Modification and crosslinking of gelatin-based biomaterials as tissue adhesives

Accepted Manuscript Title: Modification and Crosslinking of Gelatin-Based Biomaterials as Tissue Adhesives Authors: Yi Liu, Sai Cheong NG, Jiashing Yu, Wei-Bor Tsai PII: DOI: Reference:

S0927-7765(18)30772-0 https://doi.org/10.1016/j.colsurfb.2018.10.077 COLSUB 9761

To appear in:

Colloids and Surfaces B: Biointerfaces

Received date: Revised date: Accepted date:

28 April 2018 29 September 2018 27 October 2018

Please cite this article as: Liu Y, Cheong NG S, Yu J, Tsai W-Bor, Modification and Crosslinking of Gelatin-Based Biomaterials as Tissue Adhesives, Colloids and Surfaces B: Biointerfaces (2018), https://doi.org/10.1016/j.colsurfb.2018.10.077 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Modification and Crosslinking of Gelatin-Based Biomaterials as Tissue Adhesives

IP T

Yi Liu1, Sai Cheong NG1, Jiashing Yu1,* and Wei-Bor Tsai1,*

1. Department of Chemical Engineering, National Taiwan University, No. 1, Sec. 4,

SC R

Roosevelt Rd., Taipei 106, Taiwan

A

CC E

PT

ED

M

A

N

U

Graphical abstract

Highlights  Tissue adhesives are desired with low cytotoxicity and high tissue adhesion.  Catechol- and phenol-modified gelatin were synthesized respectively.  Adhesives were prepared via photochemistry or mussel-inspired chemistry.  Adhesives were evaluated by mechanical properties, tissue adhesion and toxicity.

1

Abstract Tissue adhesives have been developed to overcome the difficulties of conventional wound closure techniques (e.g. sutures and staples), such as the potential for collateral damage and difficulty of stopping body fluid and gas. At the same time,

IP T

it provides advantages such as simpler implementation, less painful, and does not require removal. However, representative adhesives such as cyanoacrylates and fibrin

SC R

glues are plagued by cytotoxicity and low adhesion. In this study, we choose instead

gelatin as the backbone of adhesive, due to its biocompatibility, biodegradability, and low cost. Firstly, catechol-modified gelatin and phenol-modified gelatin were

U

synthesized via an EDC/NHS chemistry. Then, gelatin-based adhesives were prepared

N

via ruthenium-based photochemistry, including photo-crosslinked gelatin (PG),

A

phenol-modified gelatin (PPG), and catechol-modified gelatin (PCG). We also

M

compared the photo-crosslinked adhesives to the recently reported ion-crosslinked catechol-modified gelatin. Our results indicate that gelatin-based adhesives

ED

demonstrate lower swelling index, great degradability, and low cytotoxicity. This

CC E

healing.

PT

shows that gelatin-based adhesives demonstrate great potential for wound closure and

Keywords: tissue adhesive, gelatin, dopamine, phloretic acid, photo-crosslink, ion-

A

crosslink.

1. Introduction Wound closure and healing are essential for millions of surgical patients each year. Conventional techniques such as sutures and staples have many limitations such 2

as the need for a relatively long application time, the possibility of causing further tissue damage and inflammatory responses, and inefficiency in stopping body fluid or gas for larger scale wounds [1-3]. Tissue adhesives offer comparative advantages, including simpler implementation, shorter time, less painful, and no need for removal [1]. Cyanoacrylates and fibrin sealants are representative types of commercial

IP T

adhesives. Cyanoacrylates polymerize rapidly for hemostasis and achieve a strong bond to tissue [4]. Fibrin sealants form a stable clot in a short time and are usually

SC R

applied to small blood vessels [5]. However, the toxicity of degradation byproducts

for cyanoacrylates and low adhesive strength of fibrin sealants limit their applications,

U

respectively [4, 6]. To design the ideal tissue adhesive, biocompatibility and high

N

adhesive tissue strength are desired.

A

A ruthenium-based photochemistry has been used to form crosslinked proteins as

M

tissue adhesives, employing methods described by Kodadek and colleagues in 1999 [7]. The method was used to form tissue adhesives using various proteins, such as

ED

resilin [8], fibrinogen [9, 10], collagen [11, 12], gelatin [13], and keratin [14], through the photo-crosslinking using tris(2,2’-bipyridyl)dichlororuthenium(II) hexahydrate

PT

(Ru(bpy)3Cl2) and persulfate in the presence of ~450 nm light. The cell viability was maintained with very few Ru(bpy)3Cl2 and persulfate, and the irradiation of visible

CC E

light. To enhance the mechanical and adhesive properties, phenol and catechol can be conjugated with protein to increase the crosslinking density photo-crosslinked

A

adhesives. In addition, phenol and catechol can react with near nucleophilic side chain (such as amino and thiol) through the photochemical crosslinking mechanism [7], leading to an increase in tissue surface adhesiveness. Gelatin, is a protein obtained by partial hydrolysis of collagen, the most abundant protein in the animal body [15]. In addition to biodegradability, biocompatibility, 3

lower immunogenicity and commercial availability at low cost, gelatin can be crosslinked or modified with the inclusion of other materials to significantly alter its mechanical and biochemical properties [16]. Photo-crosslinked gelatin have been developed for tissue adhesive in recent years. Elvin et al. prepared photo-crosslinked gelatin and introduced extra phenolic side chains using Bolton-Hunter modification,

IP T

leading to photo-crosslinked products with excellent adhesive properties [13].

Vuocolo et al. have developed photo-crosslinked gelatin with high elasticity and

SC R

strong adhesion for wound sealing [17].

