J. Phys. Chem. B 2007, 111, 2727-2735
2727
Interfacial Behavior of N-Nitrosodiethylamine/Bovine Serum Albumin Complexes at the Air-Water and the Chloroform-Water Interfaces by Axisymmetric Drop Tensiometry J. Jua´ rez,† J. G. Galaz,‡ L. Machi,† M. Burboa,†,§ L. E. Gutie´ rrez-Milla´ n,†,§ F. M. Goycoolea,| and M. A. Valdez*,†,‡ Departamento de InVestigacio´ n en Polı´meros y Materiales, Departamento de Fı´sica, and Departamento de InVestigaciones Cientı´ficas y Tecnolo´ gicas, UniVersidad de Sonora, Rosales y TransVersal, 83000 Hermosillo Sonora, Me´ xico, and Departamento de Farmacoloxı´a, UniVersidad de Santiago de Compostela, Campus UniVersitario Sur s/n Pabello´ n C, Santiago de Compostela, A Corun˜ a 15782, Galicia, Espan˜ a ReceiVed: September 16, 2006; In Final Form: January 16, 2007
Interfacial properties of N-nitrosodiethylamine/bovine serum albumin (NDA/BSA) complexes were investigated at the air-water interface. The interfacial behavior at the chloroform-water interface of the interaction product of phospholipid 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), dissolved in the chloroform phase, and NDA/BSA complex, in the aqueous phase, were also analyzed by using a drop tensiometer. The secondary structure changes of BSA with different NDA concentrations were monitored by circular dichroism spectroscopy at different pH and the NDA/BSA interaction was probed by fluorescence spectroscopy. Different NDA/ BSA mixtures were prepared from 0, 7.5 × 10-5, 2.2 × 10-4, 3.7 × 10-4, 5 × 10-4, 1.6 × 10-3, and 3.1 × 10-3 M NDA solutions in order to afford 0, 300/1, 900/1, 1 500/1, 2 000/1, 6 000/1, and 12 500/1 NDA/BSA molar ratios, respectively, in the aqueous solutions. Increments of BSA R-helix contents were obtained up to the 2 000/1 NDA/BSA molar ratio, but at ratios beyond this value, the R-helix content practically disappeared. These BSA structure changes produced an increment of the surface pressure at the air-water interface, as the R-helix content increased with the concentration of NDA. On the contrary, when R-helix content decreased, the surface pressure also appeared lower than the one obtained with pure BSA solutions. The interaction of DPPC with NDA/BSA molecules at the chloroform-water interface produced also a small, but measurable, pressure increment with the addition of NDA molecules. Dynamic light scattering measurements of the molecular sizes of NDA/BSA complex at pH 4.6, 7.1, and 8.4 indicated that the size of extended BSA molecules at pH 4.6 increased in a greater proportion with the increment in NDA concentration than at the other studied pH values. Diffusion coefficients calculated from dynamic surface tension values, using a short-term solution of the general adsorption model of Ward and Tordai, also showed differences with pH and the NDA concentration. Both, the storage and loss dilatational elastic modulus were obtained at the air-water and at the chloroform-water interfaces. The interaction of NDA/BSA with DPPC at the chloroform-water produced a less rigid monolayer than the one obtained with pure DPPC (1 × 10-5 M), indicating a significant penetration of NDA/BSA molecules at the interface. At short times and pH 4.6, the values of the storage elastic modulus were larger and more sensible to the NDA addition than the ones at pH 7.1 and 8.4, probably due to a gel-like network formation at the air-water interface.
Introduction Investigation of macromolecules’ (proteins and DNA) adducts has become one of the most important research fields due to the direct or indirect connection with the carcinogenesis phenomenon. The protein and DNA adducts measurement is a form of chemical dosimetry, which can be used to know the history of exposure to genotoxic substances.1 Albumin is one of the most investigated proteins, and albumin adducts have been investigated for a long time. Bahl and Gutman in 1964 performed in vivo experiments to analyze the binding of rat serum albumin with N-2 fluorenylacetamide.2 Because of the * To whom correspondence should be addressed. E-mail valdez@ fisica.uson.mx. Fax: +6622592109. † Departamento de Investigacio ´ n en Polı´meros y Materiales, Universidad de Sonora. ‡ Departamento de Fı´sica, Universidad de Sonora. § Deparamento de Investigaciones Cientificas y Tecnolo ´ gicas, Universidad de Sonora. | Universidad de Santiago de Compostela.
