Interaction of a Rhodococcus sp. Trehalose Lipid Biosurfactant with

Article pubs.acs.org/Langmuir

Interaction of a Rhodococcus sp. Trehalose Lipid Biosurfactant with Model Proteins: Thermodynamic and Structural Changes Ana Zaragoza,† José A. Teruel,† Francisco J. Aranda,† Ana Marqués,‡ María J. Espuny,‡ Á ngeles Manresa,‡ and Antonio Ortiz*,† †

Departamento de Bioquímica y Biología Molecular-A, Facultad de Veterinaria, Universidad de Murcia, E-30100 Murcia, Spain Laboratorio de Microbiología, Facultad de Farmacia, Universidad de Barcelona, Joan XXIII s/n, E-08028 Barcelona, Spain

‡

ABSTRACT: One major application of surfactants is to prevent aggregation during various processes of protein manipulation. In this work, a bacterial trehalose lipid (TL) with biosurfactant activity, secreted by Rhodococcus sp., has been identified and purified. The interactions of this glycolipid with selected model proteins have been studied by using differential scanning calorimetry (DSC), Fourier-transform infrared (FTIR) spectroscopy, isothermal titration calorimetry (ITC), and fluorescence spectroscopy. Bovine serum albumin (BSA) and cytochrome c (Cyt-c) have been chosen because of their quite different secondary structures: BSA contains essentially no β-sheets and an average 66% α-helix, whereas Cyt-c possesses up to 25% β-sheets and up to 45% α-helical structure. Differential scanning calorimetry shows that addition of TL to BSA at concentrations below the critical micelle concentration (cmc) shifts the thermal unfolding temperature to higher values. FTIR indicates that TL does not alter the secondary structure of native BSA, but the presence of TL protects the protein toward thermal denaturation, mainly by avoiding formation of β-aggregates. Studies on the intrinsic Trp fluorescence of BSA show that addition of TL to the native protein results in conformational changes. BSA unfolding upon thermal denaturation in the absence of TL makes the Trp residues less accessible to the quencher, as shown by a decrease in the value of Stern−Volmer dynamic quenching constant, whereas denaturation in the presence of the biosurfactant prevents unfolding, in agreement with FTIR results. In the case of Cyt-c, interaction with TL gives rise to a new thermal denaturation transition, as observed by DSC, at temperatures below that of the native protein, therefore facilitating thermal unfolding. Binding of TL to native BSA and Cyt-c, as determined by ITC, suggests a rather nonspecific interaction of the biosurfactant with both proteins. FTIR indicates that TL slightly modifies the secondary structure of native Cyt-c, but protein denaturation in the presence of TL results in a higher proportion of β-aggregates than in its absence (20% vs 3.9%). The study of Trp fluorescence upon TL addition to Cyt-c results in a completely opposite scenario to that described above for BSA. In this case, addition of TL considerably increases the value of the dynamic quenching constant, both in native and denatured protein; that is, the interaction with the glycolipid induces conformational changes which facilitate the exposure of Trp residues to the quencher. Considering the structures of both proteins, it could be derived that the characteristics of TL interactions, either promoting or avoiding thermal unfolding, are highly dependent on the protein secondary structure. Our results also suggest the rather unspecific nature of these interactions. These might well involve protein hydrophobic domains which, being buried into the protein native structures, become exposed upon thermal unfolding.

■

INTRODUCTION The term biosurfactant is commonly used to describe any amphiphilic compound of biological origin with the capacity of lowering the surface tension of water and aqueous solutions. Currently, biosurfactants are defined as surface-active compounds of microbial origin. Biosurfactants present several advantages over surfactants of a chemical origin. As a general rule, biosurfactants are more biocompatible, more easily biodegradable, and therefore less toxic and less harmful to the environment, and more active at lower concentrations. In addition, some of them display interesting biological activities. Some of these compounds can be produced through biotechnological processes in which the carbon source is © 2011 American Chemical Society

provided by waste materials or side products from several industries, constituting an added value not only from an economic point of view but also for ecological reasons. Thus, it is of great interest to characterize new biosurfactants to evaluate their use as potential alternatives to chemically synthesized compounds.1−4 The chemical nature of biosurfactants is diverse, including glycolipids, lipoaminoacids and lipopeptides, polymers, and phospholipids, mono- and diacylglycerols, and fatty acids.5 A group of Received: October 3, 2011 Revised: November 24, 2011 Published: December 15, 2011 1381