In this study, catechol-modified gelatin (Gel-Ca) and phenol-modified gelatin

U

(Gel-Ph) were synthesized via an EDC/NHS chemistry and crosslinked using a

N

ruthenium-based photochemistry. The gelatin-based adhesives were tested in detail

A

regarding swelling, degradation, rheology, tissue adhesion, and cytocompatibility. In

M

many recent studies, mussel-inspired gelatin-based adhesives have been developed using ion-crosslinked catechol-modified gelatin, and also compared with photo-

A

CC E

PT

ED

crosslinked adhesives [16, 18-21].

4

2. Materials and methods 2.1. Synthesis of Modified Gelatin The conjugation of dopamine hydrochloride (DA, Sigma, Germany) or phloretic acid (PA, Alfa Aesar, Great Britain) to gelatin (from porcine skin, Sigma,

ethylcarbodiimide hydrochloride (EDC, Fluka, United Kingdom) and N-

IP T

USA) was carried out via chemical reaction using N-(3-Dimethylaminopropyl)-N′-

SC R

Hydroxysuccinimide (NHS, Sigma-Aldrich, USA). In the case of preparing catecholmodified gelatin, 1.0 g of gelatin was dissolved in degassed 100 mL of MES (2-(N-

Morpholino)ethanesulfonic acid hydrate, Sigma, USA) buffer (pH 4.5, 100 mM) at 37

U

°C. Then, EDC (575.1 mg) and NHS (345.3 mg) was added to the mixed solution.

N

After 20 min stirring, dopamine hydrochloride (568.9 mg) dissolved in 3 mL of MES

A

buffer (pH 3.3, 100 mM) was added to the mixture. The mixture was allowed to react

M

at 37 °C in dark with shaking 100 rpm for 24 hours. The resulting solution was purified in dialysis membrane (MWCO 3500 Da, Cellu Sep, USA) against acidified

ED

deionized water (pH ~3) four times and finally against deionized water (pH ~7) for 2

PT

hours under vigorous stirring. The final product was lyophilized and stored at -20 °C. For conjugation of phloretic acid to gelatin, 1.0 g of gelatin was added in 20 mL of

CC E

MES buffer (pH 4.5) at 50 °C to dissolve. Phloretic acid (50.0 mg), EDC (58.0 mg) and NHS (35.0 mg) was dissolved in degassed 80 mL of MES buffer (pH 4.5) at 37 °C. After 20 min stirring, solution of gelatin was added to the mixture. The mixture

A

was allowed to react at 37 °C in dark with shaking 100 rpm for 24 hours. The resulting solution was purified by dialysis (MWCO 3500 Da) against acidified deionized water (pH ~3) four times and finally against deionized water (pH ~7) for 2 hours under vigorous stirring. The final product was lyophilized and stored at -20 °C. 2.2. Characterization of Modified Gelatin 5

Proton nuclear magnetic resonance (1H NMR, AVIII-500MHz FT-NMR, Bruker, USA) analysis was used to confirm the successful conjugation of dopamine or phloretic acid. Samples dissolved in deuterium oxide (D2O, Aldrich, USA) were transferred into a 5 mm NMR tube and recorded on an NMR spectrometer. Furthermore, the modified gelatin was assessed by UV-Vis spectroscopy. Samples

IP T

dissolved in deionized water were scanned, respectively, at wavelengths from 200 nm to 500 nm using a UV-visible spectrophotometer (Cary 300, Agilent, USA).

SC R

The catechol content and phenol content were quantified by Arnow’s method

[22]. The gelatin, catechol-modified gelatin, and phloretic acid-modified gelatin were

U

dissolved in deionized water to 5 mg/mL. 200 µL of 0.5 N hydrochloric acid

N

(J.T.Baker, USA) and 200 µL of a nitrite molybdate reagent containing 100 mg/mL of

A

sodium nitrite (J.T.Baker, USA) and 100 mg/mL of sodium molybdate (Sigma-

M

Aldrich, USA) were added to 200 µL of the samples, and then the mixtures were shaken for 5 min. After the reactions, the mixtures were mixed with 200 µL of 1 N

ED

sodium hydroxide (Sigma-Aldrich, Sweden). The absorbance was measured in a 96well plate at 490 nm using a microplate reader (ELx800, BioTek, USA). The

PT

dopamine hydrochloride was used as a standard. To analyze the quantity of phenol groups, each sample was mixed with 150 mg/mL of mercury sulfate (Sigma-Aldrich,

CC E

India) in 5 N sulfuric acid (Aldrich, USA), and the tube with mixture was immersed in a boiling water for 10 minutes. The mixture was cooled to room temperature and

A

was then mixed with 2 mg/mL of sodium nitrite in deionized water, and centrifuged at 10000 rpm for 5 min. The supernatant was collected, and the absorbance was measured at 490 nm. The phloretic acid was used as a standard. 2.3. Preparation of Photo-Crosslinked Adhesives Tris(2,2’-bipyridyl)dichlororuthenium(II) hexahydrate (Ru(bpy)3Cl2, Aldrich, 6

USA) and sodium persulphate (SPS, Sigma-Aldrich, Germany) were prepared as 50 mM and 1 M stock solutions in deionized water. Gelatin (or modified gelatin) was dissolved in TBS (pH 7.4) to 100 mg/mL, and was mixed with 1 mM Ru(bpy)3Cl2 and 20 mM SPS. The mixture was irradiated for 5–20 s with an LED high power lamp (440–460 nm, ~8 W, PAR20, VITALUX, ROC) from about 5 cm.

