Injectable Nanoengineered Adhesive Hydrogel for Treating Enterocutaneous Fistulas

Abstract

Enterocutaneous fistula (ECF) is a severe medical condition where an abnormal connection forms between the gastrointestinal tract and skin. ECFs are, in most cases, a result of surgical complications such as missed enterotomies or anastomotic leaks. The constant leakage of enteric and fecal contents from the fistula site leads to skin breakdown and increases the risk of infection. Despite advances in surgical techniques and postoperative management, ECF accounts for significant mortality rates, estimated between 15-20%, and causes debilitating morbidity. Therefore, there is a critical need for a simple and effective method to seal and heal ECF. Injectable hydrogels with combined properties of robust mechanical properties and cell infiltration/proliferation have the potential to block and heal ECF. Herein, we report the development of an injectable nanoengineered adhesive hydrogel (INAH) composed of a synthetic nanosilicate (Laponite^®) and a gelatin-dopamine conjugate for treating ECF. The hydrogel undergoes fast cross-linking using a co-injection method, resulting in a matrix with improved mechanical and adhesive properties. INAH demonstrates appreciable blood clotting abilities and is cytocompatible with fibroblasts. The adhesive properties of the hydrogel are demonstrated in ex vivo adhesion models with skin and arteries, where the volume stability in the hydrated internal environment facilitates maintaining strong adhesion. In vivo assessments reveal that the INAH is biocompatible, supporting cell infiltration and extracellular matrix deposition while not forming fibrotic tissue. These findings suggest that this INAH holds promising translational potential for sealing and healing ECF.

Graphical Abstract

1. Introduction

An enterocutaneous fistula (ECF) is a pathological link between the gastrointestinal tract and skin that significantly reduces a patient’s quality of life [1]. While most ECFs result from surgical complications, others are caused by Crohn’s disease, radiation treatment, diverticular disease, dysregulated erosion of tubes, cancer, injury, and abdominal sepsis [2]. Importantly, Crohn’s disease patients are particularly susceptible, with approximately 40% developing a fistula [1]. Despite the improvement in mortality rates in recent decades, recent reports demonstrate an overall ECF mortality rate of 15-20% due to sepsis, nutrition deficiencies, and electrolyte imbalances [3–5].

Accumulating clinical evidence indicates an unfavorable classification of ECF for spontaneous closure: a fistula tract less than 2 cm and an enteral defect greater than 1 cm [6]. The treatment of ECF involves controlling sepsis, replenishing fluid, electrolytes, and nutrition, safeguarding the skin from enteric discharge, determining anatomy, and formulating a definitive action plan [7]. Medical management to reduce fistula includes medications such as antidiarrheals or somatostatin and dressings such as ostomy appliances or negative-pressure wound therapy [7]. Non-operative treatments such as fibrin glue, endoscopic clips, and fistula plugs have also been explored for ECF treatment; however, they have shown limited success. Although fibrin glue has been demonstrated to be efficient in shortening the time to fistula closure, this conclusion is based on small case studies [8–10]. Fibrin glue frequently fails due to its liquid consistency, which results in leaking out post-recovery; a seminal study reported a failure rate of 85% [11]. Endoscopic clip closure has garnered increased interest in treating enteric fistulas and has long-term success rates of 40%. However, it is unsuitable for chronic fistulas or those not reached through endoscopy [12, 13]. Fistula plugs, such as the Fistula Plug^® manufactured by Cook Medical in Indiana, have been tested for ECF with some initial success. However, the data is based on small case series, and the outcomes are based on short-term results [13].

These treatments may fail for several reasons, including dislodgment of the material and recurring sepsis. Fistula plugs are unsuitable for complex fistulas with multiple branches, blind endings, and secondary openings. Hence, definitive surgery is the viable option owing to 82-86% healing rates, although recurrences are reported at 21-32% with postoperative morbidity of 86% [6, 14, 15]. Moreover, the choice of surgical approach depends on the anatomy of the fistula and the patient’s ability to tolerate the procedure [6]. The high prevalence, complexity, and current limitations of ECF treatment demand an innovative solution with minimal adverse effects. Hence, a minimally invasive biomaterial is desired to minimize bacterial infection, facilitate sustained sterile closure of the fistula tract, and promote accelerated healing.

Recently, with the advancement of biomaterials, the application of hydrogels to treat fistulas has expanded dramatically. For instance, Piantanida et al. developed an injectable hydrogel based on hyaluronic acid to treat esophageal fistulas [16]. Although it provides an injectable platform for blocking the fistula tract, the material does not offer adhesion to host tissue, and no antibacterial tests were performed to prove their platform as a drug-eluting material. In another example, Qu et al. developed a bilayer flexible hydrogel patch based on acrylamide-acrylic acid/cellulose nanocrystal for treating enteroatmospheric fistulas [17]. In this example, the adhesion to host tissue is addressed. However, the final material is a patch that needs to be implanted endoscopically, not allowing for a non-invasive approach such as injection. In addition, no antibacterial evaluation was demonstrated. Therefore, there is still a lack of an injectable biomaterial for non-invasive fistula treatments that can provide significant mechanical stability, adhesion to host tissue, drug-elution capabilities for antibacterial control, and high regeneration properties.

In developing innovative biomaterials for regenerative medicine, selecting constituent materials plays a pivotal role in determining the therapeutic efficacy and versatility of the final formulation. In this study, two key components, nanosilicate (Laponite^® and gelatin-dopamine conjugate, were carefully chosen for their unique properties and synergistic contributions to the final formulation. Nanosilicate, a synthetic nano clay, was selected for its structural benefits, offering a robust scaffold for tissue regeneration while providing a platform for the controlled release of bioactive molecules [18–21]. The gelatin-dopamine conjugate was also chosen for its adhesive properties, which enhance cell attachment and mechanical stability, which is crucial for tissue repair [22–26]. These materials have been thoughtfully integrated to harness their complementary characteristics, resulting in a versatile and promising hydrogel platform for ECF treatment.

Here, we report an Injectable Nanoengineered Adhesive Hydrogel (INAH) for occlusion and healing of the fistula tract (Figure 1). We have previously reported nanocomposite shear-thinning hydrogels based on gelatin to treat internal injuries [27–33]. In this work, we further modified this hydrogel for adhesive applications by conjugating dopamine to gelatin. The in situ cross-linking through a reaction with sodium periodate allows for adhesive properties. This chemical cross-linking involves the formation of covalent bonds between polymer chains (gelatin-dopamine conjugate) within the gel and between the polymer chain and the host’s tissue. These covalent bonds create a 3D network structure, stabilizing the gel and giving it characteristic properties. Chemical cross-linking is commonly used in preparing hydrogels and other gel systems because it precisely controls the gel’s mechanical properties, stability, and degradation rate.

Figure 1. Injectable nanoengineered adhesive hydrogel for treating enterocutaneous fistula.

