Biocompatibility,hemostatic properties,and wound healing evaluation of tilapia skin collagen sponges

Abstract

Dialyzed tilapia skin collagen sponge (DTSCS) and self-assembled tilapia skin collagen sponge (STSCS) were prepared by freeze-drying. The raw components used in the fabrication of DTSCS and STSCS were separated and purified from tilapia fish skin. It is anticipated that these collagen sponges could be developed into medical dressings for hemostasis and wound healing. The aim of the present research was to explore the possibility of DTSCS and STSCS as medical dressings and compare their differences by scanning electron microscopy (SEM), water absorption measurement, differential scanning calorimetry (DSC), measurement of porosity, cytotoxicity, hemolysis, in vivo biocompatibility, and evaluation of hemostatic performance and wound healing. The results indicate that DTSCS and STSCS are suitable materials for use in medical applications with a loose and porous structure, high water absorption, high porosity, and high thermal stability. The materials also displayed good biocompatibility, including excellent blood compatibility, a lack of cytotoxicity, with no apparent rejection following implantation. STSCS exhibited rapid hemostasis and promoted healing, with slightly greater efficacy than DTSCS. The hemostatic properties and promotion of healing in DTSCS was similar to that of commercial bovine collagen sponge. Therefore, DTSCS and STSCS both represented excellent potential candidate materials for use as hemostatic agents and wound dressings.

Keywords

tilapia skin collagen collagen sponge biocompatibility wound healing hemostatic properties

Introduction

Collagen is a biological macromolecule and key structural protein, the principal component of the connective tissues in animals and the most widely-distributed functional extracellular matrix protein, representing approximately 30% of the total protein in an animal.^1,2 As an ideal matrix molecule, collagen is widely used in biomedical materials. Of all types of known collagens, the most common form is Type I collagen, predominating in animal tissues and an important structural protein in the extracellular matrix.³ Through continuous remodeling, it plays a key role in maintaining the dynamic balance in tissues, with biological and structural mechanical integrity.⁴ Collagen has many advantages, including good physical and chemical properties. Furthermore, collagen has outstanding biological properties, including being nonantigenic, biodegradable, non-toxic, biocompatible, and with outstanding hemostatic and wound healing properties, and proliferative and antioxidant effects.^2,5 Its superior properties can fulfill a variety of biomedical applications.

Collagen can be fabricated into multiple physical forms, such as sponges, films, gels, nanofibers, and powders.⁴ As the principal component of the extracellular matrix, collagen is not only affected by its inherent physiological characteristics but also influenced by its structure and stability. A three-dimensional structural collagen sponge can be constructed from only collagen molecules. Collagen sponge materials retain their inherent triple helical structure and have the biological characteristics of collagen molecules.⁶ Soft spongy collagen can attach to different types of wound, absorb excess tissue exudate, reduce pain, and can promote the formation of epithelial and granulation tissue in a wound.⁷ Over the past few years, collagen and collagen-based biomaterials have been widely used for rapid hemostasis, wound dressings, tissue engineering, and matrices for antibiotic delivery and have been successfully used clinically.^8,9 Many collagen-based wound dressing products have been developed, a number of which have been approved and are commercially available for hemostasis and the treatment of trauma, burns, chronic ulcers, and other wounds.^7,10

Collagen is principally sourced from the skin and bones of mammals such as cattle and pigs. However, a number of constraints have hindered the development of novel collagen products. There is a high risk of contracting a disease from human-animal cross-infection, for example bovine spongiform encephalopathy (BSE), transmissible spongiform encephalopathy (TSE), and foot and mouth disease (FMD).^1,4,11 In addition, religious issues/beliefs have reduced the demand for collagen of terrestrial animal origin. Both Judaism and Islam ban pork products, while Hindus do not eat bovine products. Therefore, a more acceptable and safer alternative source of collagen would be of particular benefit.¹¹

Over recent years, marine collagen has attracted increasing attention and has become an alternative source of collagen in biomedical fields to replace land mammal sources.¹ In comparison, there are no religious restrictions and it represents no risk of zoonosis, so safety may be superior to that of mammalian collagen.^12,13 In particular, fish collagen for use in wound dressings may be safer and more widely applicable.^14,15 The skin, scales, and bones of fish are rich in collagen and it has become a developing trend to extract collagen from aquatic animals and amphibians.⁶ Where fish skin, scales, and bones are the by-products in the seafood processing industry they have traditionally been considered waste and have simply been discarded, if not incorporated into animal feed. Thus, the cost of sourcing these by-products is extremely low. In comparison, the quantity of mammalian products used in animal husbandry is considerably less than that of aquatic products. As the growth cycle of mammals is longer, the raw materials from bovine and swine skin are more expensive with limited availability. In addition, the process of extraction of collagen from mammalian skin is more complex and more expensive than that of marine collagen, which is water-soluble and can be easily extracted.¹⁶ Hence, collagen obtained from fish skin, scales, and bones represents a superior choice which reduces cost, resources, waste, and environmental pollution, resulting in a better economic lifecycle.¹⁷ Tilapia skin is a high yield source of collagen of good quality, the use of which may be economically beneficial.¹

