- Email: [email protected]
Atomic force microscope studies on the interactions of Candida rugosa lipase and supported lipidic bilayers
Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
Atomic force microscope studies on the interactions of Candida rugosa lipase and supported lipidic bilayers Nuria Prim a,b,∗ , Lars Iversen b , Pilar Diaz a , Thomas Bjørnholm b a
Department of Microbiology, Faculty of Biology, University of Barcelona, Av. Diagonal 645, 08028 Barcelona, Spain b NanoScience Center, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark Received 16 May 2006; accepted 19 May 2006 Available online 2 June 2006
Abstract Using the Langmuir–Blodgett technique we prepared substrate supported well-defined lipid/phospholipid (1-mono-palmitoyl-rac-glycerol (MPG)/l,2-dipalmitoyl-sn-glycerol-3-phosphocholine (DPPC)) bilayers in which the MPG lipid leaflet was exposed to the aqueous phase. Hydrolysis of MPG performed by Candida rugosa lipase (CRL) on the upper MPG layer of these supported bilayers on mica was imaged by real time atomic force microscope (AFM) using a liquid cell, so that the area increase of the initial structural defects could be followed over time. Our data strongly suggest that the edges of the initial structural defects are the preferred activation sites for CRL once the enzyme is adsorbed onto these interfaces. When a 2.5 nM bulk concentration of CRL was assayed on this planar lipid substrate, we found a long lag phase before a sharp increase of catalytic activity. The lag–burst kinetic behaviour was related to the interfacial activation phenomenon although we propose that it is also dependent on the gel-phase state of this interface. © 2006 Elsevier B.V. All rights reserved. Keywords: Lipase; Candida rugosa; Langmuir–Blodgett; Atomic force microscope (AFM)
1. Introduction Lipases (EC 3.1.1.3, triacylaglycerol hydrolases) cleave the ester bonds of long-chain lipids like triacylglycerides [1]. The catalytic activity of lipases on soluble substrates is very low, and it sharply increases when the substrate is aggregated forming a lipid/water interface, a phenomenon which is known as interfacial activation [2]. This process has been associated with a conformational change of a helical structure (lid) covering the catalytic centre of most lipase. When the enzyme is adsorbed to a lipidic interface, this lid opens making the hydrophobic active site accessible to the substrate [3]. Amphiphilic molecules, like certain lipids, can be assembled as monomolecular layers at the air–water interface [4–6]. For decades lipid monolayers have been used as a substrate for lipolytic enzymes, their main advantage being the possibility of controlling the quality of the interface [7]. The overall
∗
Corresponding author. Tel.: +34 934064675; fax: +34 934034629. E-mail address: [email protected] (N. Prim).
0927-7765/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2006.05.018
catalytic hydrolysis involves adsorption of the enzyme from the bulk aqueous phase to the surface, followed by catalytic action at the interfacial plane [8]. Additionally, these lipid monolayers can be transferred to a solid support by dipping, which allows also the formation of bilayers and multilayers [9]. In the last few years, atomic force microscopy (AFM) has become a key technique in biophysics and biochemistry for the characterization of supported lipid films, in part due to its unique capacity to probe their local surface structure in real time and in an aqueous environment [10]. AFM studies on the hydrolytic performance on supported bilayers prepared by the Langmuir–Blodgett technique have been so far reported for only two lipolytic enzymes in the literature, snake venom phospholipase A2 from Agkistrodon piscivorus piscivorus [11,12] and Humicola lanuginosa lipase [13–15]. The objective of the present work was to study the catalytic performance of Candida rugosa lipase on 1-mono-palmitoyl-rac-glycerol (MPG) by means of in situ AFM imaging. These studies were performed for the first time on hybrid supported DPPC/MPG bilayers, using CRL lipase.