biological importance of serum albumin and the facility to obtain it, many investigations of serum albumin adducts have been performed.3,4 Recently Boysen and Hecht have reviewed DNA and protein adducts of benzo(a)pyrene.5 Protein interacting with water soluble carcinogens has also been focus of much attention since long time ago; Bemis et al. in 1966 investigated the effect of N-nitrosodiethylamine (NDA) and N-nitrosomethylamine in protein denaturation by using optical rotation, circular dichroism (CD), and light-scattering techniques.6 NDA is present in foods, beverages, tobacco smoke, herbicides, pesticides, drinking water, and industrial pollution.7 Gorsky and Hellenberg8 demonstrated the preferential labeling of albumin secreted from isolated rat hepatocytes during the metabolism of NDA. Most of the studies on protein adducts have aimed to understand their biochemical and physicochemical properties in solution,9-11 and far less attention has been paid to the surface and interfacial properties of these complexes. On the contrary, investigations of surface properties of pure proteins, such as bovine serum albumin (BSA),12 R-lactalbumin, β-lac-
10.1021/jp066061m CCC: $37.00 © 2007 American Chemical Society Published on Web 02/22/2007
2728 J. Phys. Chem. B, Vol. 111, No. 10, 2007 toglobulin,13 lysozyme, and others,14 have received reasonable attention. The influence of pH on the surface properties of human serum albumin (HSA) performed by axisymmetric shape analysis were investigated by Hansen and Myrvold15 and demonstrated that, at lower pH, HSA is strongly adsorbed at the air-water interface, probably due to its expanded structure. The changes of BSA conformation due to the influence of pH and temperature have been investigated by several authors.16 Also the rheological properties of BSA at the air-water interface are strongly dependent on the pH as was demonstrated by Cascao-Pereira et al.12 They showed that the dilatational elastic modulus is higher at pH near the BSA isoelectric point (IP) in comparison with the modulus above and below the IP. The BSA adsorption onto the hydrophilic silica-water interface is decreased at pH above and below the BSA IP, as was demonstrated by Su et al.17 On the other hand, the interaction of proteins with lipids and phospholipids at interfaces has been investigated by several authors17-20 due to their role as structural elements in biological membranes.21,22 The interaction of BSA23,24 and HSA22,25-27 with phospholipids as membrane models has also been investigated due to its importance in deepening the knowledge of the serum albumin-membrane interaction. In light of the above, we considered it important to understand the influence of adducts on the surface properties of proteins and the interactions of protein adducts with different phospholipids at interfaces. In this paper we have, therefore, addressed the effect of the carcinogen NDA on the interfacial properties of BSA at the interface air-water and investigated the interaction of the complexes NDA/BSA with the phospholipid 1,2dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) at the chloroform-water interface. To this end, we performed axial symmetric drop tensiometry measurements of the complexes NDA/BSA with different NDA concentrations. Because of the fact that BSA takes different conformations at different pH,12,16 we analyzed the short time behavior of the influence of NDA on the BSA adsorption at the air-water interface and measured the complex dilatational elastic modulus of the adsorbed complexes at the interface for different pH. By dynamic light scattering (DLS), we compared the change of hydrodynamic radius of the NDA/BSA complexes at different pH. CD of the NDA/BSA complexes was performed to investigate the change of BSA secondary structure in pure water and in presence of a buffer solution. The type of bonding in the NDA/BSA complex was studied by fluorescence measurements and the Van’t Hoff equation.11 At times when both the surface tension and the elastic modulus were practically constant, we measured the effect of NDA on the BSA adsorption at the air-water interface and at the chloroform-water interface. At the chloroform-water interface, we analyzed the interaction of NDA/BSA with DPPC dissolved in the chloroform phase. The characterization of the phospholipid/protein adsorption at the aqueous solution/ chloroform interface was analyzed by Wu et al.21 with different proteins and demonstrated the important role of the kind of the lipid head group and the lipid concentration. However, they did not measure the elastic modulus of the interaction between lipids and the adsorbed proteins. In this work we used the Maxwell model28 and determined the relaxation times of the NDA/BSA complexes at both interfaces. Experimental BSA was obtained from BD Biosciences (99%, delipidized, and globulins free), and DPPC was obtained from Avanti Polar Lipids. Both were used without further treatment. NDA was