IP T

2.4. Preparation of Fe3+-Crosslinked Adhesives

The catechol-modified gelatin was dissolved in pH 7.4 TBS at 37 °C to 200

SC R

mg/mL, and FeCl3 (Fluka, Germany) was dissolved in pH 7.4 TBS to form 20 mM

solution. To prepare an injectable Fe3+-crosslinked tissue adhesive, catechol-modified

U

gelatin solution was mixed with equal volume FeCl3 solution to form adhesive in

N

seconds.

A

2.5. Swelling and Degradation

M

For measuring swelling and degradation of the gelatin-based adhesives, each adhesive was prepared as mentioned above in 2 mL centrifuge tube from 0.2 mL of

ED

pre-adhesive, and then was lyophilized. The swelling of adhesives was determined by examining water uptake capacity. Dried hydrogels were incubated in pH 7.4 TBS at

PT

37 ºC with shaking 50 rpm. The wet weight of each sample was measured at several time points during the incubation. The swelling ratio was calculated using the

CC E

following equation: (Ws - Wd)/Wd, where Ws represents the weight of the swollen adhesive at each time point and Wd represents the weight of the dried adhesive. Four

A

samples for each group (n = 4) were used in the swelling test. To determine the degradation rate of gelatin-based adhesives, the dried hydrogels

were preincubated in pH 7.4 TBS for 3 days, and then treated with Trypsin-EDTA solution (10 ×, Sigma, USA) diluted 1/200 with pH 7.4 TBS at 37 ºC with shaking 50 rpm. At the indicated time points, adhesives were removed and weighed during the 7

enzymatic treatment. The weight remaining was calculated using the following equation: Wr/Wi × 100, where Wr represents the weight of the remaining adhesive at each time point and Wi represents the weight of the hydrogel after preincubation. Three samples for each group (n = 3) were used in the degradation study. 2.6. Rheological Analysis

IP T

Rheological analysis was carried out using a modular compact rheometer (MCR-102, Anton Paar, Canada) equipped with 20 mm diameter stainless steel

SC R

parallel plate geometry and electrical heated plate. The operation temperature was maintained at 37 °C in the following measurements. The elastic moduli of the

adhesives were measured in a frequency sweep mode of 0.1-1 Hz to recorded as

A

freshly prepared before the frequency sweep.

N

U

storage modulus (G’) and loss modulus (G”) when strain was set at 1%. Samples were

M

2.7. Tissue Adhesion Test

The tissue adhesion was determined by tensile test using a motorized test stand

ED

(FGS-50VB-H, NIDEC-SHIMPO, Japan) with a 50 N digital force gauge (FGP-5, NIDEC-SHIMPO, Japan). Due to the biological similarity to human tissue surface,

PT

fresh egg membrane was chosen as adherend. Two 2 mL of glass vials were fixed on the motorized test stand and the digital force gauge by custom-designed jigs to hold

CC E

egg membranes. The fresh egg membranes were separated from egg shells, and washed using deionized water to remove residual albumen. The membrane was placed

A

over the end (surface area = 1.1 cm2) of a 2 mL of glass vial and held in place by an oring. Each adhesive was prepared from 50 mL of solution which combined two surfaces fully. The two surfaces were pulled apart at a rate of 40 mm/min. The stress at the breakpoint is defined as adhesive strength. 2.8. MTS Assay 8

Cytotoxicity of gelatin-based adhesives was studied using L929 cells according to ISO10993 standard test. Each adhesive was prepared as mentioned above in each 2 mL of centrifuge tube from 0.2 mL of solution, and was extracted using 1 mL of serum-free medium comprised of minimum essential medium alpha medium (Gibco, USA), 2-Mercaptoehanol (Sigma-Aldrich, USA), Penicillin Streptomycin (Gibco,

°C with 50 rpm. Followed by incubation, the final extract solutions were

IP T

USA), Gentamicin solution (GE, USA). The extraction proceeded for 48 hours at 37

SC R

supplemented with 10% fetal bovine serum (BI, USA) and diluted using culture

medium to 100%, 50%, and 10% for culturing L929 cells. The culture medium was prepared by supplementing serum-free medium with 10% fetal bovine serum.

N

U

L929 cells were seeded in 96-well plate at a density of 6 × 103 cells per well, and

A

cultured in 100 µL of culture medium for 1 day. Thereafter, the culture medium was replaced with 100 µL of prepared dilution, and the plate was incubated for 24 hours.

M

20 μL of CellTiter 96® AQueous One Solution Reagent (Promega, USA) in 100 μL of

ED

culture medium replaced the solution of each well, and the plate was incubated for 3 hours. Finally, absorbance was measured at 490 nm using a microplate reader. The

PT

results were compared with cells treated using culture medium with or without 0.64% phenol (Sigma-Aldrich, USA).

CC E

2.9. Live/Dead Assay

Cell viability of gelatin-based adhesives was studied using L929 cells. L929

A

cells were seeded in 96-well plate at a density of 6 × 103 cells per well, and cultured in 100 µL of culture medium for 1 day. Thereafter, the gelatin-based adhesive was prepared freshly from 40 µL of pre-adhesive covering the cells, and then 100 µL culture medium was added in each well. The plate was incubated for 24 hours. After that, the LIVE/DEAD Cell Imaging Kit (Thermo Fisher Scientific, USA) was used to 9

determine the cell viability. The fluorescence images were taken using a microscope (IX71S1F-3, Olympus, Japan). 2.10. Statistical Analysis The data was presented as means ± standard deviation (SD). The statistical

A

CC E

PT

ED

M

A

N

U

SC R

p < 0.05 was considered statistically significant difference.

IP T

analyses between different groups were determined with Student’s t-tests. A value of

Fig. 1. Modification of gelatin via an EDC/NHS chemistry. A schematic representation of synthesizing (a) catechol-modified gelatin (Gel-Ca) and (b) phenol-modified gelatin (Gel-Ph). (c) 1H NMR spectrum of gelatin, Gel-Ca, and Gel-Ph.

