Influence of Enzymatic Specificity on the Behavior of Ephemeral Gels

Nov 30, 2007 - Julien Picard, Sébastien Giraudier, and Véronique Larreta-Garde* ... fax: +33 134 256 552; e-mail: [email protected]...
0 downloads 0 Views 1MB Size
Biomacromolecules 2008, 9, 13–20

13

Articles Influence of Enzymatic Specificity on the Behavior of Ephemeral Gels Julien Picard, Sébastien Giraudier, and Véronique Larreta-Garde* Laboratoire ERRMECe UFR Sciences et Techniques Université de Cergy-Pontoise, 2 avenue Adolphe Chauvin BP222 95302, Pontoise cedex, France Received May 30, 2007; Revised Manuscript Received October 4, 2007

Ephemeral gels, called Enzgels, successively undergo sol–gel and then gel-sol transition under the action of two antagonistic enzymes, transglutaminase and protease. Molecular and macroscopic properties of Enzgels are directly dependent on the enzymatic activities and their ratios. This work studies the characteristics of Enzgels according to the specificity of three different proteases: thermolysin, trypsin, and collagenase. The experiments are conducted using three types of gelatin networks, one created only by triple helices, one only by covalent bonds, and the last network by both triple helices and covalent bonds. Rheology and polarimetry measurements show that the evolution of Enzgels is directly dependent on the specificity of the protease used. Moreover, gelatin network conformation has different influences according to this proteolytic specificity. Collagenase is not very sensitive to gelatin conformation, whereas trypsin is very limited by the presence of covalent bonds. This study considerably expands the knowledge of Enzgel properties.

Introduction The gelation of biopolymers depends on their chemical nature and structural properties, as well as medium parameters such as temperature, pressure, or pH.1–3 The study of the related phase transitions is of interest not only in order to master their industrial applications (food science, health, etc.) but also to expand the understanding of complex biological processes such as tumor invasiveness or wound healing.4–6 In a previous work, Giraudier and Larreta-Garde described the production of an ephemeral gel called Enzgel.7 These gels consist of a gelling protein and two antagonistic enzymes, one generating and the other cleaving covalent bonds. Alternate sol–gel and gel-sol transitions may occur once within such a system, generating transient gel phases. The various gels obtained are programmed to dissolve after a determined time without any change in temperature or medium composition. The two antagonistic enzymes are directly included in the liquid sample which contains gelatin as the protein. The enzymes used are a transglutaminase, which creates isopeptidic bonds between the lateral chains of lysine and glutamine residues of gelatin8 to form a covalent network, and a protease, which can hydrolyze peptide bonds. Thus, transglutaminase is involved in gel network formation and the protease (thermolysin) in the gel network destruction. The final properties of the gel, as well as the double phase transition kinetics, directly depend on the concentrations of and the ratio between the two enzymes. The originality of the system resides in the fact that these antagonistic enzymes are not only able to modify the proteins on a molecular scale but can act as phase transition catalysts. The system dynamics may contribute to our understanding of biological phenomena such as matrix remodelling involved in fibrosis or cancer dissemination.9,10 * Corresponding author tel.: +33 (0)1 34 25 66 05; fax: +33 134 256 552; e-mail: [email protected].

By using gelatin as a gelling protein, different types of gels can be obtained, such as physical, chemical, and physical-chemical gels, each one presenting its own molecular characteristics and macroscopic properties. The physical gel forms only from collagen-like triple helices;11 when a gelatin solution is cooled, protein coils locally assemble into triple helices and a network is formed, giving rise to a gel. This temperature-dependent gel is not in an equilibrium state and continuously evolves with time. The chemical gel forms only from covalent isopeptidic bonds, catalyzed by transglutaminase, and is irreversible with temperature. Lastly, the physical-chemical gel is comprised of the two kinds of bonds, which give it higher viscoelastic properties than the two other gels.12 These gels present different supramolecular organizations leading to a variety of ultrastructures.12 Various proteases are available which show different mechanisms and specificities for cleavage sites. Thermolysin, a bacterial metallo-protease, recognizes the amino part of peptide bonds that involve hydrophobic large lateral chain residues (Ile and Phe).13 In a previous study, it was shown that, as thermolysin cleavage sites are rare in helices, and the protease should preferentially act on the random coil part of the gel chains, the hydrolysis of the physical gel hence occurred faster than that of the covalently linked gels, whether they contained triple helices or not.12 Trypsin is a serine-protease that cleaves arginyl and lysyl bonds.14 Lysines are also a substrate for transglutaminase, but in that case, the lateral chain is implied. The use of this trypsin will give additional information about the influence of the lateral chain rigidity upon the recognition and hydrolysis of the protein main chain by a protease. Finally, collagenase is specific for the triplet Gly-X-Y and cleaves the peptidic chain before the glycyl bond.15,16 This sequence is characteristic of the part of collagen chains giving rise to triple helices and is also found in gelatin and where transglutaminase-catalyzed covalent bonds are rare.

