Effect of Transglutaminase on Reconstruction and Physicochemical

denaturation temperature than collagen from land verte- brates.6,7,10,11 Among aquatic type I collagens, shark collagen is potentially important, beca...
7 downloads 0 Views 133KB Size
Download PDF
Biomacromolecules 2001, 2, 105-110

105

Effect of Transglutaminase on Reconstruction and Physicochemical Properties of Collagen Gel from Shark Type I Collagen Yoshihiro Nomura,* Shinzi Toki, Yasuhiro Ishii, and Kunio Shirai Applied Protein Chemistry, Faculty of Agriculture, Tokyo University of Agriculture and Technology, 3-5-8, Saiwai-cho, Fuchu, Tokyo 183-8509, Japan Received August 14, 2000; Revised Manuscript Received October 9, 2000

The effects of microbial transglutaminase (MTGase) on type I collagen self-assembly and properties of reconstructed collagen fibrils from shark were investigated. Collagen self-assembly was accelerated with the addition of MTGase in dependence on that concentration. The relative amount of reconstructed collagen slightly decreased with MTGase. The diffusion coefficient of collagen gel was reduced by treatment with MTGase, and that suggested the reduction of mobility of the whole collagen network. At a high temperature, used to denature the collagen, MTGase-treated collagen gel remained as aggregates. By differential scanning calorimetry, the denaturation temperature of MTGase-treated collagen gel was about 2 °C higher than that of nontreated collagen gel. Treatment with MTGase yielded thermally stable cross-links in collagen molecules. 1. Introduction Type I collagens from land animals have been extensively utilized in food, industrial products, medical treatment, and cosmetic materials.1,2 Collagen is used in the form of natural collagen, chemically modified collagen, telopeptide-digested collagen, gelatin, and physiologically active peptides. The sources of these collagens are mainly bovine or pig skins. In recent reports, Sakaguti et al.3 showed that collagen peptide can be a food allergen. As an approach to develop new collagen materials that have a modified allergenic property, it is considered that aquatic animal collagens could be a bright prospect for collagen sources and are present abundantly. For type I collagen from aquatic animals, however, there are a limited number of studies, mainly in comparative biochemistry.4-7 A few papers have suggested the potential of fish gelatins for food and photographic uses.8,9 Type I collagen from aquatic animals may provide an alternative collagen source, since it has the unique features of containing fewer imino acid residues and having a lower denaturation temperature than collagen from land vertebrates.6,7,10,11 Among aquatic type I collagens, shark collagen is potentially important, because a great number of sharks are collected in connection with tuna fishery and the meat and a part of the skin serve as food in Japan. For purpose of developing a new collagen material, we have started to investigate the structural and physicochemical properties of aquatic animal collagen and gelatin.11-15 In this connection, our previous paper reported that shark type I collagen has basically the same ability of self-assembly to form fibril as that of pig type I collagen.12 However, there must be some differences in the detailed molecular structure and the physicochemical property between the both collagens.13 In * To whom correspondence should be addressed. E-mail: [email protected]. Telephone: +81-42-367-5790. Fax: +81-42-3643327.

particular, shark collagen has the disadvantage of lower thermal stability in practical applications. Therefore, the authors tried to overcome the disadvantage by cross-linking of shark collagen with microbial transglutaminase from StreptoVerticillium species (MTGase). MTGase is a Ca2+independent enzyme that catalyzes an acyl transfer reaction between a γ-carboxyamide group of glutamine residue in protein and a primary amine. It is therefore in close relation to transglutaminase (TGase) which is widely distributed in mammals, birds, marine life, crustacean, and plants.16-18 Moreover, intermolecular -(γ-glutamyl)lysine cross-links are recognized to occur naturally in several food proteins.19,20 The effects of MTGase as protein cross-linker are reported in soy protein isolate, sodium caseinate, gelatin, egg yolk, and egg white.21 Therefore, we tried to introduce the MTGase cross-link in shark collagen and examined the thermal stability of MTGase-treated collagen. 2. Materials and Methods 2.1. Materials. Type I collagen preparations were obtained from fresh skin corium of great blue shark (Prionace glauca) as described in our previous papers.11 MTGase was obtained from Ajinomoto Co. Inc.22 2.2. Collagen Self-Assembly in the Presence of MTGase. The collagen self-assembly experiment was done as described previously.23 Shark collagen was dissolved in 67 mM phosphate buffer at pH 6.0 and mixed with MTGase at a ratio of 0:400 U/g collagen, and then warmed at 25 °C. This temperature was adopted as the upper limit to avoid the gelatinization of shark collagen during the self-assembly experiment. The collagen self-assembly process was monitored by the absorbance at 310 nm using UV-spectrophotometer (UV-2000, Shimadzu, Tokyo). After 24 h, the amount of reconstructed collagen fibril was estimated as described previously.23