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
2. Experimental 2.1. Lipids and enzyme l,2-Dipalmitoyl-sn-glycerol-3-phosphocholine (DPPC), 1mono-palmitoyl-rac-glycerol (MPG) and C. rugosa lipase (CRL) were purchased from Sigma. CRL was diluted to a concentration of 2.5 nM in 50 mM Tris–HCl buffer (pH 7.4) supplemented with 1.5 mM CaCl2 . 2.2. Bilayer preparation Hybrid DPPC/MPG bilayers were prepared by means of the Langmuir–Blodgett technique on a commercial LB trough (KSV 5000, KSV Ltd., Finland) and using freshly cleaved mica as a support. DPPC was dissolved in HPLC grade n-hexane to a concentration of 0.6 mg/ml, while MPG was dissolved in chloroform to a concentration of 1 mg/ml. To prepare the bottom layer, DPPC was spread with a Hamilton microsyringe on a pure MilliQ water surface. Ten minutes were allowed to elapse for solvent evaporation before compression was carried out until a target pressure of 35 mN/m. After monolayer stabilization for 30 min, vertical transfer was performed onto mica at a rate of 1 mm/min. After cleaning the trough, the MPG solution was spread on a pure Milli-Q water surface and compression was performed to the same target pressure as before. The MPG monolayer was stabilized for 2–3 h before horizontally transferring onto the first DPPC layer at a transfer arte of 0.5 mm/min. 2.3. AFM imaging The supported bilayers were transferred onto a liquid cell of the AFM (Nanoscope IIIa, Digital Instruments, Santa Barbara, CA) using a standard silicon O-ring in order to seal the chamber. The liquid cell was flushed in with 50 mM Tris–HCl buffer (pH 7.4) supplemented with 1.5 mM CaCl2 and the sample was allowed to thermally equilibrate before sucking in the enzyme solution prepared as described above. Scanning was performed using oxide sharpened silicon-nitride tips (NanoProbes, Santa Barbara, CA). Imaging was carried out in tapping mode at room temperature. 2.4. Analysis of images Image analysis to measure the area and perimeter of the growing bilayer defects was performed by using the public domain ImageJ 1.33u software (http://rsb.info.nih.gov/ij/). 3. Results and discussion Due to the special nature of lipases and their interfacial activation, during the kinetic studies of these enzymes it is extremely important to have well-defined substrates. Simple lipids like 1-mono-palmitoyl-rac-glycerol have only a weak amphiphilic behaviour. This was overcome by using a strong amphiphilic phospholipid (1,2-dipalmitoyl-sn-glycerol3-phosphocoline, DPPC) as a supporting first layer onto mica, as
139
it had been described before for other lipidic bilayers [15]. Onto this first hydrophobic supporting layer, a gel-phase monolayer of MPG was easily transferred. The resulting hybrid bilayer has a well-defined planar top layer of MPG in which the headgroups of the lipids are exposed to the aqueous solution containing the enzyme. C. rugosa lipase acts on the ester bonds of a broad range of substrates, mainly mono-, di- and triacylglycerides [16]. In the present work, these hybrid DPPC/MPG bilayers supported on mica have been proved to be also a good substrate for this enzyme, allowing in situ tapping mode AFM imaging of the hydrolysis process performed inside the liquid cell. All DPPC/MPG bilayers showed some initial structural defects (Fig. 1) with a height difference between the top part and the mica (dark domains) of 5.5 ± 0.3 nm (Fig. 1a, inset), a value that is in agreement with previous data on the monolayer thickness for both molecules [15,17]. In control experiments, these bilayers were imaged with the AFM over 2 h in order to check for stability, revealing no tip–sample interaction effects. The study of enzyme catalysis at lipid–water interfaces is very complex, mainly because of the difficulty of exactly determining the concentration of enzyme acting at the interface [18]. Lipase activity is preceded by the binding of the enzyme from the bulk phase to the interface, this being followed by its activation. It is established that in the order of nanomolar concentrations of enzymes, reaching an equilibrium value at the interface would need an elapse of some minutes [18]. Therefore, we chose a low concentration of CRL in order to perform reproducible analysis and understand this lipase behaviour at this interface. After thermal equilibration of the liquid cell, a 2.5 nM solution containing the enzyme was flushed into the cell, and hydrolytic desorption of the bilayer was imaged over time. As it was initially presumed that the enzyme would initiate hydrolysis of MPG at these structural defects of the bilayer, activity could thus be easily monitored by analysing the area increase of uncovered mica due to the cleaving of palmitic acid and subsequent fatty acid release from the MPG upper layer. When hydrolysis was followed for a long period of time (80–120 min), “new” holes appeared on some of these bilayers due to the presence of initial nanoscale structural defects that could not be easily observed in the usual 25 m2 AFM scan area. As the existence of these “new” growing holes was at very definite spots and the bilayer degradation was not randomly performed in the presence of CRL, these results confirmed the requirement of a structural defect for this enzyme to first attach to the interface and get into an active conformation, so that the hydrolysis process on the MPG could take place. A time-course set of images from one of these experiments is shown in Fig. 1. The first image (Fig. 1a) was taken prior to flushing a 2.5 nM solution of CRL into the liquid cell, while the rest of the images (Fig. 1b–f) show the changes produced on the bilayer by the enzymatic action. For quantification of the kinetic rates, all holes from the AFM images recorded along the first 80 min for each experiment were analysed in respect of the desorbed area increase (A), the initial perimeter length (P0 ) and the change in time (t). As described before by Balashev et al. [15], the initial perimeter length of the holes may
140
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
Fig. 1. Time set of AFM images showing the MPG hydrolysis when a 2.5 nM CRL solution was flushed into the liquid cell. The dark areas are the bilayer structural defects, which expand after adding the enzyme due to MPG hydrolysis and subsequently desorption of the DPPC bottom layer. (a) Prior to flushing the enzyme into the cell; (b) 12 min after flushing the enzyme into the cell; (c–f) selected images 44–82 min after flushing the enzyme. Inset in ‘a’ shows the height measure (white line) of the bilayer, corresponding to 5.5 nm depth.