10.1021/bm700601n CCC: $40.75  2008 American Chemical Society Published on Web 11/30/2007

14

Biomacromolecules, Vol. 9, No. 1, 2008

Picard et al.

Scheme 1

Therefore, the three tested proteases are characterized by the part of gelatin they hydrolyze: thermolysin and trypsin are mainly efficient in the random coils but do not cleave the triple helix,17 while collagenase can hydrolyze both parts of the protein. These data are summarized in Scheme 1. In this study, we focus on the importance of protease specificity and the influence of the network organization on the Enzgels’ properties and behavior.

Materials and Methods Materials. Gelatin and Transglutaminase. Gelatin was provided by SIGMA (G2500). It is extracted from pigskin by an acid process (pI ) 8.8) and presents a bloom of 300. The concentration of each sample was 7% (w/v) in a 50 mM Tris HCl buffer, at pH 7.4. Transglutaminase from microbial origin was produced by Ajinomoto, Japan, under the tradename of Activa, with a nominal activity of 100 U/g of powder, determined according to de Macédo et al.18 Proteases. Thermolysin (protease type X from Bacillus Thermoproteolyticus rokko, from Sigma P-1512) is a zinc metalloprotease. Its activity was determined on N-(3-[2-furyl]acryloyl)-Gly-Leu-amide from Sigma (P1512) at both 27 and 40 °C. The Vi40°C/Vi27°C ratio was 1.66. Trypsin (from bovine pancreas, Sigma T-1426) is a serine protease. Its activity was determined on N-R-p-tosyl-L-arginine methyl ester hydrochloride (SIGMA, T-4626) at both 27 and 40 °C. The Vi40°C/ Vi27°C ratio was 1.54. Collagenase is a type IA zinc metalloprotease isolated from Clostridium histolyticum (Sigma, C-9891). Its activity was determined on N-[3-(2-furyl)acryloyl]-L-leucyl-glycyl-L-prolyl-L-alanine (Fluka, 48173) at both 27 and 40 °C. The Vi40°C/Vi27°C ratio was 1.74. All enzymes were stored at -20°, and enzyme solutions were made fresh before each reaction in a 50 mM Tris-HCl buffer, at pH 7.4. Methods. Protocols for Gel Preparation. For all preparations, gelatin powder was swelled for 15 min in a buffer at 4 °C at 11.66% (w/v). The solution was incubated at 40 °C, for 15 min, for gelatin resolubilization. A solution of buffer containing or not containing the enzyme was added, and then the solution was diluted to obtain a 7% gelatin solution. If an enzyme solution was used, the stock solution was 5 times more concentrated than the final solution. Physical Gel. For this gel, the sample was cooled from 40 to 27 °C with a temperature ramp of 0.5 °C min-1. Gel time was measured from the beginning of the cooling ramp. Raising the temperature back leads to gel-sol transition.

Chemical Gel. The gelation was performed at 40 °C with 1.5 U mL-1 transglutaminase. Under these conditions, no triple helix was formed. This gel is irreversible with temperature. Physical-Chemical Gel. A total of 1.5 U mL-1 transglutaminase was added to a 7% gelatin solution at 40 °C. The sample was immediately cooled with the same temperature ramp as the physical gel. Enzgels. In this case, a protease was added directly in gelatin solution. Various temperature protocols were applied according to the gel type. Rheology. Rheology measurements were performed with a Rheostress 150 from Thermoelectron, operating in the oscillatory mode, with a strain of deformation of 0.05 and a frequency of 1 Hz. A cone/plate geometry with a cone of 60 mm/2° was used. Storage modulus G′ and loss modulus G′′ were recorded as a function of time. Temperature was controlled by a Cryostat F6 from Thermoelectron. Temperature ramps of 0.5 °C min-1 were applied. Gel time was estimated when G′ ) G′′ for every gel. For the physical Enzgel resolubilization time, the same determination was used. But, for chemical and physical-chemical Enzgels, it was determined as the time where G′ < 1. Optical Rotation. Measurements were performed on a P-1010 polarimeter from Jasco. Temperature control was performed by a Julabo F25 bath. Cooling and heating ramps of 0.5 °C min-1 were applied. The helix amount χ is derived from specific optical rotation as

