- Email: [email protected]
Rhodanese incorporated in Langmuir and Langmuir–Blodgett films of dimyristoylphosphatidic acid: Physical chemical properties and improvement of the enzyme activity
Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Rhodanese incorporated in Langmuir and Langmuir–Blodgett films of dimyristoylphosphatidic acid: Physical chemical properties and improvement of the enzyme activity Felipe Tejada de Araújo, Luciano Caseli ∗ Institute of Environmental, Chemical and Pharmaceutical Sciences, Federal University of São Paulo, Diadema, SP, Brazil
a r t i c l e
i n f o
Article history: Received 14 December 2015 Received in revised form 15 January 2016 Accepted 19 January 2016 Available online 22 January 2016 Keywords: Rhodanese Air–water interface Langmuir monolayers Langmuir–Blodgett Enzyme activity Cyanide
a b s t r a c t Preserving the catalytic activity of enzymes immobilized in bioelectronics devices is essential for optimal performance in biosensors. Therefore, ultrathin films in which the architecture can be controlled at the molecular level are of interest. In this work, the enzyme rhodanese was adsorbed onto Langmuir monolayers of the phospholipid dimyristoylphosphatidic acid and characterized by surface pressurearea isotherms, polarization-modulated infrared reflection-absorption spectroscopy (PM-IRRAS), and Brewster angle microscopy (BAM). The incorporation of the enzyme (5% in mol) in the lipid monolayer expanded the film, providing small surface domains, as visualized by BAM. Also, amide bands could be identified in the PM-IRRAS spectra, confirming the presence of the enzyme at the air–water interface. Structuring of the enzyme into ␣-helices was identified in the mixed monolayer and was preserved when the film was transferred from the liquid interface to solids supports as Langmuir–Blodgett (LB) films. The enzyme-lipid LB films were then characterized by fluorescence spectroscopy, PM-IRRAS, and atomic force microscopy. Measurements of the catalytic activity towards cyanide showed that the enzyme accommodated in the LB films preserved more than 87% of the enzyme activity in relation to the homogeneous medium. After 1 month, the enzyme in the LB film maintained 85% of the activity in contrast to the homogeneous medium, which 24% of the enzyme activity was kept. The method presented in this work not only points to an enhanced catalytic activity toward cyanide, but also may explain why certain film architectures exhibit an improved performance. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The immobilization of enzymes on solid supports as Langmuir–Blodgett (LB) films is a recent approach to obtain devices whose architectures can be controlled at the molecular level. For that, monomolecular films are first formed at the air–water interface and transferred to solid supports during their vertical passage through the floating monolayer. Although pure enzymes can be spread on the air–water interface [1,2], their incorporation in pre-formed phospholipid monolayers may preserve the conformation of the enzyme [3–5]. As a result, a well-ordered nanostructured system can be constructed and the enzyme may serve as a recognizing element. If an optical or electrical signal is generated due to a recognizing chemical reaction, the film is a potential candidate for a biosensor.
∗ Corresponding author. E-mail address: [email protected] (L. Caseli). http://dx.doi.org/10.1016/j.colsurfb.2016.01.037 0927-7765/© 2016 Elsevier B.V. All rights reserved.
Particularly, rhodanese is a mitochondrial enzyme that detoxifies cyanide, converting it into thiocyanate [6]. This reaction takes place in two steps: in the first step, thiosulfate reacts with the thiol group to form a disulfide. In the second step, the disulfide reacts with cyanide to produce thiocyanate, and is converted again to thiol. This reaction is environmentally important because it leads to a reduced exposure of cyanide since the less toxic thiocyanate is formed. The use of a thiosulfate solution as an antidote for cyanide poisoning is based on the activation of the enzymatic cycle. Thus, the immobilization of rhodanese in ultrathin films is interesting since it can provide a device capable of remedying the amount of cyanide in a controlled manner. LB films seem to be a suitable and innovative strategy for the construction of nanostructured films since monomolecular films can be deposited on solid supports with control over the chemical composition and surface density. This approach has already been performed for some enzymes [5,7,8], but not found in the literature for rhodanese. In this paper, emphasis on the interaction of the enzyme with the phospholipid dimyristoylphosphatidic acid (DMPA) was placed. As
60
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
we want to consider further the ability of such architecture to recognize cyanide, the enzymatic activity was investigated for mixed lipid-enzyme films supported as LB films.
2. Materials and methods DMPA was purchased from Sigma–Aldrich (purity higher than 99%) and dissolved in chloroform (Sytnh) to a concentration of 0.5 mg/mL. This lipid was chosen considering the ease to transfer it to solid supports as shown in previous studies [7,8]. Water used in all experiments was purified using a Milli-Q® system (resistivity of 18.2 cm−1 , pH 6.0). Rhodanese from Sigma–Aldrich (purity higher than 99% obtained from bovine liver) was dissolved in an aqueous buffer solution of K2 HPO4 (Sigma–Aldrich) and KH2 PO4 (Sigma–Aldrich) with a salt concentration of 0.01 mol/L and pH 7.0. The final concentration of the enzyme solution was 1 mg/mL. A Langmuir trough (KSV Instruments, Helsinki—Finland, model: Mini, 36.5 × 7.5 cm) was employed for the preparation of the Langmuir and LB films. It was filled with water and then 45–65 L of the DMPA solution was spread on the air–water interface to obtain an area per molecule of ∼90–120 Å2 . After the evaporation of chloroform for 15–20 min, the interface was compressed with two movable barriers at a rate of 10 cm min−1 (5 Å2 molecule−1 s−1 ), and the surface pressure values were monitored with a Wilhelmy plate made of filter paper that intercepted the air–water interface. For mixed enzyme-lipid monolayers, pre-determined aliquots of the enzyme solution (16–80 L) were carefully injected below a pre-formed DMPA monolayer into the aqueous subphase. After allowing the surface pressure to stabilize during 30 min., the interface was compressed and the surface pressure monitored. Measurements through polarization-modulation infrared reflection absorption spectroscopy (PM-IRRAS) were done with a KSV PMI 550 instrument (KSV Instruments Ltd., Helsinki—Finland). The monolayers were compressed to reach the surface pressure of 30 mN/m, and this value was kept constant by using mobile barriers while the spectra were obtained. The incidence angle to the normal was 80◦ and a minimum of 600 scans were obtained for each spectrum. The incoming light was continuously modulated between the p and s polarization, allowing simultaneous measurements of the spectra for both polarizations. The difference between the two absorbance signals gives surface-specific information and the sum provides the reference spectrum. Since the spectra are measured simultaneously, the effect of isotropic vibrations (water vapor and carbon dioxide) is largely reduced. Brewster angle microscopy (KSV-Nima Instruments, model: micro BAM3) was employed in order to obtain images of the monolayer at desired values of surface pressure. Solid glass supports were inserted in the aqueous subphase of the Langmuir films. The films were then spread on the air–water interface and compressed to reach the surface pressure of 30 mN/m. The supports were vertically withdrawn across the air–water interface with a speed of 5 mm min−1 keeping the surface pressure of 30 mN/m constant during the passage of the support. Transfer ratio values of about 0.95–1.05 were required for further analysis. LB films with 1 single layer were characterized by PMIRRAS and fluorescence spectroscopy (Spectrophotometer model RF-5301PC, Shimadzu) with an excitation wavelength of 285 nm with the glass directly placed in the fluorimeter holder. Atomic force microscopy (AFM) was also employed for further characterization, and the images were obtained in the tapping mode, employing a resonance frequency of approximately 300 kHz, a scan rate of 1.0 Hz, and scanned areas of 5.0 × 5.0 m on films deposited on mica. Nanogravimmetry through a quartz crystal microbalance (SRS—Stanford Research Systems model QCM200) was employed to estimate the mass of film deposited on a surface bounded by
Fig. 1. Surface pressure-area isotherms for DMPA and rhodanese (relative proportions in mol of the enzyme injected in the subphase are indicated in the inset).
