- Email: [email protected]
Transient amidolytic activity changes of plasmin adsorbed onto carbons
Biomolecular Engineering 19 (2002) 281 /285 www.elsevier.com/locate/geneanbioeng
Transient amidolytic activity changes of plasmin adsorbed onto carbons S. Hourdin a, S. Longchamp a, O. Gallet b, J.-M. Nigretto a,* a
b
Universite´ de Cergy-Pontoise, LECMA (EA2528), 5 mail Gay Lussac, 95031 Cergy-Pontoise Cedex, France Universite´ de Cergy-Pontoise, ERRMECe (EA1391), 2 avenue A. Chauvin, BP122, 95302 Cergy-Pontoise Cedex, France
Abstract The amidolytic activity of plasmin (Pln) that spontaneously adsorbs from solutions to modified-graphite (GR) and glassy carbon (GL) surfaces was studied in the 10 /45 8C temperature range in the presence of a chromogenic substrate. Surfaces were modified with a coating of fibrinogen, either electrochemically oxidized or not. The effect of an additional modification via deposition of Langmuir /Blodgett films of behenic acid (BA-LB) onto the former surfaces, thus leading to either hydrophobic or hydrophilic surfaces according to the number of deposed layers, was examined. In all cases results showed the occurrence of a first order transition which strongly and transiently increases the surface activity of Pln. At BA-LB surfaces, the most prominent change compared with other coatings was a significant enhancement of the critical temperature Tc that characterizes the beginning of the transition. When fibrinogen was present in the solution, the transition was no longer observable up to 37 8C as shown by the linear kinetics exhibited by surfaces bearing oxidized fibrinogen. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Plasmin; Fibrinogen; Graphite and glassy carbon; Contact angles; Langmuir /Blodgett
1. Introduction Carbons are excellent candidates for uses as coatings on implants due to inertness and biocompatible chemical surface composition [1,2]. One strategy to improve biocompatibility of man-made medical devices recommends developing materials that preferentially adsorbs plasminogen (Plg), in the belief that such surfaces may have clot-lysing properties [3,4]. Whether Plg keeps its biological functionality following its adsorption onto carbons is crucial for this task. In a recent work [5], we described the thermal behavior of plasmin (Pln) following its spontaneous adsorption to bare glassy carbon (GL), graphite (GR) and fibrinogen-modified graphite (GR/Fbg) in the 10 /45 8C temperature range with the aid of a chromogenic substrate. The favorable effect exerted by an electrochemical oxidative treatment of the * Corresponding author. Tel.: /33-1-34-257049; fax: /33-1-34257071. E-mail address: [email protected] (J.-M. Nigretto).
fibrinogen under-layer protein was recognized. The aim of this paper is to explore the activity changes undergone by surface Pln following deposition onto the protein layer of Langmuir /Blodgett (LB) ordered films composed of behenic acid (BA). This technique indeed affords the possibility of rendering the protein surface either hydrophobic or hydrophilic (thus leading, respectively, to the so-called GR/Fbg-ox/BA-LBo or GR/Fbgox/BA-LBi surfaces) according to the number of deposed chemical layers. In addition, the effect of the presence of Fbg in the substrate solution to approach as far as possible biological conditions was also examined.