χ)

[R]helix - [R]coil λ λ [R]collagen - [R]coil λ λ

(1)

where λ is the wavelength (λ ) 435nm), [R]λ ) R/lc is the specific optical rotation of the protein in solution, c is the concentration (grams per cubic centimeter), l is the optical path length (0.1 decimeter), R is the optical rotation angle (degrees) measured experimentally, [R]collagen λ is the specific optical rotation of native soluble collagen (χ ) 1), which contains only triple helices, and [R]coil is the specific optical rotation λ of the coils (χ ) 0). [R]collagen ) -800 deg cm3 g-1 dm-1 at 27 °C λ and [R]coil ) -274 deg cm3 g-1 dm-1 at pH ) 7.4 and 40 °C. λ The additivity of the signals due to both transglutaminase and gelatin was checked. As the signal from the enzyme is constant, it was subtracted from the apparent angle. Protease-Catalyzed Surface Degradation of the Gels. The different gels (100 µL) were directly formed in 1.6 mL cuvettes. After 2.5 h of gel formation time, transglutaminase (when used) was inactivated by heat shock (10 min at 80 °C). The gels were covered in the cuvette with 900 µL of Tris-HCl buffer, at pH 7.4, at 40 °C for chemical gel and at 27 °C for the others. Under these conditions, the gels do not spontaneously solubilize. After 45 min of storage at the desired

Influence of Enzymatic Specificity

Biomacromolecules, Vol. 9, No. 1, 2008 15

Figure 1. Hydrolysis kinetics of physical (P, black circle), chemical (C, grey square), and physical-chemical (P-C, open triangle) gels by trypsin (A) and collagenase (B).

temperature, 100 µL of protease (at the same activity for collagenase and trypsin) was added to the upper liquid phase. Regular mixing was performed. As the gel hydrolysis proceeded, solubilized products diffused from the bottom gel to the upper liquid phase. The solubilization kinetics were monitored at 280 nm with a Uvikon spectrophotometer by measuring the appearance of hydrolysis products in the liquid phase. Gelatin contains only a few aromatic residues, but the signal is high enough to be significant.

Results Protease Capability To Hydrolyze Gels. Enzymatic degradation of the gels by thermolysin has previously been described;12 the same procedure was used here with the three proteases to characterize the different enzymes through their capability to hydrolyze the gels from the outside. Peptides released from the gel were observed in the solution phase through OD280nm. After the completion of gel hydrolysis, the OD280nm became constant. To abolish the difference due to temperature on protease activity, the theoretical Vi at 27 °C of the chemical gel hydrolysis (performed at 40 °C) was calculated using the Vi40°C/ Vi27°C ratio specific to each enzyme. Results similar to those obtained with thermolysin are observed with trypsin (see Figure 1). Hydrolysis of the physical gel occurred faster than that of the transglutaminase covalently linked gels; the difference is even higher with the serine protease. The presence of covalent bonds slows down hydrolysis, while triple-helix organization of the protein does not modify the enzyme–substrate recognition. Collagenase does not show the same behavior as the two other proteases, as it hydrolyzes the three types of gels at the same rate. This enzyme is able to recognize its cleavage sites whether they are included in a triple helix structure or not. Here, contrary to what is usually observed,19 covalent bonds do not protect the gel against collagenase hydrolysis. Moreover, with this protease, large clusters are released from the gel which can explain the lower absorbance of the liquid phase. The interesting point is that collagenase acts on a peptidic sequence involved in the helices while trypsin and thermolysin react on the coil parts of gelatin where covalent bonds are mainly located. This first set of results shows that enzyme specificity directly controls the way the network is hydrolyzed. Influence of Protease Specificity on Enzgels. Enzgels were recently described as ephemeral gels, presenting successive and preprogrammed sol–gel and then gel-sol transitions due to the