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
Fig. 2. Desorbed lipid area corrected by the initial perimeter of the defects vs. time after flushing a 2.5 nM solution of enzyme into the liquid cell.
enter the rate equation as an indicative parameter for effective substrate concentration. Therefore, we refer to the initial rate of hydrolysis as v0 = A/t and the normalised rate of hydrolysis as A/t × P0 . Plots of the desorbed area/perimeter versus time (Fig. 2) revealed the existence of a long lag phase (38 ± 2 min) after flushing the enzyme, followed by a burst of hydrolytic activity until reaching a practically zero hydrolysis rate. Lipase hydrolytic performance generates water soluble fatty acids that may remain transiently at the interface [12]. Accumulation of hydrolysis products and DPPC may cause inhibition of CRL [12,19], which would explain the progressive decrease of the hydrolysis rates after the burst of activity and the big variation observed at higher times. Rao and Damodaran [18] showed that when the CRL reaction on dicaprin monolayers was allowed to proceed indefinitely in the presence of -cyclodextrin by removing the reaction products, the surface pressure decreased to zero because of complete hydrolysis of the layer. In our case, bilayers were not completely desorbed after imaging for 3–4 h, indicating that the aforementioned practically zero activity rate is reached because of the presence of the catalysis products and DPPC. Therefore, only initial rates (v0 ) for each hole, calculated from the two consecutive AFM images where the first sharp increase of activity was detected, were considered for the kinetic measurements. The average normalised initial rate A/t × P0 for CRL catalysis (2.5 nM) in all the experiments performed on these supported bilayers was found to be 11 × 10−5 ± 7 × 10−5 m/min. When we plotted the initial rate v0 versus the initial perimeter P0 of the holes (Fig. 3), the high standard deviation was translated into a big spread in the distribution of the different experiments, a fact previously reported also for low concentrations of H. lanuginosa lipase acting on similar systems [15]. For most of the experiments, we found initial rates of 6 × 10−5 ± 1.5 × 10−5 m2 /min, independently of the initial perimeter of the structural defect. Additionally, in a number of experiments, the initial rates clustered into three other groups, displaying higher v0 values of 12.3 × 10−5 ± 1.5 × 10−5 m2 / min, 18.3 × 10−5 ± 2 × 10−6 m2 /min and 23 × 10−5 ± 2.8 × 10−5 m2 /min, respectively. This scattering in initial
141
Fig. 3. Initial rates of hydrolysis (v0 ) corresponding to a single experiment plotted vs. the initial perimeter (P0 ). The patterns show the clustering observed in the rates, which would correspond to one (䊉), two (), three () and four () enzyme molecules.