2. Materials and methods 2.1. Reagents and proteins The pH of Tris 0.15 M containing buffered solutions was adjusted either to 7.7 to achieve optimal adsorption of Pln, or to 8.5 to assess optimal amidolytic activity. Lyophilized Pln from Fluka (France) was aliquoted in
1389-0344/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 9 - 0 3 4 4 ( 0 2 ) 0 0 0 3 8 - 2
282
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
Tris 0.15 M in 120 ml volumes at 2.5 U/ml concentration. Lyophilized Fbg from Biogenic (France) was adjusted at 0.7 mg/ml concentration before spontaneous adsorption. Analytical grade behenic acid (BA, C11H44O2) (Sigma) was used without further purification. 2.2. Carbonaceous substrates Polycristalline GR (ref. 9900-207) or GL (ref.45210601, V10) used as sorbent surfaces were purchased from Le Carbone Lorraine (France) in the form of cylindrical rods (diameter 3 mm). A preconditioning treatment consisting of several soakings in alcoholic KOH, rinsing and equilibration in pH 7.7 Tris buffer was applied before use. 2.3. Modification of GR by Fbg and oxidation of the protein GR/Fbg surfaces were prepared upon spontaneous adsorption of Fbg to the carbon in 120 ml of 7 /10 4 mg/ml Fbg solution for 10 min, then carefully rinsed and equilibrated at pH 7.7. To get GR/Fbg-ox surfaces, the protein was mildly oxidized electrochemically for 60 s at pH 7.7 under /0.25 V versus the open circuit potential. 2.4. LB film preparation
circular edge of the cylinder was polished. The carbon shafts were adapted vertically to the chuck of an electric motor and rotated at 500 rpm in a semi-micro polystyrene cuvette (1 cm path length) placed in the cell holder of an UVIKON spectrophotometer thermostated at the desired temperature (9/0.5 8C). The amidolytic activity of surface Pln was monitored at different but constant temperatures as a function of time at 405 nm following cleavage of S-2251 (H-D-Val-Leu-Lys-paranitroanilide, 2 HCl), a chromogenic substrate from Biogenic (France) at 50 mM concentration in the cell for a kinetic of 30 min. For the experiments carried out with Fbg in the assay solution, 400 mM of S-2403 (pyroGlu-Phe-Lys-pNA, HCl) was preferred to enhance the sensitivity of the detection. We verified that these final concentrations were still in excess after completion of the kinetics. In all cases, the reliability of the measurements was judged sufficient given a signal/noise ratio larger than five.
3. Results
3.1. Contact angle measurements 2
BA films were prepared by using a 10 M solution of BA dissolved in chloroform and spread on the water surface of a Nima 611D trough. After the chloroform evaporated, the monolayer was compressed at a surface pressure of 30 mN/m. The graphite rod was immersed in the subphase prior to spreading and compressing the monolayer. To obtain contrasted wettability, three or four multilayers of BA were successively transferred to the surface at 10 mm/min rate. The amount of BA transferred was approximately the same for each layer. 2.5. Contact angle measurements The wettability of the modified surfaces was estimated by contact angle measurements. The experiments were performed with 1 ml drops of ultra-pure water carefully deposed onto the upper part of the modified carbon cylinders disposed horizontally. Angles were estimated manually from printed micrographs obtained with a camera equipped with a zoom. Due to the form of the rods, only relative values contact angle values would be taken for pertinent discussion. 2.6. Immobilization and amidolytic assay of Pln Spontaneous adsorption of Pln was achieved upon immersion of the surfaces in 120 ml of Pln containing aliquots for 10 min. After rinsing in pH 8.5 buffer, the
Contact angle (U) measurements were undertaken to estimate the relative hydrophilic character of the studied GL, GR, GR/Fbg, GR/Fbg-ox and LB coated surfaces (Table 1). As the drop of water spread onto the cylinder right after its deposition, the most hydrophilic surface was GL. The same property was exhibited by GR/Fbgox following inclusion of four successive layers of BALB films (GR/Fbg-ox/BA-LBi surfaces). More stable measurements could be performed on the other surfaces. The least hydrophilic surfaces were GR, GR/Fbg and GR/Fbg-ox after transfer of three layers of BA films (GR/Fbg-ox/BA-LBo). Interestingly, the wettability of these latter surfaces were about the same. Intermediate contact angles were obtained at GR/Fbg-ox surfaces. Table 1 Surface properties of the carbon surfaces Surface
Tc (8C)
Ea (kJ/mol)
U (8)
GL GR GR/Fbg GR/Fbg-ox GR/Fbg-ox/BA-LBo GR/Fbg-ox/BA-LBi
19 19 14 17 29 29
76 102 39 45 104 119
0 75 72 45 72 0
Tc is the critical temperature that initiates the transition, Ea is the activation energy and U is the contact angle.
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
283
3.3. Kinetics at BA-LB- modified GR/Fbg-ox surfaces
Fig. 1. (A) Amidolytic rates for Pln-modified GR/Fbg-ox surfaces in the presence of 50 mM S-2251 Tris 0.15 M at increasing temperatures. (B) (a) same in presence of 400 mM S-2403 at 37 8C, (b) and 2.7/ 10 7 M Fbg.