concomitant activity of two antagonistic enzymes.7 The system contains a transglutaminase, which binds soluble protein molecules into a protein network, leading to an increase in viscoelasticity, and a protease, which solubilizes both the bound and soluble proteins and leads to a decrease in viscoelasticity. Here, Enzgels were generated with three different proteases on three different types of gels (physical, chemical, and physicalchemical). The viscoelastic properties and the helix content were tracked by rheology and polarimetry, respectively. These results are usually related, as described by Djabourov and Papon,20 which allows for a description of the dynamics of the formation/ degradation of the network relative to the protease specificity. A. Physical Enzgels. Rheology. The emergence of the physical network due to hydrogen bonding and triple helix association upon a temperature decrease was first tracked through the viscoelastic properties of the gel. These markers also reveal the gel’s destruction properties (see Figure 2 and Table 1). Whatever the protease used, the double transition (sol–gel, then gel-sol) follows the same evolution with enzyme concentration. The production of Enzgels is only observed in a very narrow enzyme concentration range, confirming that they constitute a precise phenomenon. This also shows that the balance between triple helix formation and enzymatic hydrolysis of the network is very subtle. Slight variations of protease concentrations can lead the system to a sol or to a gel phase. The influence of protease specificity is mainly observed through the enzyme concentrations used to generate the characteristic double transition. A total of 10 times less collagenase than thermolysin or trypsin is needed to obtain physical Enzgels; this is due to the higher number of cleavage sites for collagenase on gelatin chains.21 Moreover, collagenase is not limited by the formation of helices, contrary to the two other enzymes, which act only on the coil part of the protein. Polarimetry. The helix content evolution throughout the Enzgels kinetics was tracked by polarimetry. In each case, whatever the protease, the effect of the enzyme on helix formation is important as the physical gel usually contains 25.5% triple helices at 900 min, while in the presence of the hydrolytic enzyme, this value is reduced to 16-19% (see Figure 3). Proteases are thus able to influence gel rheologial properties but also may modify the supramolecular structure of the sample. Interestingly, the master curves relating G′ and the triple helix previously described22 do not apply for physical Enzgels. Polarimetry does not show a decrease in helix content

16

Biomacromolecules, Vol. 9, No. 1, 2008

Picard et al.

Figure 2. Evolution of visco-elastic properties of physical Enzgels (G′, squares; G′′, triangles) as a function of time for various proteases and protease concentrations. (A) Thermolysin: 2 × 10-3 U mL-1 (black symbols), 3.25 × 10-3 U mL-1 (grey symbols), 4.5 × 10-3 U mL-1 (open symbols). (B) Trypsin: 1.65 × 10-3 U mL-1 (black symbols), 2.45 × 10-3 U mL-1 (grey symbols), 3.3 × 10-3 U mL-1 (open symbols). (C) Collagenase: 0.95 × 10-4 U mL-1 (black symbols), 1.1 × 10-4 U mL-1 L (grey symbols), 1.3 × 10-4 U mL-1 (open symbols). Table 1. Viscoelastic Parameters Characterizing Physical Enzgel Evolution for Three Proteases and for Different Enzyme Concentrationsa [protease] (U mL-1) 0 2 × 10-3 3.25 × 10-3 4.5 × 10-3 0 1.65 × 10-3 2.45 × 10-3 3.3 × 10-3 0 9.5 × 10-5 1.15 × 10-4 1.3 × 10-4 a

gel time (min)

G′ max (Pa)

36 43 57

Thermolysin 816 62 8

36 43 47

Trypsin 816 85 19

36 41 40

Collagenase 816 68 37

G′′ max (Pa)

solubilization time (min)

49 22 6

never more than 760 361

49 25 9

never more than 760 537

49 18 12

never more than 760 549

no gelation

no gelation

no gelation

Gel time was estimated as the first G′ ) G′′ time and solubilization time as the second G′ ) G′′ time.