The figure shows the components of the two hydrogels; for elective procedures, the gelatin-dopamine conjugate hydrogel (A) and nanosilicate hydrogel (B) are stored at 4 °C. The hydrogels can be injected through a mixing tip or catheter to seal the fistula channel (C). Upon injection, the hydrogel mix undergoes fast cross-linking between gelatin-dopamine conjugate molecules (D i and ii) and the patient’s tissues (D iii and iv). The improved mechanical and adhesive properties prevent material leakage with the patient’s movement. Furthermore, the gelatin-dopamine conjugate hydrogel and nanosilicate hydrogel provide excellent options for loading drugs for antibacterial control and improved healing.

Our nanoengineered adhesive hydrogel was produced through a co-injection method using 1) sodium periodate in nanosilicate hydrogel and 2) gelatin-dopamine conjugate hydrogel, displaying rapid in situ cross-linking. The nanosilicate and gelatin-dopamine conjugate combination improved mechanical and adhesive properties. It displayed effective blood clotting and cytocompatibility with fibroblasts. Ex vivo tests involving skin and artery tissues confirmed its adhesive properties. The hydrogel demonstrated volume stability in humid internal environments, preserving its strong adhesion strength. Studies with antimicrobial drugs proved the material’s capability to load and release drugs functionally active to prevent and fight bacterial infections. Furthermore, the in vivo evaluation in a mouse subcutaneous implantation model showed that the reorganization of ECM with a concomitant increase in cellular infiltration suggests neotissue formation. Our findings indicate that the INAHs can provide an effective solution for tissue adhesion and potentially meet the critical clinical requirement for fast sealing and healing ECF.

2. Experimental Section

Synthesis of Gelatin-Dopamine Conjugate:

Dopamine hydrochloride (Sigma, H8502) was conjugated to gelatin-A from porcine skin (Sigma, G1890) in the presence of N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) (TCI Chemicals, D1601 and H0623). Briefly, 1 L of a 1.4% (w/v) gelatin in deionized water reacted with EDC (7.042 g) and NHS (6.015 g) for 15 min before adding 3.179 g of dopamine dissolved in deionized water. After 24 h mixing at room temperature and protection from light, the solution dialyzed against deionized water (12,000 – 14,000 Da dialysis membrane, Fisher Scientific, 08-667E). The dialyzed solution was then lyophilized and stored at −20 °C.

Fourier-transform infrared spectroscopy and dopamine conjugation degree:

The Fourier Transform Infrared (FTIR) spectra of gelatin, dopamine, and gelatin-dopamine conjugate were accessed to identify functional groups and the success of synthesis. The samples were measured by the Attenuated Total Reflection (ATR-FTIR) method using a TENSOR 27 (Bruker^®, Germany) [34]. The dopamine conjugation degree was assessed by Arnow’s method [35].

Cross-linking Time:

The cross-linking time of gelatin-dopamine conjugate hydrogel and injectable nanoengineered adhesive hydrogels were assessed by a Rheometer (Anton Paar, MCR 302) following previous protocols [33]. Briefly, storage and loss moduli were evaluated with an 8 mm sandblasted parallel plate geometry. Oscillatory stress was recorded at 1 Hz.

Tissue Adhesive Testing:

Lap shear tests using porcine skin (local butcher) as a model were performed to evaluate the adhesive abilities of the gelatin-dopamine conjugate. The lap shear test proceeded according to the standard of ASTM F2255. Gelatin-dopamine conjugate (5%, 7.5%, and 10%) and sodium periodate (NaIO₄, Sigma, 311448) were applied to the inner surface of rectangular pieces of freshly cut porcine skin and incubated at 37 °C for 2 hours before mechanical tests at 1 mm/min tensile rate. Fibrin sealant (OMRIX 63713-390-55 EVICEL Fibrin Sealant) was tested as a control.

Preparation of Injectable Nanoengineered Adhesive Hydrogel:

The injectable nanoengineered adhesive hydrogels were prepared by a co-injection method. Nanosilicate (Laponite XLG, BYK Additives Ltd) hydrogels with 6%, 8%, and 10% (w/v) were combined with a 10% (w/v) gelatin-dopamine conjugate hydrogel in a 1:1 ratio utilizing two 3cc syringes (BD Luer-Lok^™ Syringe) connected to a mixing tip (Ivoclar Vivadent Inc., 645951). The two hydrogels are mixed at 1:1 upon injection through the mixing tip. The reaction with sodium periodate takes place for fast hydrogel cross-linking and tissue adhesion. The final concentration of sodium periodate in the adhesive hydrogels is 0.05% (w/w). This concentration was determined according to available literature [36].

Injectability testing:

For injection force assessment, nanoengineered adhesive hydrogels were loaded into 1 cc syringes and injected through mixing tips 1:1. The nanoengineered hydrogels were examined on a mechanical tester (Instron 5943, Instron Int. Ltd., MA, USA) using the Bluehill version 3 software with a 100 N load cell and an injection rate of 50 mm/s.

Tissue channel burst pressure testing:

The tissue channel burst pressure test utilized porcine aortas (Sierra for Medical Science, PAO). A piece of the wet, fresh porcine aorta (7 cm length × 2 cm diameter) is connected to a water manometer and a peristaltic pump. A 3 mL INAH was injected into the aorta and incubated at 37 °C for 2 h before testing. Then, PBS (Gibico, 14-190-250) was injected, and the peak pressure was recorded as burst. Gelatin-dopamine conjugate at 10% w/v and fibrin sealant were tested as a control.

In vitro Swelling and Degradation:

The swelling ratio of injectable nanoengineered adhesive hydrogels was investigated through a swelling equilibrium experiment. The INAHs were lyophilized and weighted as W_o before immersing in PBS at 37 °C. The weight of the swelled samples was recorded as W_s, and the swelling ratio was calculated using Equation (1):

$$\mathit{Swelling\; ratio}\%=\frac{\left(Ws-Wo\right)}{Wo}\times 100$$ (1)

Dry weights were recorded to investigate the in vitro enzymatic degradation behavior of INAHs. The samples were incubated at 37 °C in PBS containing 5.1 U mL⁻¹ collagenase type II (Gibico, 17-101-015) that refreshed every 48 h. The dry weights of the initial and degraded hydrogels were recorded as W_do and W_dt, respectively. The degradation kinetics was calculated using Equation (2):

$$\mathit{Massloss}\%=\frac{\left(\mathit{Wdo}-\mathit{Wdt}\right)}{\mathit{Wdo}}\times 100$$ (2)

Compression testing:

The compressive strength of INAHs was obtained at a rate of 2 mm min⁻¹ using a mechanical testing machine, Instron (5943, USA).

Cytotoxicity testing:

Cytotoxicity of INAHs was tested in vitro using PrestoBlue (Fisher Scientific, A13262) and live/dead assays (Fisher Scientific, L3224) on human dermal fibroblast (HDF) cells (PCS-201-012TM, ATCC). Briefly, 200 mg of INAHs were immersed in 1 mL of culture media (Dulbecco’s modified eagle medium, Cytiva, SH3024.01), incubated for 24 h, and then filtered for sterilization (0.2 μm syringe filter). HDFs seeded at 5 × 10³ cells/48-well were incubated with the leaching solutions for 24 h before metabolic activity and live/dead assays.

Clotting time assay:

Human sodium citrate whole blood (ZenBio, SER-WB) was utilized to test the hemocompatibility of INAHs. Clotting time trials (n=8) were conducted as previously described [37].