The aim of the present study was the development of collagen sponge materials that are effective medical dressings displaying rapid hemostasis and promoting wound healing. Thus, dialyzed tilapia skin collagen sponge (DTSCS) and self-assembled tilapia skin collagen sponge (STSCS) were prepared by freeze-drying, without chemical cross-linking agents or other chemically synthesized macromolecules. These factors improved the biocompatibility profile of the materials. Their micromorphology, water absorption, porosity, and thermal stability were investigated in detail. Safety and biocompatibility of DTSCS and STSCS were evaluated by hemolysis, cytotoxicity, and in vivo biocompatibility testing. Their capability to cause rapid hemostasis and promote wound healing were investigated in a whole blood clotting assay, hepatic trauma model, femoral artery bleeding model, and full-thickness skin defect model in rats.

Materials and methods

Materials

Purified tilapia skin collagen was provided by Shandong International Biotechnology Park Development Co., Ltd., Yantai, China. A commercial bovine collagen sponge (BCS, Kejibang^® Medical Collagen Sponge, Biot Biology Technology Co., Ltd, China) was used as a control material for comparison in various tests.

Sprague-Dawley (SD) rats, weighing 200–250 g, 6–7 weeks of age, were provided by Luye Pharma Experimental Animal Center, Yantai, China. All animal procedures were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee. The rats were individually housed with ad libitum access to food and water and maintained at 22°C, 60%–70% humidity in a 12 h light/dark cycle.

Preparation of DTSCS and STSCS

Dialyzed tilapia skin collagen sponges (DTSCS) and self-assembled tilapia skin collagen sponges (STSCS) were constructed from purified tilapia skin collagen, as shown in Figure 1. The method of preparation has been described in detail previously.^3,4,18 Briefly, purified tilapia skin collagen was dissolved in acetic acid, which was dialyzed until the pH had risen close to 7, then freeze-dried to obtain DTSCS. STSCS was prepared using a self-assembly process. Collagen molecules were self-assembled to form collagen microfibers, with thicker and longer microfibers forming between the microfibers. The acidic solution of collagen was mixed with PBS and the pH adjusted to 7. A self-assembled collagen gel was obtained after standing for one hour. Collagen microfibers were precipitated after thorough stirring, collected in a sieve, resuspended then freeze-dried, in order to obtain STSCS.

Characterization

SEM analysis

To observe the microstructure of DTSCS and STSCS, collagen sponge samples were mounted on a metal stub with conductive tape and sputter-coated with a thin layer of gold. The morphology of the collagen sponges was captured by scanning electron microscopy (EVO 15LS, Zeiss, Germany) at an accelerating voltage of 5 kV.

Water absorption

Water absorption of DTSCS and STSCS samples (0.02 g) was immersed in 20 mL PBS (0.01 M, pH 7.4) then removed after water absorption had completed. Superficial water was removed from the samples using filter paper which were then weighed immediately.¹¹ Experiments were performed in triplicate. Water absorption was calculated using equation (1).

Water absorption = \frac{W_{s} - W_{i}}{W_{i}}

(1)

W_i: initial weight,

W_s: weight after reaching swelling equilibrium.

Measurement of porosity

The porosity of DTSCS and STSCS was measured using the liquid substitution method.¹⁹ Ethanol can easily penetrate polymers and does not cause shrinkage or expansion. The samples were soaked in ethanol of volume V0 for 5 min. A series of brief evacuation and re-pressurization cycles were conducted to remove any air bubbles in the samples. After recording the total volume Va, the sample was removed from the ethanol and the volume remaining was recorded as Vb. The porosity of the sponges was calculated using equation (2). The experiment was repeated three times.

Porosity = \frac{(V_{a} - V_{b}) - (V_{a} - V_{0})}{V_{a} - V_{b}} = \frac{V_{0} - V_{b}}{V_{a} - V_{b}}

(2)

DSC analysis

Differential scanning calorimetry (DSC) was employed to evaluate the thermal properties and thermal denaturation temperature (T_d) of the DTSCS and STSCS.²⁰ The analysis was performed using a differential scanning calorimeter (DSC214, Netzsch, Germany). Experiments were conducted in triplicate on samples (5−10 mg) within an atmosphere of nitrogen by applying a constant heating rate of 5°C/min from 28°C to 92°C.