when G′ values decrease due to the gel resolubilization. This was predictable for thermolysin and trypsin, which cleave gelatin outside the triple-helix sequence. For collagenase, which hydrolyzes gelatin inside the triple helix, this means that the molecular rupture of a peptide bond on one chain of the protein does not interfere with the three-chain structure of the triple helix (see Figure 3). B. Chemical Enzgels. These Enzgels are generated on the basis of transglutaminase-catalyzed covalent bonds. The 40 °C temperature does not permit helices to form. When chemical Enzgels are formed, rheology profiles do not show that G′ < G′′; however, macroscopic observations of the preparations reveal that clear solutions are obtained after a long

reaction time, indicating that a gel/sol transition has occurred. Thus, the second phase transition was thus considered when G′ values decreased below 1 Pa. Indeed, at this value, the rheometer does not control deformation any more, as the system is a liquid phase (see Figure 4). All the results in Figure 4 show that a chemical gel may spontaneously liquefy when an adequate quantity of protease is added to the liquid preparation. As for the physical gel, a balance is created between the activity of transglutaminase (which creates covalent bonds) and the protease (which hydrolyzes covalent bonds). These two enzyme activities can be considered as antagonistic. As previously described,7 obtaining the double transition is a delicate phenomenon. A slight change

Influence of Enzymatic Specificity

Figure 3. Evolution of triple helices as a function of time for physical Enzgels realized with thermolysin (black square), trypsin (grey square), and collagenase (open triangle).

in the enzyme concentrations results in large variations in the macroscopic properties of the mixture. The overall behavior of the preparation is also strictly dependent on the protease specificity. A comparison of the properties of the Enzgels reveals that, for a double transition occurring in 150 min, the maximum value of G′ with trypsin is only 5 Pa, while it is 52 Pa when collagenase is used. For identical mechanical properties (G′ ) 200 Pa and G′′ ) 3 Pa), the Enzgel containing thermolysin solubilizes over 800 min, while the one with collagenase will turn to a liquid phase in 300 min. Moreover, for a comparable

Biomacromolecules, Vol. 9, No. 1, 2008 17

maximum elasticity (G′ ) 220 Pa for trypsin and G′ ) 160 Pa for thermolysin), a 25% increase in thermolysin concentration induces a G′ decrease by 11 times, while a 12.6% trypsin increase leads to a G′ decrease by 44 times: trypsin has more influence on maximum Enzgel elasticity than thermolysin. To enlarge the range of concentrations giving rise to chemical Enzgels, a constant Tgase/protease ratio was used with lower enzyme quantities. In each case, changing the enzyme concentration modifies the chemical Enzgel viscoelasticity. With both trypsin and collagenase, the less enzymes used, the less the two antagonistic activities are expressed, so the two transitions are delayed. Moreover, the maximum G′ value is directly dependent on the quantity of transglutaminase employed. For thermolysin, an opposite behavior is observed; when a small amount of enzymes is used, a higher elasticity is detected and the liquefaction of the gel is delayed (see Figure 5). C. Physical-Chemical Enzgels. The enzymatic reaction is performed at 27 °C where coils undergo a conformational transition and form triple helices, so that a gel due to both weak interactions and covalent bonds is obtained. Under all of the conditions tested, no physical-chemical Enzgel was obtained with trypsin within 900 min. Two different cases are observed without a double phase transition: either no gel is obtained, or the gel does not liquefy. The maximum elasticity of the Enzgel obtained with collagenase progressively decreases when the protease concentration increases (from 110 to 23 Pa). Similarly, the liquefaction time is reduced when more collagenase is used (from 450 to 250

Figure 4. Influence of protease concentration on the kinetics of viscoelastic properties (G′, squares; G′′, triangles) of chemical Enzgels. (A) Thermolysin: 5.2 × 10-3 U mL-1 (black symbols), 6.5 × 10-3 U mL-1 (grey symbols), 7.8 × 10-3 U mL-1 (open symbols). (B) Trypsin: 2.95 × 10-3 U mL-1 (black symbols), 3.8 × 10-3 U mL-1 (grey symbols), 4.35 × 10-3 U mL-1 (open symbols). (C) Collagenase: 1.8 × 10-4 U mL-1 (black symbols), 2.05 × 10-4 U mL-1 (grey symbols), 2.3 × 10-4 U mL-1 (open symbols).

18

Biomacromolecules, Vol. 9, No. 1, 2008

Picard et al.