3.2. Kinetics at GL, GR and Fbg-modified GR surfaces In the preceding paper (3), we established the existence of a first order transition affecting the amidolytic behavior of Pln following its immobilization onto GL, GR and GR/Fbg surfaces, whether oxidized or not. As typically shown in Fig. 1 at GR/Fbg-ox surfaces, the kinetics exhibited two regimes according to the temperature of the monitoring. Provided the temperature is kept below a critical value termed Tc (for instance the curve recorded at 15 8C in Fig. 1A), the amidolytic rates were low (DA per min B/1.5 /103 with a mean of 1 /103 DA per min) and constant for at least 30 min. Past Tc, the rates increased markedly, up to 6 /10 3 DA per min for some selected samples. Thereafter, a gradual decay affects the rates. Once the transient variation is elapsed, rates tends to stabilize (see for instance the curve recorded at 31 8C in Fig. 1A). In the Tc /45 8C range, the slopes of the stabilized rates diminish in proportion of the increase of the temperature (see the curve at 45 8C in Fig. 1A). Depending of the nature of the sorbent surface, Tc values (Table 1) spread in a restricted range from 14 8C (for GR/Fbg modified surface) to 19 8C (for bare GL or GR).
The deposition of BA in the form of controlled amounts and ordered multi-layers onto pre-oxidized Fbg-modified GR surfaces was chosen as a means to change the wettability of the protein layer. The results confirmed the occurrence of the transient amidolytic activity changes undergone by Pln at surfaces observed formerly with no LB modification. To quantify the transient kinetic changes, two parameters had been used [5,7]: the initial rate at zero time, noted V0, and the Relative Activity, calculated 30 min after zero time and defined as RA /V30/V0. Plots of RA versus temperature for GR/Fbg-ox/BA-LBo and GR/Fbg-ox/BA-LBi surfaces are given in Fig. 2 along with those of GR/Fbg-ox surfaces for the sake of comparison. Whatever the modification of the surface, Pln retains still 10 /20% of its initial activity at 35 8C. However, whereas Tc was not significantly altered by surfaces solely modified with Fbg (either oxidized or not), the beginning of the transient kinetic was pushed up from 17 to 29 8C for both BA-LBi and BA-LBo following deposition of amphiphilic films. Considering a first order mechanism to depict the intermediate decay of Pln activity [5], we calculated the corresponding rate constant k and the activation energy using the slopes of the Ln (Vt/V0) versus time and Ln k versus (1/T ) Arrhenius plots, respectively, (Fig. 3A and B). Results presented in Table 1 indicated that the activation energies are similar for GR/Fbg-ox/BA-LBo and GR/Fbg-ox/BA-LBi but differ significantly if compared with that of the initial GR/Fbg-ox surface with no coating. 3.4. Kinetics at GR/Fbg-ox surfaces in Fbg solutions Additional experiments were performed at 37 8C in S-2403 assay solutions containing 2.7 /10 7 M Fbg
Fig. 2. RA (see text) of Pln adsorbed to GR/Fbg-ox, GR/Fbg-ox/BALBo and GR/Fbg-ox/BA-LBi surfaces in the presence of S-2251 as a function of the temperature.
284
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
allotropic form of carbon used) and requiring activation energies of about 70/100 kJ/mol. During this process, it is important to seize that amidolytic surface activity exalts (though RA decreases), whereas the absolute rates decrease uniformly at the hydrophobic and highly heterogeneous former carbon paste surfaces. The behavior of the LB-modified surfaces of the present study follow the same pattern. The transient kinetics should, therefore, be modeled by the same basic sequential reaction scheme: (Pln)ads1 0 (Pln)adsi 0 (Pln)ads2
Fig. 3. (A) Logarithmic transform of RA vs. time according to the first-order thermal transition of Pln adsorbed to GR/Fbg-ox/BA-LBo (B) Arrhenius plot accounting for this kinetics.
(about 1/30th of the plasmatic level) to examine on the surface activity of Pln the effect of the presence of this protein in the medium in conjunction with that of an underlying coating. In those conditions, the transient kinetics was no longer observed as shown by the linear absorbance versus time plot of Fig. 1B. That means that in this case only the first kinetic regime occurring for T B/Tc for the surfaces described can be detected.