gold electrodes in the thin disk of quartz as substrate. The mass of the deposited film was determined according to the Sauerbrey equation [9]. The catalytic activity of the enzyme was estimated according to a method previously described in the literature [10]. For this purpose, the LB film was inserted in a solution containing cyanide, and the enzyme activity measured due to the evolution of the optical density of the solution measured at 470 nm (UV–vis Hitashi, model U2001). The enzymatic activity was also measured in a homogeneous environment (enzyme dissolved in aqueous solution) for comparison. All enzymatic activities were measured again after 30 days with the sample kept at low temperatures (5–10 ◦ C). All experiments were performed at a temperature of 25 ± 1 ◦ C. 3. Results and discussion 3.1. Langmuir monolayers Firstly, it is important to emphasize that DMPA was chosen because of its easiness to be transferred to solid supports as a LB film and also because this lipid is reported as a suitable matrix for immobilizing enzymes [7,8]. DMPA monolayers present a typical surface pressure-area isotherm (Fig. 1), in agreement with the literature [11]. Inserting 1% in mol of rhodanese, a shift of the curve to smaller lipid areas is observed as a consequence of the condensation of the monolayer. With 5% in mol of the enzyme, the isotherm is shifted to higher areas, which may be a consequence of the incorporation of the enzyme inside the lipid chains, causing the lipid monolayer expansion. With higher concentrations of rhodanese, the isotherm is no longer shifted to higher areas, indicating saturation. This saturation effect can be attributed to a probable enzyme aggregation in the aqueous subphase as previously reported for other macromolecules adsorbing on lipid monolayers [12,13]. As the isoelectric point of this enzyme varies between 5.6 and 7.15 [14], rhodanese is zwitterionic under the experimental conditions (pH around 6.0). Being DMPA negatively charged at this pH, dipole-ion attractions between the lipid and the enzyme must influence the adsorption of the biomacromolecule and the subsequent monolayer expansion. Rheological properties of the monolayer can be accessed by means of its compressional modulus (E), which is defined as −A(∂/∂A)T [15], where A is the molecular area, is the surface pressure of the monolayer, and T is the temperature. The highest values of E for DMPA were of ca. 300 mN/m, featuring the liquidcondensed state (Fig. 2A and B). DMPA mixed with 1% of rhodanese presents higher values of E (550 mN/m) as a consequence of the condensation of the monolayer caused by small amounts of the enzyme. With 5% of rhodanese, the maximum values of E are similar to those obtained for pure DMPA. It is probable that the penetration of the enzyme into the lipid monolayer makes the film more
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
A
A
B
B
Fig. 2. Compressional modulus-area isotherms for DMPA and rhodanese (relative proportions in mol indicated in the inset).
rigid. This kind of behavior is reported in the literature for other macromolecules interacting with lipid Langmuir monolayers and the effect is reported as a “piston-like effect” [16], where the protein insertion forces the monolayer compression, providing a more packed interfacial structure. Also, Fig. 2B shows that between 30 and 40 mN/m the maximum variation of surface elasticity takes place, which is the range that corresponds approximately to the lateral pressure of cell membranes [17]. PM-IRRAS spectra for DMPA and mixed rhodanese-DMPA monolayers are shown in Fig. 3. It is important to mention that these spectra were obtained at the surface of 30 mN/m because of the parameters obtained with the surface elasticity measurements, and because we intended to prepare these films in bio-inspired conditions. Moreover, this surface pressure was the value employed for the transfer of the floating monolayers to solid supports as LB films, as shown later in this paper. It is important to highlight that since the bands are positive to the baseline of the spectra, the vibration moments lie parallel to the air/water interface [18]. The bands centered at 2918 and 2850 cm−1 correspond to CH2 antisymmetric and symmetric stretching modes, respectively. The band centered at 2956 cm−1 corresponds to the CH3 stretching mode. With the insertion of the enzyme, a slight shift is observed from 2918 to 2921 cm−1 and from 2850 to 2846 cm−1 . These changes are not relevant considering the resolution of the instrument (8 cm−1 ). Also the band at 2956 cm−1 becomes a shoulder upon the incorporation of the enzyme. For the mixed enzyme-lipid monolayer, a significant change is observed for the ratio between the maximum intensity of the antisymmetric and symmetric bands, which varies from 1.6 for pure DMPA to 2.4 for the mixed monolayer. This ratio is usually attributed to the order parameter of the film [19], and indicates that the mixed monolayer is in a higher ordered structure when compared to that for the pure lipid monolayer. This fact could be attributed to the incorporation of the enzyme into the lipid matrix, which provides a film with a close packed structure.
61
Fig. 3. PM-IRRAS spectra for DMPA (pure or mixed with 5% of rhodanese as indicated in the inset) monolayers at 30 mN/m.
