Biocatalytic Langmuir–Blodgett assemblies based on penicillin G acylase

Biocatalytic Langmuir–Blodgett assemblies based on penicillin G acylase

Colloids and Surfaces B: Biointerfaces 23 (2002) 357– 363 Biocatalytic Langmuir –Blodgett assemblies based on penici...

179KB Sizes 3 Downloads 29 Views

Colloids and Surfaces B: Biointerfaces 23 (2002) 357– 363

Biocatalytic Langmuir –Blodgett assemblies based on penicillin G acylase L. Pastorino a, T.S. Berzina b,*, V.I. Troitsky b, M.P. Fontana b, E. Bernasconi c, C. Nicolini a a

Department of Biophysical M&O Sciences and Technologies, Uni6ersity of Genoa, Corso Europa 30, 16132 Genoa, Italy b Department of Physics and INFM, Uni6ersity of Parma, 6iale delle Scienze, 43100 Parma, Italy c Antibioticos, 6ia Winckelmann 1, 20146 Milan, Italy Received 29 August 2000; accepted 21 November 2000

Abstract The recently developed ‘protective plate’ method offers the possibility to include protein layers into a Langmuir– Blodgett (LB) assembly without contact of protein molecules with the air– water interface thus avoiding their denaturation. In the present work, this technique was applied for the deposition of biocatalysts with active layers of penicillin G acylase (PGA), an enzyme widely used for medicine production. Easy selection of LB and adsorbed layers resulted in the creation of appropriate environments for the preservation of PGA functions. Two structures were tested regarding such performances as the enzymatic activity value and the level of PGA detachment in aqueous solutions. It was shown that they satisfy the requirements for biocatalytic applications. The enzymatic activity of PGA monolayer incorporated into the film reached 25– 30% of the activity value of the equivalent amount of protein in the solution, which is a good result for an immobilized enzyme. Further modification of the deposition procedure resulted in increasing the effective activity per unit of the substrate surface due to adsorption of a thicker protein layer in one cycle. Probably, a three-dimensional frame-like structure was formed, which allowed the substrate molecules to penetrate into the film. The enzymatic activity of such films per unit of the substrate surface was 20 – 25 times higher than that of the assemblies with one adsorbed monolayer. Finally, the method is proposed of biocatalytic LB assembly deposition onto flexible supports of practically unlimited length without the exposure of protein layer to air medium. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Langmuir– Blodgett film; Adsorption; Biocatalysis; Immobilization; Penicillin G acylase

1. Introduction

* Corresponding author. Tel.: + 39-0521-905251; fax: + 390521-905223. E-mail address: [email protected] (T.S. Berzina).

Langmuir –Blodgett (LB) and related methods were used in numerous works to incorporate biologically active molecules into organized molecular films. Considerable efforts were applied to make sensing media for the developments of

0927-7765/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 6 5 ( 0 1 ) 0 0 2 5 3 - 3


L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363

biosensors and laboratory bioanalytic techniques [1 –3]. In reality, the goal of application of other methods such as layer-by-layer adsorption to biological objects [4,5] was the same. Although biocatalytic properties of LB assemblies prepared in different ways were studied [6– 8], to our knowledge no approach was proposed for the development of biocatalysts based on LB films suitable for practical applications. The difficulty is obvious. One of the requirements of industrial biocatalysis is the immobilization of enzymes onto the carriers of very large effective surface. On the other hand, if such an approach is proposed, the application of the LB technique and combinations of the latter with the other methods of deposition may offer new possibilities for the production of biocatalysts with the improved efficiency and stability. In the present study, we demonstrate how to produce highly active and structurally stable biocatalytic films with a previously proposed method [9] based on the combination of LB deposition and adsorption procedures and protection of protein from denaturation by a thin layer of water. Industrially important enzyme, penicillin G acylase (PGA), widely used for the production of medicines was selected for our experiments. FTIR spectroscopy was used for the determination of relative amount of adsorbed PGA. Data were compared with the results of enzymatic activity measurements. Then we propose the method of continuous LB assembly deposition onto a flexible support of practically unlimited length, which allows the incorporation of undenatured protein molecules into the film.