10.1021/bm000085w CCC: $20.00 © 2001 American Chemical Society Published on Web 11/30/2000

106

Biomacromolecules, Vol. 2, No. 1, 2001

2.3. Dynamic Light Scattering. The network structure of collagen gel was evaluated by dynamic light scattering (DLS-7000, Ootska Electronic, Tokyo). A 0.05% collagen solution with added MTGase at a ratio of 0 to 400 U/g collagen in 67 mM phosphate buffer at pH 6.0 was poured into glass cell of 12 mm o.d. for DLS use and warmed at 25 °C. The analytical procedures were described in our previous paper.13 Two correlator modes were used: time interval (TI) and time domain (TD) (TI, sampling time 4 µs, correlator channel 512; TD, sampling time 800 µs, correlator channel 512). 2.4. Electron Microscopy. Reconstructed collagen fibril was observed under a high-resolution scanning electron microscope (SEM; JSM-6000F, JEOL, Tokyo), as described previously.13 The specimen for SEM was fixed with 2% glutaraldehyde and dehydrated with alcohol, dried at the critical point of carbon dioxide, and finally coated with gold by ion spatter (JVC-5000, JEOL, Tokyo). The SEM apparatus was operated at 3.0 kV and a magnification of 30 000. 2.5. Mechanical Strength of MTGase-Treated Collagen Gel. The mechanical strength of MTGase-treated collagen gel (at 0.05-0.2% collagen concentration) was measured by a creep meter (RE-33005, Yamaden, Tokyo), as described previously.13 MTGase at 0 to 400 U/g collagen was added to collagen solution and warmed at 25 °C for 24 h on tissue culture plate (48 wells, Becton Dickinson and Co., NJ) to set a gel. The breaking strength of the formed gel was measured in a room at 60% humidity and 25 °C using a cylindrical probe of 5 mm o.d. moving against the gel at a speed of 5 mm/min. The penetration of the probe was stopped at a point halfway through the whole thickness of the gel specimen. The first peak top or plateau point of stress-strain curve was defined as the breaking point of a gel specimen. 2.6. Thermal Stability of MTGase-Treated Collagen Gel. Collagen gel formed at 25 °C as described above was heated quickly to 40 °C, and the dissolution of gel was monitored by the absorbance at 310 nm using a UV spectrophotometer (UV-2000, Shimadzu, Tokyo). 2.7. Measurement of Dynamic Viscoelasticity by Rheometer. A dynamic viscoelasticity measurement was done on a rheometer (ARES viscoelastic measurement system, Rheometric Scientific) as described previously.15 A portion of the collagen solution (0.5 mL) was placed between the parallel plates with 1 mm space. Time-course measurement of dynamic viscoelasticity was started after transferring the sample to 25 °C at a constant frequency (1 Hz) and a constant shear strain (5%). The storage modulus (G′) was determined. The denaturation process (transition from gel to sol) of the gel by heating was also studied by a dynamic viscoelasticity measurement. The collagen concentration was adjusted to 2.0 mg/mL at pH 6.0, and MTGase at 400 U/g collagen was added. Rheological properties were measured by a dynamic temperature ramp mode at a range of 25-50 °C at a constant rate of 1.0 °C/min. 2.8. Differential Scanning Calorimetry of MTGaseTreated Collagen Gel. The denaturation temperature of MTGase-treated collagen gel was measured by differential scanning calorimetry (DSC). DSC was carried out on a DSC apparatus (Seiko, SSC5000, Tokyo) coupled with a thermal

Nomura et al.

