is feasible to reverse the phase behavior of gelatin—transitioning from a sol at ambient conditions to a gel at physiological temperatures—without chemical modifications. The hypothesis explored in this study, which addresses the limitations of conventional gelatin systems, draws inspiration from candy and ice. Observations of these items reveal potential insights: for example, when exposed to warmer temperatures, the surface of candy softens rather than fully melting, allowing individual pieces to cohere into a larger candy agglomerate (Figure 1A). Similarly, ice cubes at room temperature exhibit surface melting, with the resulting water layer enabling adhesion between cubes; yet refreezing stabilizes their connection (Figure 1B). Both examples illustrate how partial surface melting under thermal fluctuations can generate interconnected networks while preserving structural integrity (Figure 1C).

FIGURE 1.

(A) An image of separated candy particles in ambient condition (left) and a large candy agglomerate when exposed to heat (right). (B) An image of single ice cubes on room temperature (left) and its agglomerate when exposed to heat (right). (C) Conceptual image of temperature‐dependent candy adhesion: individual candy pieces maintain separation at low temperatures (top) but undergo surface‐mediated adhesion creating interconnected networks at elevated temperature (bottom).

In this study, we demonstrated that gelatin, without any chemical modification, can achieve a sol‐to‐gel transition as the temperature increases from ambient to physiological levels. We termed this gelatin “rev‐Gelatin,” to emphasize its reverse‐directional phase transition properties. rev‐Gelatin is engineered to mimic the partial melting and agglomeration observed in candy, remaining a sol state (i.e., fluid) at room temperature yet forming a stable gel at physiological temperatures. Using a water‐in‐oil emulsion technique, rev‐Gelatin microgels (∼30 µm) were produced. To investigate its thermal responsiveness, we evaluated the temperature‐dependent phase transition behavior of rev‐Gelatin, confirming its sol‐to‐gel shift upon heating. Under increasing temperatures, with 28°C as the critical point, conventional gelatin exhibits gel‐to‐sol phase behavior above this threshold, whereas rev‐Gelatin demonstrates a sol‐to‐gel transition. Rheological analysis further characterized the thermo‐responsive behavior and mechanical properties of rev‐Gelatin, supporting its suitability for biomedical applications. Functional evaluations include hemostasis tests in a rat liver hemorrhage model and submucosal injection agent (SIA) tests in a porcine gastrointestinal model. In the hemostasis tests, rev‐Gelatin rapidly gelled upon contact with warm blood, significantly reducing bleeding time and blood loss compared to controls. In SIA tests, rev‐Gelatin was easily injected through an 18G syringe, temporarily elevated tissue for surgical procedures, and then dissolved naturally without residues. These findings underscore rev‐Gelatin's potential as a versatile biomaterial, adaptable to dynamic physiological environments, and demonstrate its capacity for reliable structural support and functional stability. Consequently, rev‐Gelatin holds significant promise for a wide range of biomedical applications.

2. Result and Discussion

2.1. Fabrication and Characterization of Gelatin Microgels Exhibiting Reverse Directional Phase Transition

The synthesis of gelatin microgels was achieved using a water‐in‐oil emulsion technique. A 20 wt% gelatin solution was prepared by dissolving 200 g of gelatin in 800 mL of deionized distilled water (DDW) at 60°C for 60 min. This solution was then gradually introduced dropwise into a sorbitan oleate‐paraffin oil mixture (70 mL:7 L) under continuous stirring at 450 rpm at 65°C. The emulsion was stirred for 30 min to facilitate the formation of uniform gelatin microgels, followed by cooling to 4°C for structural stabilization. Once the stirring ceased, the water‐containing gelatin microgels settled at the bottom due to a density‐driven phase separation. The suspension was maintained at 4°C for an additional 60 min to ensure complete gelation of the microgels for subsequent experiments. The microgels were then washed sequentially with acetone and ethanol and stored at 25°C for 24 h under 25% relative humidity (Figure 2A).

FIGURE 2.

(A) Schematic illustrations of the preparation process for rev‐Gelatin microgels. This image was created with BioRender (https://biorender.com/). (B) Photographic image (left), SEM image (middle), and optical microscopy image (right) of rev‐Gelatin. (C) Thermo‐responsive behavior of conventional gelatin (top) and rev‐Gelatin (bottom), showing phase transitions between hydrogel and sol as temperature rises. (D) Flowability test of conventional gelatin (top) and rev‐Gelatin (bottom) at 25°C (left) and 30°C (right). (E) A scratch test comparing rev‐Gelatin at 25°C and 30°C (left top and bottom) and conventional gelatin at 25°C and 30°C (right top and bottom).