Biocompatibility evaluation

Cytotoxicity evaluation

The cytotoxicity of collagen sponges was determined using L929 cells using a 3-(4,5-dimethylthiazol-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT) assay in accordance with ISO 10993-5.²¹ Extracts of BCS, DTSCS, and STSCS were prepared using MEM culture medium (Gibco) supplemented with 10% fetal bovine serum (FBS) (v/v) (Gibco) and 1% penicillin-streptomycin solution (v/v) (Solarbio) (hereafter referred to as “complete medium”) at a ratio of 6 cm²/mL at 37°C for 24 h. L929 cells were cultured in 96-well plates with complete medium at 37°C for 24 h. The culture medium was then removed and fresh complete medium containing 100%, 75%, 50%, and 25% extract of each sample then cultured in a CO₂ incubator at 37°C for 24 and 48 h. Extracts of sterile PE using complete medium represented the negative control group. Complete medium containing 0.64% phenol represented the positive control group. Complete medium alone was used as the blank control group. Cell viability was measured after 24 and 48 h using MTT and 10% dimethyl sulfoxide. Absorbance at 570 nm and 650 nm was measured in a microplate reader (Epoch, BioTek, USA). Relative cell growth rate (RGR) was calculated as shown in equation (3).

RGR = \frac{{OD}_{570 e} - {OD}_{650 e}}{{OD}_{570 b} - {OD}_{650 b}} \times 100 %

(3)

OD_570e: absorbance of the experimental group,

OD_570b: absorbance of the blank group.

Hemolysis test

Hemolysis testing was conducted to evaluate blood compatibility using a method reported previously.²² Extracts of BCS, DTSCS, and STSCS were created using normal saline at 37°C for 72 h. Blood was collected from the ear vein of a New Zealand rabbits into a tube containing 2% potassium oxalate as anticoagulant. A volume of 5 mL BCS, DTSCS, or STSCS extract was added to each tube and then 0.1 mL of blood diluted with normal saline (4/5, v/v) was added and incubated at 37°C for 60 mins. Blood treated with deionized water acted as a positive control and blood with added normal saline represented the negative control. All samples were centrifuged after incubation and the absorption at 545 nm of the supernatants recorded in a UV/Vis spectrophotometer (UV2450, Shimazu, Japan). Hemolysis was calculated using equation (4).

Hemolysis rate (%) = \frac{{(A}_{1} - A_{2})}{(A_{3} - A_{2})} \times 100 %

(4)

A₁: absorbance of the collagen sponge group,

A₂: absorbance of the negative control group,

A_3: absorbance of the positive control group.

In vivo biocompatibility

The in vivo biocompatibility of DTSCS and STSCS was evaluated in SD rats (female, 200–250 g, 6–7 week) in accordance with ISO 10993-6.²³ Two transverse incisions were created in the dorsal skin of the rats on the left and right sides approximately 1.5 cm long. Four subcutaneous bursae were isolated by blunt separation above and below the incision. A sample (1 cm × 1 cm) was inserted into each subcutaneous bursa and the wound sutured following implantation. All animals were maintained within suitable conditions until the end of the experiment. Samples were removed for histological examination after 7 and 14 days after implantation. Implants were harvested by carefully cutting the surrounding tissue and fixing in 10% formaldehyde. Fixed tissues were embedded in paraffin, sectioned then stained with hematoxylin-eosin (H&E) prior to histopathological examination. The extent of fibrosis and inflammatory cell infiltration into the implants were scored relative to those of a commercial dressing as follows: 0.0–2.9: non-irritant; 3.0–8.9: slight irritant; 9.0–15: moderate irritant; greater than 15: severe irritant. Materials scoring none to slightly irritant are considered safe for use as a medical device.

Evaluation of hemostatic performance

Whole blood clotting assay

The in vitro hemostatic performance of DTSCS and STSCS was evaluated using a whole blood coagulation assay.¹¹ Whole blood was obtained from the ear vein of a New Zealand rabbit and collected into a tube containing sodium citrate. Approximately 10 mg of each sample was placed separately into centrifuge tubes and preheated to 37°C. Anticoagulated blood (450 µL) and CaCl₂ solution (0.2 M, 50 µL) were added to the centrifuge tubes so that each sample was completely immersed. The tubes were then gently shaken and incubated at 37°C for 5 min. Deionized water (40 mL) was then added and incubated for a further 5 mins. Absorbance of the supernatant was measured at 545 nm using a UV/Vis spectrophotometer (UV2450, Shimazu, Japan). The absorbance at 545 nm of 450 µL anticoagulated blood and CaCl₂ solution (0.2 M, 50 µL) mixed with 40 mL deionized water was also measured. The relative degree of blood clotting was evaluated using a blood coagulation index (BCI), calculated using equation (5):

BCI (%) = \frac{A_{1}}{A_{2}} \times 100 %

(5)

A₁: absorbance value of blood with experimental sample,

A₂: absorbance value of anticoagulated blood mixed with deionized water.