dx.doi.org/10.1021/la203879t | Langmuir 2012, 28, 1381−1390

Langmuir

Article

When required, BSA and Cyt-c were denatured by incubation at 70 and 90 °C for 1 h, respectively. The buffer used throughout the work was 150 mM NaCl, 5 mM Hepes, pH 7.4, unless otherwise indicated. Water and all other solutions used in this work were filtered through 0.2 μm filters prior to use. Trehalose Lipid Production and Purification. Strain 51T7 was isolated from an oil-contaminated soil sample after culture enrichment with kerosene and was identified as Rhodococcus sp. This strain was maintained by fortnightly cultures on Trypticase Soy Agar (Pronadisa, Spain) and preserved in cryovials at −20 °C. Biosurfactants were produced, purified, and characterized as described before.30 Briefly, cultures were incubated at 27 °C and aeration was achieved with strong agitation of the culture medium on a shaker (100 rpm). Inoculum was prepared by incubation of the strain in mineral medium with 10 g L−1 of C13 paraffin for 36 h. A 2% v/v of this culture was used as inoculum in 500 mL volumetric baffled flasks with 50 mL medium. The surface active compounds were isolated by liquid−liquid extraction. The isolated products were analyzed by thin layer chromatography on silica gel G60, using chloroform/methanol/water (65:25:4) as mobile phase, and visualized with iodine vapor. GS/MS allowed determination of the succinoyl, heptanoyl, decanoyl, nonanoyl, and undecanoyl residues. The main fraction with surface activity was identified with 1H NMR and 13C NMR as the 2,3,4,2′-trehalose tetraester. Sample Preparation. TL aqueous samples were prepared by dispersion of the required amount of the biosurfactant in the appropriate buffer, as indicated. Briefly, the desired amount of TL was dissolved in chloroform/methanol (2:1), and the solvent was gently evaporated under a stream of dry N2, to obtain a thin film at the bottom of a glass tube. The last traces of solvent were removed by a further minimum 3 h desiccation under high vacuum. The appropriate buffer was added to the dry samples, and these were shaken at room temperature until a visually homogeneous solution or suspension was obtained. Stock proteins solutions were freshly prepared just before the experiments, at a concentration of 1.5 mM in the appropriate filtered buffer, by gentle stirring at 25 °C. Concentration was checked by ultraviolet absorption using ε = 44 720 M−1 cm−1 for BSA and ε = 106 100 M−1 cm−1 for Cyt-c. The stock solution was diluted to the desired concentration in the same buffer, prior to the experiments. High Sensitivity Differential Scanning Calorimetry. Thermal denaturation of BSA and Cyt-c was monitored by differential scanning calorimetry (DSC) using a VP-DSC high sensitivity differential scanning calorimeter from MicroCal (Northampton, MA). Thermograms were recorded between 40 and 90 °C for BSA and between 40 and 110 °C for Cyt-c, at a scan rate of 30 °C h−1. It is known that there is some dependence of the transition temperature upon scanning rate; thus, the same scan rates were used for pure proteins and proteins in the presence of TL, in order to make proper comparisons. Furthermore, the scan rate used here has been widely used for this type of studies with these and other proteins.31−33 Protein concentration was 0.05 mM (molecular weights 66 300 and 12 327 for BSA and Cyt-c, respectively), unless otherwise indicated. The calorimetric data were analyzed using the Origin software provided with the equipment, to obtain ΔH and Tm values. The curve fitting procedure carried out with some thermograms was performed with the same software using a Gaussian function. Isothermal Titration Calorimetry. High sensitivity isothermal titration calorimetry (ITC) measurements were performed in a VPITC titration calorimeter from MicroCal (Northampton, MA). The mixing cell had a volume of 1.442 mL. The data were processed using the Origin software provided with the equipment. The experiments were carried out as follows. The calorimeter syringe was filled with a 2 mM solution of TL (about 20 times larger than the critical micelle concentration, cmc). The injection of this solution into the protein solution (0.1 mM) contained in the calorimeter cell resulted in release of heat upon binding of trehalose lipid to the protein. Experiment was conducted at 25 °C. The data were corrected for TL heats of dilution, which were determined in a separate set of experiments. Infrared Spectroscopy. Samples for the infrared measurements were prepared essentially as described above, in a D2O buffer containing

glycolipid biosurfactants is formed by trehalose-containing glycolipids.6 These trehalose lipids are mainly produced by rhodococci and present interesting physicochemical and biological properties.7 Trehalose lipids significantly reduce the surface tension of water8 and form microemulsions.9 Thus, a number of possible applications for these compounds have been proposed.7 In addition, succinoyl trehalose lipids have been found to induce differentiation of leukemia cell lines10 and to inhibit protein kinase activity.11 Albumins are extraordinary molecules because of their properties and applications, and BSA has been widely used as a general model to study this type of interactions of surfactants with globular proteins.12−16 Albumins, as transport proteins, can bind a wide number of organic compounds of medium size, including fatty acids, amino acids, steroids, or surfactants. BSA is a singlechain protein with 581 amino acids and a molecular weight of 66 300, and its structure and physicochemical properties have been thoroughly described.17−21 The properties and efficacy of many pharmaceutical drugs can be modified on the basis of their interactions with BSA. Cytochrome c (Cyt-c) is a small globular protein which consists of 104 amino acid residues (molecular weight 12 327) in a single polypeptide chain (N-terminal acetylated), to which a protoporphyrin IX heme prosthetic group is bound via two thio-ether linkages. Out of its 104 residues, 24 are lysine, arginine, and histidine. It is an essential redox protein found in the mitochondria of all eukaryotic as well as prokaryotic cells, functioning as an electron transporter in the energy yielding respiratory chain.22−25 More recently, Cyt-c has been identified as an important mediator in apoptotic pathways,26,27 and it has been widely used as a model for both the kinetic and thermodynamic studies of protein folding. Surfactants are essential compounds for the study of the structure and function of membrane proteins.28 The studies on the specific and nonspecific interactions of surfactants and proteins are of wide relevance because of the various industrial and research laboratory applications of surfactant/protein systems. Thus, applications in cosmetics and detergency can be found, as well as in various biochemical laboratory methods.29 Given the growing interest in the development of new applications for biosurfactants,3 the aim of this work is to present a detailed molecular level study of the interactions of a succinoyl bacterial trehalose lipid (TL) biosurfactant produced by Rhodococcus sp. (Figure 1) with model proteins, namely,

Figure 1. Chemical structure of Rhodococcus sp. trehalose lipid.

BSA and Cyt-c, in order to characterize basic features of this compound and to extend its applicability to proteins studies.

■

EXPERIMENTAL METHODS

Chemicals. All reagents used were of the highest purity available. Purified water was deionized in Milli-Q equipment from Millipore (Millipore, Bedford, MA) and had a resistivity of 18 MΩ. Bovine serum albumin (BSA; essentially fatty acid and globulin free; min. 99%) and Cyt-c (min. 95%) were from Sigma Chemical Co. (St. Louis, MO). 1382