Panel B for Fig. 3 shows some bands that are attributed to rhodanese, indicating the adsorption of the enzyme. Amide II (C N and N H vibrations) bands are observed in the wavelength range between 1500 and 1600 cm−1 , while amide I (C O stretching modes) bands are observed in the range between 1600 and 1700 cm−1 . The band centered at 1743 cm−1 may be attributed to the C O stretching mode of DMPA. The band centered at 1676 cm−1 can be attributed to the secondary structure of the enzyme polypeptide moiety without any defined order, and the band centered at 1560 cm−1 is attributed to amide II vibrations. The bands centered at 1464 cm−1 is attributed to C H bends, and increases their relativity intensity with the incorporation of the enzyme. The bands centered at 1269 and 1112 cm−1 are attributed to P O stretches in the phosphate group of DMPA. The relative intensity of these bands decreased significantly for the enzyme-DMPA mixed monolayer, suggesting that the enzyme interacts with the polar head-groups of the lipid. In this region of the spectra, it is important to emphasize that an amide I/amide II intensity ratio higher than 1.0 indicates the non-denaturated form of a protein [20]. Since this ratio is 1.9 in the spectra shown in Fig. 3B, it is probable that the incorporation of rhodanese in the lipid monolayer must prevent the enzyme from denaturation by maintaining at least part of its molecular conformation. Fig. 4 shows the BAM imagens for the monolayers with and without the enzyme. As expected for a pure DMPA monolayer, a homogenous pattern is observed at 30 mN/m, regarding that there is no phase segregation, and a single liquid condensed phase is noted. For the mixed monolayer, however, some small spots of bright domains are observed along the surface. Other higher bright domains can be observed as a consequence of aggregates induced by the presence of rhodanese. This fact indicates that the presence of the enzyme slightly changes the morphology of the lipid film. 3.2. LB films For the building of LB films, a preliminary test was carried out to check if the enzyme molecules present in the bulk of the
62
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
Fig. 4. BAM images (3600 × 4000 m) for DMPA (pure or mixed with 5% of rhodanese as indicated in the inset) monolayers at 30 mN/m.
A
Fig. 6. Fluorescence spectra for rhodanese-DMPA LB film.
B
Fig. 5. PM-IRRAS spectra for rhodanese-DMPA LB film.
aqueous subphase adsorbs on the immersed solid support. For that, the solid support was immersed in a solution containing the enzyme, but with no lipid at the air–water interface. After 60 min, which corresponds to the time of permanence of the solid support in the aqueous subphase during the LB deposition, the plate was removed, and measurements of fluorescence, PM-IRRAS and catalytic activity were carried out. No fluorescence signal was observed, neither amide bands in the vibrational spectra. Also, no catalytic activity was detected, which means that no significant amount of enzyme is adsorbed from the bulk of the aqueous subphase onto the solid support. Therefore, the adsorption of the enzyme on the solid support must be attributed exclusively to the passage of the solid support through the mixed enzyme-lipid monolayer. The enzyme-lipid films were then transferred to solid supports by using the LB technique with a relative amount of enzyme of 5% in the system. For all depositions, a transfer ratio close to 1.0 was obtained for all passages of the substrate. The co-incorporation of rhodanese was then investigated with PM-IRRAS (Fig. 5).
Antisymmetric and symmetric bands for CH2 are centered at 2925 and 2850 cm−1 , respectively. A band at 2890 cm−1 , attributed to CH3 stretching modes, is noticeable. The ratio between the maxima for the antisymmetric and the symmetric bands is 1.7 for the pure lipid, and 2.3 for the mixed film. These values are slightly different from the relative ratio values observed for the floating films, which indicates some alteration in the packing degree of the lipids when transferred to the solid support. Normally, the deposited layers are affected by the underlying solid support structure and by the loss of water molecules during the deposition procedure. Amide I and II bands are also present (Panel B for Fig. 4), confirming the transfer of the enzyme to the solid support, denoting thinner bands when compared to the spectra obtained for the air–water interface. The amide I band for ␣-helices is present at 1659 cm−1 and a small band for -sheets is observed at 1620 cm−1 . These facts indicate some conformational changes of rhodanese during the transfer, and may be also related to the loss of water molecules during the procedure. All the shifts of the maxima of the bands in relation to the monolayer on the air–water interface reflect the interaction of the film components with the solid substrate. Fig. 6 shows the fluorescence spectrum for the mixed LB films and confirms the presence of the enzyme in the structure since DMPA should not present fluorescence. The tryptophan group, present in some amino-acid residues of rhodanese, was excited at 285 nm, and fluorescence emission was observed with a maximum at 331 nm. Other studies with mixed enzyme-lipid LB films also confirmed the presence of the enzymes by using fluorescence spectroscopy [6,21]. The morphology of the mixed enzyme-lipid LB films can be accessed by AFM images (Fig. 7). For pure DMPA LB films, a flat pattern with no distinction of aggregates is always observed [22]. With rhodanese, irregularities, such as ridges, protuberances and holes are observed, being a consequence of the enzyme
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
63
Fig. 7. AFM images of rhodanese-DMPA LB film transferred at 30 mN/m.
incorporation. Aggregates with diameter varying from 50 to 200 nm are observed, but maximum highness of the aggregates is 3–3.5 nm. Also the RMS (root mean square roughness) increased from 0.3 nm, for the pure DMPA LB film, to 3.2, for the enzyme-lipid LB film, which indicates the increase of roughness of the film with the introduction of rhodanese. This pattern of morphology can be related to the accommodation of the enzyme in the lipid matrix, revealing that the co-transfer of the macromolecule affects the level of organization of DMPA transferred as a LB film. On the other hand, this film is relatively homogeneous when compared to other AFM images for protein-lipid LB films [22]. Table 1 shows the relative enzyme activity for the LB films, indicating that they are sensitive to the presence of cyanide, presenting enzyme activity. This is a preliminary indication for the possible use of such films as a cyanide sensor. The enzyme activity for LB films is 87% compared to the activity obtained for the homogeneous medium. These lower values, as well as the longer lag time, are common for enzymes immobilized on solid supports, as frequently reported in the literature [5,23]. These facts can be attributed to restrictions of molecular mobility for the polypeptide structure. The literature shows, for instance, that alkaline phosphatase co-immobilized with DMPA LB films does not present enzyme activity higher than 12% in relation to the homogeneous environment, with the longer lag time attributed to diffusive processes. Consequently, the catalytic substrate (cyanide) may present some restrictions to access the catalytic site of the enzyme . Also, a supposed higher amount of the enzyme in the homo-