Jua´rez et al. obtained from TCI America (99%) and chloroform (highperformance liquid chromatography grade) and all other chemical reagents (reactive grade) were purchased from Sigma-Aldrich (U.S.A.). Water used throughout the experiments was deionized by an Easy pure/Barnstead instrument with a resistivity of 18.3 MΩ cm. BSA solutions at pH 4.6, 7.1, and 8.4 were prepared in a 10 mM phosphate buffered solution. The final pH was adjusted with diluted NaOH and HCl solutions. DLS Measurements. The size of BSA molecules and NDA/ BSA complexes was estimated with the DLS technique. Measurements were performed using an ALV-5000 digital correlator system (Langen-GmbH, Germany) fitted with a temperature control set at 25 ( 0.1 °C. The incident light was vertically polarized with a λ0 ) 632 nm argon laser (30 mW), and the scattered light was measured at 90°. The hydrodynamic radius, RH, was obtained for diluted samples through the Stokes-Einstein relation D0 ) kBT/6πηRH, where kB is the Boltzmann constant, T the absolute temperature, η the viscosity of the solvent, and D0 the diffusion coefficient at infinite dilution. Average hydrodynamic radius of BSA diluted solutions and two different NDA/BSA proportions were determined at pH 4.6, 7.1, and 8.4. All measurements were conducted in triplicate, each one recorded in a time interval of 100 s and averaged with the software of the instrument (ALV 5000/E/WIN Software). The average size of the NDA/BSA molecules was measured about 30 min after the NDA solution was added to the BSA solution. All solutions were filtered with a 0.2 µm filter. Surface Tension and Rheology Measurements. Pendant bubble tensiometry was used to determine the dynamic interfacial tension and the interfacial dilatational rheology measurements at the air-water interface with the air bubble in the upward direction. The bubble was formed at the tip of a U-shaped stainless steel needle (0.5 mm inside diameter (i.d.)) immersed in the aqueous BSA or the NDA/BSA complexes solutions. The equipment used was a Tracker tensiometer (I. T. Concept, France) capable of real-time surface tension with an accuracy of 0.1 mN/m. For the analysis of the chloroformwater interface, a vertical stainless steel needle (0.5 mm i.d.) mounted on a syringe produces a pendent drop and allows changing the drop size by a mechanically controlled system. The needle is immersed into a quartz cuvette (103.051F-Og, 20-10, Hellma, Germany) and filled with the BSA or the NDA/ BSA complex solutions to produce the chloroform/water interface. The measurement of the interfacial tension and the rheological measurements are based on the digital profile of the drop image and the solution of the Gauss-Laplace equation. The software used for the axisymmetric drop shape analysis is the Win Drop software (I. T. concept, France). Temperature was kept constant at 25 ( 0.1 °C. BSA concentration was kept constant (2.5 × 10-7 M) in all performed experiments. Different NDA/BSA mixtures were prepared from NDA solutions of concentration 0, 7.5 × 10-5, 2.2 × 10-4, 3.7 × 10-4, 5 × 10-4, 1.6 × 10-3, and 3.1 × 10-3 M in order to afford 0, 300/1, 900/1, 1 500/1, 2 000/1, 6 000/1, and 12 500/1 NDA/BSA molar ratios, respectively, in the aqueous solutions. For the analysis of the interfacial tension (γ), the dilatational elastic modulus (E′), and viscous modulus (E′′) at the air-water interface, the measurements were started after 1800 s. The used frequencies were 1, 0.5, 0.25, 0.1, 0.05, and 0.033 s-1. After this time, all parameters were measured in a time interval of 400 s by oscillating drops with a 10% maximum drop volume increment. All measurements were repeated at least three times.