rates can be explained by a stochastic distribution of single enzyme molecules in the bilayer defects because of the very low bulk concentration of enzyme. We hypothesize from these data that the lower initial rates might be the result of one single molecule of enzyme, while the higher rates could be due to two, three and four molecules, respectively. The presence of a lag phase is a very common phenomenon in the kinetics of lipases, having usually been related to the interfacial activation of these enzymes [3]. Lag–burst kinetics has been reported before for CRL acting on dicaprin monolayers using the zero order trough technique [18] and was also observed by means of AFM for snake venom phospholipase A2 (PLA2 ) on DPPC interfaces [13]. Curiously no lag phase was reported for H. lanuginosa lipase (HLL) on MOG/DPPC bilayers using AFM [15] despite being usually known to perform interfacial activation [20], most probably because of interference in the detection of a short lag phase by the elapse of time in the AFM imaging. The importance of the conformational and interfacial properties of lipids in the catalytic activity of lipases has been reported before [8,20]. In this sense, Rao and Damodaran [18] observed a decrease in the rate of CRL adsorption to monolayers with increasing surface pressure. We propose that differences in the absence/presence of a lag phase in the case of HLL [15], PLA2 [13] and CRL (this work) could not only be related to the interfacial phenomenon itself but also to the physical state of the different supported lipid bilayers analysed. Thus, the long lag phase observed for CRL on these bilayers may also be due to a highly structured gel-phase MPG layer, which would make it more difficult for the enzyme molecules to adsorb onto the interface and therefore get into an active conformation. 4. Conclusions C. rugosa lipase activity on MPG has been imaged in situ by means of AFM for the first time. Qualitative analysis showed that this enzyme needed to be adsorbed at the edge of the initial structural defects present at this interface in order to get into the
142
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
active conformation and initialize hydrolysis. We hypothesize that the long lag phase behaviour observed may be related not only to the interfacial activation phenomenon itself, but also the gel-phase state of the MPG layer. The burst of hydrolysis on this substrate after the lag phase was followed by a practically zero hydrolysis rate most probably due to enzyme inhibition by both the released fatty acids and DPPC constituting the bottom layer. Acknowledgements This work was partially financed by a nanotechnology grant from the Catalan Government Research Foundation (Dursi). The authors would like to thank Dimitrios Stamou, Martin Gudman, Tue Hassenkam and Pedro Nunes for fruitful discussions. References [1] [2] [3] [4] [5]
R. Gupta, N. Gupta, P. Rathi, Microbiol. Biotechnol. 64 (2004) 763–781. L. Sarda, P. Desnuelle, Biochim. Biophys. Acta 30 (1958) 513–521. L. Arpigny, K.E. Jaeger, Biochem. J. 343 (1999) 177–183. A. Pockels, Nature (London) 43 (1891) 437. I. Langmuir, J. Am. Chem. Soc. 39 (1917) 1848–1906.
[6] I. Langmuir, Trans. Faraday Soc. 15 (1920) 62. [7] I. Douchet, G. Haas, R. Verger, Chirality 15 (2003) 220–226. [8] G.H. Peters, S. Toxvaerd, N.B. Larsen, T. Bjørnholm, K. Schaumburg, K. Kjaer, Struct. Biol. 2 (1995) 395–401. [9] K.G. Blodgett, Films built by depositing successive monomolecular layers on a solid surface, J. Am. Chem. Soc. 57 (1935) 1007–1022. [10] Y.F. Dufrˆene, G.U. Lee, Biochim. Biophys. Acta 1509 (2000) 14–41. [11] M. Grandbois, H. Clausen-Schaumann, H. Gaub, Biophys. J. 74 (1998) 2398–2404. [12] M. Ivanova, A. Svedsen, R. Verger, R.I. Panaiotov, Colloids Surf. B: Biointerfaces 26 (2002) 301–314. [13] L.K. Nielsen, J. Risbo, T.H. Callisen, T. Bjørnholm, Biochim. Biophys. Acta 1420 (1999) 266–271. [14] K. Balashev, T.R. Jensen, K. Kjaer, T. Bjørnholm, Biochimie 83 (2001) 387–397. [15] K. Balashev, M. Gudman, L. Iversen, T.H. Callisen, A. Svedsen, T. Børnholm, Biochim. Biophys. Acta 1615 (2003) 93–102. [16] B. Sailas, P. Ashok, Yeast 14 (1998) 1069–1987. [17] J.M. Rodriguez Patino, C. Carrera Sanchez, Langmuir 20 (2004) 4530–4539. [18] C.H. Rao, S. Damodaran, Langmuir 18 (2002) 6294–6306. [19] C.I. Lopez-Amaya, A.G. Marangoni, Colloids Surf. B: Biointerfaces 32 (2003) 263–274. [20] Y. Cajal, A. Svendsen, J. De Bolos, S.A. Patkar, M.A. Alsina, Biochimie 82 (2000) 1053–1062.