geneous medium could boost the enzyme activity. However the value of 87% obtained in this paper is higher when compared to other enzymes immobilized in LB films [5,7,8,23]. It is important to emphasize that understanding the reasons by which the enzyme activity is conserved in lipid matrices is not straightforward. Although it is expected a decrease of the enzyme activity in relation to the homogeneous environment after immobilization in solid supports, there are cases where the enzyme activity is even higher than in homogeneous as for horseradish perenvironment oxidase [24], where the close packed structure facilitates the transfer of electrons to the heme group present in the enzyme. It is important to mention the mass estimated for the enzyme using quartz crystal microbalance was 485 ng for the total area of the solid substrate (about 1 cm2 ) and total amount of enzyme in the aqueous solution was 0.5 g. As a result, the activity per quantity of enzyme may not vary considerably since the total amount of rhodanese was approximately the same for both systems (homogeneous and immobilized as LB film). After 1 month, the enzyme activity was measured again for all films, and the persisting activity was about 85%, while for the homogenous medium this value was reduced to 25%. This fact indicates that the enzyme-lipid layers present a higher stability in relation to the homogenous medium. Finally, rhodanese could be incorporated in DMPA Langmuir films and transferred to solid supports as LB films, with a high
Table 1 Enzyme activity estimated for rhodanese in a homogeneous medium or immobilized in DMPA LB films. Each value is an average of 4 measurements with 4 different films. The error bar displays the standard deviation from four independent experiments. System LB film Homogeneous(0.5 g/mL of enzyme)
Lag time (min) 35 ± 3 5±1
Relative activity (Abs470nm /min) −3
50 ± 2 × 10 57 ± 2 × 10−3
% of relative activity after 30 days 85% 24%
64
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
degree of preservation of its relative catalytic activity, which may be related to the secondary structure of the enzyme, partially kept. The enzyme activity of the rhodanese-DMPA LB film was then exploited to detect cyanide in a proof-of-concept approach, which permits the use of the Langmuir–Blodgett methodology for future development of stable colorimetric cyanide sensors with control over the molecular structure. Evidently, the performance of this methodology can be enhanced by changing parameters such as number of layers, surface pressure of deposition, method of co-immobilization of the enzyme with the lipid, temperature, dipping rate, etc. Also the chemical nature of the lipid can be changed in order to provide an ideal matrix not only for the transfer of a floating monolayer to a solid support, but also to provide a better environment to conserve the enzyme activity. It is important to emphasize that improved organic and bioinspired devices, such as biosensors, can be only achieved if fundamental studies on interfaces are carried out. 4. Conclusions We demonstrated that rhodanese can be incorporated in DMPA Langmuir monolayers as detected with tensiometry, vibrational spectroscopy, and Brewster angle microscopy measurements. The enzyme could be transferred along with DMPA onto solid supports as LB films, being confirmed by PM-IRRAS, fluorescence spectroscopy, and atomic force microscopy. The enzyme activity could be detected for mixed enzyme-lipid LB films, thus demonstrating that this architecture is suitable for sensing cyanide. Performance for1-layer rhodanese—DMPA LB films could be compared to that for the enzyme in homogeneous medium, resulting in systems with a higher stability after one month. Consequently, the lipid matrix may have provided a favorable environment for preserving the catalytic activity of the enzyme, which is associated to the interaction of the polypeptide structure with the phospholipid. Such interactions facilitate the access of the analyte to the catalytic site of the enzyme, and allow for catalyzing the conversion of cyanide to other products, which may cause an impact on the environment. The novelty of this paper lies in the fact that the enzyme activity of rhodanese could be better preserved when immobilized as DMPA LB film in a proof-of-concept experiment.
Acknowledgements We are grateful to FAPESP (2013/10213-1), CNPq-Universal (470890/2012-6), Rede nBioNet: Films and Sensors, and CNPqINCT-INEO for the financial support. References [1] L. Dziri, S. Boussad, S.P. Wang, R.M. Leblanc, J. Phys. Chem. B 101 (1997) 6741–6748. [2] S.V. Mello, M. Maabrouki, X.H. Cao, R.M. Leblanc, T.C. Cheng, J.J. DeFrank, Biomacromolecules 4 (2003) 968–973. [3] A.P. Girard-Egrot, S. Godoy, L.J. Blum, Adv. Colloid Interf. Sci. 116 (2005) 205–225. [4] S. Godoy, B. Leca-Bouvier, P. Boullanger, L.J. Blum, A.P. Girard-Egrot, Sens. Actuator B-Chem. 107 (2005) 82–87. [5] N.C.M. Zanon, O.N. Oliveira Jr., L. Caseli, J. Colloid. Interface Sci. 373 (2012) 69–74. [6] A. Koj, J. Frendo, L. Wojtczac, FEBS Lett. 57 (1981) 42–46. [7] L. Caseli, D.C. Masui, R.P.M. Furriel, F.A. Leone, M.E.D. Zaniquelli, Thin Solid Films 515 (2007) 4801–4807. [8] F.A. Scholl, L. Caseli, Colloids Surf. B 126 (2015) 232–236. [9] G.Z. Sauerbrey, Z. Phys. 155 (1959) 206–222. [10] D.R. Singleton, D.W. Smith, Appl. Environ. Microbiol. 54 (1988) 2866–2867. [11] M. Losche, C. Helm, H.D. Mattes, H. Möhwald, Thin Solid Films 133 (1985) 51–64. [12] T.F. Schmidt, L. Caseli, T.M. Nobre, M.E.D. Zaniquelli, O.N. Oliveira Jr., Colloid Surf. A: Physicochem. Eng. Asp. 321 (2008) 206–210. [13] L. Caseli, M.E.D. Zaniquelli, R.P.M. Furriel, F.A. Leone, Colloid Surf. B: Biointerf. 30 (2003) 273–282. [14] C. Cannella, M. Costa, B. Pensa, G. Ricci, L. Pecci, D. Cavallini, Eur. J. Biochem. 119 (1981) 491–495. [15] J.T. Davies, E.K. Rideal, Interfacial Phenomena, 2nd ed., Academic Press, New York, 1963. [16] L. Caseli, R.P.M. Furriel, F.A. Leone, M.E.D. Zaniquelli, Colloids Surf. B 30 (2003) 273–282. [17] A. Brume, Biochim. Biophys. Acta 557 (1979) 32–44. [18] D. Blaudez, T. Buffeteau, J.C. Cornut, B. Desbat, N. Escafre, M. Pezolet, J.M. Turlet, Thin Solid Films 242 (1994) 146–150. [19] C. Huang, J.R. Lapides, I.W. Levin, J. Am. Chem. Soc. 104 (1982) 5926–5930. [20] F. Dousseau, M. Pézolet, Biochemistry 29 (1990) 8771–8779. [21] L. Caseli, M.L. Moraes, V. Zucolotto, M. Ferreira, T.M. Nobre, M.E.D. Zaniquelli, U.P. Rodrigues, O.N. Oliveira Jr., Langmuir 22 (2006) 8501–8508. [22] N.C. de Souza, W. Caetano, R. Itri, C.A. Rodrigues, O.N. Oliveira, J.A. Giacometti, M. Ferreira, J. Colloid Interf. Sci. 297 (2006) 546–553. [23] S. Paddeu, A. Fanigliulo, M. Lanzi, T. Dubrovsky, C. Nicolini, Sens. Actuators B: Chem. 25 (1995) 876–882. [24] T.F. Schmidt, L. Caseli, T. Viitala, O.N. Oliveira Jr., Biochim. Biophys. Acta 1778 (2008) 2291–2297.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Rhodanese incorporated in Langmuir and Langmuir–Blodgett films of dimyristoylphosphatidic acid: Physical chemical properties and improvement of the enzyme activity Felipe Tejada de Araújo, Luciano Caseli ∗ Institute of Environmental, Chemical and Pharmaceutical Sciences, Federal University of São Paulo, Diadema, SP, Brazil
a r t i c l e
i n f o
Article history: Received 14 December 2015 Received in revised form 15 January 2016 Accepted 19 January 2016 Available online 22 January 2016 Keywords: Rhodanese Air–water interface Langmuir monolayers Langmuir–Blodgett Enzyme activity Cyanide