In vivo hemostatic capability

To evaluate the hemostatic potential of DTSCS and STSCS in vivo, a hepatic trauma model and femoral artery bleeding model (SD rats, 200–250 g, 6–7 weeks, male) were employed.²⁴ Briefly, for the hepatic trauma model, rats were anesthetized with chloral hydrate (10%) and immobilized on a surgical corkboard. The rat liver was exposed through an abdominal incision. Tissue fluid was gently wiped away and a piece of gauze was placed under the liver to prevent any subsequent movement. Bleeding from the liver was induced by the creation of an X-shaped incision 1 cm × 1 cm × 0.3 cm using a scalpel. In the femoral artery bleeding model, a 3 cm-long incision was carefully created along the medial side of the rear leg close to the femoral artery. The connective tissue and muscle were removed layer-by-layer to expose the femoral artery. Any tissue effusion or hemorrhage of capillaries on the surface of the femoral artery was removed. One-third of the femoral artery was cut laterally with surgical scissors to induce bleeding. The initial outflow of blood was wiped away and then a layer of experimental sponge was immediately overlaid onto the hemorrhaging area and the hemostatic time recorded.

In vivo evaluation of wound healing

Fifteen SD rats (200–250 g, 6–7 weeks) were divided into three groups: 7 days group, 14 days group, and 21 days group. Animals were anesthetized with chloral hydrate (10%) during surgery. Surgery was conducted in accordance with standard aseptic operating procedures. After removal of the dorsal fur, the epidermis was cleaned with iodophor and four wounds were created on the left and right sides of the back. Each wound had a diameter of 1.5 cm and were 2 cm apart. The depth of each wound was sufficient to reach the fascia layer. Samples of DTSCS and STSCS were secured to the wound with surgical sutures after irradiation sterilization. Bovine collagen sponge (BCS) and gauze were stitched to the other two wounds. After applying wound dressings, each rat was housed separately and fed carefully. Each group of rats was anesthetized at either day 7, 14, or 21, as appropriate. Images of the wounds were recorded to analyze the different wound dressings. Wound area was analyzed by ImageJ software. The rate of wound closure was calculated using equation (6). New skin tissue and normal skin surrounding the wounds in the 14 days and 21 days groups were harvested and analyzed by H&E staining in accordance with previous research.²⁵

\begin{array}{l} Wound closure rate (%) \\ = \frac{A_{0} - A_{t}}{A_{0}} \times 100 % \end{array}

(6)

A₀: wound area on day 0,

A_t: wound area on day 7, 14, or 21.

Statistical analysis

The experimental results are expressed in terms of means ± standard deviation (SD) and analyzed using SPSS v19.0 software (IBM SPSS Statistics, Ehningen, Germany).

Results and discussion

Characterization

SEM analysis

Collagen has many potential applications in biomedical engineering, such as wound dressings, hemostatic agents, or matrices for cell proliferation. The microstructure of the material has a considerable influence on its cellular properties. Collagen sponge materials with an ideal pore size allow solutes to permeate through interconnected pores that promote cell migration, nutrient flow, and vascular growth.¹⁸ An ideal structure for tissue repair should be characterized by large pores with high porosity. Collagen sponges with pore sizes ranging from 63 μm to 150 μm may be suitable for the adhesion and growth of fibroblasts.²⁶ Images of DTSCS (Figure 2(a)) and STSCS (Figure 2(b)) show that these freeze-dried sponges were white and with a surface that was flat and uniform, by direct observation. At a magnification of 200×, the structure of DTSCS (Figure 2(c)) indicated dense collagen fibers distributed in strips. STCSC (Figure 2(e)) had a loose structure with collagen fibers that were flaky. At a magnification of 1000×, the size of pores on the surface of DTSCS (Figure 2(c)) was shown to be small and evenly distributed, at approximately 15–40 μm. STSCS (Figure 2(f)) displayed irregular lamellar folded structures with larger pores (50–150 μm) widely distributed and containing a microfiber structure not found in DTSCS. The results indicate that the two sponge types had different porosities and reticular structures. The structure and pore size of STSCS appeared more suitable for cell growth and biomedical applications.

Water absorption and porosity measurements

The water absorption and porosity of the collagen sponges refer directly to the wettability and plasma adsorption properties while functioning as hemostatic materials.²⁷ The porosity of collagen-based sponges for biomedical applications should higher than 90%.⁶ The internal surface area of the collagen sponge increases with increased porosity, providing greater capability to absorb a large quantity of blood from a wound quickly, aggregating platelets and promoting the concentration of blood coagulation factors to accelerate the hemostasis process. Additionally, a good pore structure allows the collagen sponge to provide more space for cell attachment, growth, and proliferation during tissue repair. Water absorption of DTSCS and STSCS was 33.6- and 20.5-fold greater than previously, meeting the requirements of specification YY/T 1511-2017 Collagen Sponge (The water absorbed by samples should not be less than 20-fold greater than its weight). The porosity of DTSCS and STSCS was 91.25% and 97.38%, respectively. Thus, the water absorption and porosity of DTSCS and STSCS satisfied the requirements of ideal tissue engineering materials.