The studied structures (Section 3) were composed of adsorbed layers of PGA, glutaraldehyde, poly(diallyldimethylammonium) chloride (pDADMAC), and LB monolayers of octadecylamine and stearic acid, which were step by step deposited onto the substrate and adsorbed to realize required sequences. The PGA was adsorbed from the solution in 100 mM phosphate buffer with the concentration of 2.7 mg/ml at 4 °C and pH 7.5. Time of adsorption varied from 5 to 20 min. Adsorption of p-DADMAC was performed from 5% aqueous solution during 15 min at room temperature and pH 7.5. The concentration of glutaraldehyde in water (pH 6.0, room temperature) was in the range of 0.5– 10% and the adsorption time interval was equal to 5 –20 min. The monolayers of stearic acid and octadecylamine were spread from the solution in benzene (0.5 mg/ml) at the surface of pure water (pH of about 6.0), compressed up to the surface pressure of 32 and 30 mN/m, respectively, and transferred onto the substrate at a rate of 5 mm/min. The PGA activity was measured with 6-nitro-3(phenylacetamido)benzoic acid (NIPAB) method of assay [11], which uses NIPAB (0.25 mM solution at pH 7.5 and temperature of 37 °C) as substrate in the reaction of hydrolysis catalyzed by PGA. The product, 3-amino-6-nitrobenzoic acid, was monitored spectrophotometrically at 405 nm. Reference values on PGA activity in the solutions of different concentrations were obtained proceeding from evaluated amount of protein in closely packed monolayer (0.41 mg on silicon solid supports with the dimensions of 7× 15 mm2).

2. Experimental The deposition of biocatalytic films was performed with an LB apparatus described elsewhere [10]. Pure water was prepared by using a Milli-Q purification system. IR spectra were recorded with Jasco FT-IR spectrometer model 420 in a transmission mode. Morphology of the deposited films was observed with Carl Zeiss optical microscope Axiotech. Jasco spectrophotometer model 7800 was used for the determination of enzymatic activity.

3. Results and discussion

3.1. Film structure Films of two types were deposited and investigated (Fig. 1). Bottom layers composed of either octadecylamine and glutaraldehyde (a) or pDADMAC and glutaraldehyde (b) sublayers ensure high adhesion of PGA to the substrate, which is achieved due to interaction of aldehyde

L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363

groups with amino groups of the protein molecules. This is important to avoid detachment of protein during reaction implementation as well as for the purpose of multiple utilization of biocatalyst. The use of p-DADMAC in the bottom layer can also result in a preferential orientation of the PGA molecules and in an increased efficiency of adsorption because this polymer is positively charged at pH 7.5. Further results show that the stable structure was surprisingly obtained also in the case of a bottom layer with the octadecylamine monolayer. This monolayer is bound to the hydrophobic surface of silicon solid support by weak Van der Waals forces and poor adhesion was initially expected. However, probably due to the interaction through glutaraldehyde molecules the strong coating cross-linked in lateral directions is formed. On the one hand, separate molecules of octadecylamine cannot be detached in this case from the solid support. On the other hand, water cannot penetrate between highly hydrophobic surfaces of silicon and octadecylamine monolayer to detach macroscopic pieces of the film.

Fig. 1. Structure of biocatalytic LB assemblies (1 – octadecylamine monolayer, 2 – glutaraldehyde adsorbed layer, 3 – PGA adsorbed layer, 4 – stearic acid monolayer, 5 – p-DADMAC adsorbed layer).


The monolayer of stearic acid was deposited over the adsorbed layer of PGA for the purpose of protection of the latter. Functional characteristics of the enzyme are better preserved in this case during storage because the protein layer is in direct contact with the hydrophilic surface. The protective monolayer of stearic acid is completely removed from the surface when the sample is immersed in some aqueous solution. This was confirmed for similar films by AFM study [12]. Thus, an effective biocatalytic activity of the film does not change.