a b s t r a c t Preserving the catalytic activity of enzymes immobilized in bioelectronics devices is essential for optimal performance in biosensors. Therefore, ultrathin films in which the architecture can be controlled at the molecular level are of interest. In this work, the enzyme rhodanese was adsorbed onto Langmuir monolayers of the phospholipid dimyristoylphosphatidic acid and characterized by surface pressurearea isotherms, polarization-modulated infrared reflection-absorption spectroscopy (PM-IRRAS), and Brewster angle microscopy (BAM). The incorporation of the enzyme (5% in mol) in the lipid monolayer expanded the film, providing small surface domains, as visualized by BAM. Also, amide bands could be identified in the PM-IRRAS spectra, confirming the presence of the enzyme at the air–water interface. Structuring of the enzyme into ␣-helices was identified in the mixed monolayer and was preserved when the film was transferred from the liquid interface to solids supports as Langmuir–Blodgett (LB) films. The enzyme-lipid LB films were then characterized by fluorescence spectroscopy, PM-IRRAS, and atomic force microscopy. Measurements of the catalytic activity towards cyanide showed that the enzyme accommodated in the LB films preserved more than 87% of the enzyme activity in relation to the homogeneous medium. After 1 month, the enzyme in the LB film maintained 85% of the activity in contrast to the homogeneous medium, which 24% of the enzyme activity was kept. The method presented in this work not only points to an enhanced catalytic activity toward cyanide, but also may explain why certain film architectures exhibit an improved performance. © 2016 Elsevier B.V. All rights reserved.
1. Introduction The immobilization of enzymes on solid supports as Langmuir–Blodgett (LB) films is a recent approach to obtain devices whose architectures can be controlled at the molecular level. For that, monomolecular films are first formed at the air–water interface and transferred to solid supports during their vertical passage through the floating monolayer. Although pure enzymes can be spread on the air–water interface [1,2], their incorporation in pre-formed phospholipid monolayers may preserve the conformation of the enzyme [3–5]. As a result, a well-ordered nanostructured system can be constructed and the enzyme may serve as a recognizing element. If an optical or electrical signal is generated due to a recognizing chemical reaction, the film is a potential candidate for a biosensor.
∗ Corresponding author. E-mail address: [email protected] (L. Caseli). http://dx.doi.org/10.1016/j.colsurfb.2016.01.037 0927-7765/© 2016 Elsevier B.V. All rights reserved.
Particularly, rhodanese is a mitochondrial enzyme that detoxifies cyanide, converting it into thiocyanate [6]. This reaction takes place in two steps: in the first step, thiosulfate reacts with the thiol group to form a disulfide. In the second step, the disulfide reacts with cyanide to produce thiocyanate, and is converted again to thiol. This reaction is environmentally important because it leads to a reduced exposure of cyanide since the less toxic thiocyanate is formed. The use of a thiosulfate solution as an antidote for cyanide poisoning is based on the activation of the enzymatic cycle. Thus, the immobilization of rhodanese in ultrathin films is interesting since it can provide a device capable of remedying the amount of cyanide in a controlled manner. LB films seem to be a suitable and innovative strategy for the construction of nanostructured films since monomolecular films can be deposited on solid supports with control over the chemical composition and surface density. This approach has already been performed for some enzymes [5,7,8], but not found in the literature for rhodanese. In this paper, emphasis on the interaction of the enzyme with the phospholipid dimyristoylphosphatidic acid (DMPA) was placed. As
60
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
we want to consider further the ability of such architecture to recognize cyanide, the enzymatic activity was investigated for mixed lipid-enzyme films supported as LB films.
2. Materials and methods DMPA was purchased from Sigma–Aldrich (purity higher than 99%) and dissolved in chloroform (Sytnh) to a concentration of 0.5 mg/mL. This lipid was chosen considering the ease to transfer it to solid supports as shown in previous studies [7,8]. Water used in all experiments was purified using a Milli-Q® system (resistivity of 18.2 cm−1 , pH 6.0). Rhodanese from Sigma–Aldrich (purity higher than 99% obtained from bovine liver) was dissolved in an aqueous buffer solution of K2 HPO4 (Sigma–Aldrich) and KH2 PO4 (Sigma–Aldrich) with a salt concentration of 0.01 mol/L and pH 7.0. The final concentration of the enzyme solution was 1 mg/mL. A Langmuir trough (KSV Instruments, Helsinki—Finland, model: Mini, 36.5 × 7.5 cm) was employed for the preparation of the Langmuir and LB films. It was filled with water and then 45–65 L of the DMPA solution was spread on the air–water interface to obtain an area per molecule of ∼90–120 Å2 . After the evaporation of chloroform for 15–20 min, the interface was compressed with two movable barriers at a rate of 10 cm min−1 (5 Å2 molecule−1 s−1 ), and the surface pressure values were monitored with a Wilhelmy plate made of filter paper that intercepted the air–water interface. For mixed enzyme-lipid monolayers, pre-determined aliquots of the enzyme solution (16–80 L) were carefully injected below a pre-formed DMPA monolayer into the aqueous subphase. After allowing the surface pressure to stabilize during 30 min., the interface was compressed and the surface pressure monitored. Measurements through polarization-modulation infrared reflection absorption spectroscopy (PM-IRRAS) were done with a KSV PMI 550 instrument (KSV Instruments Ltd., Helsinki—Finland). The monolayers were compressed to reach the surface pressure of 30 mN/m, and this value was kept constant by using mobile barriers while the spectra were obtained. The incidence angle to the normal was 80◦ and a minimum of 600 scans were obtained for each spectrum. The incoming light was continuously modulated between the p and s polarization, allowing simultaneous measurements of the spectra for both polarizations. The difference between the two absorbance signals gives surface-specific information and the sum provides the reference spectrum. Since the spectra are measured simultaneously, the effect of isotropic vibrations (water vapor and carbon dioxide) is largely reduced. Brewster angle microscopy (KSV-Nima Instruments, model: micro BAM3) was employed in order to obtain images of the monolayer at desired values of surface pressure. Solid glass supports were inserted in the aqueous subphase of the Langmuir films. The films were then spread on the air–water interface and compressed to reach the surface pressure of 30 mN/m. The supports were vertically withdrawn across the air–water interface with a speed of 5 mm min−1 keeping the surface pressure of 30 mN/m constant during the passage of the support. Transfer ratio values of about 0.95–1.05 were required for further analysis. LB films with 1 single layer were characterized by PMIRRAS and fluorescence spectroscopy (Spectrophotometer model RF-5301PC, Shimadzu) with an excitation wavelength of 285 nm with the glass directly placed in the fluorimeter holder. Atomic force microscopy (AFM) was also employed for further characterization, and the images were obtained in the tapping mode, employing a resonance frequency of approximately 300 kHz, a scan rate of 1.0 Hz, and scanned areas of 5.0 × 5.0 m on films deposited on mica. Nanogravimmetry through a quartz crystal microbalance (SRS—Stanford Research Systems model QCM200) was employed to estimate the mass of film deposited on a surface bounded by
Fig. 1. Surface pressure-area isotherms for DMPA and rhodanese (relative proportions in mol of the enzyme injected in the subphase are indicated in the inset).