Langmuir

Article

100 mM NaCl, 100 mM phosphate, pD 7.4 (pH 7.0). NaCl was lowered to 100 mM to keep a similar osmolarity to that in the buffer used in other experiments. The use of Hepes buffer was avoided for these measurements because it presented infrared bands that could interfere with those of the proteins. Thus, the use of phosphate buffer avoided the presence of infrared bands that could interfere on BSA or Cyt-c structural determination. An aliquot of the sample (approximately 30 μL of a 0.2 mM solution) was placed between two CaF2 windows using 50 μm Teflon spacers, and the set was mounted in a thermostatted Symta cell holder. Infrared spectra were acquired in a Nicolet 6700 Fourier-transform infrared (FTIR) spectrometer (Madison, WI). Each spectrum was obtained by collecting 256 interferograms with a nominal resolution of 2 cm−1. The equipment was continuously purged with dry air to minimize the contribution peaks of atmospheric water vapor. The sample holder was thermostatted to 25 °C using a Peltier device (Proteus system from Nicolet). Spectra were collected for native and denatured proteins in the presence and absence of TL. The D2O buffer spectra were subtracted interactively using either Omnic or Grams (Galactic Industries, Salem, NH) software. A synthetic spectrum for the contribution of the Gln and Asn side chains was constructed using the data published by Chirgadze et al.34 for spectra in D2O, taking into account the primary sequences of BSA35 and Cyt-c.36 This spectrum was subtracted from the protein spectra, considering protein concentration (0.2 mM) and path length (50 μm). Derivation and Fourier self-deconvolution were applied in order to resolve the component bands of the amide I′ region of the spectrum.37,38 The secondary structure of the proteins was quantitatively determined by curve-fitting analysis of the amide I′ band, using Grams software. During the fitting procedure, the maxima of the bands, determined from deconvolution as explained above, were allowed to move ±2 cm−1. Fluorescence Spectroscopy. Fluorescence measurements were carried out in a PTI Quantamaster spectrofluorometer (Photon Tecnology, NJ). In the acrylamide quenching experiments, aliquots of a 10 M acrylamide stock solution were added to a protein stock solution to get the desired acrylamide concentration. The final concentration of protein was 0.02 mM, and the final concentration of TL was 0.2 mM. The excitation was set at 295 nm in order to excite tryptophan fluorescence only, and the emission spectrum was recorded between 300 and 400 nm. Slit widths were set at 2 nm for BSA and 5 nm for Cyt-c. The experimental data on the quenching of the tryptophan fluorescence were analyzed by conventional methods to distinguish between static quenching, which is due to formation of a nonfluorescent complex of the quencher with the fluorophore, and dynamic quenching, which is due to collisions between the fluorophore and the quencher. Since the fluorescence in most proteins is heterogeneous, a convenient way to represent fluorescence quenching data is the following form of the Stern−Volmer equation:

⎛ n ⎞−1 fi F0 ⎜ ⎟ =⎜∑ Vi[Q ] ⎟ F (1 K [ Q ]) e + ⎝ i=1 ⎠ i

Figure 2. DSC thermograms for the termal denaturation of BSA in the absence and presence of TL. The experiment was carried out under the conditions explained in Experimental Methods. The concentration of BSA was 0.05 mM. Numbers on the curves give the TL concentration (mM). Scan rate was 30 °C h−1.

induced unfolding with a midpoint temperature Tm centered at 62.9 °C and an average unfolding ΔH of 187 ± 1 kcal mol−1. This transition, as well as a new transition observed in the presence of TL and described below, were totally irreversible, as checked by successive scans after cooling the sample. Our data were in good agreement with others reported before,15,31,40,41 taking into consideration the differences in the experimental conditions. Concentrations of TL below 0.1 mM (TL/protein molar ratio 2:1) did not alter this transition. However, upon reaching a 4:1 molar ratio (0.2 mM TL), the endothermic transition was shifted to 65.9 °C, and this higher temperature endotherm was the only one present at a 6:1 molar ratio (0.3 mM TL) and higher TL concentrations, suggesting that it corresponded to the thermal unfolding of BSA interacting with TL. Further increasing TL concentration above a 4:1 molar ratio shifted the transition to 67 °C, and even a shoulder in the high temperature side became apparent. It is likely that the high temperature endothermic transition (66−67 °C) corresponded to the unfolding of BSA interacting with TL, since it became more prominent as the concentration of the biosurfactant was increased. Figure 3 shows DSC thermograms of native Cyt-c in the absence and presence of increasing concentrations of TL. Pure Cyt-c presented an endothermic thermally induced unfolding with a midpoint temperature Tm centered at 80.1 °C and an average unfolding ΔH of 123.8 ± 1 kcal mol−1, in good agreement with literature data.42−44 This transition, as well as the new transitions observed in the presence of TL, was totally irreversible as checked by successive scans after cooling the sample. Upon addition of a 2:1 TL/protein molar ratio (0.1 mM TL), a second endothermic transition centered at 63.3 °C became apparent, and reaching a 4:1 molar ratio (0.2 mM TL) resulted in the prevalence of the lower temperature endotherm at the expense of the higher temperature one. It is likely that the endotherm with higher Tm corresponded to the unfolding of pure Cyt-c molecules, and the lower Tm endotherm to the unfolding of Cyt-c rich in interacting TL. The observed increase in noise in the presence of TL might be due to the increase of protein−biosurfactant complexes involved in

(1)

where F0 and F are the fluorescence intensities in absence and in presence of a quencher, Ki and Vi are the dynamic and static quenching constants for the fluorescence component i (may represent different Trp residues of the protein), f i is the fractional contribution of component i to the total fluorescence, and [Q ] is the molar concentration of the quencher.39 Ki, Vi, and f i were determined by nonlinear fitting of the experimental data to eq 1.

■

RESULTS AND DISCUSSION Effect of Trehalose Lipid Interaction on the Thermal Unfolding of BSA and Cyt-c. Dilute protein solutions were used through the work, both with BSA and Cyt-c, to minimize protein aggregation effects. Figure 2 shows DSC thermograms of native BSA and in the presence of increasing concentrations of TL. Native BSA presented an endothermic thermally 1383