4. Discussion To discuss the effect of LB filming on the surface activity of Pln, an examination of the results obtained previously at other modified carbon surfaces would be helpful. Overall, whatever the underlying surface used, the enzyme keeps irreversibly bound to its solid substrate [5] and provided the temperature is sufficiently low, its activity remains independent of the temperature. As the temperature of the assay solution increases, several catalytic features related to the nature of the sorbent surface occur [6]. Upon spontaneous adsorption to quite heterogeneous substrates such as pastes made of an intimate mixture of paraffin oil (nujol) with grounded graphite powder, we assessed that Pln undergoes an irreversible thermally-induced degradation process starting at a temperature as low as Tc :/12 8C [7]. At more ordered surfaces, however, such as bare GL or GR or after spontaneous adsorption of Fbg */further electrochemically oxidized or not */onto these [5], Pln does not only retain its activity up to ca. 45 8C, but we evidenced the existence of a first-order thermal transition, beginning at Tc :/14 /19 8C (depending on the
(1)
According to Eq. (1), the Pln molecules, initially adsorbed in a native and moderately active state denoted (Pln)ads1 at T B/Tc, are believed to pass intermediately through one or a series of transient forms (noted (Pln)adsi in Eq. (1)), endowed with increased activity compared with (Pln)ads1, before evolving toward a final state (Pln)ads2 having an activity again comparable to (Pln)ads1. Decreasing rates observed after Tc above 45 8C had been attributed to the thermal denaturation of (Pln)ads2. The fact that Tc was similar for all surfaces appears intriguing, in that Pln seems remaining sensitive to the presence of GR under the layer of Fbg. Following the transfer of LB layers, a marked change affects the catalytic properties of surface Pln as shown by a gain of about 10 8C for Tc with no variation of the amplitude of the thermal transition. The enhancement of Tc is strongly dependent on the molecular order of the transferred films since there was no change when BA was let to adsorb randomly to the surface from a solution of this solute (unpublished results). As shown by the relative contact angle variations, GR/Fbg-ox surfaces starting with an intermediate hydrophilicity can be readily endowed with contrasted wettability upon deposition of few LB layers. Three transfers are sufficient to render the surface clearly hydrophobic (GR/Fbg-ox/BA-LBo) whereas four transfers make it hydrophilic (GR/Fbg-ox/BA-LBi). Interestingly, the enhancement of Tc appears unrelated to this surface property as the same value of 29 8C is observed irrespective of the nature of the coating. Prior to the transfer, LB films are known to form well-defined and highly organized structures. But once in the presence of the protein-modified substrate, i.e. at a surface bearing some bulk roughness (SEM micrographs not shown) and a mean wettability, it would be unrealistic to assume that the primitive molecular arrangement remains unaffected while being transferred. Nevertheless the contrasted final contact angles suggest that some molecular integrity is yet preserved after transfer. It is probable that the first LB layer more or less penetrates the swollen protein layer and so the subsequent coatings would contribute to mask gradually the physical heterogeneity of the protein coating.
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
Otherwise, the activation energy characterizes the energy required by Pln to overcome the barrier of the rate-limiting step, i.e. the transition whereby amidolysis accelerates due to more favorable exposure of the catalytic site to the approaching substrate. Intuitively, increased Tc should reflect strong binding of the enzyme to the surface. On this basis, Ea and Tc should not be related each other, as shown by the corresponding variations represented in Table 1. This remark leads to the conclusion that the presence of Fbg in the solution reinforces the binding of Pln to GR/Fbg-ox since no transition could be detected before 37 8C, thus preventing the enzyme from the subsequent conformational changes resulting in increased activity. In the recent literature, much efforts have been reported for assessing pertinent parameters responsible for the expression of enzymatic surface activity. As the existence of specific molecular interactions have not been recognized insofar, the results of this study plead in favor of the relevance of the physical homogeneity of the sorbent surface rather than of its wettability.
285
Acknowledgements This work was supported by ANVAR (Nr A9812238Z/AT). The authors thank O. FICHET for assistance to the LB film preparation.
References [1] Bockros JC, LaGrange LD, Schoen J. In: Walker PL, Jr, Thrower PA, editors. Chemistry and Physics of Carbon, vol. 9. New York: Dekker, 1973:103. [2] Haubold AD, Shim HS, Bockros JC. In: William DF, editor. Biocompatibility of Clinical Implant Materials, vol. 2. Boca Raton: CRC Press, 1981:3. [3] Packham MA. Proc Soc Exp Biol Med 1988;189:261 /74. [4] McClung WG, Clapper DL, Hu SP, Brash JL. J Biomed Mater Res 2000;490:409 /14. [5] Hourdin S, Longchamp S, Gallet O, Nigretto J-M. J Colloid Interf Sci 2001;236:132 /40. [6] Sadana A. Chem Rev 1992;92:1799 /818. [7] Longchamp S, Randriamahazaka HN, Nigretto J-M. J Colloid Interf Sci 1994;166:444 /50.