DSC analysis

Because some materials are unstable or degrade slowly at physiological temperatures, the thermal stability of collagen sponges is worth investigation so that they meet the requirements of medical dressings.¹⁵ T_d is important for the clinical applications of biomaterial scaffolds in regenerative medicine. While heating, the triple helical structure of collagen can break. The typical endothermic peak is often called T_d.²⁰ DSC curves for STSCS and DTSCS can be viewed in Figure 3. The T_d of both DTSCS and STSCS was higher than the physiological temperature, demonstrating that DTSCS and STSCS would not be denatured over a short period of time following contact with the wound surface.

Biocompatibility evaluation

Hemolysis assay

The most important factor for medical dressing materials is patient safety. Although tilapia skin collagen itself may be safe, there may be residues of solvent or possible contamination by microorganisms during the process of extraction of the raw materials and preparation of the sponges. Hemolysis refers to the release of hemoglobin from red blood cells (RBCs) into plasma due to their rupture.²⁸ Low rates of hemolysis indicate that fewer RBCs have been destroyed, suggesting a material with good blood compatibility. Hemolysis was clearly apparent in the deionized water group, while hemolysis was not apparent in the normal saline, BCS, DTSCS, or STSCS groups. The results demonstrate that BCS, DTSCS, and STSCS displayed excellent hemocompatibility with a hemolytic ratio considerably lower than 5% (Table 1), within the range allowable for biomaterials used for medical applications.

Table 1.

The absorption value and hemolysis rate of different groups.

Groups	Absorption value	Hemolysis rate (%)
Normal saline	0.019 ± 0.006	/
Deionized water	0.803 ± 0.021	/
BCS	0.022 ± 0.003	0.38
DTSCS	0.024 ± 0.008	0.64
STSCS	0.021 ± 0.003	0.26

SD: standard deviation; BSC: bovine collagen sponge; DTSCS: dialyzed tilapia skin collagen sponge; STSCS: self-assembled tilapia skin collagen sponge.

Date presented mean ± SD (n = 3).

Cytotoxicity evaluation

Biological safety assessment is essential for a new biomedical material. Cytotoxicity testing is used as a screening test in the early stages of biological safety assessment of medical materials because of its high sensitivity, simplicity, and rapidity. The test can be categorized as either an indirect or direct contact method, or extraction method. At present, the majority of cytotoxicity testing of medical biomaterials is conducted by testing extracts. In this evaluation, L929 cells were used to evaluate the cytotoxicity of BCS, DTSCS, and STSCS by extraction. Cell viability after 24 h and 48 h following contact of the extracted fluid with BCS, DTSCS, and STSCS is presented in Table 2. The RGR values of the concentration groups (25%, 50%, 75%, and 100%) for DTSCS and STSCS were substantially higher than 70%. Therefore, it was clear that DTSCS and STSCS did not exhibit potential cytotoxicity.

Table 2.

The RGR value of different groups after 24 and 48 h.

Groups	Extract concentration (%)	RGR (%)
Groups	Extract concentration (%)	24 h	48 h
Blank control	/	100	100
Negative control	/	87.6	85.3
Positive control	/	4.52	3.21
BCS	100	91.29	93.14
	75	96.58	96.86
	50	98.82	97.33
	25	99.85	99.73
DTSCS	100	96.36	93.43
	75	98.08	96.64
	50	101.75	98.09
	25	102.34	98.31
STSCS	100	89.82	93.61
	75	93.42	97.72
	50	92.39	96.29
	25	96.46	101.14

BCS: bovine collagen sponge; DTSCS: dialyzed tilapia skin collagen sponge; STSCS: self-assembled tilapia skin collagen sponge.

In vivo biocompatibility test

Subcutaneous implantation refers to placing a test material of a defined size under the skin of an animal to observe any macroscopic and histopathological changes in the host tissues over a range of implantation durations to evaluate the biological response around the implant. It is a principal method for evaluating biocompatibility and safety. Following implantation, wound dressings may cause inflammation in skin tissue. Depending on the stage of inflammation, polymorphonuclear (PMN) cells, especially neutrophils, are responsible for acute inflammation whereas mononuclear cells, such as lymphocytes, macrophages, or giant cells are involved in chronic inflammation. Inflammatory cells and their secretions can remove foreign bodies and may also lead to a variety of histopathological changes, especially fibrosis, fat infiltration, and necrosis.²⁹