Figure 1. Type I collagen self-assembly in the presence of MTGase. Table 1. Effect of MTGase on the Reconstructed Content of Shark Collagen Gel (%) MTGase concn (U/g) concn (%)

control

100

200

400

0.05 0.1 0.2

94.3 98.3 96.8

95.7 98.6 94.2

94.0 95.8 89.2

91.4 94.5 83.5

data analysis system, DSC 100, as described previously.23 MTGase-treated collagen gel was compressed to a pellet by centrifugation at 3000 rpm (800 g) for 30 min and washed three times with 67 mM phosphate buffer at pH 6.0. A portion of the pellet (about 13 mg in wet weight) was placed in an Ag cell (75 µL) and subjected to DSC. 3. Results 3.1. Process of Collagen Self-Assembly in the Presence of MTGase. In vitro, type I collagen from shark is able to self-assemble spontaneously to form fibril under physiological conditions.12 MTGase was added to the collagen solution, and then the mixture was set to form the gel. Collagen selfassembly was monitored by the absorbance change at 310 nm (Figure 1). The incubation temperature of shark collagen was set at 25 °C so as not to denature it. With an increase in the addition of MTGase, the rate of collagen self-assembly became faster. This lag phase and growth phase of collagen self-assembly was shortened with increased amount of MTGase added. The yield of reconstructed collagen fibril was only slightly decreased with the addition of MTGase (Table 1). The process of collagen self-assembly was also monitored by dynamic viscoelasticity measurement (elastic modulus G′) as shown in Figure 2. The change in G′ (B) in the time course indicated a pattern similar to that in optical density change. It is notable that the phase of G′ change coincides with that of the absorbance change (A). That is, G′ of MTGase added to collagen took off the baseline significantly faster than that of the control. This implies that the network formation in the reconstructed MTGase-treated collagen as indicated by G′ change proceeds simultaneously with the nucleation and propagation in collagen self-assembly as indicated by the absorbance change.

Transglutaminase-Treated Shark Type I Collagen

Figure 2. Absorbance and dynamic rheological parameter changes of type I collagen during self-assembly in the presence of MTGase: [A] optical density curves; [B] elastic modulus (G′) curves.

3.2. SEM Observation of MTGase-Treated Collagen Gel. SEM observation of MTGase-treated collagen gel demonstrated a well-developed fibril network with the diameter at about 40-140 nm (Figure 3). The average diameter of the control fibril was about 87.3 nm, while that of MTGase-treated collagen was slightly thinner, about 74.2 nm. The unusual fibrils in MTGase-treated collagen fibrils were not observed. 3.3. Collagen Network Structure of MTGase-Treated Collagen Gel by DLS. According to our previous paper,13

Biomacromolecules, Vol. 2, No. 1, 2001 107

collagen network treated with MTGase was characterized by DLS (Figure 4). In this report, the fluctuation in the light scattering from collagen network due to Brownian motion was translated to a diffusion coefficient, D, that presents information about particle size of entangled fibrils. For this purpose, two modes of correlator are used time interval (TI) and time domain (TD). TI is effective for analysis of an internal structure corresponding to smaller size particles, whereas TD is effective for analysis of an internal structure corresponding to larger size particles. D of collagen gel for the both modes of TI and TD was reduced with the concentration of MTGase. This reduction in D reflects the enlargement of particle size, namely an increase the mesh of collagen fibrils. 3.4. Mechanical Strength of MTGase-Treated Collagen Gel. The stress-strain curve of MTGase-treated collagen gel was shown in Figure 5, part I. With increasing MTGase concentration, the mechanical strength of collagen gels fell to about 25% in 0.2% collagen concentration (Figure 5, part II). In 0.1 or 0.05% collagen concentration, the mechanical strength of collagen gels was constant, independent of the amount of MTGase added. 3.5. Thermal Stability of MTGase-Treated Collagen Gel. The thermal stability of MTGase-treated collagen gel was examined by changes in the optical density (Figure 6), elasticity modulus G′ (Figure 7), and endotherm (Figure 8). When collagen gel formed at 25 °C was heated to 40 °C, control gel was quickly dissolved into solution within 12 min of heating, whereas MTGase-treated collagen gel dissolved more slowly (Figure 6). A part of the latter gel was resistant to the dissolution with heating at 40 °C. This suggests that cross-linking collagen was caused by MTGase.