3.2. Enzymatic acti6ity One result of the present work important for biocatalysis is the development of deposition procedure, which considerably increases the enzymatic activity referred to the unit of the substrate surface. In the case of structure (a) in Fig. 1, initially used procedure can be described as follows. (For more details on the deposition technique see Refs. [9,10]). A hydrophobic substrate is dipped down to deposit the monolayer of octadecylamine. The surface is protected with a hydrophilic plate to hold thin layer of water between the two surfaces. The protected sample is transferred first into the compartment with glutaraldehyde solution for adsorption and then into the compartment with water to wash out physically adsorbed glutaraldehyde. In the same way, the PGA active layer is adsorbed. Finally, the protected substrate with the adsorbed protein is immersed into the compartment with a stearic acid monolayer, the protective plate is removed, and the monolayer is deposited over the protein layer. Glutaraldehyde was adsorbed during 15 min from 5% aqueous solution. Duration of PGA adsorption was equal to 10 min. An average enzymatic activity for the films deposited in this way was equal to 0.00173 mmol/min (substrate of 7× 15 mm2) while the detachment of protein molecules during measurements was practically absent. Substitution of the octadecylamine monolayer by the adsorbed layer of p-DADMAC gave the enzymatic activity value of 0.00186 mmol/min. It should be noted that further successive adsorption of PGA over the same samples through the inter-


L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363

mediate layers of glutaraldehyde does not result in the change of activity values within the experimental errors caused by insufficient reproducibility. Thus, we can suppose that only the film surface is responsible for the biocatalytic process and reported above values can be attributed to the activity of approximately one closely packed monolayer. Comparison with the results obtained for an equivalent amount of the enzyme in the solution shows that at least 25–30% of native activity was preserved. This is a rather high value for immobilized enzyme, which cannot be achieved if the ordinary LB and LS techniques of protein deposition are applied. However, the described above deposition procedure can be modified to further increase the biocatalyst efficiency. As it was mentioned, simple addition of enzyme adsorbed layers onto the sample surface does not increase the activity. In one attempt, we have excluded the procedure of sample washing after glutaraldehyde adsorption. As the result, an enzymatic activity of 0.0467 mmol/ min was measured (the sample with the p-DADMAC sublayer). The protein detachment as before was negligible and the activity was not changed after several repeated measurements during a few days. Thus, the enzymatic activity per unit of the sample surface was 25 times increased in comparison with that of one closely packed monolayer. This behaviour can be probably explained by formation of a bulk frame of the PGA molecules cross-linked by glutaraldehyde (instead of protein layer 3 in Fig. 1(b)). In this structure, however, the PGA molecules should not be closely packed as before and small substrate molecules of NIPAB and water can penetrate into the film. We have also varied conditions of film preparation to understand our possibilities for the development of the process of continuous film deposition (Section 3.3). A typical morphology of the deposited film is shown in Fig. 2. IR spectra in the region of amid bands (Fig. 3) were recorded and values of film activity were measured. Intensities of CO stretching bands at 1650– 1660 cm − 1 were measured to compare samples with each other. Five sets of data (Table 1) were obtained, which gave the dependences of absorbance at

Fig. 2. Microphotographs of film surface: (a) perfect deposition and (b) defect in the film (Series 1 in Table 1, CGA =5%).

1650–1660 cm − 1 (Abs) and normalized activity (Arel) on variable parameters shown in Tables 2 and 3. A normalized activity Arel is defined as a ratio of the activity value of the sample with the dimensions of 7× 15 mm2 to that of the solution of equivalent amount of protein in one closely packed monolayer on the same surface (Section 2). This parameter should be introduced because the experiments are time consumable and absolute data measured during long period may differ considerably. The Abs value was accepted to be proportional to the amount of adsorbed protein. One can see that the amount of adsorbed PGA qualitatively corresponds to the film activity. However, a quantitative analysis should be done with definite precautions. The dependence obtained within one selected series is rather reliable, i.e. the activity variations are caused by the variations of controlled variable parameters, but not by occasional reasons. However, from the comparison of data between various sets wrong con-

L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363


Table 2 Dependence of absorbance at 1650–1660 cm−1 (Abs) and normalized activity (Arel) on the concentration of glutaraldehyde solution (CGA) CGA (%) Series 1

0.5 1 2 5 10

Fig. 3. Transmission IR spectra of films deposited at CGA of (1) 0.5%, (2) 2%, and (3) 10% (Series 4 in Table 1).

clusions can be made. The problem is that the Arel values measured for the films prepared in different successive experiments and with different portions of the PGA solution differ from each other because numerous uncontrolled factors usually result in partial inhibition of the enzymatic activity (at least, in the conditions of our experiments). For example, in the present work, we have obtained the highest Arel value for the Series 3. However, in the next attempt they can appear to be some lower for the same amount of the ad-