Drug release study:

6 g of nanoengineered adhesive hydrogels loaded with vancomycin hydrochloride (Sigma, SBR00001), metronidazole (Sigma, M3761), or ciprofloxacin hydrochloride (Sigma, PHR1044) (1,667 μg of drug /g of hydrogel) were immersed in tubes containing 20 mL of PBS (pH 7.4) at 37 °C. 10 μL of the solution in the tube was removed to test the absorbance with an ultraviolet spectrophotometer (DeNovix, DS-11 FX) at the wavelengths of 280 nm (vancomycin hydrochloride and ciprofloxacin hydrochloride) and 313 nm (Metronidazole). Another 10 μL of the same solution was supplemented in the tube at predetermined times. The absorbance of the background was tested using the same solution at 280 and 313 nm. The drug release was calculated through the standard curves.

Antibacterial Testing by Diffusion Method:

Bacteria used in this study included Escherichia coli ATCC 25922 and Staphylococcus aureus BH1CC (methicillin-resistant S. aureus (MRSA) clinical isolate). All bacterial strains were cultured in lysogeny broth (LB) and Lennox agar (LA) and incubated aerobically at 37°C throughout the study.

The susceptibility screening of E. coli and S. aureus to the hydrogel formulations was performed using standard technique (CLSI, 2012b). Aliquots of 100 μL freshly prepared suspension of each bacterial strain corresponding to 10⁸ colony-forming units per mL were used. Filter discs were impregnated with 30 μg of antimicrobial drugs; an 8 mm biopsy punch was used to create a well for injecting drug-loaded INAHs and incubated for 24 h at 37°C.

Biocompatibility Studies in rodent models:

Seven-week-old C57BL/6J mice (2 males and 2 females/group, weight of 20–25 g) were purchased from The Jackson Laboratory (Maine, USA) and acclimatized for 1 week before the experimentation. All the animal experiments were carried out following the “Guide for the Care of Laboratory Animals” after the animal protocol was approved by the Lundquist Institute Animal Research Committee (Protocol # 22747-02). The mice were kept in pathogen-free conditions under a 12 h light/dark cycle at 25 °C. Moreover, standard sterile laboratory pellets and purified water were used to feed the animals. Isoflurane (1.5% in O₂) was used for general anesthesia, and Carprofen subcutaneous injection was used for analgesia. After anesthetizing the animals, the back of the animals was shaved and sanitized with chlorhexidine. After that, an incision (1.5 cm) was created in the upper part of the dorsal back of the animals. The skin was then held with surgical forceps, and two subcutaneous pockets were created on each side of the incision line by dissection using blunt surgical scissors. Next, the sterile scaffolds were placed in the subcutaneous pockets with dressing forceps, and then the incision edges were stapled to close the wound. Finally, the animals were maintained on a warm pad during the postoperative period. After 3 and 14 days, animals were sacrificed by CO₂ inhalation, and samples were harvested with the surrounding full-thickness skin. In addition, vital organs (heart, lungs, kidneys, liver, and spleen) were harvested. All samples were collected in 10% formalin for histological section preparation.

Histological Staining:

The fixed tissues were dehydrated in serial ethanol concentrations (60%–100%) and embedded in paraffin wax. Tissue sections (4 μm thick) were stained using H&E and Movat Pentachrome following previously published protocols [38]. The stained sections were imaged using an optical microscope (Amscope, USA).

Statistical Analysis:

A minimum of three separate test sessions were conducted for all experiments. The collected data are presented as mean values with corresponding standard deviations. Statistical analysis was performed using GraphPad Prism software. To assess variances among multiple groups, one-way ANOVA with the Tukey multiple comparison test was employed. On the other hand, differences between two groups were analyzed using two-tailed unpaired t-tests. In the quantitative images, p-values less than 0.05 were considered statistically significant, while “ns” indicated no statistically significant difference.

3. Results and Discussion

Synthesis and characterization of gelatin-dopamine conjugate

A gelatin-dopamine conjugate was synthesized through EDC/NHS 2-step coupling as one of the components of the nanoengineered hydrogel (Figure 1A). The structure of dopamine closely resembles the 3,4-dihydroxy-L-phenylalanine (dopa) amino acid in mussels, employed in creating biomimetic materials demonstrating substantial adhesive potency [39]. When subjected to oxidation, the activated catechol moiety of dopamine undergoes a reduction of two electrons and two protons, leading to a reactive quinone. This quinone can react with thiol and amine groups through Michael-type addition or Schiff-base formation reactions [36]. In addition, quinones can chemically bind with nucleophilic groups, such as the side chains of lysine, cysteine, and histidine. This interaction leads to the formation of interfacial covalent bonds, resulting in robust adhesion to the substrate (hydrogel matrix and host’s tissue) [40].

The chemical structure of the synthesized gelatin-dopamine conjugate was assessed by Proton Nuclear Magnetic Resonance (H¹ NMR) and Fourier-transform infrared (FTIR) spectroscopy to investigate the successful substitution of dopamine to the gelatin. NMR spectra had a series of peaks at δ 6.5 – 6.8 ppm corresponding to the aromatic system of dopamine, which is conjugated to the gelatin macro chain (Figure 2A-i). Peaks at 2.75 ppm correspond to protons of the methylene group close to the phenyl group in dopamine. This is also observed in the spectrum of gelatin-dopamine conjugate but does not appear in the spectrum of pure gelatin (Figure 2A-ii). Similar NMR spectra for gelatin-dopamine synthesis were reported before [36, 41, 42].

Figure 2. Synthesis and characterization of gelatin-dopamine conjugate hydrogels.

(A) H¹ NMR of synthesized gelatin-dopamine conjugate. The -NH- bond at 7.2 ppm and 6.7 ppm represents aromatic benzene peaks. Peaks at 2.75 ppm correspond to protons of the methylene group close to the phenyl group in dopamine. (B) FTIR spectra of gelatin-dopamine conjugate samples compared with dopamine and gelatin. The panel on the right presents a picture of the lyophilized gelatin-dopamine conjugate. (C) Cross-linking kinetics of varying concentrations of gelatin-dopamine conjugate hydrogels when exposed to NaIO₄. (D) Digital picture of gelatin-dopamine conjugate at different concentrations before and after cross-linking. (E-G) Mechanical lap-shear test for assessing adhesiveness of gelatin-dopamine conjugate at different concentrations (F) and INAHs formulations (G) to skin. *p < 0.05, **p < 0.01, ***p < 0.001.

In the FTIR spectra (Figure 2B), a peak corresponding to an sp² single bond -C-H stretching was found at 3030 cm⁻¹. This corresponds to the aromatic functional group found in dopamine. The detection of amide stretching at 1640 cm⁻¹ confirms the formation of an HN single bond to CO (-HN-CO), thus providing evidence for the chemical bonding between dopamine and gelatin. Additionally, an OH stretch in the range of 3550-3200 cm⁻¹ is observed in pure gelatin, dopamine samples, and the gelatin-dopamine conjugate, indicating the presence of OH groups.