gold electrodes in the thin disk of quartz as substrate. The mass of the deposited film was determined according to the Sauerbrey equation [9]. The catalytic activity of the enzyme was estimated according to a method previously described in the literature [10]. For this purpose, the LB film was inserted in a solution containing cyanide, and the enzyme activity measured due to the evolution of the optical density of the solution measured at 470 nm (UV–vis Hitashi, model U2001). The enzymatic activity was also measured in a homogeneous environment (enzyme dissolved in aqueous solution) for comparison. All enzymatic activities were measured again after 30 days with the sample kept at low temperatures (5–10 ◦ C). All experiments were performed at a temperature of 25 ± 1 ◦ C. 3. Results and discussion 3.1. Langmuir monolayers Firstly, it is important to emphasize that DMPA was chosen because of its easiness to be transferred to solid supports as a LB film and also because this lipid is reported as a suitable matrix for immobilizing enzymes [7,8]. DMPA monolayers present a typical surface pressure-area isotherm (Fig. 1), in agreement with the literature [11]. Inserting 1% in mol of rhodanese, a shift of the curve to smaller lipid areas is observed as a consequence of the condensation of the monolayer. With 5% in mol of the enzyme, the isotherm is shifted to higher areas, which may be a consequence of the incorporation of the enzyme inside the lipid chains, causing the lipid monolayer expansion. With higher concentrations of rhodanese, the isotherm is no longer shifted to higher areas, indicating saturation. This saturation effect can be attributed to a probable enzyme aggregation in the aqueous subphase as previously reported for other macromolecules adsorbing on lipid monolayers [12,13]. As the isoelectric point of this enzyme varies between 5.6 and 7.15 [14], rhodanese is zwitterionic under the experimental conditions (pH around 6.0). Being DMPA negatively charged at this pH, dipole-ion attractions between the lipid and the enzyme must influence the adsorption of the biomacromolecule and the subsequent monolayer expansion. Rheological properties of the monolayer can be accessed by means of its compressional modulus (E), which is defined as −A(∂/∂A)T [15], where A is the molecular area, is the surface pressure of the monolayer, and T is the temperature. The highest values of E for DMPA were of ca. 300 mN/m, featuring the liquidcondensed state (Fig. 2A and B). DMPA mixed with 1% of rhodanese presents higher values of E (550 mN/m) as a consequence of the condensation of the monolayer caused by small amounts of the enzyme. With 5% of rhodanese, the maximum values of E are similar to those obtained for pure DMPA. It is probable that the penetration of the enzyme into the lipid monolayer makes the film more
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
A
A
B
B
Fig. 2. Compressional modulus-area isotherms for DMPA and rhodanese (relative proportions in mol indicated in the inset).
rigid. This kind of behavior is reported in the literature for other macromolecules interacting with lipid Langmuir monolayers and the effect is reported as a “piston-like effect” [16], where the protein insertion forces the monolayer compression, providing a more packed interfacial structure. Also, Fig. 2B shows that between 30 and 40 mN/m the maximum variation of surface elasticity takes place, which is the range that corresponds approximately to the lateral pressure of cell membranes [17]. PM-IRRAS spectra for DMPA and mixed rhodanese-DMPA monolayers are shown in Fig. 3. It is important to mention that these spectra were obtained at the surface of 30 mN/m because of the parameters obtained with the surface elasticity measurements, and because we intended to prepare these films in bio-inspired conditions. Moreover, this surface pressure was the value employed for the transfer of the floating monolayers to solid supports as LB films, as shown later in this paper. It is important to highlight that since the bands are positive to the baseline of the spectra, the vibration moments lie parallel to the air/water interface [18]. The bands centered at 2918 and 2850 cm−1 correspond to CH2 antisymmetric and symmetric stretching modes, respectively. The band centered at 2956 cm−1 corresponds to the CH3 stretching mode. With the insertion of the enzyme, a slight shift is observed from 2918 to 2921 cm−1 and from 2850 to 2846 cm−1 . These changes are not relevant considering the resolution of the instrument (8 cm−1 ). Also the band at 2956 cm−1 becomes a shoulder upon the incorporation of the enzyme. For the mixed enzyme-lipid monolayer, a significant change is observed for the ratio between the maximum intensity of the antisymmetric and symmetric bands, which varies from 1.6 for pure DMPA to 2.4 for the mixed monolayer. This ratio is usually attributed to the order parameter of the film [19], and indicates that the mixed monolayer is in a higher ordered structure when compared to that for the pure lipid monolayer. This fact could be attributed to the incorporation of the enzyme into the lipid matrix, which provides a film with a close packed structure.
61
Fig. 3. PM-IRRAS spectra for DMPA (pure or mixed with 5% of rhodanese as indicated in the inset) monolayers at 30 mN/m.