Atomic force microscope studies on the interactions of Candida rugosa lipase and supported lipidic bilayers Nuria Prim a,b,∗ , Lars Iversen b , Pilar Diaz a , Thomas Bjørnholm b a
Department of Microbiology, Faculty of Biology, University of Barcelona, Av. Diagonal 645, 08028 Barcelona, Spain b NanoScience Center, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark Received 16 May 2006; accepted 19 May 2006 Available online 2 June 2006
Abstract Using the Langmuir–Blodgett technique we prepared substrate supported well-defined lipid/phospholipid (1-mono-palmitoyl-rac-glycerol (MPG)/l,2-dipalmitoyl-sn-glycerol-3-phosphocholine (DPPC)) bilayers in which the MPG lipid leaflet was exposed to the aqueous phase. Hydrolysis of MPG performed by Candida rugosa lipase (CRL) on the upper MPG layer of these supported bilayers on mica was imaged by real time atomic force microscope (AFM) using a liquid cell, so that the area increase of the initial structural defects could be followed over time. Our data strongly suggest that the edges of the initial structural defects are the preferred activation sites for CRL once the enzyme is adsorbed onto these interfaces. When a 2.5 nM bulk concentration of CRL was assayed on this planar lipid substrate, we found a long lag phase before a sharp increase of catalytic activity. The lag–burst kinetic behaviour was related to the interfacial activation phenomenon although we propose that it is also dependent on the gel-phase state of this interface. © 2006 Elsevier B.V. All rights reserved. Keywords: Lipase; Candida rugosa; Langmuir–Blodgett; Atomic force microscope (AFM)
1. Introduction Lipases (EC 3.1.1.3, triacylaglycerol hydrolases) cleave the ester bonds of long-chain lipids like triacylglycerides [1]. The catalytic activity of lipases on soluble substrates is very low, and it sharply increases when the substrate is aggregated forming a lipid/water interface, a phenomenon which is known as interfacial activation [2]. This process has been associated with a conformational change of a helical structure (lid) covering the catalytic centre of most lipase. When the enzyme is adsorbed to a lipidic interface, this lid opens making the hydrophobic active site accessible to the substrate [3]. Amphiphilic molecules, like certain lipids, can be assembled as monomolecular layers at the air–water interface [4–6]. For decades lipid monolayers have been used as a substrate for lipolytic enzymes, their main advantage being the possibility of controlling the quality of the interface [7]. The overall
∗
Corresponding author. Tel.: +34 934064675; fax: +34 934034629. E-mail address: [email protected] (N. Prim).
0927-7765/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2006.05.018
catalytic hydrolysis involves adsorption of the enzyme from the bulk aqueous phase to the surface, followed by catalytic action at the interfacial plane [8]. Additionally, these lipid monolayers can be transferred to a solid support by dipping, which allows also the formation of bilayers and multilayers [9]. In the last few years, atomic force microscopy (AFM) has become a key technique in biophysics and biochemistry for the characterization of supported lipid films, in part due to its unique capacity to probe their local surface structure in real time and in an aqueous environment [10]. AFM studies on the hydrolytic performance on supported bilayers prepared by the Langmuir–Blodgett technique have been so far reported for only two lipolytic enzymes in the literature, snake venom phospholipase A2 from Agkistrodon piscivorus piscivorus [11,12] and Humicola lanuginosa lipase [13–15]. The objective of the present work was to study the catalytic performance of Candida rugosa lipase on 1-mono-palmitoyl-rac-glycerol (MPG) by means of in situ AFM imaging. These studies were performed for the first time on hybrid supported DPPC/MPG bilayers, using CRL lipase.