The ratio of dopamine conjugated to gelatin was determined by Arnow’s method [35, 43]. The substitution degree of catechol groups in the synthesized gelatin-dopamine conjugate was 1,056 μmol/g. Gowda et al. used EDC-NHS chemistry for gelatin dopamine conjugation. They demonstrated substitution degrees varying from 16.78 μmol/g to 87.25 μmol/g [36]. In another study using EDC-NHS chemistry for dopa addition to gelatin backbone, the modification was 237 μmol/g [44]. The degree of substitution in the current study is much higher than in the previous examples.

Most EDC/NHS reactions described in the literature conjugate dopamine to gelatin under acidic pH using MES buffer (pH 4.5) and elevated temperatures around 37°C or higher [36, 43, 45]. In contrast, we perform the reaction in deionized water under neutral pH and low temperatures around 20°C. High temperature can promote the hydrolysis of EDC, O-acylurea intermediate, NHS-ester, and acyl transfer from O-acylurea to N-acylurea, decreasing the activation efficiency [46]. Previous studies have shown that the preferential temperature for activation is 4-25°C [47]. Additionally, since the amine group of dopamine exhibits poor nucleophilic capability to NHS-ester due to forming a protonated amino group at low pH, the preferential pH for the immobilization of dopamine is in the range of 7-8 [48].

The gelatin-dopamine conjugate was also assessed for its cross-linking kinetics upon reaction with sodium periodate and its adhesive properties with porcine skin. Sodium periodate oxidation of the dopa generates an ortho quinone, an important intermediate, coupling the amino acid residues of the gelatin [49]. Liu et al. showed that the curing rate and duration were a function of the peroidate to dopamine ratio [49]. The assessments were performed by varying the concentration of gelatin-dopamine conjugate (5%, 7.5%, and 10%, w/v). It was observed that the higher the concentration of the gelatin-dopamine conjugate, the lower the cross-linking time required (Figure 2C). These results agree with the previous studies that related dopa amount to the cross-linking rate [50].

Solutions of 5% gelatin-dopamine conjugate initiated a detectable increase in the storage modulus (G’) after 45 min of mixing with a sodium periodate solution (gelatin-dopamine conjugate: 600 μL in PBS; and crosslinker: 40 μL of 25 mg/mL NaIO₄). In contrast, 7.5% and 10% gelatin-dopamine conjugate solutions presented a shorter time for detecting an increase in G’, with 25 min for 7.5% and 6 min for 10%. After a detectable increase in the storage modulus was recorded, all the gelatin-dopamine conjugate concentrations presented a close to full cross-linking in about 5 min (5%: 45-50 min, 7.5%: 25-30 min, 10%: 6-11 min).

A lap-shear test with porcine skin to assess adhesion to tissues demonstrated no significant difference upon increasing the concentration of gelatin-dopamine conjugate when tested at 5%, 7.5%, and 10% (Figure 2E and F). On the other hand, the adhesion of gelatin-dopamine conjugate was comparable to fibrin sealant. A study demonstrated that the dopa formed strong intramolecular bonds, resulting in higher substituted gelatin samples and lower lap shear stress values [36].

These results demonstrate that cross-linking time correlates with increased gelatin-dopamine conjugate concentrations. At the same time, the adhesive properties have no significant changes in the gelatin-dopamine conjugate concentrations tested.

Production of injectable nanoengineered adhesive hydrogels and tissue adhesion testing

After synthesizing and characterizing the gelatin-dopamine conjugate, gelatin-dopamine conjugate was combined with nanosilicates to produce the INAHs (Figure 3A). Different ratios of the nanosilicate hydrogel with sodium periodate to the gelatin-dopamine conjugate were prepared and tested for their tissue adhesion, swellability, degradation, and mechanical properties. Nanosilicate (Laponite^®), a type of 2D synthetic clay, has a disc-shaped morphology and a dual-charged surface. Due to their charged surfaces and strong structural anisotropy, nanosilicate discs can interact with various synthetic and natural polymers [18, 19]. Specifically, strong electrostatic interactions between nanosilicates and polymers are caused by abundant cationic and anionic charges on the nanoparticle’s surface. These interactions cause the matrix to form transient interactions, dissociating under shear forces and exhibiting shear-thinning properties [51, 52].

Figure 3. Production of injectable nanoengineered adhesive hydrogels and injectability and burst pressure testing.

(A) Gelatin-dopamine conjugate and nanosilicate hydrogels were used to compose INAHs. (B) The hydrogels were designed to be injected and mixed through a co-injection/mixing approach. Figure shows a 3D model of the proposed injection system with digital pictures of a real injection system (left) and the INAH after injection (right). (C) Injectability testing. When injected, the hydrogels are mixed in a 1:1 ratio to block the fistula channel. The light-colored dotted lines represent SD (n=4). (D and E) Tissue channel burst testing with pig aorta. *p < 0.05, **p < 0.01, ***p < 0.001.

Nanosilicate hydrogel was chosen for its mechanical and shear-thinning properties, good biocompatibility, and antimicrobial properties [20]. Due to the fast cross-linking, the 10% gelatin-dopamine conjugate solution was selected to compose Hydrogel 1 (H1) of a co-injection system (Figure 3B left syringe). Nanosilicate hydrogels with varying concentrations (6%, 8%, and 10%, w/w) and a fixed concentration of sodium periodate (3.33 mg/mL) were used to prepare the Hydrogel 2 (H2) of the co-injection system (Figure 3B right syringe). H1 and H2 were co-injected utilizing two syringes connected to a mixing tip. The two hydrogels are mixed at 1:1 upon injection through the mixing tip. The reaction with sodium periodate takes place for fast hydrogel cross-linking and tissue adhesion. The co-injection approach showed high reproducibility for mixing and cross-linking the dual-hydrogel system (INAH).

A lap-shear test with porcine skin to assess INAHs’ adhesion to tissues demonstrated that increasing the concentration of nanosilicates in the INAH formulations decreases the adhesion when tested in lap-shear mode (Figure 2G). On the other hand, the adhesion of INAHs with nanosilicate concentrations around 6 and 8% was comparable to fibrin sealant.

Injectability holds great importance for embolizing hydrogels. INAH was no exception, and injectability defines the easy applicability during fistula embolization or intervention. We assessed the injectability of various INAH compositions (Figure 3C). All compositions could be delivered via co-injection with a 1 cc syringe/mixing tip combination. Hydrogel compositions containing higher nanosilicate concentrations presented higher levels of injection force, showing that the nanosilicate component contributes majorly to the mechanical properties [29, 32]. Our previous work and others showed nanosilicate contribution to injectability [20, 28–30, 33, 53]. While the typically required force for clinical applications is usually below 20 N, the injectability of a substance is influenced not only by force exerted on the syringe but also by the ergonomic design of both the syringe and the mixing tip. These factors can collectively impact the overall applied force during the injection process [54]. The mean downward thumb-pushing strength is 184 N for males and 135 N for females aged 21-30 [55]. Our results showed that all INAH compositions presented an injection force below 35 N, considerably lower than the upper limits of human strength [55]. Hence, all hydrogel compositions tested can be hand-injected.