10

3. Results 3.1. Modification of Gelatin The catechol-modified gelatin (Gel-Ca) and phenol-modified gelatin (Gel-Ph) were synthesized by conjugating dopamine and phloretic acid to the gelatin backbone

IP T

(Fig. 1a-b) via a carbodiimide coupling reaction using EDC and NHS. The yields of Gel-Ca and Gel-Ph were 53.75% and 73.96%, respectively. As shown in (Fig. S1a),

SC R

dopamine and Gel-Ca display the maximum absorbance at 280 nm, and gelatin

displays less absorbance value at 280 nm due to the presence of few aromatic amino acid residues, indicating that catechol groups were successfully conjugated to gelatin.

U

In addition, the absent of peaks for Gel-Ca at wavelengths longer than 300 nm

N

indicates that the conjugated catechol was not oxidized during preparation of Gel-Ca

A

[23]. The UV-visible spectral analysis of phloretic acid, gelatin, and Gel-Ph shows

M

that phloretic acid displays the maximum absorbance value at 275 nm, and Gel-Ph displays more absorbance value at 275 nm in comparison with gelatin, demonstrating

ED

the successful incorporation of phenol groups (Fig. S1b). From the 1H NMR

PT

spectrum, Gel-Ca displays the chemical shift values at δ 6.6, 6.7 and 6.8 ppm, and Gel-Ph displays the chemical shift values at δ 6.7 and 7.0 ppm, confirming the

CC E

effective conjugation of catechol groups and phenol groups to the gelatin (Fig. 1c). The content of catechol and phenol group in gelatin, Gel-Ca and Gel-Ph were

determined by Arnow’s method [22], and the results were shown in Table 1 and Table

A

S1. Gelatin contains 89.97 ± 0.96 µmol/g of phenol groups and 11.30 ± 0.40 µmol/g of catechol groups due to the tyrosine and few oxidized tyrosine in gelatin, respectively. Calculating from the catechol and phenol contents, gelatin contains about 0.2% tyrosine, similar to the result Cobbett et al. reported [24]. The catechol group content and phenol group content in gelatin are 11.30 ± 0.40 µmol/g and 89.97 11

± 0.96 µmol/g. The catechol content of Gel-Ca is 39.32 ± 0.61 µmol/g which is much higher than the catechol content of gelatin or Gel-Ph. The Gel-Ph has the highest phenol group content which is 238.41 ± 3.79 µmol/g. The phenol group content of Gel-Ca is similar to the phenol group content of gelatin. However, due to a small number of phenol groups turning to catechol groups by oxidation during the

IP T

preparation of Gel-Ph, the catechol group content of Gel-Ph is 12.81 ± 0.26 µmol/g

ED

M

A

N

U

SC R

which is significantly higher than gelatin (p < 0.05).

Fig. 2. Physical characterization of gelatin-based adhesives. (a) A photograph of the gelatin-based

PT

adhesives. (b) Swelling (n = 4) and (c) degradation (n = 3) of gelatin-based adhesives, including photocrosslinked gelatin (PG), photo-crosslinked phenol-modified gelatin (PPG), photo-crosslinked

CC E

catechol-modified gelatin (PCG), and ion-crosslinked catechol-modified gelatin (ICG).

3.2. Characterization of Gelatin-Based Adhesives

A

The gelatin-based adhesives were fabricated as photo-crosslinked gelatin (PG),

photo-crosslinked phenol-modified gelatin (PPG), photo-crosslinked catecholmodified gelatin (PCG), and ion-crosslinked catechol-modified gelatin (ICG) (Fig. 2a). All adhesives were formed rapidly seconds after illumination with blue light or mixture with Fe3+ ion. The swelling properties of the gelatin-based adhesives were investigated by 12

measuring changes in water content during 7 days of incubation in physiological conditions (i.e., in pH 7.4 TBS) at 37 °C with shaking 50 rpm) (Fig. 2b). There is light weight loss of ICG after 3 days, due to deformation of weak ion-crosslinks. No significant weight loss and structural deformations were displayed in the other adhesives. All adhesives reach an equilibrium state of swelling after 3 days of

IP T

incubation, and remain constant during the rest of the incubation period. The average swelling ratio at 3 days of incubation are 11.41 ± 0.26, 5.99 ± 0.26, 6.96 ± 0.20, and

SC R

10.66 ± 0.26 for PG, PPG, PCG, and ICG, respectively. The gelatin-based adhesives

show no significant swelling and even lower swelling index compared to their freshly

U

prepared adhesives. The high swelling ratio of tissue adhesive may build up extreme

N

pressure on the surrounding tissues, especially when used in closed cavities [25].

A

However, the slight swelling or low water content of gelatin-based adhesives would

in stopping fluid leakage [26].

M

not produce the pressure effect on the surrounding tissues, while being more efficient

ED

As to the investigation of degradation properties of gelatin-based adhesives, adhesives were exposed to enzymatic degradation by trypsin-EDTA in pH 7.4 TBS at

PT

37 °C with shaking 50 rpm, and weighed at different points in time (Fig. 2c). The complete degradation of ICG takes the least time, about 2.5 hours, and then the

CC E

complete degradation of PG takes 3 hours. Complete degradation of PCG and PPG takes about 4 and 12.5 hours respectively. In practical applications, the

A

biodegradability of gelatin-based adhesives in enzyme solution can avoid the need for secondary surgery for removal. This result also shows that the degradation time for the adhesives can be tunable by using different modifications and crosslinking mechanisms.