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
2. Experimental 2.1. Lipids and enzyme l,2-Dipalmitoyl-sn-glycerol-3-phosphocholine (DPPC), 1mono-palmitoyl-rac-glycerol (MPG) and C. rugosa lipase (CRL) were purchased from Sigma. CRL was diluted to a concentration of 2.5 nM in 50 mM Tris–HCl buffer (pH 7.4) supplemented with 1.5 mM CaCl2 . 2.2. Bilayer preparation Hybrid DPPC/MPG bilayers were prepared by means of the Langmuir–Blodgett technique on a commercial LB trough (KSV 5000, KSV Ltd., Finland) and using freshly cleaved mica as a support. DPPC was dissolved in HPLC grade n-hexane to a concentration of 0.6 mg/ml, while MPG was dissolved in chloroform to a concentration of 1 mg/ml. To prepare the bottom layer, DPPC was spread with a Hamilton microsyringe on a pure MilliQ water surface. Ten minutes were allowed to elapse for solvent evaporation before compression was carried out until a target pressure of 35 mN/m. After monolayer stabilization for 30 min, vertical transfer was performed onto mica at a rate of 1 mm/min. After cleaning the trough, the MPG solution was spread on a pure Milli-Q water surface and compression was performed to the same target pressure as before. The MPG monolayer was stabilized for 2–3 h before horizontally transferring onto the first DPPC layer at a transfer arte of 0.5 mm/min. 2.3. AFM imaging The supported bilayers were transferred onto a liquid cell of the AFM (Nanoscope IIIa, Digital Instruments, Santa Barbara, CA) using a standard silicon O-ring in order to seal the chamber. The liquid cell was flushed in with 50 mM Tris–HCl buffer (pH 7.4) supplemented with 1.5 mM CaCl2 and the sample was allowed to thermally equilibrate before sucking in the enzyme solution prepared as described above. Scanning was performed using oxide sharpened silicon-nitride tips (NanoProbes, Santa Barbara, CA). Imaging was carried out in tapping mode at room temperature. 2.4. Analysis of images Image analysis to measure the area and perimeter of the growing bilayer defects was performed by using the public domain ImageJ 1.33u software (http://rsb.info.nih.gov/ij/). 3. Results and discussion Due to the special nature of lipases and their interfacial activation, during the kinetic studies of these enzymes it is extremely important to have well-defined substrates. Simple lipids like 1-mono-palmitoyl-rac-glycerol have only a weak amphiphilic behaviour. This was overcome by using a strong amphiphilic phospholipid (1,2-dipalmitoyl-sn-glycerol3-phosphocoline, DPPC) as a supporting first layer onto mica, as
139
it had been described before for other lipidic bilayers [15]. Onto this first hydrophobic supporting layer, a gel-phase monolayer of MPG was easily transferred. The resulting hybrid bilayer has a well-defined planar top layer of MPG in which the headgroups of the lipids are exposed to the aqueous solution containing the enzyme. C. rugosa lipase acts on the ester bonds of a broad range of substrates, mainly mono-, di- and triacylglycerides [16]. In the present work, these hybrid DPPC/MPG bilayers supported on mica have been proved to be also a good substrate for this enzyme, allowing in situ tapping mode AFM imaging of the hydrolysis process performed inside the liquid cell. All DPPC/MPG bilayers showed some initial structural defects (Fig. 1) with a height difference between the top part and the mica (dark domains) of 5.5 ± 0.3 nm (Fig. 1a, inset), a value that is in agreement with previous data on the monolayer thickness for both molecules [15,17]. In control experiments, these bilayers were imaged with the AFM over 2 h in order to check for stability, revealing no tip–sample interaction effects. The study of enzyme catalysis at lipid–water interfaces is very complex, mainly because of the difficulty of exactly determining the concentration of enzyme acting at the interface [18]. Lipase activity is preceded by the binding of the enzyme from the bulk phase to the interface, this being followed by its activation. It is established that in the order of nanomolar concentrations of enzymes, reaching an equilibrium value at the interface would need an elapse of some minutes [18]. Therefore, we chose a low concentration of CRL in order to perform reproducible analysis and understand this lipase behaviour at this interface. After thermal equilibration of the liquid cell, a 2.5 nM solution containing the enzyme was flushed into the cell, and hydrolytic desorption of the bilayer was imaged over time. As it was initially presumed that the enzyme would initiate hydrolysis of MPG at these structural defects of the bilayer, activity could thus be easily monitored by analysing the area increase of uncovered mica due to the cleaving of palmitic acid and subsequent fatty acid release from the MPG upper layer. When hydrolysis was followed for a long period of time (80–120 min), “new” holes appeared on some of these bilayers due to the presence of initial nanoscale structural defects that could not be easily observed in the usual 25 m2 AFM scan area. As the existence of these “new” growing holes was at very definite spots and the bilayer degradation was not randomly performed in the presence of CRL, these results confirmed the requirement of a structural defect for this enzyme to first attach to the interface and get into an active conformation, so that the hydrolysis process on the MPG could take place. A time-course set of images from one of these experiments is shown in Fig. 1. The first image (Fig. 1a) was taken prior to flushing a 2.5 nM solution of CRL into the liquid cell, while the rest of the images (Fig. 1b–f) show the changes produced on the bilayer by the enzymatic action. For quantification of the kinetic rates, all holes from the AFM images recorded along the first 80 min for each experiment were analysed in respect of the desorbed area increase (A), the initial perimeter length (P0 ) and the change in time (t). As described before by Balashev et al. [15], the initial perimeter length of the holes may
140
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
Fig. 1. Time set of AFM images showing the MPG hydrolysis when a 2.5 nM CRL solution was flushed into the liquid cell. The dark areas are the bilayer structural defects, which expand after adding the enzyme due to MPG hydrolysis and subsequently desorption of the DPPC bottom layer. (a) Prior to flushing the enzyme into the cell; (b) 12 min after flushing the enzyme into the cell; (c–f) selected images 44–82 min after flushing the enzyme. Inset in ‘a’ shows the height measure (white line) of the bilayer, corresponding to 5.5 nm depth.