A tissue channel burst pressure test was performed to evaluate INAH composition performance. A porcine vessel section was connected to the three-way valve (Figure 3D). From one line, the adhesive hydrogel was applied to the lumen of the vessel and cross-linked. From the other line, burst pressure was applied using a peristaltic pump. The highest pressure level that adhesive hydrogels can bear before cracking and leaking fluid is known as burst pressure [56]. Hydrogels are frequently subjected to severe forces from underlying tissues or biological fluids when utilized as sealants, adhesives, embolizing agents, or hemostats. Therefore, measuring the burst pressure in an ex vivo environment enables evaluating the materials’ ability to seal incisions/wounds under similar conditions in vivo. As is shown in Figure 3E, increased concentrations of the nanosilicate led to improved burst pressures.

Additionally, gelatin-dopamine conjugate and fibrin sealants were tested as control samples. However, tissue channel blocking was unsuccessful for all trials due to the liquid consistency of fibrin sealant and gelatin-dopamine conjugate and tissue channel dimensions. Our results show that although the gelatin-dopamine conjugate hydrogel is responsible for the adhesive properties in our nanoengineered hydrogels, it is not always the case that larger gelatin-dopamine conjugate ratios will result in larger adhesive strength, as proved in the tissue channel burst tests—the gelatin-dopamine conjugate forms strong intramolecular hydrogen bonds with cohesiveness [57]. The increased nanosilicate ratio in the tissue channel pressure significantly increases cohesiveness, channel blocking, and pressure resistance. This happened because of the shear-thinning and “house of cards” self-assembling properties and ionic interactions of nanosilicates with protein systems, which led to improved channel blocking.

Due to the advantages of gelatin-dopamine conjugate with reduced cost for synthesis, no risk for blood-related diseases, and the capabilities of nanosilicates to block varying tissue channel dimensions, the INAHs we propose are a promising replacement for fibrin glue in treating ECF. Plugging or embolizing the ECF is a relatively new approach where surgical reconstruction is accepted as the gold standard method, with a recurrence rate of 20% and longer hospital stays [58]. Several case reports are in the literature for plugging ECF using minimally invasive methods. One study demonstrated using collagen fistula plugs with a 33% recurrence rate [59]. In one case study, Onyx 34 polymer was injected into the fistula tract and successfully treated the enteric leak [60]. Fibrin glues and derivatives showed efficiency only in low-output fistulas (<200 ml/day) [61]. Other commonly used materials for low-output fistulas are cyanoacrylate glues and derivatives [62]. High burst pressure values of the INAH hydrogel described in the current study could be a good alternative even for high-output fistulas (>500 ml/day).

In vitro Swelling, Degradation, and Mechanical Properties of Injectable Nanoengineered Adhesive Hydrogels

Hydrogels, by nature, swell in aqueous environments, and the swelling behavior can dictate the suitability of the material for a desired application. In addition, assessing the swelling degree is valuable for determining drug-releasing ability. We examined the swelling ratio of INAHs in phosphate-buffered saline (PBS) solution (pH 7.4) at 37 °C (Figure 4A). The high swelling ratio of INAHs (D10L6: 13.88; D10L8: 11.68; and D19L10: 11.03) provides an understanding regarding drug release, with hydrogels with high swelling ratios presenting faster release rates [63]. The swelling also contributes to the increased hydrostatic pressure after injection and better INAH-ECF tract epithelium interactions.

Figure 4. In vitro swelling, degradation, and mechanical properties of injectable nanoengineered adhesive hydrogels.

(A) The representative images and the swelling ratio of cross-linked adhesive hydrogels. (B) In vitro enzymatic degradation of cross-linked adhesive hydrogels. (C) Storage and loss modulus of cross-linked adhesive hydrogels in oscillatory frequency sweep. (D) Compressive mechanical testing and average compression modulus of cross-linked adhesive hydrogels. *p < 0.05, **p < 0.01, ***p < 0.001.

The chronic nature of ECF requires low-degrading materials for tract embolization to prevent recanalization and ECF recurrence. As shown in Figure 4B, all formulations containing nanosilicates (D10L6: 13 days; D10L8: 24 days; and D10L10: >30 days) had slower degradation when compared with hydrogels without nanosilicate (D10L0: 24h), proving that increased concentrations of nanosilicates lower enzymatic degradation of the composite hydrogels.

Adding nanosilicates to biomaterials can decrease the in vitro degradation rate due to interactions between the nanosilicate, the biomaterial matrix, and the environment [19]. Nanosilicates can form a physical barrier within the biomaterial matrix that limits the diffusion of water, enzymes, and other degradation-promoting molecules, slowing down the degradation process [64]. In addition, nanosilicate nanoparticles can reinforce the mechanical properties of the biomaterial [20, 21]. By forming a network within the biomaterial matrix, nanosilicates can increase their overall strength and stability; enhanced mechanical integrity can slow down the breakdown of the biomaterial and make it more resistant to degradation [65–67]. Furthermore, nanosilicates have a high surface area-to-volume ratio due to their nanoscale dimensions [68–71]. This large surface area can interact with surrounding molecules and ions, influencing degradation kinetics. It is important to note that the exact mechanisms affecting degradation rates can depend on the specific biomaterial, the type of nanosilicate used (such as Laponite), and the environmental conditions.

Golafshan et al. developed a hydrogel composed of Laponite, polyvinyl alcohol, and alginate for wound healing applications [72]. The study demonstrated that adding Laponite enhanced cross-linking density and reduced 1.2 times the degradation ratio. In another study, Ren et al. developed an injectable hydrogel based on quaternized chitosan, gelatin, and dopamine to deliver drugs for Parkinson’s disease [73]. According to the study, injectable hydrogels with high dopamine content have a fast degradation rate because increasing the dopamine content decreases the chemical cross-linking density of the hydrogels. Our results align with the literature when an increase in the nanosilicate ratio in the biomaterial composition increases the time required for degradation. It is worth mentioning that this in vitro degradation method provides an estimation of the behavior of the biomaterials after implantation, however, in vivo experiments are required for validation. In our in vivo experiments (see in vivo biocompatibility section), the implanted biomaterials were still visible 14 days after subcutaneous implantation.

We applied dynamic rheology to study the cross-linking reaction of INAHs after co-injection as a function of time with a constant oscillation frequency (6.28 rad s⁻¹) (Supplementary Figure 1). As the cross-linking reaction proceeded, the storage modulus (G’) increased, indicating increased stiffness. The INAHs reached 50 kPa G’ at 53.5 min for D10L6, 22.5 min for D10L8, and 16.5 min for D10L10. Further, to study the effect of nanosilicates on the mechanical behavior in the non-destructive deformation range of the cross-linked hydrogels, the storage modulus (G’) was studied as a function of angular frequency ω) (Figure 4C). A frequency sweep from 0.628 to 100 rad s⁻¹ was performed. At the ω studied, G’ was constant for all hydrogels up to 50 rad s⁻¹, revealing the stability of the cross-linked structures. In addition, G’ levels were increased together with the nanosilicate concentration. The cross-linked hydrogels with varying concentrations of nanosilicates presented significantly different G’ after cross-linking, D10L6 with 5.1 kPa G’, D10L8 with 10.5 kPa G’, and D10L10 with 11.6 kPa G’ (Figure 4C-iv). Previous studies on the phase diagram of synthetic nanosilicates have shown that at concentrations above 3% (w/w), nanosilicates form liquid crystals characterized as a tightly packed “house-of-cards” structure with minimal Laponite movement [68–71, 74, 75]. Hence, the G’ of cross-linked hydrogels can be tuned by increasing nanosilicate concentration.