13

IP T SC R

Fig. 3. Rheometric analysis of the gelatin-based adhesives. (a) Rheometric analysis of the gelatin-based adhesives in a frequency sweep mode. Black symbols are storage moduli (G′), and white symbols are

U

loss moduli (G′′). (b) Average storage modulus and (c) average loss modulus of gelatin-based

N

adhesives (n = 3), including photo-crosslinked gelatin (PG), photo-crosslinked phenol-modified gelatin (PPG), photo-crosslinked catechol-modified gelatin (PCG), and ion-crosslinked catechol-modified

M

A

gelatin (ICG). ** and *** represent p < 0.01 and 0.001 in comparison with PG.

ED

The viscoelastic properties of gelatin-based adhesives were characterized by rheometric analysis. Each sample of gelatin-based adhesives was freshly prepared,

PT

and measured in a frequency sweep mode of 0.1-1 Hz. The strain was set at 1% and temperature was maintained at 37 °C, and the storage modulus (G’) and loss modulus

CC E

(G”) were recorded. As shown in (Fig. 3a), the gelatin-based adhesives are stable viscoelastic solids except ICG. The modulus of ICG appeared to increase gradually over time in the frequency sweep, and turned to a viscoelastic solid when frequency is

A

more than 0.3 Hz. This result might be due to either the unreacted catechol groups in the Gel-Ca, which may react further during the mechanical test, or to reversible bonds, such as hydrogen bonding, which result in variations in the relaxation of GelCa [23]. The average storage moduli of gelatin-based adhesives measured at 1 Hz are 1503.3 ± 23.3 Pa, 3470.0 ± 191.4 Pa, 1566.7 ± 6.7 Pa, and 36.2 ± 8.6 Pa for PG, PPG, 14

PCG, and ICG, respectively (Fig. 3b), indicating PPG was the most robust adhesive, and ICG was the weakest adhesive. PG and PCG have similar high elasticity. The average loss moduli of gelatin-based adhesives measured at 1 Hz are 4.1 ± 0.7 Pa, 14.1 ± 3.5 Pa, 6.4 ± 0.5, and 22.1 ± 3.7 Pa for PG, PPG, PCG, and ICG, respectively (Fig. 3c). ICG showed the highest viscosity in all adhesives, while the viscosity of

M

A

N

U

SC R

IP T

PPG was higher than PG and PCG significantly.

Fig. 4. Tissue adhesive strength of the gelatin-based adhesives. (a) The images of the tissue adhesive

ED

strength measurement of gelatin-based adhesives to egg membrane. (b) Average adhesive strength of gelatin-based adhesives, including photo- crosslinked gelatin (PG, n = 10), photo-crosslinked phenol-

PT

modified gelatin (PPG, n = 10), photo-crosslinked catechol-modified gelatin (PCG, n = 10), and ion-

CC E

crosslinked catechol-modified gelatin (ICG, n = 6). *** represents p < 0.001 in comparison with PG.

3.3. Tissue Adhesion As discussed previously, the tissue adhesive strength of gelatin-based adhesive

A

was measured by recording detachment stress of adhesive from the egg membranes [27] (Fig. 4a). Each type of adhesives remained on the egg membrane after fracture of adhesive, demonstrating that sufficient and stable adhesion exist between the adhesive and tissue surface. The adhesive strength of the hydrogels depended on the phenol content of gelatin 15

(Fig. S2a). The adhesive strength of the PPG hydrogel with phenol content of 89.26 ± 3.84 µmol/g was 64.65 ± 0.5 kPa, and reached to a maximum of 77.0 ± 1.5 kPa at 145.76 ± 3.60 µmol/g. The adhesive strength also depended on the hydrogel precursor concentrations (Fig. S2b). 2.5% (w/v) PG was not sufficient to form strong enough adhesion. 5% (w/v) PG had a value of 32.58 ± 4.0 kPa adhesive strength, smaller than

IP T

the value of 64.3 ± 1.2 kPa for 10% (w/v) PG. In the other hand, the high viscosity

and sol-gel transition which results in difficulties of operation. 10 % polymer solution

SC R

is chosen for all the experiments due to acceptable tissue adhesion and practical operability.

U

The trend of adhesive strength was similar to storage modulus for various adhesives.

N

PG, PPG, PCG and ICG show 64.3 ± 1.2 kPa, 77.0 ± 1.5 kPa, 61.8 ± 1.8 kPa, and 7.5

A

± 0.6 kPa adhesive strength, respectively (Fig. 4b), indicating that the photo-

M

crosslinked adhesives were more robust than the ion-crosslinked adhesives. In comparison with PG, PPG showed significantly higher adhesive strength, and PCG

ED

showed similar adhesive strength.

Our laboratory has compared the adhesion strength difference between porcine skins

PT

and egg membrane (Fig. S3a). We found that the adhesion strength of porcine skins was much less than that of using egg membranes. This might be explained by the

CC E

opaque of porcine skins, in comparison with the transparent egg membranes (Fig. S3b). That is the reason why bovine amnions [27] are usually used in tissue adhesion

A

tests. To compared with bovine amnions, egg membrane is a cheap and easily available material for tissue adhesion tests. Therefore, egg membrane was used as our tissue model for adhesive tests.

16

a

b

TCPS

PG

PCG

PCG

ICG

IP T

Fig. 5. In vitro cytotoxicity of gelatin-based adhesives. (a) The viability of L929 cells cultured in

100%, 50%, and 10% of the extract solutions from gelatin-based adhesives was determined by MTS

SC R

assay. The cell viability of cells treated with culture medium represents 100%. * and ** represent p <

0.05 and 0.01 in comparison with medium. (b) L929 cells were stained using live/dead kit for live cells (green) and dead cells (red), which were cultured on 96-well plates, and covered with gelatin-based

U

adhesives for 1 day. Scale bar: 500 μm. The gelatin-based adhesives used in these assays contained

N

photo-crosslinked gelatin (PG), photo-crosslinked phenol-modified gelatin (PPG), photo-crosslinked

A

catechol-modified gelatin (PCG), and ion-crosslinked catechol-modified gelatin (ICG).