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
Fig. 2. Desorbed lipid area corrected by the initial perimeter of the defects vs. time after flushing a 2.5 nM solution of enzyme into the liquid cell.
enter the rate equation as an indicative parameter for effective substrate concentration. Therefore, we refer to the initial rate of hydrolysis as v0 = A/t and the normalised rate of hydrolysis as A/t × P0 . Plots of the desorbed area/perimeter versus time (Fig. 2) revealed the existence of a long lag phase (38 ± 2 min) after flushing the enzyme, followed by a burst of hydrolytic activity until reaching a practically zero hydrolysis rate. Lipase hydrolytic performance generates water soluble fatty acids that may remain transiently at the interface [12]. Accumulation of hydrolysis products and DPPC may cause inhibition of CRL [12,19], which would explain the progressive decrease of the hydrolysis rates after the burst of activity and the big variation observed at higher times. Rao and Damodaran [18] showed that when the CRL reaction on dicaprin monolayers was allowed to proceed indefinitely in the presence of -cyclodextrin by removing the reaction products, the surface pressure decreased to zero because of complete hydrolysis of the layer. In our case, bilayers were not completely desorbed after imaging for 3–4 h, indicating that the aforementioned practically zero activity rate is reached because of the presence of the catalysis products and DPPC. Therefore, only initial rates (v0 ) for each hole, calculated from the two consecutive AFM images where the first sharp increase of activity was detected, were considered for the kinetic measurements. The average normalised initial rate A/t × P0 for CRL catalysis (2.5 nM) in all the experiments performed on these supported bilayers was found to be 11 × 10−5 ± 7 × 10−5 m/min. When we plotted the initial rate v0 versus the initial perimeter P0 of the holes (Fig. 3), the high standard deviation was translated into a big spread in the distribution of the different experiments, a fact previously reported also for low concentrations of H. lanuginosa lipase acting on similar systems [15]. For most of the experiments, we found initial rates of 6 × 10−5 ± 1.5 × 10−5 m2 /min, independently of the initial perimeter of the structural defect. Additionally, in a number of experiments, the initial rates clustered into three other groups, displaying higher v0 values of 12.3 × 10−5 ± 1.5 × 10−5 m2 / min, 18.3 × 10−5 ± 2 × 10−6 m2 /min and 23 × 10−5 ± 2.8 × 10−5 m2 /min, respectively. This scattering in initial
141
Fig. 3. Initial rates of hydrolysis (v0 ) corresponding to a single experiment plotted vs. the initial perimeter (P0 ). The patterns show the clustering observed in the rates, which would correspond to one (䊉), two (), three () and four () enzyme molecules.