Compression tests further assessed the mechanical properties of INAHs. A mechanical tester was used to compress samples (10 mm diameter x 4 mm height) to obtain the stress-strain curve at a 1 mm/min compression rate (Figure 4D-i). Then, the corresponding compressive modulus was calculated according to the selected strain interval (0-0.5 mm), Figure 4D-ii. Compressive stress-strain curves demonstrated a positive correspondence for nanosilicate concentrations and compressive moduli, with 0.0005 MPa for D10L0, 0.0010 MPa for D10L6, 0.0026 MPa for D10L8, and 0.0034 MPa for D10L10 formulation (Figure 4D-iii). The mean failure strain, failure strength, and fracture energy for each INAH formulation are as follows. D10L0: 0.78 mm/mm, 0.0540 MPa, 0.034 N; D10L6: 0.64 mm/mm, 0.0439 MPa, 0.141 N; D10L8: 0.76 mm/mm, 0.0374 MPa, 0.458 N; and D10L10: 0.53 mm/mm 0.0210 MPa, 4.922 N.

Tan et al. fabricated a Laponite cross-linked hydrogel with gradient thermoresponsive and strong anisotropic mechanical properties. The study demonstrated that the gradient distribution of Laponite enhances the compressive strength due to having higher cross-linking density [67]. In a different example, Li et al. utilized Laponite to improve collagen hydrogels’ mechanical properties and thermostability [76]. Their results demonstrated that increased Laponite ratios could reinforce the mechanical properties of collagen hydrogels. Collagen hydrogels with up to 10% Laponite enhanced 45-fold the compressive modulus compared with hydrogels without Laponite.

Regarding hydrogels composing dopamine, Rajabi et al. fabricated a polydopamine functionalized Laponite gelatin hydrogel for surgical sealant. The compressive stress-strain tests of the hydrogel demonstrated that the incorporation of 1% wt of polydopamine-Laponite significantly enhances the compressive strength by 2.8 times owing to the strong bonding between polydopamine-Laponite and the hydrogel network. However, incorporating higher concentrations reduces the compressive strength due to the aggregation of polydopamine-Laponite nanoparticles [65]. In another example integrating dopamine-modified biomaterials, Gan et al. developed a gelatin methacryloyl (GelMA)-dopamine conjugate hydrogels for cartilage regeneration applications. Their results demonstrated that increased dopamine substitution in GelMA-dopamine conjugate hydrogels enhances compression modulus. Although dopamine substitution or gelatin-dopamine conjugate concentrations were not assessed in this work, it is worth mentioning that nanosilicate hydrogel alone is a shear-thinning biomaterial that becomes fluid when exposed to external forces; therefore, the inclusion of the gelatin-dopamine conjugate provides improved stability of the biomaterial after injection.

In vitro cytocompatibility and Hemostasis Experiment

In vitro cytotoxicity studies were conducted using human dermal fibroblasts (HDF) to prove the potential applicability of the INAHs (Figure 5). The PrestoBlue testing was employed to evaluate the effect of the hydrogels on HDFs. The INAHs presented slightly lower metabolic activities than those of the groups using culture media alone (Figure 5D), which indicates that the cells exposed to the hydrogels had a slower proliferation than the control group. A live/dead testing was performed to quantify the total number of live and dead cells (Figure 5B–C). In the samples with increased concentrations of nanosilicates, a high number of live cells with a minimal number of dead cells was shown, proving the nanoengineered adhesive hydrogels are non-toxic but slightly decreasing cell proliferation. The decrease in cell proliferation is a desired property since fistula histology shows granulomatous infiltrations and inflammation [77] and sometimes transforms into cancerous lesions [78]. Less proliferation observed with INAHs could work towards controlling the uncontrolled proliferation in the ECF tract.

Figure 5. In vitro cytocompatibility and hemostasis.

(A) Schematic representation of in vitro cytocompatibility testing. (B) Fluorescence live/dead images of fibroblast cells exposed to nanoengineered hydrogels. Scale bar: 200 μm. (C) Cell viability quantification. (D) Metabolic activity quantification. (E and F) Clotting assay for human blood samples exposed to INAHs. *p < 0.05, **p < 0.01, ***p < 0.001.

A serious and fatal consequence in people with ECF is intra-abdominal hemorrhage. Wu et al. reported 10.6% intra-abdominal bleeding in hospitalized ECF patients [79]. Coagulative properties are desired in hydrogels for ECF management. In general, effective hemostatic materials demonstrate the ability to facilitate rapid and efficient blood clotting while minimizing damage to blood cells. To evaluate the blood coagulation capacity of the INAHs, we conducted experiments using human whole blood treated with sodium citrate. The clotting time of the whole blood was assessed after introducing it to a 48-well plate coated with hydrogels. In the control group, where the wells were left uncoated, the human blood took approximately 12.8 minutes to coagulate (Figure 5E). Blood clot formation was significantly accelerated in the wells coated with INAHs or gelatin-dopamine conjugate hydrogels. In contrast, There was no notable disparity observed in the clotting time of the blood samples added to wells coated with either INAH or gelatin-dopamine conjugate hydrogels (Figure 5F, nanoengineered adhesive hydrogels: D10L6: 9 min; D10L8: 10.4 min; and D10L10: 10.4 min vs. gelatin-dopamine conjugate hydrogel: 9.4 min).

Gaharwar et al. introduced gelating-based hydrogel with Laponite to promote the hemostatic behavior of gelatin. The fabricated hydrogel reduced the clotting time by 77% [30]. Applying polydopamine in gelatin has also been demonstrated to improve blood clotting. A study on polydopamine and gelatin cryogels has demonstrated to stop noncompressible bleeding [80]. The study showed that adding 8 mg/ml of dopamine can enhance the blood clotting ability of gelatin. In our results, all tested compositions reduced the clotting time to an average of 76% compared to untreated human blood. Although increased nanosilicate concentrations were provided for different hydrogel compositions, no significant difference was observed.

Incorporating the catechol group, a key structural moiety found in dopamine, into biomaterials has demonstrated a remarkable ability to induce hemostasis through a multifaceted mechanism [81]. One crucial aspect of this mechanism involves the ability of the catechol group to constrict blood vessels upon contact with blood. This vasoconstrictive effect helps reduce blood flow at the application site, facilitating stable clot formation [82]. Moreover, the catechol group imparts a negative charge to the modified biomaterials [24]. This negative charge is essential in attracting and interacting with positively charged blood components, such as platelets and coagulation factors [81, 83]. This interaction enhances the adhesion and aggregation of platelets, initiating the clotting cascade more effectively and expediting the clot formation process. Furthermore, the unique chemical properties of the catechol group enable it to bind to various proteins present in the blood plasma and at the wound site [24, 80]. This protein binding not only reinforces the stability of the developing clot but also promotes the recruitment of additional clotting factors, accelerating the overall hemostatic response.