The size of the gelatin microgels in their dry state was analyzed using scanning electron microscopy (SEM), which showed an average diameter of 31.1 µm ± 12.9 µm (Figure 2B, middle). Upon hydration, the microgels expanded to an average diameter of 79.4 ± 43.6 µm (Figure 2B, right), representing a 2.6‐fold increase in diameter and a 16.6‐fold increase in volume. Considering that polyethylene glycol (PEG)‐based gels typically show a volume increase exceeding tenfold upon swelling, this value can be deemed reasonable [33, 34, 35].

The low‐temperature gelation mechanism is attributed to the coil‐to‐coil self‐assembly of individual gelatin chains [36, 37, 38, 39, 40, 41]. When thermal energy is applied, these self‐assembled chains undergo a process of decoiling, resulting in a phase transition to a solution state. As hypothesized in Figure 1, while each individual microgel behaves as a gel on its own at low temperatures (25°C–27°C), the overall system appears as a solution at a macroscopic level due to microgel dispersion. However, as the temperature exceeds 28°C, the gel surface of each microgel begins to melt, leading to gel‐to‐gel adhesion, which results in a transition toward a gel state at a macroscopic level. This phenomenon is comparable to candies at low temperatures, where each piece remains separate without sticking. However, as the temperature rises, the surface of the candy begins to melt, increasing molecular mobility and causing the pieces to adhere (Figure 1C). Unlike conventional gelatin, which typically undergoes a gel‐to‐sol phase transition, prefabricated microgels exhibit a reverse “sol‐to‐gel” phase transition, showing LCST‐like behavior. These unique properties led to the designation of these microgels as “rev‐Gelatin”. Notably, further temperature elevation (33–35°C) induces complete chain mobility throughout the microgels, ultimately leading to a transition to the sol state (Figure 2C, bottom). This behavior would positively influence degradation kinetics under in vivo conditions.

To demonstrate the distinctive properties of rev‐Gelatin, solutions of both conventional gelatin and rev‐Gelatin were prepared and loaded into 3 mL syringes to observe their flow properties. rev‐Gelatin exhibited a slurry‐like sol state with fluidity at 25°C, transitioning to a non‐flowing gel state at 30°C (Figure 2D, bottom). In contrast, conventional gelatin showed the opposite behavior, remaining in a gel state at 25°C and transitioning to a solution state at 30°C (Figure 2D, top). To further illustrate this behavior, a scratch test was conducted by placing rev‐Gelatin in a petri dish. As the rev‐Gelatin remains in a liquid state at 25°C, it exhibits self‐healing properties after scratching, due to its fluid characteristics. In contrast, the gel state of rev‐Gelatin at 30°C did not exhibit healing (Figure 2E, left). Conversely, gelatin, which is a gel state at 25°C, did not show self‐healing after being scratched (Figure 2E, right).

2.2. Physical Property Characterization of rev‐Gelatin

To investigate the inter‐particle adhesion characteristics, SEM imaging was conducted. The results showed that, similar to melted candies, the microgels partially retained their structure while forming agglomerates with surface adhesion (Figure 3A). To investigate the reverse thermo‐responsive behavior of rev‐Gelatin, we conducted temperature‐dependent rheological analysis. As a reference, conventional gelatin (20 wt%) exhibited a typical gel‐to‐sol transition, with a crossover between elastic (G′) and viscous (G″) moduli at 27°C (Figure 3B), consistent with previous studies [42, 43].

FIGURE 3.

(A) A photographic (top) and SEM (bottom) image of rev‐Gelatin. (B) Rheological analysis of 20 wt% conventional gelatin at a constant strain (λ = 0.5%) from 20°C to 40°C (Black line: elastic modulus G′; Black dots: viscous modulus G″). (C) Rheological analysis of 5 wt% rev‐Gelatin at a constant strain (λ = 0.5%) from 20°C to 40°C (Dark red line: elastic modulus G′; Dark Red dots: viscous modulus G″; Red arrow: G′ peak point; Inset Photo: 5 wt% rev‐Gelatin at 20°C). (D) Rheological analysis of 20 wt% rev‐Gelatin at a constant strain (λ = 0.5%) from 20°C to 40°C (Red line: elastic modulus G′; Red dots: viscous modulus G″; Red arrow: G′ peak point).

In contrast to conventional gelatin, the rev‐Gelatin microgel suspension exhibits a unique sol–gel–sol phase behavior at the macroscopic level, driven by reversible temperature‐dependent, surface‐mediated interparticle interactions. Despite its macroscopically sol‐like appearance at room temperature, rheological measurements reveal that the material exhibits a gel state—particularly at higher concentrations—due to increased interparticle adhesion.