Figure 3. Scanning electron micrographs of MTGase-treated collagen fibrils. Magnifications: 30 000×. Bars represent 1 µm.

108

Biomacromolecules, Vol. 2, No. 1, 2001

Figure 4. Diffusion coefficient of MTGase-treated type I collagen gel. Bar represents standard deviation of triplicate measurements.

Nomura et al.

Figure 7. Dynamic rheological parameter of MTGase-treated type I collagen gels. G′: elastic modulus curves.

Figure 8. Differential scanning calorimetry curves of MTGase-treated type I collagen gels.

Figure 5. Mechanical strength of MTGase-treated type I collagen gel.

linkage of collagen peptides resisted breaking down of the collagen network by heating. DSC curves of MTGase-treated collagen gel showed an endothermal peak (Figure 8). Difference in peak temperature between MTGase cross-linked and control gels were only approximately 2 °C, indicating that the cross-linking of shark collagen makes no significant contribution to the stability of helix conformation in collagen molecule under the condition adopted in the present experiment. 4. Discussion

Figure 6. Absorbance change of MTGase-treated type I collagen gels incubating at 40 °C.

When the temperature was above 40 °C, the elastic modulus (G′) of the control collagen gel fell quickly, indicating gel melting (Figure 7). In contrast to this, G′ of MTGase-treated collagen gel did not show such a quick fall up to 50 °C. These results showed that thermostability was given to collagen fibrils by MTGase treatment. The cross-

The utilization of type I collagen is extended to a wide field, covering food materials, medical supplies, and biomaterials. The main sources of type I collagen have been acid- or pepsin-solubilized materials from bovine or pig skin. As an approach to develop new collagen, we paid attention to a collagen from aquatic species. Our previous studies of shark collagen showed that its framework of molecular structure and ability of fibrilogenesis are substantially equal to those of the land mammalian collagen, irrespective of the peculiarity in the primary structure of shark collagen such as a reduced content of imino acid.11-13,24 This implies that shark collagen can provide a substitute for the land mammal collagen. However, the lower denaturation temperature of

Transglutaminase-Treated Shark Type I Collagen

shark collagen can be a disadvantage in general applications. In the present study, the authors used transglutaminase, which is widely distributed in animal and plant tissues. Hence MTGase can be a biofriendly modifier for shark collagen. In the presence of MTGase, shark collagen self-assembly unambiguously was rapid (Figures 1 and 2). The present experiment adopted a reaction system where the enzymatic reaction of MTGase and collagen self-assembly could proceed simultaneously. Therefore, the accelerated collagen fibril reconstruction is considered to be resulting from the introduction of cross-links in collagen molecule with MTGase. The shortening of the lag phase for MTGase-treated collagen means that only a limited number of cross-linkages introduced in the collagen can accelerate the nucleation for fibril reconstruction. Electron microscopic observation of MTGase-treated collagen showed no difference in morphological feature from the control (Figure 3). MTGase has no influence on the morphology of collagen fibril but changes the lag and growth phase of collagen fibril reconstruction. In the experiment of fibril reconstruction, the curves of optical density change and G′ change resembled each other (Figure 2). This implies that the growths of individual fibrils and their entanglement to form a network are synchronized. For MTGase-treated collagen, G′ curve showed no lag phase. This means that the MTGase cross-links in collagen molecule accelerates the formation of collagen oligomer in early stage of network formation of collagen fibrils. However the interfibrilar entanglement in gel structure is less matured for MTGase-treated collagen since the gel indicated the reduction in D estimated by DLS and breaking strength determined on creep meter (Figures 4 and 5, respectively) upon addition of MTGase to collagen. This suggested that an acceleration of the nucleation process in collagen self-assembly by MTGase affected competitively the fibril network formation and deceleration. After the growth of fibril and the formation of network is completed, the additional introduction of MTGase cross-links in collagen network must make no contribution to the increase in mechanical strength of the network because the site of MTGase cross-linking must be extremely restricted. If the distribution of MTGase crosslinking site is even in and among collagen molecules, the concentration of β chain and highly polymerized chains must be demonstrated by SDS-PAGE. MTGase-treated collagen showed the same SDS-PAGE pattern as the control (data not shown). This means that the MTGase cross-links are localized and limited in sites of collagen molecules. After the collagen fibril reconstruction is completed, only a limited number of functional groups are available for cross-linking due to the restriction by the special arrangement. The time cause of melting was monitored at 40 °C by the change in optical turbidity for MTGase-treated collagen gel (Figure 6). The turbidity of the control gel fell quickly to zero in about 10 min after heating. This rapid fall of optical density reflects the transformation of shark collagen fibril to gelatin, since the denaturation temperature of shark collagen gel was demonstrated to be lower than 40 °C (Figure 8). However, the fall of the optical density of MTGase-treated collagen gel was slow and did not reach zero. Taking into account the denaturation temperature of MTGase-treated collagen gel,