Drug Release and Antibacterial Testing

One of the primary goals of ECF treatment is to prevent infection. Therefore, antibacterial properties are essential to achieving this goal. The use of antibiotics can help i) prevent or treat bacterial infections that may occur in the affected area; ii) reduce the inflammation associated with infection, improving wound healing; iii) control the growth of bacteria, which may contribute to the formation of the ECF. Hence, INAHs (D10L0, D10L6, D10L8, and D10L10) were tested regarding their antibacterial drug loading and release capabilities (Figure 6A–B).

Figure 6. Drug release kinetic and antibacterial activity.

(A) Schematic representation of drug release testing. (B) Quantification of cumulative antibacterial drug release. (C) Schematic representation of antibacterial testing. (D) Agar plates showing the zones of inhibition formed by antibacterial drug-loaded INAHs. (E) The mean diameter of the zones of inhibition of different antimicrobials. *p < 0.05, **p < 0.01, ***p < 0.001.

The INAHs were loaded with 10 mg of antibacterial drugs (vancomycin, metronidazole, or ciprofloxacin) per 6 g of the hydrogel. Release studies were performed in PBS, pH 7.4, at 37°C, and quantification of cumulative release percentage was assessed by measuring absorbance against standard curves. Due to its size and more complex structure, vancomycin loaded in the INAHs presented a sustained release kinetic for over 15 days (Figure 6B). In contrast, metronidazole, with a smaller and simpler structure, showed fast-release kinetics. Interestingly, ciprofloxacin, although with a smaller and simpler structure than vancomycin and a relatively similar size to metronidazole, presented a fast but incomplete release. Ciprofloxacin contains several functional groups with acid-basic properties and is also amphoteric [84]. Ciprofloxacin presents different charges depending on the environment’s pH. Therefore, ciprofloxacin adsorption to matrices and release depends on pH and matrix charges. As mentioned, nanosilicate is negatively and positively charged, capable of interacting with drugs such as ciprofloxacin under different pH ranges according to the drug’s charge [20, 21].

Regarding the varying formulations of INAHs, the concentration of nanosilicates plays a role in drug release kinetics, with increased concentrations of nanosilicate presenting slower and more incomplete release. For instance, vancomycin-loaded nanoengineered hydrogels were shown to release in 17 days about 44.5% for D10L10, 50.9% for D10L8, and 78.9% for D10L6. In contrast, the D10L0 hydrogel, which does not integrate nanosilicates in its formulation, released the total amount of vancomycin initially loaded in 5 days. Therefore, it is important to consider the concentration of nanosilicates according to the hydrogel’s formulation, drug size, complexity, and functional groups potentially interacting with nanosilicates for antibacterial drug loading and release. Among the antibacterial drugs tested, vancomycin presented a desired long and sustained release, which could be beneficial in treating ECF.

Although INAHs can load and release different antibacterial drugs, it is important to assess if the released drugs maintain their antibacterial activity. Disc diffusion assays were carried out for initial antibacterial testing of drug-loaded and control (drug-free) hydrogels toward Gram− E. coli and Gram+ S. aureus bacteria (Figure 6D–E). Control hydrogels were devoid of antibacterial activity. In contrast, drug-loaded hydrogels presented antibacterial activity according to the loaded drug and the bacteria tested. Vancomycin, a glycopeptide antibiotic that targets the bacteria’s cell wall, is active against Gram+ bacteria such as S. aureus but not against Gram− bacteria such as E. coli. Metronidazole is primarily active against anaerobic bacteria and has limited activity against S. aureus and E. coli. Ciprofloxacin is a broad-spectrum antibiotic that is effective against many Gram− and Gram+ bacteria, including some strains of S. aureus and E. coli. Therefore, the INAHs can load and release antibacterial drugs with different properties and targets while maintaining antibacterial activity.

In work by Pacelli et al., ofloxacin antibiotics were loaded in gellan gum methacrylic combined with Laponite hydrogel for producing wound dressings. Their diffusion studies demonstrated that the concentration of 1% w/v of Laponite could provide controlled drug release to prevent the spread of infection in wounds [85]. In another example, Liang et al. reported gelatin-grafted-dopamine hydrogels loaded with doxycycline [86]. Their hydrogel exhibited sustained drug release with a diffusion pattern that released 50% of the drug in the first 20 h, and then the release rate was slowed down, and the release course lasted for 100 h. Furthermore, our previous research on Laponite-gelatin hydrogels as drug-eluting biomaterials demonstrated Laponite as a pH-responsive hydrogel. When small molecules such as the anticancer chemotherapy doxorubicin are loaded in Laponite-gelating hydrogels, the release is modulated upon Laponite concentration and environment pH, with higher release rates obtained at low Laponite ratios and acidic solutions (pH<7.0) [20]. In contrast, when more complex drugs, such as the immune checkpoint inhibitor anti-PD-1, are loaded in Laponite-gelatin hydrogels, the pH-responsiveness of Laponite might not play a significant role due to the multiple protonation states of proteins and other complex molecular structure [87]. Therefore, the findings of this work align with our previous studies and work described in the literature.

The most significant forces that our INAHs load antibacterial agents are electrostatic and Van der Waals interactions with nanosilicates. Electrostatic forces involve the attraction between positively and negatively charged molecules. In the case of our INAH, nanosilicates have a negatively charged surface due to silicate groups [19, 20]. Many antibacterial agents, especially antibiotics, carry positive charges or charged functional groups [88], resulting in electrostatic interactions that facilitate antibacterial agents’ binding and retention within the hydrogel matrix.

In contrast, Van der Waals forces are weak, short-range interactions arising from electron distribution fluctuations around atoms and molecules. Although weaker than covalent or ionic bonds, Van der Waals forces play a crucial role in the adsorption of molecules onto surfaces, including nanosilicate particles and gelatin in our INAHs [89, 90]. These forces help stabilize the antibacterial agents within the hydrogel, preventing their rapid diffusion or release and protecting them from degradation or inactivation in the surrounding biological environment.

The interactions described above contribute to controlled release mechanisms regarding the impact on pharmacological activity. This controlled release can prolong the presence of antibacterial agents at the target site, enhancing their therapeutic efficacy while minimizing potential systemic side effects. Additionally, the hydrogel’s ability to load and retain antibacterial agents can create a localized high concentration of the drug at the treatment site, which can be particularly advantageous for infections where a higher drug concentration is needed to combat bacteria effectively.

In vivo Biocompatibility Studies

The bioactive effects of compositional changes were evaluated to assess the clinical application of the INAHs. Based on mechanical testings, including burst pressure, swelling, degradation, and drug release properties, we further selected two composite hydrogels (gelatin-dopamine conjugate/nanosilicate = 5%/4% (D10L8) and 5%/5% (D10L10)) to investigate in vivo biocompatibility.