Our experiments identified a concentration‐dependent window in which this thermoreversible behavior is preserved. At 5 wt%, the system behaves as a sol both macroscopically and rheologically. The storage (G′) and loss (G″) values were nearly equal (G′ = 8.69 × 10⁻³ Pa; G″ = 1.08 × 10⁻² Pa) at 20°C, confirming a weakly structured, sol state (Figure 3C). Upon heating to 25°C, partial surface melting of the microgels initiated interparticle fusion, resulting in a sharp increase in elasticity, indicating a gel state (G′ = 4.33 × 10⁻¹ Pa). Above 32°C, complete microgel melting occurs, leading to a transition back to fluid‐like behavior (G″ = 4.61 × 10⁻² Pa > G′ = 9.19 × 10⁻³ Pa).

At a concentration of 20 wt%, rev‐Gelatin exhibits macroscopically a sol appearance. However, rheological characterization reveals gel properties at 20°C, indicated by the storage modulus exceeding the loss modulus (G′ = 1.07 × 10⁻¹ Pa > G″ = 3.83 × 10⁻² Pa) (Figure 3D). The increased microgel concentration reduces the proportion of free water, resulting in the overall mechanical properties being dominated by the microgel phase. This higher microgel content also leads to an elevated maximum storage modulus, reaching 1.04 × 10² Pa (Figure 3D, red arrow). Beyond 35°C, complete melting of the microgels restores a fluid‐like state, as indicated by the crossover in moduli (G″ = 1.06 Pa > G′ = 3.06 × 10⁻¹ Pa).

Notably, rev‐Gelatin demonstrates a distinctive LCST‐like behavior, characterized by an appearing sol state at ambient temperature that transitions into a gel state within the temperature range of 30°C–35°C via inter‐particle interactions. This unconventional inverse thermal phase transition, in contrast to typical gelatin behavior, allows for precise modulation of material properties through controlled temperature adjustments, highlighting rev‐Gelatin as a promising material for biomedical applications.

2.3. Evaluation of rev‐Gelatin's Hemostatic Properties via In Vivo Liver Hemorrhage Model

The demonstrated reverse thermo‐sensitive behavior of rev‐Gelatin, transitioning from sol‐to‐gel state with increasing temperature, offers significant advantages for hemostatic applications. Maintenance of flowability at ambient temperature enables injectability, and the rapid gelation of rev‐Gelatin constructs forms a robust physical barrier upon contact with blood at elevated temperatures. This unique phase transition behavior, combined with gelatin's inherent hemostatic properties, suggests enhanced wound (i.e., hemorrhaging sites)‐sealing capabilities through optimal surface coverage and subsequent network formation. In open surgery, the hemostatic mechanism of rev‐Gelatin leverages the temperature difference between the operating room (typically controlled at 20°C–25°C) [44, 45] and the wound site (approximately 30°C). Thus, rev‐Gelatin remains in its sol state at operating room temperature, ensuring ease of applying rev‐Gelatin solution. Subsequently, the applied rev‐Gelatin experiences temperature increases upon contact with the wound (surface temperature = ∼30°C) [46], which triggers a rapid sol‐to‐gel transition. This temperature‐induced gelation results in the formation of a physical barrier, effectively controlling bleeding and promoting hemostasis.

We evaluated the hemostatic efficacy of the materials using a standardized rat liver hemorrhage model [47, 48, 49]. A 6‐mm circular wound was created on the liver surface using a biopsy punch, and the animals were assigned to three groups: untreated control, conventional gelatin, and rev‐Gelatin (n = 4 per group) (Figure 4A). Upon application to the bleeding site, rev‐Gelatin rapidly transitioned into a gel state due to the local surface body temperature (∼30°C), forming a physical barrier that effectively suppressed bleeding (Figure 4B). As a result, the control group exhibited substantial hemorrhaging with a mean blood loss of 2.47 ± 1.52 g and a hemostasis time of 217 ± 66 s. While conventional gelatin demonstrated modest improvement with blood loss of 2.02 ± 1.17 g and hemostasis time of 230 ± 43 s, rev‐Gelatin showed superior hemostatic performance, significantly reducing blood loss to 0.15 ± 0.05 g and achieving hemostasis within 108 ± 56 s (Figure 4C,D), representing a 94% reduction in blood loss and a 50% decrease in hemostasis time compared to control. Compared to previous hemostatic studies using gelatin, such as those involving pure gelatin or varying liver wound sizes, rev‐Gelatin offers improved performance in both reducing blood loss and shortening hemostasis time [50, 51, 52]. In one study, a 1 cm linear incision in a rat liver model showed an 85% reduction in blood loss—from 326 mg in the control group down to 50 mg with the hemostatic agent [50]. In our study, a severely elevated blood loss of 2,470 mg was created by a 6 mm biopsy punch. rev‐Gelatin achieved a 94% reduction, lowering blood loss down to 150 mg, surpassing hemostatic efficacy of the previous study. Another study evaluated gelatin sponges in a rat femoral artery injury model, observing a 73% decrease in blood loss from 1,000 to 266 mg [51]; despite the different injury type and lower baseline blood loss, rev‐Gelatin demonstrated improved efficacy in a severe hemorrhaging scenario. These comparisons underscore that rev‐Gelatin not only outperforms conventional gelatin but also exceeds the performance of other gelatin‐based agents reported in the literature. rev‐Gelatin offers practical advantages, such as injectability and flowability, due to its sol state at ambient temperature, making it suitable for minimally invasive procedures. One fascinating aspect of rev‐Gelatin is that no chemical modifications were made to gelatin backbones. A catechol‐introduced gelatin chain, co‐formulated with graphene oxide, showed 80% reduction (355 to 70 mg) in a rat liver injury model [52], whereas rev‐Gelatin achieved a greater reduction in hemostasis of 94%.