Biomacromolecules, Vol. 2, No. 1, 2001 109

the residue by incubation at 40 °C was not native collagen gel, but MTGase-treated gelatin gel. Dynamic viscoelasticity modulus G′ of the control gel fell quickly at about 40 °C and reached the zero at 45 °C (Figure 7). In contrast, MTGase-treated collagen gel retained a high level of G′ up to 50 °C at least. As MTGase-treated collagen gel has been perfectly denatured at 50 °C as shown by DSC, the G′ value in a region above 50 °C must reflect the rheological property of gelatinized MTGase-treated collagen. In other words, MTGase-treated gelatin can form a gel having a thermally stable elasticity resistant to heating at 50 °C. The addition of MTGase to collagen solution accelerated the collagen fibril reconstruction, and that to collagen gel raised the denaturation temperature. By dynamic viscoelastic measurement with heating, it is considered that MTGasetreated gelatin was formed having thermal stability. The utilization of MTGase is possible to extend the utility value of collagen and gelatin. Acknowledgment. We are particularly grateful to Ajinomoto Co. Inc. for providing MTGase. This work was supported by a Grant-in-Aid for Scientific Research (No. 11760183) from the Ministry of Education, Science, Sports, and Culture of Japan. References and Notes (1) Reich, G. Die nutzung von kollagen ausserhalb der lederindustrie, I. Allegemeines und einsatzgebiete mit hoher wirtschaftlicher bedeutung. Leder 1995, 46, 2-11. (2) Reich, G. Die nutzung von kollagen ausserhalb der lederindustrie, II. Spezielle einsatzgebiete, kollagenhochveredelung. Leder 1995, 46, 18-27. (3) Sakaguti, M.; Hori, H.; Hattri, S.; Irie, S.; Imai, A.; Yanagida, M.; Miyazawa, H.; Toda, M.; Inouye, S. IgE reactivity to R1 and R2 chains of bovine type I collagen in children with bovine gelatin allergy. J. Allergy Clin. Immunol. 1998, 104, 695-699. (4) Kimura, S.; Takema, Y.; Kubota, M. Octopus skin collagen. J. Biol. Chem. 1981, 256, 13230-13234. (5) Kimura, S.; Kamimura, T.; Takema, Y.; Kubota, M. Lower vertebrate collagen evidence for type I-like collagen in the skin of lamprey and shark. Biochim. Biophys. Acta 1981, 669, 251-257. (6) Kimura, S.; Ohno, Y. Fish type I collagen: Tissue-specific existence of two molecular forms, (R 1) 2 R 2 and R 1 R 2 R 3, in alaska pollack. Comp. Biochem. Physiol. 1987, 88B, 409-413. (7) Kelly, J.; Tanaka, S.; Hardt, T.; Eikenberry, E. F.; Brodsky, B. Fibrilforming collagens in lamprey. J. Biol. Chem. 1988, 263, 980-987. (8) Berg, R. A.; Silver, F. H.; Watt, W. R.; Norland, R. E. Fish gelatin in coating applications. Image Technol. 1985, 106-109. (9) Leuenberger, B. H. Investigation of viscosity and gelation properties of different mammalian and fish gelatins. Food Hydrocolloids 1991, 5, 353-361. (10) Piez, K. A. Molecular and aggregate structures of the collagens. In Extracellular Matrix Biochemistry; Piez, K. A., Reddi, A. H., Eds.; Elsevier: New York, 1988; pp 1-39. (11) Nomura, Y.; Yamano, M.; Shirai, K. Renaturation of R 1 of collagen type I from shark skin. J. Food Sci. 1995, 60, 1233-1236. (12) Nomura, Y.; Yamano, M.; Hayakawa, C.; Ishii, Y.; Shirai, K. Structural property and in vitro self-assembly of shark type I collagen. Biosci. Biotech. Biochem. 1997, 61, 1919-1923. (13) Nomura, Y.; Ishii, Y.; Shirai, K. The physicochemical property of shark type I collagen gel and membrane. J. Agric. Food Chem. 2000, 48, 2028-2032. (14) Yoshimura, K.; Terashima, M.; Hozan, D.; Shirai, K. Preparation and dynamic viscoelasticity characterization of alkali-solubilized collagen from shark skin. J. Agric. Food Chem. 2000, 48, 685-690. (15) Yoshimura, K.; Terashima, M.; Hozan, D.; Ebato, T.; Nomura, Y.; Ishii, Y.; Shirai, K. Physical property of shark gelatin compared with pig gelatin. J. Agric. Food Chem. 2000, 48, 2023-2027.