Panel B for Fig. 3 shows some bands that are attributed to rhodanese, indicating the adsorption of the enzyme. Amide II (C N and N H vibrations) bands are observed in the wavelength range between 1500 and 1600 cm−1 , while amide I (C O stretching modes) bands are observed in the range between 1600 and 1700 cm−1 . The band centered at 1743 cm−1 may be attributed to the C O stretching mode of DMPA. The band centered at 1676 cm−1 can be attributed to the secondary structure of the enzyme polypeptide moiety without any defined order, and the band centered at 1560 cm−1 is attributed to amide II vibrations. The bands centered at 1464 cm−1 is attributed to C H bends, and increases their relativity intensity with the incorporation of the enzyme. The bands centered at 1269 and 1112 cm−1 are attributed to P O stretches in the phosphate group of DMPA. The relative intensity of these bands decreased significantly for the enzyme-DMPA mixed monolayer, suggesting that the enzyme interacts with the polar head-groups of the lipid. In this region of the spectra, it is important to emphasize that an amide I/amide II intensity ratio higher than 1.0 indicates the non-denaturated form of a protein [20]. Since this ratio is 1.9 in the spectra shown in Fig. 3B, it is probable that the incorporation of rhodanese in the lipid monolayer must prevent the enzyme from denaturation by maintaining at least part of its molecular conformation. Fig. 4 shows the BAM imagens for the monolayers with and without the enzyme. As expected for a pure DMPA monolayer, a homogenous pattern is observed at 30 mN/m, regarding that there is no phase segregation, and a single liquid condensed phase is noted. For the mixed monolayer, however, some small spots of bright domains are observed along the surface. Other higher bright domains can be observed as a consequence of aggregates induced by the presence of rhodanese. This fact indicates that the presence of the enzyme slightly changes the morphology of the lipid film. 3.2. LB films For the building of LB films, a preliminary test was carried out to check if the enzyme molecules present in the bulk of the
62
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
Fig. 4. BAM images (3600 × 4000 m) for DMPA (pure or mixed with 5% of rhodanese as indicated in the inset) monolayers at 30 mN/m.
A
Fig. 6. Fluorescence spectra for rhodanese-DMPA LB film.
B
Fig. 5. PM-IRRAS spectra for rhodanese-DMPA LB film.
aqueous subphase adsorbs on the immersed solid support. For that, the solid support was immersed in a solution containing the enzyme, but with no lipid at the air–water interface. After 60 min, which corresponds to the time of permanence of the solid support in the aqueous subphase during the LB deposition, the plate was removed, and measurements of fluorescence, PM-IRRAS and catalytic activity were carried out. No fluorescence signal was observed, neither amide bands in the vibrational spectra. Also, no catalytic activity was detected, which means that no significant amount of enzyme is adsorbed from the bulk of the aqueous subphase onto the solid support. Therefore, the adsorption of the enzyme on the solid support must be attributed exclusively to the passage of the solid support through the mixed enzyme-lipid monolayer. The enzyme-lipid films were then transferred to solid supports by using the LB technique with a relative amount of enzyme of 5% in the system. For all depositions, a transfer ratio close to 1.0 was obtained for all passages of the substrate. The co-incorporation of rhodanese was then investigated with PM-IRRAS (Fig. 5).
Antisymmetric and symmetric bands for CH2 are centered at 2925 and 2850 cm−1 , respectively. A band at 2890 cm−1 , attributed to CH3 stretching modes, is noticeable. The ratio between the maxima for the antisymmetric and the symmetric bands is 1.7 for the pure lipid, and 2.3 for the mixed film. These values are slightly different from the relative ratio values observed for the floating films, which indicates some alteration in the packing degree of the lipids when transferred to the solid support. Normally, the deposited layers are affected by the underlying solid support structure and by the loss of water molecules during the deposition procedure. Amide I and II bands are also present (Panel B for Fig. 4), confirming the transfer of the enzyme to the solid support, denoting thinner bands when compared to the spectra obtained for the air–water interface. The amide I band for ␣-helices is present at 1659 cm−1 and a small band for -sheets is observed at 1620 cm−1 . These facts indicate some conformational changes of rhodanese during the transfer, and may be also related to the loss of water molecules during the procedure. All the shifts of the maxima of the bands in relation to the monolayer on the air–water interface reflect the interaction of the film components with the solid substrate. Fig. 6 shows the fluorescence spectrum for the mixed LB films and confirms the presence of the enzyme in the structure since DMPA should not present fluorescence. The tryptophan group, present in some amino-acid residues of rhodanese, was excited at 285 nm, and fluorescence emission was observed with a maximum at 331 nm. Other studies with mixed enzyme-lipid LB films also confirmed the presence of the enzymes by using fluorescence spectroscopy [6,21]. The morphology of the mixed enzyme-lipid LB films can be accessed by AFM images (Fig. 7). For pure DMPA LB films, a flat pattern with no distinction of aggregates is always observed [22]. With rhodanese, irregularities, such as ridges, protuberances and holes are observed, being a consequence of the enzyme
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
63
Fig. 7. AFM images of rhodanese-DMPA LB film transferred at 30 mN/m.
incorporation. Aggregates with diameter varying from 50 to 200 nm are observed, but maximum highness of the aggregates is 3–3.5 nm. Also the RMS (root mean square roughness) increased from 0.3 nm, for the pure DMPA LB film, to 3.2, for the enzyme-lipid LB film, which indicates the increase of roughness of the film with the introduction of rhodanese. This pattern of morphology can be related to the accommodation of the enzyme in the lipid matrix, revealing that the co-transfer of the macromolecule affects the level of organization of DMPA transferred as a LB film. On the other hand, this film is relatively homogeneous when compared to other AFM images for protein-lipid LB films [22]. Table 1 shows the relative enzyme activity for the LB films, indicating that they are sensitive to the presence of cyanide, presenting enzyme activity. This is a preliminary indication for the possible use of such films as a cyanide sensor. The enzyme activity for LB films is 87% compared to the activity obtained for the homogeneous medium. These lower values, as well as the longer lag time, are common for enzymes immobilized on solid supports, as frequently reported in the literature [5,23]. These facts can be attributed to restrictions of molecular mobility for the polypeptide structure. The literature shows, for instance, that alkaline phosphatase co-immobilized with DMPA LB films does not present enzyme activity higher than 12% in relation to the homogeneous environment, with the longer lag time attributed to diffusive processes. Consequently, the catalytic substrate (cyanide) may present some restrictions to access the catalytic site of the enzyme . Also, a supposed higher amount of the enzyme in the homo-