We evaluated subcutaneous implantation for biocompatibility and degradation in vivo and monitored it over 14 days (Figure 7). On day 3 and day 14, histopathological analysis of subcutaneously implanted materials revealed mild inflammation with the presence of inflammatory cells such as mast cells, polymorphonuclear cells (neutrophils), and mononuclear cells, i.e., monocytes, accompanied by minimal material degradation and limited cell infiltration. On day 14, more chronic inflammatory cells, such as macrophages, were predominant at the scaffold interface. These macrophages were seen bound to the surface of the materials, suggesting active phagocytosis. The fusion of macrophages into foreign body giant cells (FBGCs) on the surface of the materials was absent; however, an association of multiple macrophages was evident. Both materials showed tissue integration without fibrous capsule formation and similar degradation. It was noticed that D10L8 provoked less cell infiltration than D10L10. In addition, D10L10 demonstrated more extracellular matrix (ECM) deposition on day 14. D10L10 also displayed an increased density of released nanosilicate particles at the interface (Figure 7B). Mucin deposition predominated along the interface tissue, suggesting the active ECM formation. However, the deposition of elastin fibers was not evident, as revealed by pentachrome staining. Also, intense collagen deposition and histological features of fibrosis were absent in both hydrogel formulations. Importantly, this immune response was comparable to the skin injury control group, which exhibited a slightly higher level of inflammation.

Figure 7. In vivo biocompatibility assessment of nanoengineered adhesive hydrogels.

(A) Representative H&E images of subcutaneously implanted D10L8 and D10L10 materials compared to skin injury control at day 3 and day 14. (B) Pentachrome staining for the mice skin tissues. Compared to the control group, hydrogel-implanted tissues showed ECM deposition and cell infiltration following the implantation. The yellow star shows immune cells, the blue star point to mucin deposition, the green star indicates tissue infiltration around the implant, the red star show collagen deposition, and the orange star points to the implant. The images were acquired in 20x magnification.

Previous studies have reported that nanosilicates exhibit low toxicity and are biocompatible [30]. The layered structure of nanosilicate particles results in limited interactions with biological tissues, which signifies the mild immune response observed in this study. In addition, the gelatin-dopamine conjugate in the hydrogel formulation may contribute to the observed low immune response. Gelatin has been reported to possess anti-inflammatory properties due to the presence of the cell adhesive motifs-RGD, which potentially mitigate immune responses, increase cellular adhesion, and reduce inflammation [91, 92]. These findings are consistent with previous studies evaluating the biocompatibility of nanosilicate-based materials. Researchers have also reported mild or limited immune responses, including reduced inflammation and the absence of fibrous capsule formation when nanosilicate-based materials are implanted in vivo [30, 33]. Furthermore, the presence of a skin injury control group in this study, showing similar levels of inflammation and inflammatory cells compared to the nanosilicate-implanted skin, further supports the notion of mild immune responses and good biocompatibility of the composite nanoengineered biomaterial.

In conclusion, the observed mild immune response with the INAHs in this study may be attributed to the inherent properties of nanosilicates as a clay mineral and the potential anti-inflammatory properties of the gelatin-dopamine conjugate. These findings are consistent with previous studies and support the biocompatibility of nanosilicate-based hydrogel materials.

Our current formulation has potential as nanosilicate and gelatin-dopamine conjugates can promote tissue healing through several mechanisms. Nanosilicates, such as Laponite, provide a structural scaffold for tissue regeneration. It has a high surface area and can interact with cells and proteins, promoting cell adhesion and proliferation [29, 33]. The nanosilicate particles create a 3D network within the hydrogel, offering mechanical support for tissue growth. Gelatin is a biocompatible and biodegradable protein that can serve as a matrix for cell growth and tissue formation. The conjugation of dopamine to gelatin can enhance its adhesive properties, facilitating the attachment of cells to the hydrogel scaffold [22, 25, 93].

Additionally, some studies suggest that nanosilicates may have anti-inflammatory properties. They can potentially modulate the immune response, reducing excessive inflammation at the injury site, which is essential for tissue healing [94]. Dopamine-functionalized hydrogels have been shown to promote angiogenesis [95]. This is critical for supplying oxygen and nutrients to regenerating tissues. The adhesive properties of the gelatin-dopamine conjugate are crucial for securing the hydrogel at the injury site. This adhesive strength ensures that the hydrogel remains in place, providing mechanical support during the healing process.

Finally, as shown in our in vivo studies, nanosilicates and gelatin-dopamine conjugates have demonstrated biocompatibility, meaning they are well-tolerated by living tissues. This biocompatibility reduces the likelihood of adverse reactions and fibrotic tissue formation, contributing to successful tissue healing.

4. Conclusion

In this work, we report on developing drug-releasing INAHs for treating ECFs, which are pathological links between the gastrointestinal tract and skin that significantly reduce a patient’s quality of life and can be caused by various factors. While current treatments have shown limited success, definitive surgery remains the most viable option for treatment despite its high recurrence and postoperative morbidity rates. Here, we sought to develop a minimally invasive biomaterial that could provide sustained sterile closure of the fistula tract and promote accelerated healing, reducing the need for surgery.

A nanocomposite shear-thinning hydrogel based on gelatin-dopamine conjugate for adhesive applications was synthesized by conjugating dopamine to gelatin, which provides adhesive properties, allowing in situ cross-linking through a reaction with sodium periodate. The resulting nanoengineered adhesive hydrogel displayed rapid in situ cross-linking, improved mechanical and adhesive properties, effective blood clotting, and cytocompatibility with fibroblasts. Ex vivo tests involving skin and artery tissues confirmed its adhesive properties, and the hydrogel demonstrated volume stability in humid internal environments, preserving its strong adhesion strength. Studies with antimicrobial drugs proved the material’s capability to load and release drugs functionally active to prevent and fight bacterial infection. Furthermore, in vivo evaluation in a mouse subcutaneous implantation model showed that the material promoted ECM deposition and neotissue formation with minimal inflammation. Moreover, the material promoted significant cell infiltration without a fibrous capsule. These properties are significant for healing ECF, once cells can infiltrate the biomaterial and secrete ECM to form new tissue and heal the ECF.

The INAHs can provide an effective solution for tissue adhesion and meet the critical clinical requirement for fast sealing and healing of ECF. This work has important implications for developing new treatments for ECF and could improve patient outcomes and quality of life.

Statement of Significance.

This research manuscript presents a groundbreaking injectable nanoengineered adhesive hydrogel (INAH) for treating Enterocutaneous Fistula (ECF). The INAH, composed of a synthetic nanosilicate and gelatin-dopamine conjugate, offers versatile implications in tissue regeneration and localized drug delivery. Acting as a scaffold, the shear-thinning hydrogel enables easy injection, forming a stable structure that supports tissue regeneration and integrates with surrounding tissues. By incorporating bioactive cues, it guides cell behavior and promotes functional tissue regeneration. The INAH also demonstrates potential for localized drug delivery, releasing therapeutic agents over time to enhance efficacy and minimize side effects. This research showcases INAH as a promising solution for ECF, with applications in tissue engineering and regenerative medicine, marking a significant advancement in the field.

Data Availability Statement

The data supporting this study’s findings are available from the corresponding author upon reasonable request.