Langmuir

Article

bears at pH 7,47 producing electrostatic repulsions with the negatively charged molecules of TL. Surfactant binding to BSA is mainly driven by either electrostatic or hydrophobic interactions, and the former result suggests that it is more likely that, in our case, binding of TL is driven by hydrophobic interactions. This is supported by the fact that binding was entropy driven (ΔS > 0), since it is generally accepted that an entropy-driven reaction mainly involves hydrophobic interactions. The actual general consensus is that albumin has up to six dominant areas of ligand association which can allocate different types of amphiphilic molecules,18,20,48 and it has been reported that BSA can bind tens of small anionic surfactants such as n-decyl phosphate49 or sodium dodecyl sulfate.32 Furthermore, the ΔH of the interaction of these anionic surfactants was ca. 70 kcal mol−1 which, when compared to the values in the order of −3 kcal mol−1 that we have obtained here, and the lower affinity constant, indicated that binding of TL was much weaker, probably indicating a nonspecific interaction. In any case, the TL binding parameters were very close to those reported before for a bacterial rhamnolipid biosurfactant,36 which means that this type of glycolipids, bearing a rather voluminous polar head as compared for instance to sodium dodecyl sulfate, do not present specific interactions with BSA, but most likely interact unspecifically with hydrophobic domains within the protein. We were not able to obtain reproducible ITC binding curves of TL to Cyt-c (results not shown), indicating that either binding was negligible or it occurred without heat evolution. It was also possible that TL/native protein interactions were rather unspecific in nature, such that a proper saturating binding curve could not be obtained. Cyt-c exhibits a cationic domain involving seven Lys residues located around the exposed heme edge, where SDS micelles most likely interact,50 and, as a soluble globular protein, does not possesses extended hydrophobic domains exposed to the surface. Nevertheless, there is a hydrophobic segment in the center of the ring of lysines that is located in the immediate vicinity of the heme, and has been suggested as a potential first target for binding of SDS monomers.50 Our results suggest that monomer TL would most likely interact with hydrophobic portions of the protein in a rather unspecific way, but giving rise to the important structural changes discussed below. Trehalose Lipid-Induced Structural Changes in BSA and Cyt-c As Determined by FTIR. The structural changes produced in both proteins as a consequence of TL interaction were first studied through evaluation of the amide I′ band of the corresponding FTIR spectra. The amide I′ is a complex infrared absorption formed by various components, but the observed amide I′ bands of proteins are usually featureless due to the extensive overlap of the broad underlying component bands, which lie in close proximity to one another and cannot be instrumentally resolved. Some minor or rare structures might interfere with the band assignments. For example, the β-turn absorption at approximately 1665 cm−1 is near the characteristic infrared band representing 310-helix. Vibrations of some amino acid side chains might make small contributions to the intensity of characteristic protein amide bands. In addition, the experimental procedure might also bring spectral error. All these complications indicate that there is no simple correlation between the infrared spectra and secondary structural components, and thus, caution has to be exercised in the interpretation of infrared spectra of proteins.51,52 In our experiments, the amide I′ band was formed by components centered at approximately 1684, 1674, 1654, 1645, 1634, and 1615 cm−1 as determined by derivation and self-deconvolution

Figure 3. DSC thermograms for the termal denaturation of Cyt-c in the absence and presence of TL. The experiment was carried out under the conditions explained in Experimental Methods. The concentration of Cyt-c was 0.05 mM. Numbers on the curves give the TL concentration (mM). Scan rate was 30 °C h−1.

protein denaturation, as it has been described for mixtures with other surfactants.45,46 Binding of TL to BSA and Cyt-c. Binding of TL to BSA and Cyt-c was quantitatively studied by ITC. The data were best fitted to a model with one single binding site for both proteins, providing the fitting parameters n (number of TL molecules bound per mole of protein), K (binding constant), and ΔH (binding enthalpy). Fitting the data shown in Figure 4 indicated that BSA became saturated with one molecule of monomer TL (n = 0.72 ± 0.41)

Figure 4. ITC determination of the binding of TL to BSA. Heats of injection per mol of injected TL as a function of the TL/BSA molar ratio in the calorimeter cell. The data correspond to a BSA concentration of 0.1 mM. Data of a representative experiment are shown. The solid line is the theoretical fit.

bound per molecule of BSA, in an essentially noncooperative manner. ΔH was −2.94 ± 2.43 kcal mol−1, ΔS = 12.7 kcal mol−1 K−1, and the affinity constant K = 1.5 × 104 ± 1.5 × 104 M−1. This low binding could be due to the net negative charge that BSA 1384


Langmuir

Article

reported values ranging between 8 and 15% β-sheet with a lower proportion of unordered structures.58−60 According to these data and to the band assignment reported in the literature,55,61,62 we assigned the band around 1674 cm−1 to a helical structure (perhaps a 310-helix),54 and the bands at 1645 and 1634 cm−1 were assigned to unordered structures (extended chains), which was more in agreement with our experimental data on the effect of TL shown below. On the other hand, it is well-known that the side chains groups of Gln and Asn absorb in the amide I′ region.34,63 Thus, as explained under Experimental Methods, the FTIR spectra were corrected for this side chains contribution by subtracting the corresponding synthetic spectra. It has been suggested that the band at 1645 cm−1 could also be due to deuteration of the Asn residues;64 however, since this band was still present after deconvolution of the spectrum corrected for Asn side chain contribution, we had to conclude that it was due to unordered structures in BSA, which was in agreement with the literature data.51,55 Figure 5 shows the effect of TL on the secondary structure of fully hydrated BSA, as determined by fitting of the amide I′ band as explained above. Fitting the spectrum of native BSA at

(see Experimental Methods). These maxima were in good agreement with those reported in the literature.53−57 Correlations between the amide I′ band positions and the secondary structure in proteins have shown that the band centered at approximately 1654 cm−1 corresponds to α-helix; the one at approximately 1674 cm−1 corresponds to turns; the component at approximately 1645 cm−1 to disordered random coils, and the one at approximately 1634 cm−1 corresponds to β-sheets. The bands centered at approximately 1615 and 1684 cm−1 can be assigned to the formation of intermolecular antiparallel β-sheets resulting from temperature-dependent protein aggregation. One exception is the case of BSA. The structure of native fully hydrated BSA possesses an average of about 66% α-helix, 31% unordered structure (extended chains), and a very low proportion of β-structure; and these values may considerably vary depending on protein concentration, pH, temperature, and so forth.18−21 There is a certain contradiction in the literature concerning the structure of BSA, particularly in relation to the assignment of β-structure and extended or unordered structures. Whereas some authors have shown that BSA contains essentially no β-structure and a proportion of unordered structures around 23%,18,20,55 others have

Figure 5. Amide I′ band of native BSA at 25 °C in the absence (A) and presence of TL (C); and BSA incubated at 70 °C in the absence (B) and presence of TL (D). Final concentrations of protein and TL were 0.2 and 0.5 mM, respectively. The measurements were performed in a D2O buffer containing 100 mM NaCl, 100 mM phosphate, pD 7.4. The original BSA spectrum (long dashed line), the synthetic Gln and Asn side chain spectrum (short dashed line), and the BSA spectrum once the amino acid contribution has been subtracted (solid line) are shown. The dotted lines correspond to the component bands obtained after band fitting as explained in the text. 1385