Series 4

Series 5







0.0065 0.0063 0.0064 0.0084 0.009

1.24 1.14 2.29 3.34 3.17

0.0056 0.0057 0.0064 0.0065 0.0068

0.9 2.18 2.78 2.51 2.47

0.0007 0.0013 0.0013 0.0015 0.0017

0.1 0.19 0.24 0.22 0.27

sorbed protein while their relative variations will be the same when the adsorption time, which is a variable parameter in this case, is changed. For thick films prepared without glutaraldehyde washing out both the Arel and Abs values increase with the variation of glutaraldehyde concentration from 0.5 to 10% (Series 1 and 4) and its adsorption time from 5 to 20 min (Series 2), but the enzymatic activity is not linearly proportional to the amount of adsorbed PGA (i.e. the dependence of Arel on Abs is non-linear). Namely, the increase of protein mass at high glutaraldehyde concentrations and long adsorption time intervals results in higher activity enhancement than it is expected from the linear dependence. This can be explained by the increase of adsorbed PGA layer density when the amount of cross-linker, i.e. glutaraldehyde, is not enough to form the optimal frame-like structure, which is well permiable for the substrate molecules. Thus, as in the case of successive adsorption of the PGA over the glutaraldehyde layer discussed above, penetration of

Table 1 Conditions of film preparation Series number


tGA (min)

tPGA (min)

CGA (%)

GA washing

1 2 3 4 5


15 Varied 15 15 15

10 10 Varied 10 10

Varied 5 5 Varied Varied

No No No No Yes

‘Sublayer’ means the bottom layer in the structures shown in Fig. 1; tGA and tPGA are the time intervals of glutaraldehyde and PGA adsorption, respectively; CGA is the concentration of glutaraldehyde solution.

L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363


the NIPAB and water molecules into the film as well as removal of the reaction product may become difficult. Anyway, these measurements show that the best results are obtained when CGA and tGA are in the range of 3– 6% and 12– 15 min, respectively. At lower values the biocatalyst efficiency is decreased while the CGA and tGA increase results in protein coagulation caused by an excess amount of cross-linker transferred in the PGA solution. This is not suitable from practical point of view in spite of the satisfactory activity. Unexpectedly, practically no dependence of the Abs and Arel values on PGA adsorption time was observed (Series 3). In some experiments, the PGA was adsorbed during up to 60 min and the results were the same. (We did not investigate the adsorption process at smaller durations because that is difficult to do precisely with our apparatus.) Regarding the instrument design for the biocatalytic film deposition, we conclude that this parameter does not put really any limitations. A suitable value can be chosen within a wide range from 5 to 60 min. To design properly the compartment for protein adsorption (Section 3.3) the other processes such as depletion and coagulation of the solution should be mainly taken into account. We have also measured an enzymatic activity and IR spectra of super thin films (Series 5) with small amounts of the adsorbed protein, as it was supposed of about one monolayer, to evaluate the PGA mass in thick films. In this case, the excess of glutaraldehyde was washed out after adsorption. However, quantitative agreement between these data cannot be achieved under the supposition of one monolayer formation. Either the mass of adsorbed closely packed film exceeds that of

one monolayer or the enzymatic activity of one monolayer is lower for some unknown reasons. Unfortunately, both these models are not reliable enough because the Abs values are measured in this particular case with low accuracy. (They are very small for such thin films while the noise in the region of amid bands is in principle high.)

3.3. Method of biocatalyst deposition onto polymeric tape There is no chance to apply directly the ‘protective plate’ technique [9] used in the present study for biocatalysis. A maximum area of the solid support is determined by that of the protective plate. Transfer of the sample from one compartment to another without protection of the enzyme layer with water will cause denaturation of the enzyme molecules. For this reason, we propose the modification of ‘protective plate’ technique to allow continuous deposition of the LB assemblies with enclosed adsorbed layers of soluble proteins onto flexible supports of unlimited length (e.g. polymeric tape) by transferring the latter between the compartments without exposure of protein molecules to the air medium. The idea of this method is shown in Fig. 4. Movable protective plate, which can close and open the sample surface when it is necessary, is substituted here by a fixed slit formed by two hydrophilic surfaces. The slit is located over the levels of aqueous solutions in adjacent compartments. Due to such a design spontaneous exchange of the solutions between the compartments is practically excluded, but the slit is capable to hold water inside by capillary forces. Preliminary tests of this stage of the ap-