Axisymmetric Drop Tensiometry of NDA/BSA
J. Phys. Chem. B, Vol. 111, No. 10, 2007 2729
For the behavior of γ and E′ vs time at different pH, we averaged five measurements with an average error of 1%. For the measurements of γ, E′, and E′′ at the chloroformwater interface, we used a 1 × 10-5 M DPPC dissolved in chloroform and the same aqueous BSA and NDA/BSA solutions mentioned above. The chloroform evaporation was not taken into account. We assumed that the software program calculates the drop area and surface tension at each time and the changes due to evaporation are negligible during each measurement. Spectroscopic Methods. Fluorescence spectra were performed with a Perkin-Elmer Luminescence Spectrometer LS 50B. The emission spectra were recorded from 300 to 500 nm (excitation wavelength 280 nm) and the obtained intensity at 340 nm, for the different samples of BSA solutions (2.5 × 10-7 M) and the different NDA/BSA complexes, was plotted for the different NDA/BSA molar ratios at different temperatures. The NDA concentrations were 7.5 × 10-5, 1.5 × 10-4, 2.2 × 10-4, and 3.7 × 10-4 M. All experiments were measured at four temperatures (293, 299, 305, and 310 K). The temperature sample was maintained by a recirculation water bath. CD measurements were performed in a JASCO J-810 spectropolarimeter at room temperature. A baseline was obtained under the same experimental conditions as the samples and subtracted from the spectra of the samples. The BSA concentration was 0.1 mg/mL, and the molar ratio of NDA/BSA was the same as in the experiments at interfaces. The molecular weight of BSA used for the secondary structure calculations was 66 500. Both sample and reference were run three times and averaged. The wavelength range used was 180-260 nm, and the cell path length was 0.1 cm. The secondary structure was determined by using the DICHROWEB software with the program CDSSTR (http://public-1.cryst.bbk.ac.uk/cdweb/html/). Theory The fluorescence intensity values were used to obtain the relative fluorescence at different temperatures and different NDA concentrations (quencher). The temperatures chosen were 293, 299, 305, and 311 K. We analyzed the intensities with the Stern-Volmer equation11
Fo ) 1 + KSV[Q] F
(1)
where F0 and F are the steady-state fluorescence intensities in absence and presence of the quencher, respectively. KSV is the Stern-Volmer constant, and [Q] is the concentration of the quencher. The thermodynamic parameters, important to determine the binding modes between the protein and the quencher, were obtained assuming that the enthalpy change (∆H0) does not vary significantly in the range of temperatures analyzed. We determined the entropy change (∆S0) with the Van’t Hoff equation
ln(KSV) ) -
∆H0 ∆S0 + RT R
(2)
where KSV is the Stern-Volmer quenching constant at the corresponding temperature and R is the ideal constant gas. The free energy change due to the protein-quencher interaction, (∆G0), is calculated from the known thermodynamic relationship
∆G0 ) ∆H0 - T∆S0
(3)
Sinusoidal perturbations of interfaces with a drop tensiometer.
For a given sinusoidal drop pulsation of frequency ω produced by the Tracker tensiometer at every time t, we obtain a surface tension value γ(t) by analyzing the axial symmetric shape of the pendent drop and the drop area A(t), which are used to calculate the complex dilatational elasticity modulus E(ω), defined as29
dγ E ) -A dA
(4)
This can also be defined as a complex function which can be written as
E ) E′(ω) + iE′′(ω)
(5)
where E′(ω) and E′′(ω) correspond, respectively, to the real and the imaginary parts of the elasticity modulus. They are related with the change of γ(t) and A(t) in the interface by the eq 4 and with the phase angle φ between dγ(t) and dA(t). The real and imaginary parts are related with the complex modulus by E′ ) E cos(φ) and E′′ ) E sin(φ), respectively. If the phase difference approaches to zero, then the surfaces can be considered elastic, otherwise, the surface shows a viscoelastic behavior.30 By assumption that the mechanical properties of the interface follows the Maxwell model, the elastic (storage) part in eq 5 is represented by
ω2τ2 E′(ω) ) E′0 + ∆E′ 1 + ω2τ2
(6a)
and the viscous (loss) part is written
1 E′′(ω) ) ∆E′τω 1 + ω2τ2
(6b)
where E0′ is the extrapolated value of the elastic modulus at the limit ω ) 0 and τ is the relaxation time characteristic of the Maxwell model28. The last two parameters are found by fitting the experimental values with the Levenberg-Marquardt (LM) nonlinear least-squares minimization routine in OriginPro 7.0 (Northampton, MA) adjusting the parameters given in eqs 6a and 6b. The short time diffusion of proteins at the air-water interface can be approached by the asymptotic solution for t f 0 of the diffusion coefficient, described by Miller et al.,31 which is given by
Da-w )
[ ( ) ]
π 1 dγ 4 RTC0 dxt
2
(7)
tf0
where Da-w is the diffusion coefficient near the air-water interface (m2/s), R the universal gas constant (J mol-1 K-1), T the temperature in K, t time in seconds, C0 the bulk protein concentration (mol/m3), and γ the surface tension in N/m. The protein adsorption at the air-water interface for different pH at a short time, t, was determined with the Ward-Tordai equation15
Γ ) 2C0(D0/π)1/2t1/2
(8)
where D0 is the bulk diffusion coefficient. The behavior in the region time t f ∞ of the function γ vs t-1/2 was obtained, assuming a diffusion controlled adsorption and is given by the Joos relationship32-34
2730 J. Phys. Chem. B, Vol. 111, No. 10, 2007
γ - γ∞ )
( )
RTΓ2 π 2C0 D0t
1/2
Jua´rez et al.