Langmuir

Article

25 °C (Figure 5A) yielded 67.2% helical structure, 32.2% extended or unordered chains, and only 0.6% β-aggregates (Table 1), in good agreement with published data.47,61,62 Upon

heating and incubating at 70 °C, two shoulders at 1684 and 1615 cm−1 became clearly apparent (Figure 5B), indicating an increase in aggregated β-structures (9.2%) at the expense of αhelix, which decreased to 58.6%. This is in agreement with previous data showing that aggregation is accompanied by development of β-sheet.65 Nevertheless, the proportion of αhelix was still rather high upon thermal denaturation, indicating that the α-helical conformation of BSA is quite resistant to denaturation, as it was earlier shown.66 The addition of 5 mM TL to native BSA at 25 °C (Figure 5C) did not essentially affect the structure of the protein, retaining 71.1% helical structure, 28.0% unordered chains, and 0.9% aggregated βsheets. However, after incubation at 70 °C (Figure 5D), the proportion of aggregated β-structures was lower (1.2%) in the presence of the biosurfactant than in its absence. It can thus be seen that, upon heating, the decrease in helical structure was concomitant to an increase in unordered and β-structures. These results clearly showed that there was a protective effect of TL against BSA thermal denaturation. Various cationic, anionic, and nonionic surfactants, including SDS and TX-100,

Table 1. Secondary Structures of Native and Denatured BSA and Cyt-c (0.2 mM) in the Absence and Presence of TL (0.5 mM) As Determined by FTIRa secondary structure (%) sample native BSA BSA + TL denatured BSA denatured BSA + TL native Cyt-c native Cyt-c + TL denatured Cyt-c denatured Cyt-c + TL a

α-helix β-sheet unordered β-turns β-aggregates 67.2 71.1 58.6 62.7 39.2 40.5 17.1 26.6

24.5 29.8 46.0 33.2

32.2 28.0 32.2 36.1 23.1 20.8 14.2 7.0

12.4 8.9 18.7 13.2

0.6 0.9 9.2 1.2 0.8 0 3.9 20

Data of one representative experiment are shown.

Figure 6. Amide I′ band of native Cyt-c at 25 °C in the absence (A) and presence of TL (C); and Cyt-c incubated at 90 °C in the absence (B) and presence of TL (D). Final concentrations of protein and TL were 0.2 and 0.5 mM, respectively. Measurements were performed in a D2O buffer containing 100 mM NaCl, 100 mM phosphate, pD 7.4. The original Cyt-c spectrum (long dashed line), the synthetic Gln and Asn side chain spectrum (short dashed line), and the Cyt-c spectrum once the amino acid contribution has been subtracted (solid line) are shown. The dotted lines correspond to the component bands obtained after band fitting as explained in the text. 1386


Langmuir

Article

likely of different nature than those involved in TL/BSA interaction. Effect of TL on Trp Intrinsic Fluorescence. The effect of TL on the structure of BSA and Cyt-c was also determined through studying the effect of TL interaction on the Trp intrinsic fluorescence of the proteins. Figure 7A shows the effect of TL on the Trp emission spectrum of native BSA. The pure protein showed an emission spectrum centered at 342 nm. Addition of increasing concentrations of TL gave rise to a progressive quenching of Trp fluorescence. Since TL cannot act as a direct Trp fluorescence quencher, this fluorescence attenuation should be the indirect consequence of the TL-induced protein conformational changes described above by FTIR. A similar behavior has been described for the interaction of Tween 80 and Tween 20 with BSA.70 In addition, as the concentration of TL was increasing from 0 to 500 μM, the maximum emission wavelength underwent a moderate blue shift from 342 to 338 nm, compatible with the movement of the Trp residues into a more hydrophobic environment, either as a consequence of conformational changes, or by direct interaction with hydrophobic portions of the biosurfactant. In a second set of experiments, acrylamide was used as quencher, as it is a nonperturbing quencher which usually does not bind to proteins. In this case, fluorescence quenching usually proceeds via physical contacts between the quenchers and the fluorophores and hence is directly dependent on the extent to which the fluorophore can be approached. Thus, the extent of accessibility of Trp residues, as well as ligand binding sites in proteins, can be probed by quenching studies. Figure 7B shows Stern−Volmer plots for acrylamide quenching of BSA Trp intrinsic fluorescence under various conditions. Experimental data were fitted according to eq 1, and in all cases the contribution of static quenching was essentially negligible, also for the Cyt-c data shown below. The plots for native BSA in the absence and presence of TL were essentially linear, whereas denatured BSA gave rise to slightly curved plots. At first sight, it could be observed that the plot corresponding to BSA denatured in the absence of TL presented a slope quite

have been shown to induce substantial changes in the helical content of BSA,15,62 indicative of a denaturing effect. The interaction of a dirhamnolipid biosurfactant secreted by Pseudomonas aeruginosa with BSA caused a similar shift in the thermal unfolding, also decreasing the proportion of β-aggregation upon denaturation,41 suggesting that this type of glycolipid compounds, including our TL, would probably interact with the same regions within the protein. The fact that interaction of TL with BSA resulted in a protection effect opened the possibility of practical applications of this biosurfactant in protein studies. According to the literature, Cyt-c possesses an average of about 35−45% α-helix, 12−39% unordered structure, 10−25% β-sheet, and 14−25% turns.51,67−69 Figure 6 shows the effect of TL on the secondary structure of Cyt-c. It can be seen that addition of TL to native Cyt-c did not essentially alter the shape of the amide I′ band (Figure 6A and C). Upon denaturation band shape changed considerably (Figure 6B), and this modification was much more pronounced in the presence of TL (Figure 6D). Fitting the spectrum of native Cyt-c at 25 °C (Figure 6A) yielded a 39.2% α-helical structure, 24.5% βstructure, 23.1% unordered, 12.4% β-turns, and just 0.8% βaggregates (Table 1), in good agreement with the published results as commented above. Upon incubation to 90 °C (Figure 6B), an increase in β-sheet (46.0%) and aggregated β-structures (3.9%) at the expense of α-helix, which decreased to 17.1%, and disordered structures, which decreased to 14.2%, was observed. The addition of 5 mM TL to native Cyt-c at 25 °C (Figure 6C) did not essentially affect the structure of the protein, which retained a 40.5% of helical structure, 20.8% unordered chains, 29.8% β-sheets, and 8.9% turns. However, upon incubation at 90 °C in the presence of TL (Figure 6D), the proportion of aggregated β-structures was higher (20%), whereas the proportions of β-sheet and turns were lower. These results showed that the presence of TL destabilized the protein, an effect opposite to that shown above for BSA, suggesting that interaction of TL with Cyt-c involves structural motifs most