Figure 5. Influence of enzyme concentration at a constant transglutaminase/protease ratio on the kinetics of viscoelastic properties (G′, squares; G′′, triangles) of chemical Enzgels for the following. (A) Thermolysin 4.35 × 10-3 U mL-1 + transglutaminase 1 U mL-1 (open symbols), thermolysin 6.5 × 10-3 U mL-1 + transglutaminase 1.5 U mL-1 (black symbols). (B) Trypsin 2.9 × 10-3 U mL-1 + transglutaminase 1 U mL-1 (open symbols), trypsin 4.35 × 10-3 U mL-1 + transglutaminase 1.5 U mL-1 (black symbols). (C) Collagenase 1.35 × 10-4 U mL-1 + transglutaminase 1 U mL-1 (black symbols), collagenase 2.05 × 10-4 U mL-1 + transglutaminase 1.5 U mL-1 (open symbols).

min). The parameters characterizing the double transition depend directly on the collagenase concentration introduced in the solution at the beginning of the process (see Figure 6). Thermolysin influence is even more important. A decrease by 25% of the protease concentration increases the liquefaction time by 200 min for collagenase and 500 min for thermolysin. Additional data were obtained with polarimetry. Depending on the protease used, three different evolutions are observed (see Figure 7). Thermolysin does not affect the triple-helix content of the Enzgel during the liquefaction phase, and trypsin has only a slight effect on this parameter. Contrarily, collagenase limits the formation of triple helices in a physicalchemical gel (9% without protease after 150 min, 7% for thermolysin, 6.2% for trypsin, and 4.8% for collagenase). Moreover, collagenase also contributes to the hydrolysis of triple helices (2.1% left after 900 min). These results indicate that, in this type of gel, the specificity of the protease not only is a crucial parameter on the molecular scale but also influences the network dynamics. Increasing the collagenase concentration accelerates gel liquefaction. The maximum viscosity of the gels also decreases relative to the protease concentration. With this protease, the balance between the network formation through triple helices and covalent bonding synthesis and its enzyme-catalyzed hydrolysis is in favor of the leakage of small peptide fragments. This particular activity is due to the recognition of the cleavage bonds independently of their recruitment into a triple helix

structure or in covalent bonds. Only a few other enzymes such as MMP-2 (gelatinase) or MMP-1 (collagenase) can hydrolyze triple helices.23,24 Thermolysin reacts in a different way toward the network organization. Hydrolysis by this protease is sensitive to the presence of covalent bonds, and due to its specificity, it does not cleave the helices. The increase in thermolysin concentration leads to an acceleration of the second transition and a decrease in the viscoelastic properties of the gel phase. Trypsin obeys the same accessibility constraints as thermolysin, but in addition, it reacts on lysine residues, which are also involved in covalent bonding by transglutaminase. When a reticulated covalent network is formed, 30% of the lysines are inaccessible to trypsin hydrolysis,19 which reduces its hydrolytic activity. A kind of competition between the two antagonistic enzymes takes place, which can explain the slow hydrolysis of covalent gels (as illustrated in Figure 1). This phenomenon is also true for Enzgels. The competition affects both gelation and hydrolysis: gelation is more difficult and hydrolysis easier than with another protease. As a result, the enzyme ratio where Enzgels are observed is very precise with trypsin. It is not surprising that two different influence regimes are observed in the double phase transition with physical-chemical Enzgels depending on the trypsin concentration. In the first range, the behavior is the same whatever the protease concentra-

Influence of Enzymatic Specificity

Biomacromolecules, Vol. 9, No. 1, 2008 19

Figure 6. Influence of protease concentration on the kinetics of viscoelastic properties (G′, squares; G′′, triangles) of physical-chemical Enzgels for the following. (A) Thermolysin: 11.5 × 10-3 U mL-1 (black symbols), 13 × 10-3 U mL-1 (grey symbols), 14.5 × 10-3 U mL-1 (open symbols). (B) Trypsin: 10.8 × 10-3 U mL-1 (black symbols), 11 × 10-3 U mL-1 (grey symbols), 11.2 × 10-3 U mL-1 (open symbols). (C) Collagenase: 3.6 × 10-4 U mL-1 (black symbols), 3.95 × 10-4 U mL-1 (grey symbols), 4.3 × 10-4 U mL-1 (open symbols).