(9)
where γ∞ is the extrapolated equilibrium interfacial tension at t f ∞, R is the ideal gas constant, T is the absolute temperature and Γ is the protein surface concentration at long times. In order to evaluate the surface tension kinetics, we tried to find the best fit of the surface pressure vs time from the experimental values to get the relaxation modes through the relationship35
Π(t) - Π∞ ) A0 exp(-t/τ1) + A1 exp(-t/τ2) Π0 - Π∞
(10)
where τ1 and τ2 are the first and the second relaxation times, respectively, and A0 and A1 are adjustable parameters. Π∞ ) γs - γ∞, Π0 ) γs - γ0, and Π(t) ) γs - γ(t), where γs is the solvent surface tension or the interfacial tension without BSA or NDA/BSA solutions and γ∞ is the extrapolated equilibrium interfacial tension at t f ∞ obtained with eq 9. Results and Discussion CD of NDA/BSA Complexes. In Figure 1 are shown the CD spectra of BSA and some NDA/BSA complexes in water in the absence of buffer. As can be noticed from the ellipticity values at 208 and 222 nm, the R-helix content of BSA increases for the 1 500/1 NDA/BSA molar ratio and it is almost lost for the 6 000 and 12 500/1 NDA/BSA molar ratios. On the contrary, for these NDA concentrations, the proportion of β sheet and the unordered structures increase. However, as can be noticed in Figure 1, the characteristic negative peak of β sheet structure around 215-217 nm does not seem to be evident. This effect has also been noticed by other authors,36,37 who account for the difficulty in the characterization of β sheets by CD due to, among other reasons, the limited solubility of polypeptides in the β conformation and to the influence of lateral chains. In Table 1 the contents of secondary structure of BSA is shown with all NDA concentrations used. We also show the secondary structure of BSA with different NDA concentrations at three studied pH. The highest R- helix content was obtained for pH 4.6, consistent with the results found by Cascao-Pereira12 for BSA near the IP. Also, the changes of BSA secondary structure at this pH were more sensitive to the NDA concentration than the ones observed at other pH. For the 1 500/1 NDA/BSA molar ratio, the R-helix content decreased more than 20%. On the contrary, the secondary structure did not show important variations at pH 7.1 and 8.4 for the same NDA concentrations. These differences could be due to the extended structure attained by the BSA molecule at pH 4.6, also found by other authors,12,14 enabling more interaction sites with NDA molecules, and, as we show below, leading to an increase in molecular size. Fluorescence of NDA/BSA Complexes. The intensity of fluorescence decreased in an approximately regular form with the concomitant increase of the NDA concentration, as analyzed with the Stern-Volmer equation11 (eq 1). The results of KSV for the different temperatures used are shown in Table 2. Contrary to the tendency for KSV to grow with increasing temperatures observed by Hu et al.,11 we found a decreasing effect, which is interpreted as a static quenching mechanism.38 By assumption that the enthalpy change (∆H0) does not vary significantly in the range of temperatures analyzed, we determined the entropy change (∆S0) with the Van’t Hoff equation (eq 2). The data and the adjusted line, showing a high linear correlation (r2 ) 0.98), are shown in Figure 2. From the slope
Figure 1. CD spectra of BSA (0.1 mg/mL) mixed with different NDA/BSA molar ratios. The secondary structure analysis was performed using the DICHROWEB software with the program CDSSTR (http://public-1.cryst.bbk.ac.uk/cdweb/html/). Measurements were performed at room temperature. Numbers indicate the NDA/BSA molar ratios.
TABLE 1: CD Secondary Structure of BSA with Different NDA/BSA Molar Ratiosa no buffer
pH 4.6 pH 7.1 pH 8.4
NDA/BSA
R-helix
β-sheets
turns
unordered
0 300 600 900 1500 2000 6000 12500 0 900 1500 0 900 1500 0 900 1500
0.56 0.57 0.57 0.61 0.62 0.57 0.05 0.06 0.59 0.49 0.45 0.55 0.56 0.57 0.55 0.54 0.52
0.11 0.12 0.11 0.11 0.1 0.15 0.47 0.37 0.16 0.06 0.11 0.15 0.15 0.14 0.17 0.14 0.15
0.1 0.11 0.11 0.9 0.9 0.9 0.15 0.18 0.10 0.17 0.17 0.11 0.09 0.09 0.10 0.11 0.11
0.23 0.19 0.22 0.20 0.19 0.19 0.30 0.38 0.16 0.28 0.26 0.19 0.20 0.19 0.17 0.20 0.21
a Analysis was performed using the DICHROWEB software with the program cdsstr (http://public-1.cryst.bbk.ac.uk/cdweb/html/).