Figure 7. Effect of TL on BSA intrinsic Trp fluorescence. (A) Native BSA emission spectra in the absence and presence of increasing concentrations of TL. BSA concentration was 20 μM. (B) Stern−Volmer plots for the acrylamide fluorescence quenching of native BSA (25 °C) and BSA incubated at 70 °C (denatured), in the absence and presence of TL. (●) Native BSA, (○) denatured BSA, (■) native BSA in the presence of TL, and (□) denatured BSA in the presence of TL. Protein and TL concentrations were 20 and 200 μM, respectively. 1387


Langmuir

Article

Tryptophans are located at positions 134 and 212. Trp-134 has been proposed to lie near the surface of BSA in the second helix of the first domain, being more solvent-exposed than Trp212.71,72 In this respect, our results suggested that the conformational changes occurred during protein aggregation upon denaturation resulted in Trp-134 becoming more buried, a process which was diminished by binding of TL to appropriate protein motifs. Similar fluorescence experiments were performed for Cyt-c. According to the literature, Cyt-c contains a single tryptophan residue (Trp-59) which is localized in the deepest part in a crevice where the heme moiety is located. Trp-59 is hydrogen bonded to one of the propionic groups of heme;73−75 thus, the fluorescence signal is weak because of the proximity of Trp-59 to the heme group.76 It was observed (Figure 8A) that addition of increasing concentrations of TL to native Cyt-c resulted in a progressive fluorescence increase. This result suggested that interaction of TL induced Cyt-c unfolding, resulting in a conformational change which increased Trp-heme distance, thus decreasing resonance energy transfer efficiency and increasing fluorescence intensity of the Trp donor. A similar fluorescence increase has been recently described upon addition of SDS to Cyt-c, and the same sort of explanation was suggested.77 Acrylamide quenching results are shown in Figure 8B. For folded Cyt-c, the corresponding Stern−Volmer plot was essentially linear, clearly indicating a homogeneous population of fluorophores. In fact, the data shown in Table 2 indicated the presence of a unique fraction with K1 = 1.1 M−1, indicating an intermediate accessibility. Protein unfolding upon denaturation in the absence of TL resulted in curved Stern−Volmer plots (Figure 8A), with a major (88%) poorly accessible fraction (K2 = 0.10 M−1), most likely due to the formation of β-aggregates shown above (Table 1). In agreement with the DSC, FTIR, and fluorescence results shown above, addition of TL to native Cyt-c gave rise to a conformation change which resulted in a predominant population of protein molecules (84%) with essentially inaccessible Trp residues (K2 = 2.76 × 10−10 M−1), although a small proportion of highly accessible fluorophores (16%) still remained. Denaturation in the presence of TL did

different from the rest. Data fitting to eq 1 allowed determination of the Stern−Volmer dynamic quenching constant, as well as the fractions of accessible fluorophores in the case of heterogeneous fluorescence (Table 2). As observed, Table 2. Dynamic Fluorescence Constants (K1, K2) and Fractional Contributions of the Corresponding Components (f1, f 2), for the Acrylamide Fluorescence Quenching of BSA and Cyt-c under Various Conditionsa sample

f1

K1 (M‑1)

f2

native BSA denatured BSA native BSA + TL denatured BSA + TL native Cyt-c denatured Cyt-c native Cyt-c + TL denatured Cyt-c + TL

1 0.18 1 0.87 1 0.12 0.16 0.13

3.70 5.80 3.30 2.70 1.10 27.40 47.30 119.50

0 0.82 0 0.13 0 0.88 0.84 0.87

K2 (M‑1) 1.6 89.50 0.10 2.76 × 10−10 0.60

a

Final concentrations of protein and TL were 0.02 and 0.2 mM, respectively. Data of one representative experiment are shown.

native BSA presented a single population of Trp fluorophores with a dynamic quenching constant of 3.7 M−1. Addition of TL to native BSA did not essentially affect the quenching parameters (K1 = 3.30 M−1); that is, Trp accessibility was not affected, indicating that the biosurfactant did not essentially modify protein structure, as also shown above by FTIR. Protein denaturation in the absence of TL resulted in the conversion of a high proportion of Trp fluorophores (82%) into a fraction with a lower quenching constant (1.60 M−1), consistent with the increase in β-aggregates observed by FTIR (Table 1), which decreased Trp accessibility to the quencher. Very interestingly, denaturation in the presence of TL maintained a high proportion of Trp residues (87%) with a rather high quenching constant (2.70 M−1), confirming the protective effect of TL in the denaturation process shown above, which decreased formation of β-aggregates. BSA contains 580 amino acid residues and is characterized by a low content of tryptophan.

Figure 8. Effect of TL on Cyt-c intrinsic Trp fluorescence. (A) Native Cyt-c emission spectra in the absence and presence of increasing concentrations of TL. Cyt-c concentration was 20 μM. (B) Stern−Volmer plots for the acrylamide fluorescence quenching of native Cyt-c (25 °C) and Cyt-c incubated at 90 °C (denatured), in the absence and presence of TL. (●) Native Cyt-c, (○) denatured Cyt-c, (■) native Cyt-c in the presence of TL, and (□) denatured Cyt-c in the presence of TL. Protein and TL concentrations were 20 and 200 μM, respectively. 1388