FIGURE 4.

(A) Schematic representation and photographs of the rat liver puncture hemostasis model: untreated (left), conventional gelatin (middle), and rev‐Gelatin (right) group. (B) Schematic illustration of the hemostatic mechanism of rev‐Gelatin in open surgery, where temperature‐induced gelation forms a physical barrier to control bleeding. (C) Blood loss (g) for each group (untreated: hatched bar; conventional gelatin: black bar; rev‐Gelatin: red bar) during the in vivo hemostasis test. (D) Hemostasis time (s) for each group (untreated: hatched bar; conventional gelatin: black bar; rev‐Gelatin: red bar) during the in vivo hemostasis test. Panels A and B were created using BioRender (https://biorender.com).

2.4. Submucosal Retention Properties of rev‐Gelatin

In surgical procedures such as endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), submucosal injection agents (SIA) are injected beneath lesions to create a protective cushion, facilitating safe and precise resection while minimizing damage to underlying tissues. An ideal SIA should exhibit three key characteristics: facile injectability, in situ gelation and thermal stability of its original structure, and hemostatic properties. A thermo‐responsive polymer, Pluronic F127 (i.e., PEO‐PPO‐PEO), has been widely used as an SIA, satisfying some of the aforementioned criteria. However, disadvantages of Pluronic F127 include a lack of hemostatic ability because of the presence of the PEO segments. Additionally, its rapid bio‐absorption characteristics, due to effective molecular dissipation and dissolution [53, 54, 55], compromise its long‐lasting cushioning function. To address these limitations, we explored rev‐Gelatin as a potential alternative SIA in the endoscopic procedures. rev‐Gelatin not only provides the desired hemostatic properties but also exhibits a unique reverse directional phase transition enabling flowable injection at ambient temperatures while maintaining structural stability under physiological conditions.

To evaluate rev‐Gelatin's potential as an SIA, we conducted comprehensive in vivo stability tests using a porcine gastric model at physiological temperature. The submucosal injection assays measured both dissociation time and mucosal elevation height for conventional gelatin and rev‐Gelatin (Figure 5A). Following 1 mL injection of either 20 wt% gelatin or rev‐Gelatin, the conventional gelatin exhibited poor stability, retaining only 44.3 ± 10.3% of its original mucosal elevation at 8 min (Figure 5C second photo, upper row). After 10 min (third photo), all gelatin materials had utterly disappeared. In contrast, rev‐Gelatin demonstrated superior stability, maintaining 89.7 ± 2.1% for 8 min (Figure 5C second photo, lower row), 87.3 ± 1.5% for 10 min (third photo), and 83.0 ± 3.0% for 15 min (fourth photo) compared to its initial elevation height. Even after extended periods (120 min) (eighth photo), rev‐Gelatin showed gradual molecular dissolution, remaining 24.3 ± 7.5% of the initial elevated height. Figure 5B showed quantitative results of the aforementioned in vivo experiments.

FIGURE 5.

(A) Schematic illustration of the submucosal injection method for stability assessment of conventional gelatin and rev‐Gelatin at physiological temperature. (B) Quantitative analysis of normalized mucosal elevation height over time following submucosal injection of conventional gelatin (black) and rev‐Gelatin (red) (n = 3, mean ± SD). (C) Time‐course images of gelatin and rev‐Gelatin dissolution following submucosal injection. Upper panel: conventional gelatin; lower panel: rev‐Gelatin. The white dotted lines indicate the initial elevation height, and the red dotted lines indicate the basal line. (D) Schematic representation of the in vitro experiment emphasizing the hindrance of heat transfer between the fused microgel particles of rev‐Gelatin. (E) Representative time‐course photographs comparing dissolution of 1 cm³ cubes of conventional gelatin, collapsed rev‐Gelatin, and rev‐Gelatin at physiological temperature. (F) Quantitative measurements of normalized height during in vitro dissolution of conventional gelatin (black), collapsed rev‐gelatin (blue), and rev‐Gelatin (red) cubes (n = 3, mean ± SD). Pannels A and D were created using BioRender (https://biorender.com).