In the present study, during the periods from day 7 to day 14 post-implantation, histomorphological changes and inflammatory reaction scores in each tested material are presented in Table 3 and Figure 4. Seven days post-implantation, all groups induced both acute and chronic (or mixed) inflammation characterized by the persistence of PMNs in association with lymphocytes and the deposition of plasma cells. On the 14th day, fibroblast proliferation and neovascularization were observed within the tissue for all three materials. The acute inflammatory phenomenon of the STSCS group weakened significantly over time and gradually reverted to a chronic inflammatory reaction. In the DTSCS and BCS groups, the reaction almost completely transitioned to a chronic inflammatory state, with a number of lymphocytes infiltrating into the visual field. The three groups of materials demonstrated no tissue degeneration, no necrosis, and no fatty infiltration. The results indicated that an abnormal histocompatibility reaction in all groups was mild, confirming that DTSCS and STSCS did not induce irritation to the surrounding tissue compared with a commercial bovine collagen sponge.

Table 3.

Cellular response score after implantation of different collagen sponges.

Groups	Week	PMN	Lymphocytes	Plasma cells	Fibroblast cells	New vessels
BCS	1	0.8 ± 0.6	2.0	1.1 ± 0.7	1.5 ± 0.3	0.0
BCS	2	0.0	1.3 ± 1.0	0.0	1.9 ± 0.8	2.0
DTSCS	1	1.2 ± 0.5	2.57 ± 1.2	0.81 ± 0.5	1.7 ± 0.8	1.0
DTSCS	2	0.0	2.0 ± 0.3	0.0	2.17 ± 0.6	2.3 ± 0.4
STSCS	1	1.5 ± 0.3	3.5 ± 2.5	0.61 ± 0.5	1.24 ± 0.8	0.0
STSCS	2	0.3 ± 0.5	2.8 ± 1.3	0.46 ± 0.2	1.71 ± 1.2	1.0 ± 0.5

DTSCS: dialyzed tilapia skin collagen sponge; STSCS: self-assembled tilapia skin collagen sponge; SD: standard deviation.

Date presented mean ± SD (n = 3).

Evaluation of hemostatic performance

Whole blood clotting assay

The whole blood clotting test was performed to determine the blood clotting capacity of DTSCS and STSCS in vitro. The blood began to coagulate as soon as the collagen sponges contacted calcified whole blood. The addition of deionized water released hemoglobin from the free red blood cells (RBCs). The concentration of hemoglobin and the number of free RBCs was reflected in the absorbance at 545 nm. Lower absorbance values and BCI suggest a stronger clotting capability of a collagen sponge. The whole blood clotting test results are displayed in Figure 5. Of all the materials, the absorbance value of the STSCS group was lowest, followed by the BCS and DTSCS groups. The absorbance values of the three groups were significantly different from the gauze group (p < 0.01), with no significant difference between the BCS, DTSCS, and STSCS groups. This demonstrates that DTSCS and STSCS displayed a good hemostatic effect in vitro compared with the BCS. The hemostatic effect of STSCS was slightly greater than that of DTSCS.

In vivo hemostatic capability

Rapid hemostasis is of great importance in a clinical setting. A perfect hemostatic agent would cause rapid hemostasis and also be convenient, safe, and economically favorable.¹¹ Collagen-based materials have a clear hemostatic effect in vivo. On one hand, collagen absorbs blood from the wound, forming a blood scab that seals a ruptured vessel. On the other hand, it can activate endogenous blood coagulation and provide a physical matrix. The promotion of platelet adhesion and coagulation factor binding will accelerate thrombosis.³⁰ The two mechanisms reinforce each other and could prevent bleeding quickly and effectively.

The hemostatic effect in vivo was measured using a rat hepatic trauma model and femoral artery bleeding model. DTSCS, STSCS, BCS, and gauze were assessed in the hemostatic experiment. The time for hemostasis in each group is displayed in Figure 6. In the hepatic trauma model in rats, no additional bleeding was observed from the liver after 140 s for the STSCS and BCS groups, and 150 s for the DTSCS group. In comparison, the liver did not stop bleeding even after applying gauze for 200 s. In the femoral artery bleeding model, the first group to stop bleeding was the STSCS group, within 60 s. The BCS and STSCS groups did not continue to bleed after 80 s and 130 s, respectively. Gauze continued to bleed at 300 s. The results indicate that the collagen sponges had hemostatic performance superior to that of gauze, with a shorter hemostasis time. Of the three types of collagen sponge, the hemostatic effect of STSCS was superior to that of BCS and DTSCS.