Transient amidolytic activity changes of plasmin adsorbed onto carbons S. Hourdin a, S. Longchamp a, O. Gallet b, J.-M. Nigretto a,* a
b
Universite´ de Cergy-Pontoise, LECMA (EA2528), 5 mail Gay Lussac, 95031 Cergy-Pontoise Cedex, France Universite´ de Cergy-Pontoise, ERRMECe (EA1391), 2 avenue A. Chauvin, BP122, 95302 Cergy-Pontoise Cedex, France
Abstract The amidolytic activity of plasmin (Pln) that spontaneously adsorbs from solutions to modified-graphite (GR) and glassy carbon (GL) surfaces was studied in the 10 /45 8C temperature range in the presence of a chromogenic substrate. Surfaces were modified with a coating of fibrinogen, either electrochemically oxidized or not. The effect of an additional modification via deposition of Langmuir /Blodgett films of behenic acid (BA-LB) onto the former surfaces, thus leading to either hydrophobic or hydrophilic surfaces according to the number of deposed layers, was examined. In all cases results showed the occurrence of a first order transition which strongly and transiently increases the surface activity of Pln. At BA-LB surfaces, the most prominent change compared with other coatings was a significant enhancement of the critical temperature Tc that characterizes the beginning of the transition. When fibrinogen was present in the solution, the transition was no longer observable up to 37 8C as shown by the linear kinetics exhibited by surfaces bearing oxidized fibrinogen. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Plasmin; Fibrinogen; Graphite and glassy carbon; Contact angles; Langmuir /Blodgett
1. Introduction Carbons are excellent candidates for uses as coatings on implants due to inertness and biocompatible chemical surface composition [1,2]. One strategy to improve biocompatibility of man-made medical devices recommends developing materials that preferentially adsorbs plasminogen (Plg), in the belief that such surfaces may have clot-lysing properties [3,4]. Whether Plg keeps its biological functionality following its adsorption onto carbons is crucial for this task. In a recent work [5], we described the thermal behavior of plasmin (Pln) following its spontaneous adsorption to bare glassy carbon (GL), graphite (GR) and fibrinogen-modified graphite (GR/Fbg) in the 10 /45 8C temperature range with the aid of a chromogenic substrate. The favorable effect exerted by an electrochemical oxidative treatment of the * Corresponding author. Tel.: /33-1-34-257049; fax: /33-1-34257071. E-mail address: [email protected] (J.-M. Nigretto).
fibrinogen under-layer protein was recognized. The aim of this paper is to explore the activity changes undergone by surface Pln following deposition onto the protein layer of Langmuir /Blodgett (LB) ordered films composed of behenic acid (BA). This technique indeed affords the possibility of rendering the protein surface either hydrophobic or hydrophilic (thus leading, respectively, to the so-called GR/Fbg-ox/BA-LBo or GR/Fbgox/BA-LBi surfaces) according to the number of deposed chemical layers. In addition, the effect of the presence of Fbg in the substrate solution to approach as far as possible biological conditions was also examined.
2. Materials and methods 2.1. Reagents and proteins The pH of Tris 0.15 M containing buffered solutions was adjusted either to 7.7 to achieve optimal adsorption of Pln, or to 8.5 to assess optimal amidolytic activity. Lyophilized Pln from Fluka (France) was aliquoted in
1389-0344/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 9 - 0 3 4 4 ( 0 2 ) 0 0 0 3 8 - 2
282
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
Tris 0.15 M in 120 ml volumes at 2.5 U/ml concentration. Lyophilized Fbg from Biogenic (France) was adjusted at 0.7 mg/ml concentration before spontaneous adsorption. Analytical grade behenic acid (BA, C11H44O2) (Sigma) was used without further purification. 2.2. Carbonaceous substrates Polycristalline GR (ref. 9900-207) or GL (ref.45210601, V10) used as sorbent surfaces were purchased from Le Carbone Lorraine (France) in the form of cylindrical rods (diameter 3 mm). A preconditioning treatment consisting of several soakings in alcoholic KOH, rinsing and equilibration in pH 7.7 Tris buffer was applied before use. 2.3. Modification of GR by Fbg and oxidation of the protein GR/Fbg surfaces were prepared upon spontaneous adsorption of Fbg to the carbon in 120 ml of 7 /10 4 mg/ml Fbg solution for 10 min, then carefully rinsed and equilibrated at pH 7.7. To get GR/Fbg-ox surfaces, the protein was mildly oxidized electrochemically for 60 s at pH 7.7 under /0.25 V versus the open circuit potential. 2.4. LB film preparation
circular edge of the cylinder was polished. The carbon shafts were adapted vertically to the chuck of an electric motor and rotated at 500 rpm in a semi-micro polystyrene cuvette (1 cm path length) placed in the cell holder of an UVIKON spectrophotometer thermostated at the desired temperature (9/0.5 8C). The amidolytic activity of surface Pln was monitored at different but constant temperatures as a function of time at 405 nm following cleavage of S-2251 (H-D-Val-Leu-Lys-paranitroanilide, 2 HCl), a chromogenic substrate from Biogenic (France) at 50 mM concentration in the cell for a kinetic of 30 min. For the experiments carried out with Fbg in the assay solution, 400 mM of S-2403 (pyroGlu-Phe-Lys-pNA, HCl) was preferred to enhance the sensitivity of the detection. We verified that these final concentrations were still in excess after completion of the kinetics. In all cases, the reliability of the measurements was judged sufficient given a signal/noise ratio larger than five.