To understand the mechanism behind this enhanced retention, we conducted a corresponding in vitro experiment. We hypothesized that the rate of heat flow might be different between bulk gelatin and microbead‐to‐bead connected gelatin microgels (rev‐Gelatin). Hindrance of heat flow can occur due to the presence of a microgel boundary by incomplete gel‐to‐gel hybridization (Figure 5D).

To test this hypothesis, one cubic centimeter (cm³) of conventional gelatin and rev‐Gelatin cubes were prepared. The conventional gelatin cubes were formed by solidifying a 20 wt% conventional gelatin solution at room temperature, whereas rev‐Gelatin cubes were prepared by casting a 20 wt% microgel solution into a glass mold and subsequently solidifying it at 30°C for 30 s. We then incubated the samples at physiological temperature and observed their structural integrity over time.

The conventional gelatin cube exhibited rapid structural meltdown, retaining only 35.0 ± 5.0% of its initial height (1 cm) at 8 min and 8.3 ± 2.9% at 10 min (Figure 5E second photo, upper row), and complete melting at 15 min. In contrast, rev‐Gelatin showed increased thermal stability, maintaining 100% of its structural integrity for the first 10 min (Figure 5E second photo, lower row), retaining 91.7 ± 2.9% at 15 min, and preserving 6.7 ± 2.9% even at 150 min before complete dissolution.

Furthermore, to investigate whether this enhanced stability was indeed due to the presence of microgel boundaries, fully melted rev‐Gelatin cubes were prepared to remove microgel boundaries. Due to the absence of these boundaries, the fully melted rev‐Gelatin exhibited significantly reduced thermal stability, retaining only 28.3 ± 2.9% of its original height after just 8 min (Figure 5E, second photo, middle row). In contrast, rev‐Gelatin cubes with intact microgel boundaries maintained their structural integrity for up to 120 min, clearly demonstrating that the presence of microgel boundaries plays a crucial role in enhancing thermal stability. Figure 5F showed quantitative results of the aforementioned in vitro experiments.

2.5. Cytotoxicity and Biocompatibility of rev‐Gelatin

We evaluated several in vitro cytotoxicity tests of rev‐Gelatin. First, LIVE/DEAD staining and CCK‐8 assays were performed using HeLa cells. Cells were incubated for 24 h under four conditions: negative control (DMEM), positive control (Latex), and media containing 10 mg/mL of either rev‐Gelatin or conventional gelatin. The LIVE/DEAD staining assay revealed a similar distribution of live cells in the rev‐Gelatin and gelatin groups, compared to the negative control, while the positive control (Latex) showed a notable increase in dead cells (Figure 6A). Consistent with these findings, the CCK‐8 assay showed that cell viability in the rev‐Gelatin group was 114.97%, and 105.83% in the gelatin group (Figure 6B), both significantly higher than the control (p < 0.005, n = 3, mean ± SD). Collectively, these results demonstrate that rev‐Gelatin is non‐cytotoxic and biocompatible, supporting its safety in a cellular context.

FIGURE 6.

(A) LIVE/DEAD staining assay for assessing cytotoxicity of conventional gelatin and rev‐Gelatin on HeLa cells after 24‐h incubation under different conditions: negative control (DMEM), positive control (Latex), and medium containing 10 mg/mL of conventional gelatin or rev‐Gelatin (green: live, red: dead, scale bar: 200 µm). (B) Quantitative analysis of cell viability (%) based on the cell counting kit‐8 (CCK‐8) assay following 24‐h incubation with negative control (Green, DMEM), positive control (Blue, Latex), conventional gelatin (Black, 10 mg/mL), and rev‐Gelatin (Red, 10 mg/mL) (n = 3, mean ± SD, **p < 0.005). (C) Hemolysis assay of 20 wt% rev‐Gelatin and control samples (Positive control: black bar; negative control: hatched bar; rev‐Gelatin: red bar; n = 3, mean ± SD, ****p < 0.00005). (D) Endotoxin assay of 20 wt% rev‐Gelatin and control samples by the endpoint chromogenic LAL assay (Blank: black bar; rev‐Gelatin: red bar; n = 3, mean ± SD).

Second, a hemolysis test was performed using positive (surfactant‐containing PVC, Y‐3) and negative (polyethylene) controls, all prepared by extracting 0.2 g/mL of each material in 0.9% saline at 37°C for 72 h. As shown in Figure 6C, the positive control exhibited 96.67 ± 1.14% hemolysis, whereas the negative control demonstrated a hemolysis rate of 0.44 ± 0.10%. In contrast, the rev‐Gelatin extracts caused only 0.33 ± 0.07% hemolysis—well below the commonly accepted threshold—indicating that rev‐Gelatin is non‐hemolytic and safe for applications involving blood contact (**p < 0.00005, n = 3, mean ± SD).