rates can be explained by a stochastic distribution of single enzyme molecules in the bilayer defects because of the very low bulk concentration of enzyme. We hypothesize from these data that the lower initial rates might be the result of one single molecule of enzyme, while the higher rates could be due to two, three and four molecules, respectively. The presence of a lag phase is a very common phenomenon in the kinetics of lipases, having usually been related to the interfacial activation of these enzymes [3]. Lag–burst kinetics has been reported before for CRL acting on dicaprin monolayers using the zero order trough technique [18] and was also observed by means of AFM for snake venom phospholipase A2 (PLA2 ) on DPPC interfaces [13]. Curiously no lag phase was reported for H. lanuginosa lipase (HLL) on MOG/DPPC bilayers using AFM [15] despite being usually known to perform interfacial activation [20], most probably because of interference in the detection of a short lag phase by the elapse of time in the AFM imaging. The importance of the conformational and interfacial properties of lipids in the catalytic activity of lipases has been reported before [8,20]. In this sense, Rao and Damodaran [18] observed a decrease in the rate of CRL adsorption to monolayers with increasing surface pressure. We propose that differences in the absence/presence of a lag phase in the case of HLL [15], PLA2 [13] and CRL (this work) could not only be related to the interfacial phenomenon itself but also to the physical state of the different supported lipid bilayers analysed. Thus, the long lag phase observed for CRL on these bilayers may also be due to a highly structured gel-phase MPG layer, which would make it more difficult for the enzyme molecules to adsorb onto the interface and therefore get into an active conformation. 4. Conclusions C. rugosa lipase activity on MPG has been imaged in situ by means of AFM for the first time. Qualitative analysis showed that this enzyme needed to be adsorbed at the edge of the initial structural defects present at this interface in order to get into the
142
N. Prim et al. / Colloids and Surfaces B: Biointerfaces 52 (2006) 138–142
active conformation and initialize hydrolysis. We hypothesize that the long lag phase behaviour observed may be related not only to the interfacial activation phenomenon itself, but also the gel-phase state of the MPG layer. The burst of hydrolysis on this substrate after the lag phase was followed by a practically zero hydrolysis rate most probably due to enzyme inhibition by both the released fatty acids and DPPC constituting the bottom layer. Acknowledgements This work was partially financed by a nanotechnology grant from the Catalan Government Research Foundation (Dursi). The authors would like to thank Dimitrios Stamou, Martin Gudman, Tue Hassenkam and Pedro Nunes for fruitful discussions. References [1] [2] [3] [4] [5]
R. Gupta, N. Gupta, P. Rathi, Microbiol. Biotechnol. 64 (2004) 763–781. L. Sarda, P. Desnuelle, Biochim. Biophys. Acta 30 (1958) 513–521. L. Arpigny, K.E. Jaeger, Biochem. J. 343 (1999) 177–183. A. Pockels, Nature (London) 43 (1891) 437. I. Langmuir, J. Am. Chem. Soc. 39 (1917) 1848–1906.
[6] I. Langmuir, Trans. Faraday Soc. 15 (1920) 62. [7] I. Douchet, G. Haas, R. Verger, Chirality 15 (2003) 220–226. [8] G.H. Peters, S. Toxvaerd, N.B. Larsen, T. Bjørnholm, K. Schaumburg, K. Kjaer, Struct. Biol. 2 (1995) 395–401. [9] K.G. Blodgett, Films built by depositing successive monomolecular layers on a solid surface, J. Am. Chem. Soc. 57 (1935) 1007–1022. [10] Y.F. Dufrˆene, G.U. Lee, Biochim. Biophys. Acta 1509 (2000) 14–41. [11] M. Grandbois, H. Clausen-Schaumann, H. Gaub, Biophys. J. 74 (1998) 2398–2404. [12] M. Ivanova, A. Svedsen, R. Verger, R.I. Panaiotov, Colloids Surf. B: Biointerfaces 26 (2002) 301–314. [13] L.K. Nielsen, J. Risbo, T.H. Callisen, T. Bjørnholm, Biochim. Biophys. Acta 1420 (1999) 266–271. [14] K. Balashev, T.R. Jensen, K. Kjaer, T. Bjørnholm, Biochimie 83 (2001) 387–397. [15] K. Balashev, M. Gudman, L. Iversen, T.H. Callisen, A. Svedsen, T. Børnholm, Biochim. Biophys. Acta 1615 (2003) 93–102. [16] B. Sailas, P. Ashok, Yeast 14 (1998) 1069–1987. [17] J.M. Rodriguez Patino, C. Carrera Sanchez, Langmuir 20 (2004) 4530–4539. [18] C.H. Rao, S. Damodaran, Langmuir 18 (2002) 6294–6306. [19] C.I. Lopez-Amaya, A.G. Marangoni, Colloids Surf. B: Biointerfaces 32 (2003) 263–274. [20] Y. Cajal, A. Svendsen, J. De Bolos, S.A. Patkar, M.A. Alsina, Biochimie 82 (2000) 1053–1062.