M

3.4. In vitro Cytotoxicity

MTS assay was used to investigate the cytotoxicity of gelatin-based adhesives

ED

(Fig. 5a). In comparison with 100% viability of cell treat with culture medium, the

PT

extract of each adhesive did not show significant cytotoxicity. In addition, some extracts of PPG, PCG and ICG even improved cell growth significantly. Cytotoxicity

CC E

of gelatin-based adhesive was also determined by live/dead assay (Fig. 5b). L929 cells were cultured on 96-well plates, and covered with freshly prepared gelatin-based adhesives for 1day. Then, cells were stained using live/dead kit for live cells (green)

A

and dead cells (red). The dead cells were hardly found in each group of gelatin-based adhesives similar to the result of the cells only treated with culture medium. These results indicated the great cytocompatibility of gelatin-based adhesives.

17

4. Discussion Four types of gelatin-based adhesives, including PG, PPG, PCG, and ICG, were fabricated to overcome the limitations of conventional tissue adhesives. Gelatin has been transformed into adhesive products using enzyme (transglutaminase) or chemical crosslinkers (EDC/NHS, glutaraldehyde, resorcinol-formaldehyde, genepin),

IP T

however, the toxicity of byproducts or low adhesive strength limited their use. The

ruthenium-based photocrosslinked and Mussel-inspired ion-crosslinked gelatin-based

SC R

adhesives have been developed with low cytotoxicity. In order to increase the tissue adhesion and mechanical properties of the adhesives, catechol and phenol were

U

conjugated to gelatin as Gel-Ca and Gel-Ph, which could form crosslinks to prepare

N

hydrogels and showed a great binding affinity to diverse nucleophiles (such as

A

amines, thiol, and imidazole) on tissue surface. In this study, PG was prepared similar

M

to Gelatin-PhotoSealTM, which demonstrated high elasticity, high adhesiveness, and no inflammation response, and was successfully used in animal models [28].

ED

Recently, ICG was also formed as a self-healing and injectable adhesive, and were used along with PG as standards to compare with the other adhesives.

PT

Catechol- and phenol-modified gelatin can form adhesives with low cytotoxicity

CC E

and high tissue adhesion. Gel-Ca and Gel-Ph also showed higher solubility at room temperature. Catechol- and phenol-modified polymers have been developed as adhesives by various crosslinking, such as NaIO4-crosslinked hyaluronic acid-

A

catechol, Fe3+-crosslinked Gel-Ca [18], Pluronic-thiol-crosslinked chitosan-catechol [29], HRP/H2O2-crosslinked PVA-phenol [30], etc. In this study, Gel-Ca and Gel-Ph were synthesized successfully and contained significantly more catechol and phenol compared to gelatin. However, the total amount of catechol and phenol are ~100 µm/g, ~130 µm/g and ~250 µm/g in gelatin, Gel-Ca and Gel-Ph, respectively, which 18

determine the crosslink content after photo-crosslinked process. This result demonstrates the similar trend of rheology and tissue adhesion of photo-crosslinked adhesives, that PPG shows the highest viscoelasticity and tissue adhesion, and the viscoelasticity and tissue adhesion of PCG is similar to PG. Finally, Gel-Ca was superior to gelatin and Gel-Ph in that it contains significantly more catechol, and

IP T

could interact with Fe3+ to form thermal stable crosslink.

In this study, photo-crosslinked adhesive (PCA) obtained tunable gelation time

SC R

and mechanical properties by changing the concentration of (Ru(bpy)3Cl2 and SPS

and the conjugation of phenol and catechol. The formation of PCA was adjusted to

U

gelation within seconds and attain robust mechanical properties, which can rapidly

N

close wound and stop fluids or air. PCAs provide much higher mechanical properties

A

and tissue adhesion in comparison with ion-crosslinked adhesive (i.e. ICG).

M

Crosslinking strength seemed to be lower in ICG than PCAs, because the covalent bonds in PCAs were much stronger than the ion-catechol interaction in ICG, and the

ED

crosslinking density in ICG might be lower. The PCAs showed lower swelling ratio and longer degradation time compared to ICG also demonstrating the higher

PT

crosslinking strength in PCAs. In the case of PCAs, PPG displayed the lowest swelling index, the longest degradation time, the highest viscoelasticity, and the

CC E

strongest tissue adhesion. However, PPG demonstrated about 2.3-fold increase in storage modulus but about 1.2-fold increase in tissue adhesion compared to PG, which

A

might be caused by different frequencies and strains in two experiments. PCG showed lower swelling ratio and longer degradation time compared to PG, but viscoelasticity and tissue adhesion of PCG was similar to PG, indicating that the crosslinking density of PCG was higher than PG, which was not enough to significantly increase the viscoelasticity and tissue adhesion. 19

ICG was formed using Gel-Ca and Fe3+ via a rapid catechol-Fe3+ coordination complexation with the highest viscosity compared to the other adhesives. However, the mechanical properties of ICG were very poor, and could not even form a viscoelastic solid at low frequency (Fig. 3a), due to weak interaction between catechol and Fe3+. Choi et al. prepared a Fe3+-crosslinked catechol-conjugated gelatin by

IP T

stirring the mixture of Fe3+ and catechol-conjugated gelatin for 24 hours [18]. The

preparation of adhesive takes a long time, so that the oxidation-crosslinking can have

SC R

sufficient time to form and improve its mechanical properties. Fan et al. prepared an

adhesive hydrogel via formation of both catechol-Fe3+ coordination complexation and

U

genipin-induced covalent crosslinking [19]. Incorporation of other crosslinking mechanism was considered as a method to improve the mechanical properties of ICG.