Third, the endotoxin level contained in rev‐Gelatin was measured using a chromogenic LAL assay (endpoint method). According to FDA guidelines (ISO10993‐11), the endotoxin limit for materials in contact with the cardiovascular or lymphatic system is 0.5 EU/mL or 20 EU/device. As shown in Figure 6D, the rev‐Gelatin extraction sample had an endotoxin concentration of 0.106 ± 0.00067 EU/mL, which was similar to the negative control (0.110 ± 0.00164 EU/mL). These results confirm that rev‐Gelatin presents negligible endotoxin risk, further supporting its safety for biomedical use. Together, these findings demonstrate that rev‐Gelatin exhibits excellent biocompatibility, including low cytotoxicity, negligible hemolytic activity, and endotoxin levels well below regulatory thresholds.

Given these safety profiles, rev‐Gelatin shows strong potential as an optimal SIA material for endoscopic procedures. Compared to conventional gelatin and other thermo‐responsive materials such as Pluronic F127, rev‐Gelatin offers improved injectability, superior thermal stability, prolonged in situ retention, and enhanced hemostatic performance. These properties could significantly enhance procedural efficiency and safety in ESD, positioning rev‐Gelatin as a promising candidate for future clinical translation.

3. Conclusions

We have successfully engineered rev‐Gelatin, a unique biomaterial with reverse thermo‐responsive properties, enabling a sol‐to‐gel transition from room to physiological temperatures without the need for chemical modification. Inspired by the partial surface melting of spherical candies and subsequent particle‐to‐particle agglomeration, rev‐Gelatin offers easy injectability at ambient temperatures, transitioning rapidly to a stable gel upon exposure to body temperature through the surface melting of microgels. Our findings from thermal, rheological, and functional analyses, including hemostasis and submucosal injection tests, underscore rev‐Gelatin's effectiveness in reducing blood loss and supporting tissue elevation without residue. These findings establish rev‐Gelatin as a promising biomaterial platform for various biomedical applications that require temperature‐responsive gelation at physiological conditions.

4. Experimental Section

4.1. Materials and Reagents

Gelatin (Type A) was sourced from Sammi Industry (Korea). 2‐(3,4‐dihydroxyoxolan‐2‐yl)‐2‐hydroxyethyl (9E)‐octadec‐9‐enoate (Sorbitan oleate) was obtained from Il Shin Wells (Korea). Paraffin oil was provided by Molytech (Korea). Ethanol and acetone were supplied by Sigma‐Aldrich. HeLa cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured according to the supplier's instructions.

4.2. Preparation and Characterization of rev‐Gelatin

rev‐Gelatin was prepared using a water‐in‐oil emulsion technique. A 20 wt% gelatin solution was created by dissolving 200 g of gelatin in 800 mL of deionized distilled water (DDW) at 60°C for 60 min. This solution was then added dropwise into a custom‐made emulsification bath containing a mixture of 70 mL sorbitan oleate and 7 000 mL paraffin oil, while stirring at 450 rpm at 65°C. The emulsion was stirred for 30 min before being cooled to 4°C while maintaining the stirring speed. The gelatin microspheres formed were subsequently separated and further cooled at 4°C for 60 min to induce phase separation. The upper oil layer was removed using a pump, followed by a thorough washing process to purify the rev‐Gelatin microspheres. The rev‐Gelatin were washed with acetone and ethanol to remove residual paraffin oil and stored for 24 h at 25°C and 25% relative humidity to remove any remaining solvents. The morphology of the microspheres was examined using scanning electron microscopy (SEM; S‐4800, Hitachi, Japan), and their size distribution was measured using a laser scattering particle size distribution analyzer (LA‐960, Horiba, Japan).

4.3. Thermo‐Responsive Behavior of Conventional Gelatin and rev‐Gelatin

The thermo‐responsive behavior of conventional gelatin and rev‐Gelatin was examined across a controlled temperature range. A total of 22 vials, each containing 1 mL of 20 wt% conventional gelatin or rev‐Gelatin solution, were prepared. Eleven vials of each sample type were incubated at different temperatures (25°C–35°C) for 10 min to reach equilibrium. Immediately upon removal from the incubator, each vial was inverted, and the state of the conventional gelatin and rev‐Gelatin was documented via photographs. This process was repeated across the temperature range to compare the temperature‐dependent phase behavior of the two materials.

4.4. Comparison of Phase Change Properties of Conventional Gelatin and rev‐Gelatin

The phase translation properties of conventional gelatin and rev‐Gelatin were compared at different temperatures. A 20 wt% solution of conventional gelatin and rev‐Gelatin was prepared and loaded into 3 mL syringes. These syringes were placed in a temperature‐controlled chamber set to either 25°C or 30°C for 10 min. After equilibration, the conventional gelatin and rev‐Gelatin solutions were dispensed into petri dishes, and the initial shape and consistency of each solution were photographed immediately after injection. Additionally, to further assess the behavior of these materials, the same 20 wt% conventional gelatin and rev‐Gelatin solutions were fully spread across the surface of petri dishes and kept at 25°C and 30°C for 10 min. After incubation, the samples were scraped off using a spatula, and the remaining residue in the petri dish was photographed to compare the physical state and adhesion properties of the two materials at the different temperatures.