3. Results
3.1. Contact angle measurements 2
BA films were prepared by using a 10 M solution of BA dissolved in chloroform and spread on the water surface of a Nima 611D trough. After the chloroform evaporated, the monolayer was compressed at a surface pressure of 30 mN/m. The graphite rod was immersed in the subphase prior to spreading and compressing the monolayer. To obtain contrasted wettability, three or four multilayers of BA were successively transferred to the surface at 10 mm/min rate. The amount of BA transferred was approximately the same for each layer. 2.5. Contact angle measurements The wettability of the modified surfaces was estimated by contact angle measurements. The experiments were performed with 1 ml drops of ultra-pure water carefully deposed onto the upper part of the modified carbon cylinders disposed horizontally. Angles were estimated manually from printed micrographs obtained with a camera equipped with a zoom. Due to the form of the rods, only relative values contact angle values would be taken for pertinent discussion. 2.6. Immobilization and amidolytic assay of Pln Spontaneous adsorption of Pln was achieved upon immersion of the surfaces in 120 ml of Pln containing aliquots for 10 min. After rinsing in pH 8.5 buffer, the
Contact angle (U) measurements were undertaken to estimate the relative hydrophilic character of the studied GL, GR, GR/Fbg, GR/Fbg-ox and LB coated surfaces (Table 1). As the drop of water spread onto the cylinder right after its deposition, the most hydrophilic surface was GL. The same property was exhibited by GR/Fbgox following inclusion of four successive layers of BALB films (GR/Fbg-ox/BA-LBi surfaces). More stable measurements could be performed on the other surfaces. The least hydrophilic surfaces were GR, GR/Fbg and GR/Fbg-ox after transfer of three layers of BA films (GR/Fbg-ox/BA-LBo). Interestingly, the wettability of these latter surfaces were about the same. Intermediate contact angles were obtained at GR/Fbg-ox surfaces. Table 1 Surface properties of the carbon surfaces Surface
Tc (8C)
Ea (kJ/mol)
U (8)
GL GR GR/Fbg GR/Fbg-ox GR/Fbg-ox/BA-LBo GR/Fbg-ox/BA-LBi
19 19 14 17 29 29
76 102 39 45 104 119
0 75 72 45 72 0
Tc is the critical temperature that initiates the transition, Ea is the activation energy and U is the contact angle.
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
283
3.3. Kinetics at BA-LB- modified GR/Fbg-ox surfaces
Fig. 1. (A) Amidolytic rates for Pln-modified GR/Fbg-ox surfaces in the presence of 50 mM S-2251 Tris 0.15 M at increasing temperatures. (B) (a) same in presence of 400 mM S-2403 at 37 8C, (b) and 2.7/ 10 7 M Fbg.