Figure 7. Influence of the protease used on the kinetics of triple helices for a physical-chemical Enzgel obtained with (black square) thermolysin, (grey square) trypsin, and (open triangle) collagenase.

tion; then, past a certain threshold, no Enzgels are obtained due to the high hydrolysis of gelatin in the liquid phase before gelation.

between the different mixtures give us a better knowledge of the influence of enzyme specificity on the nature and the assembly of ephemeral gels. We have shown that G′ max as well as the gelation time and the liquefaction time are directly related to the enzyme and to the network. The enzyme concentrations allowing the formation of Enzgels are directly dependent on the protease course of action. By creating a dynamic network of jellified proteins, we generate a good model for the study and understanding of enzymatic remodelling occurring in extracellular matrices, under both normal and pathological conditions. Moreover, the components used in this study are analogous to those of natural tissues, which render extrapolation to biological behaviors more realistic. This new type of biomaterials, named Enzgels, may be feasible with various enzymes. We have shown that the phase transition characteristics and the mechanical properties of the ephemeral gel are unique for each enzyme and for each mixture. We can assume that it is possible to create as many types of Enzgels as there are available proteases, enlarging the number of feasible ephemeral gels. In addition, other types of Enzgels can be conceived for different biopolymers subject to enzymes’ ability to modify their network. This new type of biomaterials presents properties of high interest for wound care, drug delivery, and tissue engineering applications.

Conclusion The results reported here expand and generalize the previously described model of Enzgels.7 Three different kinds of Enzgels (physical, chemical, and physical-chemical) were obtained with two additional proteases. The observations and the comparison

References and Notes (1) Nishinari, K.; Takahashi, R. Curr. Opin. Colloid Interface Sci. 2003, 8, 396–400. (2) Clark, A. H.; Ross-Murphy, S. B. AdV. Polym. Sci. 1987, 83, 57–192.

20

Biomacromolecules, Vol. 9, No. 1, 2008

(3) Gosal, S. W.; Ross-Murphy, S. B. Curr. Opin. Colloid Interface Sci. 2000, 5, 188–194. (4) Matrisian, L. M.; Sledge, G. W.; Mohla, S. Cancer Res. 2003, 63, 6105–6109. (5) Standeven, K. F.; Ariëns, R. A. S.; Grant, P. J. Blood ReV. 2005, 19, 275–288. (6) Hulmes, D. J. S. Essays Biochem. 1992, 27, 49–67. (7) Giraudier, S.; Larreta-Garde, V. Biophys. J. 2007, 93 (2), 629–636. (8) Folk, J. E.; Chung, S. I. Methods Enzymol. 1985, 113, 358–375. (9) Aeschlimann, D.; Thomazy, V. Connect. Tissue Res. 2000, 41, 1–27. (10) Hornebeck, W.; Maquart, F.-X. Biomed. Pharmacother. 2003, 57, 223– 230. (11) Djabourov, M. Contemp. Phys. 1988, 29, 273. (12) Giraudier, S.; Hellio, D.; Djabourov, M.; Larreta Garde, V. Biomacromolecules 2004, 5, 1662–1666. (13) Matsubara, H.; Singer, A.; Sasaki, R.; Jukes, T. H. Biochem. Biophys. Res. Commun. 1965, 21, 242–247. (14) Craik, C. S.; Largman, C.; Fletcher, T.; Roczniak, S.; Barr, P. J.; Fletterick, R.; Rutter, W. J. Science 1985, 228, 291–297.

Picard et al. (15) Vencill, C. F.; Rasnick, D.; Crumley, K. V.; Nishino, N.; Powers, J. C. Biochemistry 1985, 24, 3149–3157. (16) Raghuraman, G.; Rama, R.; Thirumalachari, R. Biochim. Biophys. Acta 2000, 1524, 228–237. (17) Bruckner, P.; Prockop, D. J. Anal. Biochem. 1981, 110, 360–368. (18) de Macédo, P.; Marrano, C.; Keilor, J. W. Anal. Biochem. 2000, 285, 16–20. (19) Griffin, M.; Casadio, R.; Bergamini, C. M. Biochem. J. 2002, 368, 377–396. (20) Djabourov, M.; Papon, P. Polymer 1983, 24, 537–542. (21) Eastoe, J. E. Biochem. J. 1955, 61, 589–600. (22) Joly-Duhamel, C.; Hellio, D.; Ajdari, A.; Djabourov, M. Langmuir 2002, 18, 7158–7166. (23) Sano, M.; Nishino, I.; Ueno, K.; Kamimori, H. J. Chromatogr., B 2004, 809, 251–256. (24) Ottl, J.; Gabriel, D.; Murphy, G.; Knauper, V.; Tominaga, Y.; Nagase, H.; Kroger, M.; Tschesche, H.; Bode, W.; Moroder, L. Chem. Biol. 2000, 119–132.

BM700601N