Langmuir

Article

(5) Lang, S. Curr. Opin. Colloid Interface Sci. 2002, 7, 12−20. (6) Asselineau, C.; Asselineau. J. Prog. Chem. Fats Other Lipids 1978, 16, 59−99. (7) Lang, S.; Philip, J. C. Antonie van Leeuwenhoek 1998, 74, 59−70. (8) Ramsay, B. A.; Cooper, D. G.; Margaritis, A.; Zajic, J. E. In Microbial Enhanced Oil Recovery; Zajic, J. E., Cooper, D. G., Jack, T. R., Kosaric, N., Eds.; Pann Well Books: Tulsa, OK, 1983; p 61. (9) Singer, M. E. V.; Finnerty, W. R.; Tunelid, A. Can. J. Microbiol. 1990, 36, 746−750. (10) Sudo, T; Zhao, X.; Wakamatsu, Y.; Shibahara, M.; Yokoyama, K. K. Cytotechnology 2000, 33, 259−264. (11) Isoda, H.; Kitamoto, D.; Shinmoto, H.; Matsumura, M.; Nakahara, T. Biosci. Biotechnol. Biochem. 1997, 61, 609−614. (12) Vasilescu, M; Angelescu, D.; Almgren, M.; Valstar, A. Langmuir 1999, 15, 2635−2643. (13) Valstar, A; Almgren, M.; Brown, W.; Vasilescu, M. Langmuir 2000, 16, 922−927. (14) Moriyama, Y.; Takeda, K. Langmuir 2005, 21, 5524−5528. (15) Singh, S. K.; Kishore, N. J. Phys. Chem. B 2006, 110, 9728− 9737. (16) Orioni, B.; Roversi, M.; La Mesa, C.; Asaro, F.; Pellicer, G.; D́ Errico, G. J. Phys. Chem. B 2006, 110, 12129−12140. (17) Vijai, K.; Foster, J. Biochemistry 1967, 6, 1152−1159. (18) Carter, D. C.; He, X. M. Science 1990, 249, 302−303. (19) He, X. M.; Carter, D. C. Nature 1992, 358, 209−215. (20) Carter, D. C.; Ho, J. X. Adv. Protein Chem. 1994, 45, 153−203. (21) Carter, D. C.; He, X. M.; Munson, S. H.; Twigg, P. D.; Gernert, K. M.; Broom, M. B.; Miller, T. Y. Science 1989, 244, 1195−1198. (22) Margoliash, E.; Schejter, A. Adv. Protein Chem. 1966, 21, 113− 286. (23) Bushnell, G. W; Louie, G. V; Brayer, G. D. J. Mol. Biol. 1990, 214, 585−595. (24) Naeem, A.; Hasan, R. Science 2004, 36, 2281−2292. (25) Stevens, J. M. Metallomics 2011, 3, 319−322. (26) Liu., X.; Naekyung, C.; Yang, J.; Jemmerson, R.; Wang, X. Cell 1996, 86, 147−157. (27) Cai, J.; Yang, J.; Jones, D. Biochim. Biophys. Acta 1998, 1366, 139−149. (28) Ananthapadmanabhan, K. P. Interactions of Surfactants with Polymers and Proteins; CRC Press: London, 1993. (29) Lindman, B.; Karlström, G. Polymer Surfactant Systems. In The Structure, Dynamics and Equilibrium Properties of Colloidal System, NATO ASI Ser. C; Bloor, D. M., Wyn-jones, E., Eds.; Kluwer Academic: Dordrecht, 1990; Vol. 324. (30) Aranda, F. J.; Teruel, J. A.; Espuny, M. J.; Marqués, A.; Manresa, A; Palacios-Lidón, E.; Ortiz, A. Biochim. Biophys. Acta 2007, 1768, 2596−2604. (31) Giancola, C.; De Sena, C.; Fessas, D.; Graziano, G.; Barone, G. Int. J. Biol. Macromol. 1997, 20, 193−204. (32) Deep, S.; Ahluwalia, J. C. Phys. Chem. Chem. Phys. 2001, 3, 4583−4591. (33) Grasso, d.; La Rosa, C.; Milardi, D.; Fasone, S. Thermochim. Acta 1995, 265, 163−175. (34) Chirgadze, Y. N.; Fedorov, O. V.; Trushina, N. P. Biopolymers 1975, 14, 679−694. (35) Hirayama, K.; Akashi, S.; Furuya, M.; Fukuhara, K. Biochem. Biophys. Res. Commun. 1990, 173, 639−646. (36) Bushnell, G. W.; Louie, G. V.; Brayer, G. D. J. Mol. Biol. 1990, 214, 585−595. (37) Kauppinen, J. K.; Moffatt, D. J.; Mantsch, H. H.; Cameron, D. G. Anal. Chem. 1981, 53, 1454−1457. (38) Cameron, D. G.; Moffatt, D. J. J. Test. Eval. 1984, 12, 78−85. (39) Eftink, M. R.; Ghiron, C. A. Anal. Biochem. 1981, 114, 199−227. (40) Almeida, N. L.; Oliveira, C. L.P.; Torriani, I. L; Loh, W. Colloids Surf., B 2004, 38, 67−76. (41) Sánchez, M.; Aranda, F. J.; Espuny, M. J.; Marqués, A.; Teruel, J. A.; Manresa, A.; Ortiz, A. Langmuir 2008, 24, 6487−6495. (42) Privalov, P. L.; Khechinashvili, N. N. J. Mol. Biol. 1974, 86, 665− 684.

not show the protective effect shown above in the case of BSA. Altogether, our results indicated that TL bound or interacted with Cyt-c increased the proportion of unordered structures at the expense of α-helix, facilitated β-aggregation (Table 1), and resulted in a denaturing effect as observed by the corresponding decrease in Trp accessibility (Table 2).

■

GENERAL CONCLUSIONS The relevance of the studies on protein−surfactant interactions comes from the manifold applications of protein−surfactant systems: for example, for the food and pharmaceutical industries, and in analytical biochemistry. BSA is rich in αhelical structures and contains a very low proportion of βstructure, whereas Cyt-c possesses between 10 and 25% βstructure, and both proteins have been widely used as model proteins. Our data have shown that, in the case of BSA, there is no denaturation effect and, on the contrary, protein is stabilized against thermal denaturation upon interaction with TL. The interaction is rather unspecific, as seen by ITC binding, but sufficiently intense as to produce an important stabilization of the protein against thermal denaturation. It is likely that the localization of BSA into a more hydrophobic environment by incorporation, for instance, into TL micelles, probably also contributed to the above-mentioned stabilization. On the other hand, the Cyt-c−TL interaction is also very unspecific (ITC), but yet sufficiently strong as to promote protein denaturation, an opposite effect to that observed for BSA. Taking into account the structures of both proteins, it can be derived from our data that interaction of TL with BSA, a α-helix rich protein, protects toward thermal unfolding, whereas interaction with Cyt-c, with a higher β-sheet content, facilitates thermal unfolding. The fact that TL seems to affect more strongly thermally denatured than native proteins might suggest a preferential interaction with hydrophobic domains, which are buried in the native structures and become exposed upon unfolding. In any case, the different nature of the domains involved in interaction with BSA or Cyt-c seems to be clear. Since surfactants are often added to prevent and/or minimize protein aggregation during various processes, the distinct effect of TL on BSA versus Cyt-c shown here indicates that the effects of this biosurfactant will strongly depend on the particular structural characteristics of the protein sample, and studies of the type shown here should be carried out prior to its practical application in particular cases.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +34-868-884788. Fax: +34-868-884147. E-mail: [email protected].