4.5. Rheological Characterization of Conventional Gelatin and rev‐Gelatin

The rheological properties of 20 wt% conventional gelatin solution, 5 wt%, and 20 wt% rev‐Gelatin mixture were analyzed using a rotational rheometer (MCR 102 Rheometer, Anton Paar, Austria) equipped with a parallel 25 mm plate. The temperature sweep experiments were performed at 1 Hz (γ = 0.5%) in the temperature range between 20–40°C.

4.6. Thermostability of Conventional Gelatin, Collapsed rev‐Gelatin, and rev‐Gelatin

Thermostability was evaluated by monitoring structural changes in conventional gelatin and rev‐Gelatin at physiological temperatures. For conventional gelatin, 1 cm³ cubes were formed by solidifying a 20 wt% solution at room temperature and then cutting with a surgical scalpel. For collapsed rev‐Gelatin, 1 cm³ cubes were prepared by solidifying a 20 wt% solution at room temperature after it was fully dissolved at 60°C for 1 h, followed by cutting with a surgical scalpel. rev‐Gelatin cubes (1 cm³) were formed by casting a 20 wt% solution into glass slide molds, solidified at 30°C for 30 s. Each sample was placed at the center of a petri dish, which was floated in a 37°C water bath. The dissolution of the samples was continuously observed and recorded using video and photographic methods. The sample height was measured periodically until the original form had completely disappeared.

4.7. Submucosal Elevation Retention of Conventional Gelatin and rev‐Gelatin

The submucosal elevation retention of conventional gelatin and rev‐Gelatin was investigated using a porcine gastric model. The mucosal layer of the porcine stomach was maintained at 37°C, RH 80% using a controlled temperature and humidity environment. A total of 1 mL of 20 wt% conventional gelatin or rev‐Gelatin solution was injected into the submucosal region using an 18G needle. Conventional Gelatin solutions were pre‐heated to 30°C to enhance injectability. The elevation height was measured over time with a ruler, and photographs were taken to document changes in the elevated regions. The retention ability of conventional gelatin and rev‐Gelatin was evaluated by comparing the stability and height of the elevated areas over time.

4.8. In Vivo Hemostatic Capability of rev‐Gelatin

The hemostatic capability of rev‐Gelatin was assessed using a rat liver hemorrhage model. Twelve SD rats (male, 9 weeks old, ≈300 g) were used. The rats were divided into three groups: a non‐treatment control group, a conventional gelatin group, and a rev‐Gel group. Each rat was anesthetized via intramuscular injection of Zoletil 50 (zolazepam–tiletamine, 33.3 mg/kg) and Rompun (xylazine, 7.8 mg/kg), and placed in a supine position. Following a midline laparotomy, the liver was exposed and sprayed with sterile saline. A biopsy punch (6 mm, kai Disposable Biopsy Punches) was used to create a standardized injury on the liver. In the experimental groups, 0.2 mL of either 20 wt% conventional gelatin or 20 wt% rev‐Gelatin was applied to the injury site via syringe. Blood loss was quantified by measuring the weight of pre‐weighed gauze placed beneath the liver. Hemostatic time and blood loss were recorded using a stopwatch and a precision scale, respectively. The animal procedures were conducted in accordance with the ethical guidelines of the Korean Ministry of Health and Welfare and were approved by the Institutional Animal Care and Use Committee of KAIST (Approval No. KA2023‐011).

4.9. Cell Viability Test

HeLa cells (1 × 10⁴ cells/well) were seeded in 96‐well plates containing Dulbecco's Modified Eagle Medium (DMEM, pH 7.4, Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10 vol% fetal bovine serum and 1 vol% penicillin‐streptomycin. The cells were incubated at 37°C in a humidified 5% CO₂ atmosphere for 24 h to allow for cell attachment and stabilization. Following the initial incubation, cells were exposed to experimental and control conditions for an additional 24 h under identical culture conditions. rev‐Gelatin and conventional gelatin were prepared at a final concentration of 10 mg/mL in the same complete medium used for cell culture. The negative control consisted of unmodified culture medium. For the positive control, latex extract was prepared by incubating latex in culture medium at 37°C for 24 h, following ISO 10993‐12 guidelines to ensure standardized extraction conditions. After 24 h of exposure, cytotoxicity was evaluated using both qualitative (LIVE/DEAD staining) and quantitative (CCK‐8 assay) methods. For qualitative analysis, the culture medium was aspirated, and cells were stained with calcein AM (2 µM) and ethidium homodimer (4 µM) from the LIVE/DEAD Viability/Cytotoxicity Kit (Invitrogen, Carlsbad, CA, USA) for 20 min in Dulbecco's phosphate‐buffered saline (DPBS). Fluorescence images were captured using an Eclipse Ti fluorescence microscope (Nikon, Tokyo, Japan). For quantitative analysis, the culture medium was replaced with 100 mL of fresh complete medium containing 10 mL of the Cell Counting Kit‐8 (CCK‐8) reagent. Following a 1 h incubation at 37°C in a 5% CO₂ environment, absorbance at 450 nm was measured using a Varioskan Flash reader (Thermo Fisher Scientific Inc., Waltham, MA, USA). All experiments were performed in triplicate, and cell viability was determined according to the International Organization for Standardization (ISO) 10993‐5 guidelines.