In vivo evaluation of wound healing

Wound healing is not a simple physiological process. It involves multiple overlapping stages, including the inflammatory phase, and the formation and remodeling of new tissues.¹⁹ The purpose of a wound dressing is to accelerate the process of wound healing by the prevention of bacterial infection and the acceleration of tissue regeneration. As a common natural biological macromolecule, collagen contributes to guided tissue regeneration. Collagen breakdown products are also chemotactic for the cells required for granulation tissue formation.⁴ The wound healing properties of DTSCS and STSCS in vivo were observed in rats. Wounds were imaged on days 7, 14, and 21. No visible bacterial biofilm or infection could be seen, with wound area decreasing in a time-dependent manner in all groups, as displayed in Figure 7. The gauze group showed severe adhesion to the wound, with a large volume of tissue fluid infiltration with wound healing least after 7 days of treatment of all materials tested. The wound closure rate of each group is presented in Figure 8. In contrast, the size of the wounds reduced significantly after 7 days of collagen sponge implantation, especially in the BCS and STSCS groups (p < 0.05, compared with the gauze group). The wound closure rate after 7 days in the DTSCS group was slightly lower than in the STSCS group, although there was no statistical difference. The wound closure rate in the gauze group after 14 days of treatment was significantly higher than after 7 days. The wound closure rate of every group was close to 100% after 21 days of treatment, the wounds essentially healing completely in all groups. These results demonstrate that there was a trend for STSCS to be more effective than DTSCS in the promotion of wound closure in the early stages of wound healing although there was no significant difference between them.

In addition, freshly formed skin was evaluated by H&E analysis after 14 and 21 days of treatment, respectively, as shown in Figure 9. Defective granulation tissue was replaced by collagen fibers after 14 days of treatment. All groups displayed a degree of lymphocyte infiltration, and a large number of fibroblasts appeared. The fibroblasts in the STSCS group were most widely distributed with fibrous tissue from local hyperplasia apparent. Tissue organization was second best in the BCS group, and worst in the DTSCS group. New capillaries of different sizes were visible in all three groups, and epithelium began to be repaired and form granulation tissue. The epidermis was almost complete, gradually forming a hierarchical structure composed of dermis and epidermis. Histological observation 21 days after wounding demonstrated that the numbers of capillaries in the wounds of the three groups had increased significantly, with red granulation tissue gradually transforming into connective tissue, with thickening of the epidermis. Only a very small number of lymphocytes were observed in DTSCS and BCS, while inflammatory cells had essentially disappeared in STSCS, with epithelial tissue completely covering the wound surface, and healing being most advanced of all the materials. The results confirmed that STSCS was superior at promoting wound healing, similar to that of commercial bovine collagen sponge and better than DTSCS, both in terms of wound closure rate and evaluation by H&E analysis.

Conclusion

In summary, DTSCS and STSCS prepared from tilapia skin collagen, were characterized and their biocompatibility, hemostatic, and wound healing properties tested and compared with commercial bovine collagen sponge. DTSCS and STSCS displayed a loose and porous network structure, with high water absorption, high porosity, with a T_d much higher than the physiological temperature of humans. The microstructure, water absorption, and porosity of the collagen sponges prepared by the two methods were quite different. There were also differences in the hemostatic and wound healing properties. Although there was no significant difference between DTSCS and STSCS in these respects, there was still a trend for STSCS to be more effective than DTSCS in causing rapid hemostasis and promoting wound healing compared with DTSCS. DTSCS and STSCS exhibited good biocompatibility, with a rapid hemostatic effect, and clearly promoted the healing process comparable to that of commercial bovine collagen sponge. The results demonstrate that these collagen sponges fabricated from tilapia skin collagen have the potential to be effective functional biomaterials for use as hemostatic sponges and wound dressings through the optimization of their structure and method of preparation in the future.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the "Yantai science and technology plan project" (NO.2020XDRH108 and NO.2020XDRH109).

ORCID iD

Tong Wang

References

Sun

Jiang

, et al. Nile tilapia skin collagen sponge modified with chemical cross-linkers as a biomedical hemostatic material. Colloids Surf B Biointerfaces 2017; 159: 89–96.

Tang

Jin

, et al. Physicochemical properties and biocompatibility evaluation of collagen from the skin of giant croaker (Nibea japonica). Mar Drugs 2018; 16: 222.

Gao

Wang

, et al. Evaluation of biocompatibility and immunogenicity of micro/nanofiber materials based on tilapia skin collagen. J Biomater Appl 2019; 33: 1118–1127.

Pal

Srivas

Dadhich

, et al. Accelerating full thickness wound healing using collagen sponge of mrigal fish (Cirrhinus cirrhosus) scale origin. Int J Biol Macromol 2016; 93: 1507–1518.

Barros

Aroso

Silva

, et al. Water and carbon dioxide: green solvents for the extraction of collagen/gelatin from marine sponges. ACS Sustain Chem Eng 2015; 3: 254–260.

Luo

Meng

Zheng

, et al. Fabrication and characterization of Chinese giant salamander skin composite collagen sponge as a high-strength rapid hemostatic material. J Biomater Sci Polym Ed 2018; 29: 1933–1948.

Cheng

Zhao

, et al. Sponges of carboxymethyl chitosan grafted with collagen peptides for wound healing. Int J Mol Sci 2019; 20: 3890.