■

ACKNOWLEDGMENTS This work was supported by Project No. CTQ2007-66244 (to A.O.), from the Spanish Ministry of Science and Innovation (MCINN).

■

REFERENCES

(1) Desai, J. D.; Banat, I. M. Microbiol. Mol. Biol. Rev. 1997, 61, 47−64. (2) Cameotra, S. S.; Makkar, R. S. Appl. Microbial. Biotechnol. 1998, 50, 520−529. (3) Singh, P.; Cameotra, S. S. Trends. Biotechnol. 2004, 22, 142−146. (4) Rodrigues, L.; Banat, I. M.; Teixeira, J.; Oliveira, R. J. Antimicrob. Chemother. 2006, 57, 609−618. 1389


Langmuir

Article

(43) Zhang, F.; Rowe, E. Biochim. Biophys. Acta 1994, 1193, 219− 225. (44) Makhatadze, G. I.; Privalov, P. L. Adv. Protein Chem. 1995, 47, 307−425. (45) Turro, N. J.; Lei, X. Langmuir 1995, 11, 2525−2533. (46) Andersen, K. K.; Westh, P.; Otzen, D. E. Langmuir 2008, 24, 399−407. (47) Peters, T. Adv. Protein Chem. 1985, 37, 161−245. (48) De, S.; Girigoswami, A.; Das, S. J. Colloid Interface Sci. 2005, 285, 562−573. (49) Jones, M. N.; Skinner, H. A.; Tipping, E. Biochem. J. 1975, 147, 229−234. (50) Oellerich, S.; Wackerbarth, H.; Hildebrandt, P. Eur. J. Biophys. 2003, 32, 599−613. (51) Ye, M.; Zhang, Q. L.; Weng, Y. X.; Wang, W. C.; Qiu, X. G. Biophys. J. 2007, 93, 2756−2766. (52) Kong, J.; Yu, S. Biochim. Biophys. Acta 2007, 39, 549−559. (53) Le Gal, J. M.; Manfait, M. Biochim. Biophys. Acta 1990, 1041, 257−263. (54) Jackson, M.; Mantsch, H. H. Crit. Rev. Biochem. Mol. Biol. 1995, 30, 95−120. (55) Murayama, K.; Tomida, M. Biochemistry 2004, 43, 11526− 11532. (56) Zhang, J.; Yan, Y. B. Anal. Biochem. 2005, 340, 89−98. (57) Panuszko, A.; Bruzdziak, P.; Zielkiewicz, J.; Wyrzykowski, D.; Stangret, J. J. Phys. Chem. 2009, 113, 14797−14809. (58) Sjoholm, I.; Ljungstedt, I. J. Biol. Chem. 1973, 248, 8434−844. (59) Imamura, K.; Ohyama, K.; Yokoyama, T.; Maruyama, Y.; Kazuhiro, N. J. Pharm. Sci. 2009, 98, 3088−3098. (60) Charbonneau, D. M.; Tajmir-Riahi, H. A. J. Phys. Chem. B 2010, 114, 1148−1155. (61) Wetzel, R.; Becker, M.; Behlke, J.; Billwitz, H.; Böhm, S.; Ebert, B.; Hamann, H.; Krumbiegel, J.; Lassmann, G. Eur. J. Biochem. 1980, 104, 469−478. (62) Takeda, K.; Shigeta, M.; Aoki, K. J. Colloid Interface Sci. 1987, 117, 120−126. (63) Venyaminov, S. Y.; Kalnin, N. N. Biopolymers 1990, 30, 1259− 1271. (64) Besson, F.; Bouchet, R. Spectrochim. Acta, Part A 1997, 53, 1913−1923. (65) Clark, A. H.; Saunderson, D. H. P.; Suggett, A. Int. J. Pept. Protein Res. 1981, 17, 353−364. (66) Harmsen, B. J. M.; Braam, W. J. M. Int. J. Protein Res. 1969, 1, 225−233. (67) Dousseau, F.; Pézolet, M. Biochemistry 1990, 29, 8771−8779. (68) Dong, A.; Huang, P.; Caughey, W. S. Biochemistry 1992, 31, 182−189. (69) Bushnell, G. W.; Louie, G. V.; Brayer, G. D. J. Mol. Biol. 1990, 214, 585−595. (70) Gelamo, E. L.; Tabak, M. Spectrochim. Acta, Part A 2000, 56, 2255−2271. (71) Ruiz-Peña, M; Oropesa-Nuñez, R.; Pons, T.; Louro, S. R. W.; Pérez-Gramatges, A. Colloids Surf., B 2010, 75, 282−289. (72) Samson, N. Q.; Bugante, R. V.; Gagalac, M. N. J. Health Sci. 2005, 51, 8−15. (73) Fisher, W. R.; Taniuchi, H.; Anfinsen, C. B. J. Biol. Chem. 1973, 248, 3188−3195. (74) Stellwagen, E.; Rysavy, R.; Babul, G. J. Biol. Chem. 1972, 247, 8074−8077. (75) Myer, Y. P.; MacDonald, L. H.; Verma, B. C.; Pande, A. Biochemistry 1980, 19, 199−207. (76) Schlamadinger, D. E.; Kats, D. I.; Kim, J. E. J. Chem. Educ. 2010, 87, 961−964. (77) Ahluwalia, U.; Nayeem, S. M.; Deep, S. Eur. Biophys. J. 2011, 40, 259−271.

1390


Interaction of a Rhodococcus sp. Trehalose Lipid Biosurfactant with

Recommend Documents