110

Biomacromolecules, Vol. 2, No. 1, 2001

(16) Folk, J. E. Transglutaminase. Annu. ReV. Biochem. 1980, 49, 517531. (17) Falcon, P.; Serafini-Fracassini, D.; Del Duca, S. Compaeative studies of transglutaminase activity and substrates in different organs of Helianthus tuberosus. J. Plant Physiol. 1993, 142, 263273. (18) Yasueda, H.; Kumazawa, Y.; Motoki, M. Purification and characterization of tissue-type transglutaminase from red sea bream (Pagrus mejor). Biosci. Biotech. Biochem. 1994, 58, 2041-2045. (19) Kumazawa, Y.; Seguro, K.; Takamura, M.; Motoki, M. Formation of -(γ-Glutamyl) lysine cross-link in cured horse mackerel meat induced by drying. J. Food Sci. 1993, 58, 1062-1064. (20) Sakamoto, H.; Kumazawa, Y.; Kawajiri, H.; Motoki, M. -(γGlutamyl) lysine cross-link distribution in foods as determined by improved method. J. Food Sci. 1995, 60, 416-419.

Nomura et al. (21) Sakamoto, H.; Kumazawa, Y.; Motoki, M. Strength of protein gels prepared with microbial transglutaminase as related to reaction conditions. J. Food Sci. 1994, 59, 866-871. (22) Ando, H.; Adachi, M.; Umeda, K.; Matuura, A.; Nonaka, M.; Uchio, R.; Tanaka, H.; Motoki, M. Purification and characteristics of a novel transglutaminase derived from microorganisms. Agric. Biol. Chem. 1989, 53, 2613-2617. (23) Nomura, Y., Takahashi, K.; Shirai, K.; Wada, K. Features of collagen matrix reconstructed with proteodermatan sulfate from pigskin insoluble collagen. Agric. Biol. Chem. 1989, 53, 1614-1620. (24) Nomura, Y.; Sasaki, Y.; Arai, K.; Ishii, Y.; Shirai, K. Separation of anti-shark type I collagen antibody from anti-pig type I collagen antiserum and its partial characterization. Biosci. Biotech. Biochem. 1996, 60, 1697-698.

BM000085W