Table 3 Dependence of absorbance at 1650–1660 cm−1 (Abs) and normalized activity (Arel) on the duration of glutaraldehyde (tGA) and PGA (tPGA) adsorption Series 2, tGA (min)

Abs Arel

Series 3, tPGA (min)








0.006 1.04

0.0093 3.18

0.01 4.07

0.012 4.31

0.015 4.26

0.014 4.4

0.019 5.12

L. Pastorino et al. / Colloids and Surfaces B: Biointerfaces 23 (2002) 357–363


4. Conclusions

Fig. 4. Method of film deposition onto flexible support of unlimited length (1 – polymeric tape, 2 – compartment for LB deposition, 3 – compartment for adsorption, 4 and 5 – top and bottom hydrophilic surfaces of the slit, respectively, 6 – level of water in the compartments).

paratus show that also the amount of the solution transferred with the moving tape can be minimized up to an acceptable level. The instrument may consist of any desired number of ordinary compartments for the deposition of LB monolayers and adsorption. These compartments are installed along the line in a required sequence, which determines the sequence of layer alternation for a particular task. By a simple change of a number of compartments and their combination a layered structure of the prepared film can be changed. During the film deposition a polymeric tape can pass either through the slit filled by the solution or through the air medium, if necessary. The shafts shown in Fig. 4, which support the polymeric tape, have the profiles with small projections to avoid the active film damage at the major part of the support surface. The speed of film deposition is mainly limited by the LB process. It is equal to 4– 5 cm/min in the best case. Nevertheless, during a working day (8 h) the film can be deposited onto the tape 20– 25 m long. Proceeding from this value as well as from the parameters reported in Section 3.2 the required dimensions of the compartments for adsorption can be easily evaluated. The calculations show that they are all of reasonable values. After the deposition, a compact roll with small gaps between the rounds can be made of this polymeric tape with the immobilized biocatalyst to be used in bioreactor.

The LB assemblies containing incorporated enzyme molecules and promising for biocatalysis are deposited in the present work. Due to the possibility of the LB technique and related methods to manipulate with the monolayers and adsorbed layers of different compounds high activity and stability of biocatalysts are achieved. A new method of film deposition is proposed, which can open the way for practical applications of the developed biocatalytic media.

Acknowledgements We gratefully acknowledge financial support from ‘Consilio Nazionale delle Ricerche’ of Italy within the MURST-CNR Biotechnology Program L.95/95.

References [1] M. Aizawa, K. Owaki, M. Matsuzawa, H. Shinahara, Y. Ikarigama, Thin Solid Films 180 (1989) 227. [2] S.Yu. Zaitsev, Sensors Actuators B 24 – 25 (1995) 177. [3] C.A. Nicolae, S. Cantin-Rivie`re, E. El Abed, P. Peretti, Langmuir 13 (1997) 5507. [4] Yu. Lvov, K. Ariga, I. Ichinose, T. Kunitake, J. Am. Chem. Soc. 117 (1995) 6117. [5] Yu. Lvov, K. Ariga, I. Ichinose, T. Kunitake, Thin Solid Films 284 – 285 (1996) 797. [6] S. Paddeu, V. Erokhin, C. Nicolini, Thin Solid Films 284 – 285 (1996) 854. [7] A. Petrigliano, A. Tronin, C. Nicolini, Thin Solid Films 284 – 285 (1996) 752. [8] T.S. Berzina, L. Piras, V.I. Troitsky, Thin Solid Films 327 – 329 (1998) 621. [9] V.I. Troitsky, T.S. Berzina, A. Petrigliano, C. Nicolini, Thin Solid Films 284 – 285 (1996) 122. [10] V.I. Troitsky, M. Sartore, T.S. Berzina, D. Nardelli, C. Nicolini, Rev. Sci. Instrum. 67 (1996) 4216. [11] C. Kutzbach, E. Rauenbusch, Hoppe Seyler’s Z. Physiol. Chem. 354 (1974) 45. [12] T.S. Berzina, V.I. Troitsky, A. Petrigliano, D. Alliata, A.Yu. Tronin, C. Nicolini, Thin Solid Films 284 – 285 (1996) 757.