Liu

Gohi

Zeng

, et al. Effect of collagen peptides-carboxymethyl chitosan microspheres on ultraviolet induced damages. Mater Express 2015; 5: 497–504.

Abdollahi

Rezaei

Jafarpour

, et al. Sequential extraction of gel-forming proteins, collagen and collagen hydrolysate from gutted silver carp (Hypophthalmichthys molitrix), a biorefinery approach. Food Chem 2018; 242: 568–578.

10.

Liu

Xue

, et al. Hydrolyzed tilapia fish collagen modulates the biological behavior of macrophages under inflammatory conditions. RSC Adv 2015; 5: 30727–30736.

11.

Seo

Lee

Sultan

, et al. In vitro and in vivo evaluation of the duck’s feet collagen sponge for hemostatic applications. J Biomater Appl 2017; 32: 484–491.

12.

Yamamoto

Yoshizawa

Yanagiguchi

, et al. The characterization of fish (tilapia) collagen sponge as a biomaterial. Int J Polym Sci 2015; 2015: 1–5.

13.

Wang

Zhang

, et al. Composite electrospun nanomembranes of fish scale collagen peptides/chito-oligosaccharides: antibacterial properties and potential for wound dressing. Int J Nanomed 2011; 6: 667.

14.

Lukáč

Hartinger

Mlček

, et al. A novel gentamicin-releasing wound dressing prepared from freshwater fish Cyprinus carpio collagen cross-linked with carbodiimide. J Bioact Compat Polym 2019; 34: 246–262.

15.

Chen

Jin

Yang

, et al. Physicochemical, antioxidant properties of giant croaker (Nibea japonica) swim bladders collagen and wound healing evaluation. Int J Biol Macromol 2019; 138: 483–491.

16.

Lim

Hwang

, et al. Marine collagen as a promising biomaterial for biomedical applications. Mar Drugs 2019; 17: 467.

17.

Sun

Song

, et al. Characterization of Pacific cod (Gadus macrocephalus) skin collagen and fabrication of collagen sponge as a good biocompatible biomedical material. Process Biochem 2017; 63: 229–235.

18.

Zhu

Yuan

Yin

, et al. Self-assembly of collagen-based biomaterials: preparation, characterizations and biomedical applications. J Mater Chem B 2018; 6: 2650–2676.

19.

Muthukumar

Prabu

Ghosh

, et al. Fish scale collagen sponge incorporated with Macrotyloma uniflorum plant extract as a possible wound/burn dressing material. Colloids Surf B Biointerfaces 2014; 113: 207–212.

20.

Zhao

Gao

, et al. Collagen/polyethylene oxide nanofibrous membranes with improved hemostasis and cytocompatibility for wound dressing. Appl Sci 2018; 8: 1226.

21.

ISO 10993-5. Biological evaluation of medical devices-part 5: tests for in vitro cytotoxicity.

22.

Wang

, et al. Nanocomposite sponges of sodium alginate/graphene oxide/polyvinyl alcohol as potential wound dressing: in vitro and in vivo evaluation. Compos B Eng 2019; 167: 396–405.

23.

ISO 10993-6. Biological evaluation of medical devices. Part 6: tests for local effects after implantation.

24.

Sundaram

Kaliannagounder

Selvaprithiviraj

, et al. Bioadhesive, hemostatic, and antibacterial in situ chitin-fibrin nanocomposite gel for controlling bleeding and preventing infections at mediastinum. ACS Sustain Chem Eng 2018; 6: 7826–7840.

25.

Vigneswari

Murugaiyah

Kaur

, et al. Biomacromolecule immobilization: grafting of fish-scale collagen peptides onto aminolyzed P(3HB-co-4HB) scaffolds as a potential wound dressing. Biomed Mater 2016; 11: 055009.

26.

Alagha

Nourallah

Hariri

Characterization of dexamethasone loaded collagen-chitosan sponge and in vitro release study. J Drug Deliv Sci Technol 2020; 55: 101449.

27.

Cheng

Liu

, et al. Reinforced collagen with oxidized microcrystalline cellulose shows improved hemostatic effects. Carbohydr Polym 2017; 165: 30–38.

28.

Liu

Yang

Effects of l-arginine immobilization on the anticoagulant activity and hemolytic property of polyethylene terephthalate films. Appl Surf Sci 2010; 256: 3977–3981.

29.

Ampawong

Aramwit

A study of long-term stability and antimicrobial activity of chlorhexidine, polyhexamethylene biguanide, and silver nanoparticle incorporated in sericin-based wound dressing. J Biomater Sci Polym Ed 2017; 28: 1286–1302.

30.

Kumar

Taylor

Jalan

, et al. A nanostructured synthetic collagen mimic for hemostasis. Biomacromolecules 2014; 15: 1484–1490.