3.2. Kinetics at GL, GR and Fbg-modified GR surfaces In the preceding paper (3), we established the existence of a first order transition affecting the amidolytic behavior of Pln following its immobilization onto GL, GR and GR/Fbg surfaces, whether oxidized or not. As typically shown in Fig. 1 at GR/Fbg-ox surfaces, the kinetics exhibited two regimes according to the temperature of the monitoring. Provided the temperature is kept below a critical value termed Tc (for instance the curve recorded at 15 8C in Fig. 1A), the amidolytic rates were low (DA per min B/1.5 /103 with a mean of 1 /103 DA per min) and constant for at least 30 min. Past Tc, the rates increased markedly, up to 6 /10 3 DA per min for some selected samples. Thereafter, a gradual decay affects the rates. Once the transient variation is elapsed, rates tends to stabilize (see for instance the curve recorded at 31 8C in Fig. 1A). In the Tc /45 8C range, the slopes of the stabilized rates diminish in proportion of the increase of the temperature (see the curve at 45 8C in Fig. 1A). Depending of the nature of the sorbent surface, Tc values (Table 1) spread in a restricted range from 14 8C (for GR/Fbg modified surface) to 19 8C (for bare GL or GR).
The deposition of BA in the form of controlled amounts and ordered multi-layers onto pre-oxidized Fbg-modified GR surfaces was chosen as a means to change the wettability of the protein layer. The results confirmed the occurrence of the transient amidolytic activity changes undergone by Pln at surfaces observed formerly with no LB modification. To quantify the transient kinetic changes, two parameters had been used [5,7]: the initial rate at zero time, noted V0, and the Relative Activity, calculated 30 min after zero time and defined as RA /V30/V0. Plots of RA versus temperature for GR/Fbg-ox/BA-LBo and GR/Fbg-ox/BA-LBi surfaces are given in Fig. 2 along with those of GR/Fbg-ox surfaces for the sake of comparison. Whatever the modification of the surface, Pln retains still 10 /20% of its initial activity at 35 8C. However, whereas Tc was not significantly altered by surfaces solely modified with Fbg (either oxidized or not), the beginning of the transient kinetic was pushed up from 17 to 29 8C for both BA-LBi and BA-LBo following deposition of amphiphilic films. Considering a first order mechanism to depict the intermediate decay of Pln activity [5], we calculated the corresponding rate constant k and the activation energy using the slopes of the Ln (Vt/V0) versus time and Ln k versus (1/T ) Arrhenius plots, respectively, (Fig. 3A and B). Results presented in Table 1 indicated that the activation energies are similar for GR/Fbg-ox/BA-LBo and GR/Fbg-ox/BA-LBi but differ significantly if compared with that of the initial GR/Fbg-ox surface with no coating. 3.4. Kinetics at GR/Fbg-ox surfaces in Fbg solutions Additional experiments were performed at 37 8C in S-2403 assay solutions containing 2.7 /10 7 M Fbg
Fig. 2. RA (see text) of Pln adsorbed to GR/Fbg-ox, GR/Fbg-ox/BALBo and GR/Fbg-ox/BA-LBi surfaces in the presence of S-2251 as a function of the temperature.
284
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
allotropic form of carbon used) and requiring activation energies of about 70/100 kJ/mol. During this process, it is important to seize that amidolytic surface activity exalts (though RA decreases), whereas the absolute rates decrease uniformly at the hydrophobic and highly heterogeneous former carbon paste surfaces. The behavior of the LB-modified surfaces of the present study follow the same pattern. The transient kinetics should, therefore, be modeled by the same basic sequential reaction scheme: (Pln)ads1 0 (Pln)adsi 0 (Pln)ads2
Fig. 3. (A) Logarithmic transform of RA vs. time according to the first-order thermal transition of Pln adsorbed to GR/Fbg-ox/BA-LBo (B) Arrhenius plot accounting for this kinetics.
(about 1/30th of the plasmatic level) to examine on the surface activity of Pln the effect of the presence of this protein in the medium in conjunction with that of an underlying coating. In those conditions, the transient kinetics was no longer observed as shown by the linear absorbance versus time plot of Fig. 1B. That means that in this case only the first kinetic regime occurring for T B/Tc for the surfaces described can be detected.