geneous medium could boost the enzyme activity. However the value of 87% obtained in this paper is higher when compared to other enzymes immobilized in LB films [5,7,8,23]. It is important to emphasize that understanding the reasons by which the enzyme activity is conserved in lipid matrices is not straightforward. Although it is expected a decrease of the enzyme activity in relation to the homogeneous environment after immobilization in solid supports, there are cases where the enzyme activity is even higher than in homogeneous as for horseradish perenvironment oxidase [24], where the close packed structure facilitates the transfer of electrons to the heme group present in the enzyme. It is important to mention the mass estimated for the enzyme using quartz crystal microbalance was 485 ng for the total area of the solid substrate (about 1 cm2 ) and total amount of enzyme in the aqueous solution was 0.5 g. As a result, the activity per quantity of enzyme may not vary considerably since the total amount of rhodanese was approximately the same for both systems (homogeneous and immobilized as LB film). After 1 month, the enzyme activity was measured again for all films, and the persisting activity was about 85%, while for the homogenous medium this value was reduced to 25%. This fact indicates that the enzyme-lipid layers present a higher stability in relation to the homogenous medium. Finally, rhodanese could be incorporated in DMPA Langmuir films and transferred to solid supports as LB films, with a high
Table 1 Enzyme activity estimated for rhodanese in a homogeneous medium or immobilized in DMPA LB films. Each value is an average of 4 measurements with 4 different films. The error bar displays the standard deviation from four independent experiments. System LB film Homogeneous(0.5 g/mL of enzyme)
Lag time (min) 35 ± 3 5±1
Relative activity (Abs470nm /min) −3
50 ± 2 × 10 57 ± 2 × 10−3
% of relative activity after 30 days 85% 24%
64
F.T. de Araújo, L. Caseli / Colloids and Surfaces B: Biointerfaces 141 (2016) 59–64
degree of preservation of its relative catalytic activity, which may be related to the secondary structure of the enzyme, partially kept. The enzyme activity of the rhodanese-DMPA LB film was then exploited to detect cyanide in a proof-of-concept approach, which permits the use of the Langmuir–Blodgett methodology for future development of stable colorimetric cyanide sensors with control over the molecular structure. Evidently, the performance of this methodology can be enhanced by changing parameters such as number of layers, surface pressure of deposition, method of co-immobilization of the enzyme with the lipid, temperature, dipping rate, etc. Also the chemical nature of the lipid can be changed in order to provide an ideal matrix not only for the transfer of a floating monolayer to a solid support, but also to provide a better environment to conserve the enzyme activity. It is important to emphasize that improved organic and bioinspired devices, such as biosensors, can be only achieved if fundamental studies on interfaces are carried out. 4. Conclusions We demonstrated that rhodanese can be incorporated in DMPA Langmuir monolayers as detected with tensiometry, vibrational spectroscopy, and Brewster angle microscopy measurements. The enzyme could be transferred along with DMPA onto solid supports as LB films, being confirmed by PM-IRRAS, fluorescence spectroscopy, and atomic force microscopy. The enzyme activity could be detected for mixed enzyme-lipid LB films, thus demonstrating that this architecture is suitable for sensing cyanide. Performance for1-layer rhodanese—DMPA LB films could be compared to that for the enzyme in homogeneous medium, resulting in systems with a higher stability after one month. Consequently, the lipid matrix may have provided a favorable environment for preserving the catalytic activity of the enzyme, which is associated to the interaction of the polypeptide structure with the phospholipid. Such interactions facilitate the access of the analyte to the catalytic site of the enzyme, and allow for catalyzing the conversion of cyanide to other products, which may cause an impact on the environment. The novelty of this paper lies in the fact that the enzyme activity of rhodanese could be better preserved when immobilized as DMPA LB film in a proof-of-concept experiment.
Acknowledgements We are grateful to FAPESP (2013/10213-1), CNPq-Universal (470890/2012-6), Rede nBioNet: Films and Sensors, and CNPqINCT-INEO for the financial support. References [1] L. Dziri, S. Boussad, S.P. Wang, R.M. Leblanc, J. Phys. Chem. B 101 (1997) 6741–6748. [2] S.V. Mello, M. Maabrouki, X.H. Cao, R.M. Leblanc, T.C. Cheng, J.J. DeFrank, Biomacromolecules 4 (2003) 968–973. [3] A.P. Girard-Egrot, S. Godoy, L.J. Blum, Adv. Colloid Interf. Sci. 116 (2005) 205–225. [4] S. Godoy, B. Leca-Bouvier, P. Boullanger, L.J. Blum, A.P. Girard-Egrot, Sens. Actuator B-Chem. 107 (2005) 82–87. [5] N.C.M. Zanon, O.N. Oliveira Jr., L. Caseli, J. Colloid. Interface Sci. 373 (2012) 69–74. [6] A. Koj, J. Frendo, L. Wojtczac, FEBS Lett. 57 (1981) 42–46. [7] L. Caseli, D.C. Masui, R.P.M. Furriel, F.A. Leone, M.E.D. Zaniquelli, Thin Solid Films 515 (2007) 4801–4807. [8] F.A. Scholl, L. Caseli, Colloids Surf. B 126 (2015) 232–236. [9] G.Z. Sauerbrey, Z. Phys. 155 (1959) 206–222. [10] D.R. Singleton, D.W. Smith, Appl. Environ. Microbiol. 54 (1988) 2866–2867. [11] M. Losche, C. Helm, H.D. Mattes, H. Möhwald, Thin Solid Films 133 (1985) 51–64. [12] T.F. Schmidt, L. Caseli, T.M. Nobre, M.E.D. Zaniquelli, O.N. Oliveira Jr., Colloid Surf. A: Physicochem. Eng. Asp. 321 (2008) 206–210. [13] L. Caseli, M.E.D. Zaniquelli, R.P.M. Furriel, F.A. Leone, Colloid Surf. B: Biointerf. 30 (2003) 273–282. [14] C. Cannella, M. Costa, B. Pensa, G. Ricci, L. Pecci, D. Cavallini, Eur. J. Biochem. 119 (1981) 491–495. [15] J.T. Davies, E.K. Rideal, Interfacial Phenomena, 2nd ed., Academic Press, New York, 1963. [16] L. Caseli, R.P.M. Furriel, F.A. Leone, M.E.D. Zaniquelli, Colloids Surf. B 30 (2003) 273–282. [17] A. Brume, Biochim. Biophys. Acta 557 (1979) 32–44. [18] D. Blaudez, T. Buffeteau, J.C. Cornut, B. Desbat, N. Escafre, M. Pezolet, J.M. Turlet, Thin Solid Films 242 (1994) 146–150. [19] C. Huang, J.R. Lapides, I.W. Levin, J. Am. Chem. Soc. 104 (1982) 5926–5930. [20] F. Dousseau, M. Pézolet, Biochemistry 29 (1990) 8771–8779. [21] L. Caseli, M.L. Moraes, V. Zucolotto, M. Ferreira, T.M. Nobre, M.E.D. Zaniquelli, U.P. Rodrigues, O.N. Oliveira Jr., Langmuir 22 (2006) 8501–8508. [22] N.C. de Souza, W. Caetano, R. Itri, C.A. Rodrigues, O.N. Oliveira, J.A. Giacometti, M. Ferreira, J. Colloid Interf. Sci. 297 (2006) 546–553. [23] S. Paddeu, A. Fanigliulo, M. Lanzi, T. Dubrovsky, C. Nicolini, Sens. Actuators B: Chem. 25 (1995) 876–882. [24] T.F. Schmidt, L. Caseli, T. Viitala, O.N. Oliveira Jr., Biochim. Biophys. Acta 1778 (2008) 2291–2297.