4.10. Hemolytic Evaluation of rev‐Gelatin Extracts

The hemolysis assay was conducted in accordance with the ethical guidelines established by the Korean Ministry of Health and Welfare and was approved by the Institutional Animal Care and Use Committee (IACUC) of the Korea Testing & Research Institute (KTR) under approval number MGK‐2023‐001173. The study was further conducted in compliance with international standards ISO 10993‐4:2017 and ASTM F756‐17. rev‐Gelatin samples were prepared by extracting 0.2 g/mL in 0.9% saline at 37°C for 72 h, following the ISO 10993‐12 guidelines. Whole blood was obtained from specific‐pathogen‐free (SPF) New Zealand White rabbits, purchased from DooYeol Biotech (Seoul, Korea). The collected blood was confirmed to have a plasma‐free hemoglobin (PFH) level below 2.0 mg/mL prior to use. The blood was centrifuged at 800 × g for 15 min, and the red blood cell (RBC) pellet was washed and resuspended in phosphate‐buffered saline (PBS) to achieve a total blood hemoglobin (TBH) concentration of approximately 10 ± 1 mg/mL. For each test condition, 1 mL of either rev‐Gelatin extract (test), polyethylene extract (negative control), surfactant‐containing PVC extract (positive control), or reagent control (blank) was combined with 7 mL of the diluted RBC suspension. The mixtures were incubated at 37°C for 3 h under gentle agitation, followed by centrifugation at 800 × g for 15 min. Then, 1 mL of the supernatant was transferred to 1 mL of Drabkin's reagent and allowed to react for 15 min at room temperature. Absorbance was measured at 540 nm using a UV–vis spectrophotometer (UV‐1800, Shimadzu, Japan). Hemolysis (%) was calculated by: % Hemolysis = [(A_s‐A_b)/(A_t‐A_b)] × 100% where A_s is the absorbance of the test sample, A_b is the reagent‐control absorbance (indicating baseline color interference), and A_t represents fully lysed RBCs (100% hemolysis). Any intrinsic background absorbance was subtracted before the final calculation. Hemolysis below 5% was considered non‐hemolytic for materials intended for blood‐contact applications.

4.11. Endotoxin Quantification via Chromogenic LAL Assay

To determine the endotoxin level of rev‐Gelatin extracts, a chromogenic Limulus Amebocyte Lysate (LAL) assay (Pierce Chromogenic Endotoxin Quant Kit, Thermo Scientific, USA) was performed according to the manufacturer's protocol. rev‐Gelatin samples were prepared by extracting 0.2 g/mL in 0.9% saline at 37°C for 72 h, following ISO 10993‐12 guidelines. Standards (0.25, 0.5, and 1.0 EU/mL), sample extracts, and blanks (endotoxin‐free water) were each loaded in triplicate into a pre‐equilibrated 96‐well microplate (50  mL per well). After adding 50 mL of reconstituted LAL reagent, the plate was incubated at 37°C for 16 min. Next, 100  mL of chromogenic substrate solution was added and incubated for an additional 6 min. The reaction was terminated by adding 50 mL of 25% acetic acid per well. Absorbance at 405 nm was measured immediately using a Varioskan Flash Microplate Reader (Thermo Scientific, USA). A standard curve was constructed by plotting blank‐corrected absorbance values against known endotoxin concentrations (EU/mL), yielding a coefficient of determination (R²) of 0.9987. Endotoxin concentrations in the rev‐Gelatin samples were then interpolated via linear regression from this standard curve.

4.12. Statistical Analysis

Data are presented as mean ± standard deviation (SD). Statistical significance between groups was determined using Student's t‐test, with a p‐value < 0.05 considered statistically significant.

Conflicts of Interest

The authors declare no conflict of interest.

Lee Y., Nam Y. R., Kim K., et al. “rev‐Gelatin: A Gelatin with Reverse Thermo‐Responsive Behavior Inspired by Candy and Ice Cubes Phase Dynamics.” Macromolecular Bioscience 25, no. 12 (2025): e00144. 10.1002/mabi.202500144

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.