4. Discussion To discuss the effect of LB filming on the surface activity of Pln, an examination of the results obtained previously at other modified carbon surfaces would be helpful. Overall, whatever the underlying surface used, the enzyme keeps irreversibly bound to its solid substrate [5] and provided the temperature is sufficiently low, its activity remains independent of the temperature. As the temperature of the assay solution increases, several catalytic features related to the nature of the sorbent surface occur [6]. Upon spontaneous adsorption to quite heterogeneous substrates such as pastes made of an intimate mixture of paraffin oil (nujol) with grounded graphite powder, we assessed that Pln undergoes an irreversible thermally-induced degradation process starting at a temperature as low as Tc :/12 8C [7]. At more ordered surfaces, however, such as bare GL or GR or after spontaneous adsorption of Fbg */further electrochemically oxidized or not */onto these [5], Pln does not only retain its activity up to ca. 45 8C, but we evidenced the existence of a first-order thermal transition, beginning at Tc :/14 /19 8C (depending on the
(1)
According to Eq. (1), the Pln molecules, initially adsorbed in a native and moderately active state denoted (Pln)ads1 at T B/Tc, are believed to pass intermediately through one or a series of transient forms (noted (Pln)adsi in Eq. (1)), endowed with increased activity compared with (Pln)ads1, before evolving toward a final state (Pln)ads2 having an activity again comparable to (Pln)ads1. Decreasing rates observed after Tc above 45 8C had been attributed to the thermal denaturation of (Pln)ads2. The fact that Tc was similar for all surfaces appears intriguing, in that Pln seems remaining sensitive to the presence of GR under the layer of Fbg. Following the transfer of LB layers, a marked change affects the catalytic properties of surface Pln as shown by a gain of about 10 8C for Tc with no variation of the amplitude of the thermal transition. The enhancement of Tc is strongly dependent on the molecular order of the transferred films since there was no change when BA was let to adsorb randomly to the surface from a solution of this solute (unpublished results). As shown by the relative contact angle variations, GR/Fbg-ox surfaces starting with an intermediate hydrophilicity can be readily endowed with contrasted wettability upon deposition of few LB layers. Three transfers are sufficient to render the surface clearly hydrophobic (GR/Fbg-ox/BA-LBo) whereas four transfers make it hydrophilic (GR/Fbg-ox/BA-LBi). Interestingly, the enhancement of Tc appears unrelated to this surface property as the same value of 29 8C is observed irrespective of the nature of the coating. Prior to the transfer, LB films are known to form well-defined and highly organized structures. But once in the presence of the protein-modified substrate, i.e. at a surface bearing some bulk roughness (SEM micrographs not shown) and a mean wettability, it would be unrealistic to assume that the primitive molecular arrangement remains unaffected while being transferred. Nevertheless the contrasted final contact angles suggest that some molecular integrity is yet preserved after transfer. It is probable that the first LB layer more or less penetrates the swollen protein layer and so the subsequent coatings would contribute to mask gradually the physical heterogeneity of the protein coating.
S. Hourdin et al. / Biomolecular Engineering 19 (2002) 281 /285
Otherwise, the activation energy characterizes the energy required by Pln to overcome the barrier of the rate-limiting step, i.e. the transition whereby amidolysis accelerates due to more favorable exposure of the catalytic site to the approaching substrate. Intuitively, increased Tc should reflect strong binding of the enzyme to the surface. On this basis, Ea and Tc should not be related each other, as shown by the corresponding variations represented in Table 1. This remark leads to the conclusion that the presence of Fbg in the solution reinforces the binding of Pln to GR/Fbg-ox since no transition could be detected before 37 8C, thus preventing the enzyme from the subsequent conformational changes resulting in increased activity. In the recent literature, much efforts have been reported for assessing pertinent parameters responsible for the expression of enzymatic surface activity. As the existence of specific molecular interactions have not been recognized insofar, the results of this study plead in favor of the relevance of the physical homogeneity of the sorbent surface rather than of its wettability.
285
Acknowledgements This work was supported by ANVAR (Nr A9812238Z/AT). The authors thank O. FICHET for assistance to the LB film preparation.
References [1] Bockros JC, LaGrange LD, Schoen J. In: Walker PL, Jr, Thrower PA, editors. Chemistry and Physics of Carbon, vol. 9. New York: Dekker, 1973:103. [2] Haubold AD, Shim HS, Bockros JC. In: William DF, editor. Biocompatibility of Clinical Implant Materials, vol. 2. Boca Raton: CRC Press, 1981:3. [3] Packham MA. Proc Soc Exp Biol Med 1988;189:261 /74. [4] McClung WG, Clapper DL, Hu SP, Brash JL. J Biomed Mater Res 2000;490:409 /14. [5] Hourdin S, Longchamp S, Gallet O, Nigretto J-M. J Colloid Interf Sci 2001;236:132 /40. [6] Sadana A. Chem Rev 1992;92:1799 /818. [7] Longchamp S, Randriamahazaka HN, Nigretto J-M. J Colloid Interf Sci 1994;166:444 /50.