N

In this work two types of double-crosslinked Gel-Ca were prepared using Fe3+- and

M

A

H2O2-crosslinking, and Fe3+- and photo-crosslinking, which were limited by the increase of mechanical properties and difficult operation. In the most recent study,

ED

Hong et al. used ICG to repair uterine injury of mice for minimally invasive use [18], indicating the potential application of ICG in injury with less fluid and air.

PT

For the four types of adhesives, cytotoxicity was not found by MTS assay and live/dead assay. During photo-crosslinking process, the cytotoxic oxidant would be

CC E

rapid and mostly consumed to non-cytotoxic levels, and ruthenium complex was safe at the concentration used in this study [13]. It is interesting that some extract solutions

A

of adhesives promote cell viability (Fig. 5a). Except PG, the adhesive with lower crosslinking strength showed significantly higher cell viability in lower extract solution content, indicating that the components released from modified-gelatin-based adhesives could play important roles on promoting cell viability. The potential

20

mechanism of promoting cell viability of Gel-Ca and Gel-Ph should be further examined in future studies.

5. Conclusion

IP T

In this study, four types of gelatin-based adhesives, including PG, PPG, PCG, and ICG were prepared to form cyto-compatible adhesives with tunable mechanical

SC R

properties and tissue adhesion. In summary, PCAs showed much greater elasticity and tissue adhesion than ICG, and ICG. It also displayed the highest viscosity due to the different crosslinking mechanism. Furthermore, PPG showed the greatest

U

viscoelasticity and tissue adhesion amongst PCAs due to the highest crosslinking

N

strength. In addition, gelatin-based adhesives showed great tissue adhesion, low

A

swelling index, great biodegradability, and great cyto-compatibility. Consequently,

M

gelatin-based adhesives are expected to provide new methods for wound closure and

A

CC E

PT

ED

healing.

21

References

A

CC E

PT

ED

M

A

N

U

SC R

IP T

[1] N. Annabi, K. Yue, A. Tamayol, A. Khademhosseini, Elastic sealants for surgical applications, European Journal of Pharmaceutics and Biopharmaceutics, 95 (2015) 27-39. [2] A. Lauto, D. Mawad, L.J.R. Foster, Adhesive biomaterials for tissue reconstruction, Journal of Chemical Technology and Biotechnology, 83 (2008) 464472. [3] Y. Lee, C. Xu, M. Sebastin, A. Lee, N. Holwell, C. Xu, D. Miranda Nieves, L. Mu, R.S. Langer, C. Lin, J.M. Karp, Bioinspired Nanoparticulate Medical Glues for Minimally Invasive Tissue Repair, Adv Healthc Mater, 4 (2015) 2587-2596. [4] D.C. Watts, Adhesives and sealants, Biomaterials Science, 3rd Edition, (2013) 889-904. [5] M. Ryou, C.C. Thompson, Tissue adhesives: a review, Techniques in Gastrointestinal Endoscopy, 8 (2006) 33-37. [6] P.A. Leggat, D.R. Smith, U. Kedjarune, Surgical applications of cyanoacrylate adhesives: a review of toxicity, ANZ Journal of Surgery, 77 (2007) 209-213. [7] D.A. Fancy, T. Kodadek, Chemistry for the analysis of protein–protein interactions: Rapid and efficient cross-linking triggered by long wavelength light, Proceedings of the National Academy of Sciences, 96 (1999) 6020-6024. [8] C.M. Elvin, A.G. Carr, M.G. Huson, J.M. Maxwell, R.D. Pearson, T. Vuocolo, N.E. Liyou, D.C. Wong, D.J. Merritt, N.E. Dixon, Synthesis and properties of crosslinked recombinant pro-resilin, Nature, 437 (2005) 999-1002. [9] C.M. Elvin, A.G. Brownlee, M.G. Huson, T.A. Tebb, M. Kim, R.E. Lyons, T. Vuocolo, N.E. Liyou, T.C. Hughes, J.A. Ramshaw, J.A. Werkmeister, The development of photochemically crosslinked native fibrinogen as a rapidly formed and mechanically strong surgical tissue sealant, Biomaterials, 30 (2009) 2059-2065. [10] C.M. Elvin, S.J. Danon, A.G. Brownlee, J.F. White, M. Hickey, N.E. Liyou, G.A. Edwards, J.A. Ramshaw, J.A. Werkmeister, Evaluation of photo-crosslinked fibrinogen as a rapid and strong tissue adhesive, Journal of Biomedical Materials Research Part A, 93 (2010) 687-695. [11] B.P. Chan, K.F. So, Photochemical crosslinking improves the physicochemical properties of collagen scaffolds, Journal of Biomedical Materials Research Part A, 75 (2005) 689-701. [12] S. Ibusuki, G.J. Halbesma, M.A. Randolph, R.W. Redmond, I.E. Kochevar, T.J. Gill, Photochemically cross-linked collagen gels as three-dimensional scaffolds for tissue engineering, Tissue Engineering Part B: Reviews, 13 (2007) 1995-2001. [13] C.M. Elvin, T. Vuocolo, A.G. Brownlee, L. Sando, M.G. Huson, N.E. Liyou, P.R. Stockwell, R.E. Lyons, M. Kim, G.A. Edwards, G. Johnson, G.A. McFarland, J.A. Ramshaw, J.A. Werkmeister, A highly elastic tissue sealant based on photopolymerised gelatin, Biomaterials, 31 (2010) 8323-8331. [14] L. Sando, M. Kim, M.L. Colgrave, J.A. Ramshaw, J.A. Werkmeister, C.M. Elvin, Photochemical crosslinking of soluble wool keratins produces a mechanically stable biomaterial that supports cell adhesion and proliferation, Journal of Biomedical Materials Research Part A, 95 (2010) 901-911. [15] T.R. Keenan, Gelatin, Polymer science: a comprehensive reference, 10 (2012) 237-247. [16] T.K.S. Raja, T. Thiruselvi, R. Aravindhan, A.B. Mandal, A. Gnanamani, In vitro and in vivo assessments of a 3-(3,4-dihydroxyphenyl)-2-propenoic acid bioconjugated 22

A

CC E

PT

ED

M

A

N

U

SC R

IP T

gelatin-based injectable hydrogel for biomedical applications, Journal of Materials Chemistry B, 3 (2015) 1230-1244. [17] T. Vuocolo, R. Haddad, G.A. Edwards, R.E. Lyons, N.E. Liyou, J.A. Werkmeister, J.A.M. Ramshaw, C.M. Elvin, A Highly Elastic and Adhesive Gelatin Tissue Sealant for Gastrointestinal Surgery and Colon Anastomosis, Journal of Gastrointestinal Surgery, 16 (2012) 744-752. [18] S. Hong, D. Pirovich, A. Kilcoyne, C.H. Huang, H. Lee, R. Weissleder, Supramolecular Metallo-Bioadhesive for Minimally Invasive Use, Adv Mater, 28 (2016) 8675-8680. [19] C.Y. Choi, J.S. Choi, Y.J. Jung, Y.W. Cho, Human gelatin tissue-adhesive hydrogels prepared by enzyme-mediated biosynthesis of DOPA and Fe3+ion crosslinking, J. Mater. Chem. B, 2 (2014) 201-209. [20] C. Fan, J. Fu, W. Zhu, D.A. Wang, A mussel-inspired double-crosslinked tissue adhesive intended for internal medical use, Acta Biomaterialia, 33 (2016) 51-63. [21] T.K.S. Raja, T. Thiruselvi, G. Sailakshmi, S. Ganesh, A. Gnanamani, Rejoining of cut wounds by engineered gelatin-keratin glue, Biochimica et Biophysica Acta, 1830 (2013) 4030-4039. [22] L.E. Arnow, Colorimetric determination of the components of 3,4dihydroxyphenylalanine-tyrosine mixtures, The Journal of Biological Chemistry, 118 (1937) 531–537. [23] S. Hong, K. Yang, B. Kang, C. Lee, I.T. Song, E. Byun, K.I. Park, S.W. Cho, H. Lee, Hyaluronic acid catechol: a biopolymer exhibiting a pH-dependent adhesive or cohesive property for human neural stem cell engineering, Advanced Functional Materials, 23 (2013) 1774-1780. [24] W.G. Cobbett, A.W. Kenchington, A.G. Ward, The determination of the tyrosine content of gelatins, The Biochemical journal, 84 (1962) 468-474. [25] J.C. Wheat, J.S. Wolf, Jr., Advances in bioadhesives, tissue sealants, and hemostatic agents, Urol Clin North Am, 36 (2009) 265-275, x. [26] N. Artzi, T. Shazly, C. Crespo, A.B. Ramos, H.K. Chenault, E.R. Edelman, Characterization of star adhesive sealants based on PEG/dextran hydrogels, Macromolecular Bioscience, 9 (2009) 754-765. [27] M. Lu, Y. Liu, Y.C. Huang, C.J. Huang, W.B. Tsai, Fabrication of photocrosslinkable glycol chitosan hydrogel as a tissue adhesive, Carbohydr Polym, 181 (2018) 668-674. [28] L. Sando, S. Danon, A.G. Brownlee, R.J. McCulloch, J.A. Ramshaw, C.M. Elvin, J.A. Werkmeister, Photochemically crosslinked matrices of gelatin and fibrinogen promote rapid cell proliferation, J Tissue Eng Regen Med, 5 (2011) 337346. [29] J.H. Ryu, Y. Lee, W.H. Kong, T.G. Kim, T.G. Park, H. Lee, Catecholfunctionalized chitosan/pluronic hydrogels for tissue adhesives and hemostatic materials, Biomacromolecules, 12 (2011) 2653-2659. [30] S. Sakai, M. Tsumura, M. Inoue, Y. Koga, K. Fukano, M. Taya, Polyvinyl alcohol-based hydrogel dressing gellable on-wound via a co-enzymatic reaction triggered by glucose in the wound exudate, Journal of Materials Chemistry B, 1 (2013) 5067-5075.

23

Table 1. The catechol content and phenol content in gelatin, Gel-Ca and Gel-Ph (n = 4). Sample Catechol group (µmol/g) Phenol group (µmol/g) Gelatin 11.30 ± 0.80 89.97 ± 1.92 Gel-Ca 39.32 ± 1.22 94.48 ± 8.11 Gel-Ph 12.81 ± 0.52 238.41 ± 7.57

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Table R1. The catechol content and phenol content in gelatin, Gel-Ca and Gel-Ph (n = 3). Sample Catechol group (μmol/g) Phenol group (μmol/g) Gelatin 11.39 ± 0.61 83.48 ± 3.79 PG 4.33 ± 0.59 33.85 ± 3.42 Gel-Ca 40.88 ± 1.18 91.50 ± 2.61 PCG 18.51 ± 3.08 33.97 ± 1.07 Gel-Ph 13.44 ± 0.74 267.53 ± 4.51 PPG 5.74 ± 0.11 25.16 ± 1.07

24