TABLE 2: Results of the Stern-Volmer Constant (KSV) at Different Temperatures and the Termodynamic Functions Obtained from Eqs 2 and 3 in the Text T (K)
KSV (L/mol)
∆H0 (kJ/mol)
∆G0 (kJ/mol)
∆S0 (J/mol)
293 299 305 311
693.2 634.8 550.6 515.6
-10.73
-15.85 -15.95 -16.06 -16.16
17.48
and the ordinate at the origin, we obtained a negative enthalpy change and a positive entropy change. Finally, with the known thermodynamic relationship given in eq 3, we get the free energy change (∆G0). As can be deduced from the results in Table 2, the negative sign of ∆G0 means that the binding mechanism of the NDA/BSA interaction is spontaneous, the positive sign and magnitude of ∆S0 suggest that the binding between NDA and BSA is a partially entropic driven mechanism with a moderate contribution from the enthalpy factor.39,40 The entropy increase might be based on the destruction of the water iceberg structure induced by the hydrophobic interaction.39 The values of ∆H0 and ∆S0 are of the same order as the ones obtained by He et al.39 and Shaikh et al.41 As argued by these authors, hydrogen-
Axisymmetric Drop Tensiometry of NDA/BSA
Figure 2. Calculated ln(KSV) against 1/T (T, absolute temperature) and adjusted line for different temperatures. The enthalpy and entropy changes of the NDA/BSA interaction are obtained from the slope and the ordinate at the origin of the Van’t Hoff equation (eq 2).
Figure 3. Behavior of E′ vs time for the different pH values used. BSA concentration was kept constant (2.5 × 10-7 M). Dark symbols correspond to pure BSA solutions and open symbols to the 1 500/1 NDA/BSA molar ratio (3.75 × 10-4 M concentration of NDA). Squares correspond to pH 4.6, triangles to pH 7.1 and circles to PH 8.4. Measurements were performed at 25 ( 0.1 °C.
bonding forces and electrostatic interactions seem to play the most important role in the formation and stabilization of the NDA/BSA complex, along with the contribution of hydrophobic interactions. Rheology at the Air-Water Interface. The behavior of the real part of the complex elastic modulus vs time at the airwater interface of the 1 500/1 NDA/BSA molar ratio (3.7 × 10-4 M of NDA) is shown in Figure 3 for different pH and at short times. We observe the influence of the extended BSA molecules at pH 4.6 on the elastic modulus at very short times (∼100 s), when the elastic modulus reaches almost 70 mN/m. On the contrary, at pH 7.1 and 8.4, we observe only a very small elevation of E′ values increase with time. This could be the expected consequence of extended BSA molecules at pH 4.6 able to interact strongly at the air-water interface and probably leading to the build up of a protein network with large elastic modulus values. On the contrary, at pH 7.1 and 8.4, BSA molecules remain less extended and almost independent at the air-water interface. This phenomenon was also observed by Cascao-Pereyra et al.12 at pH near the IP. They compared BSA
J. Phys. Chem. B, Vol. 111, No. 10, 2007 2731 molecules with hard spheres interacting at the interface and resisting the compression yielding high storage modulus. It is also interesting to observe the influence of the NDA molecules on the elastic modulus. The elastic modulus remains almost insensible to the influence of the NDA molecules at pH 7.1 in comparison with the effect on the elastic modulus at pH 4.6 and 8.4, which means that, for these cases, the protein film at the air-water interface is more flexible in the presence of NDA molecules attached to the BSA molecules. The behavior of the elastic modulus for BSA and NDA/BSA complexes dissolved in pure water at the air-water interface is summarized in Table 3. The parameters of the Maxwell model, fitted with eqs 6a and 6b, are given for different NDA concentrations. We observe that the relaxation times do not show appreciable changes with the NDA concentration, except for the highest NDA concentration used at the air-water interface. The zero value in the second and fifth columns of Table 3 means that the behavior of the elastic modulus for BSA at the airwater interface does not follow the Maxwell model when the NDA/BSA molar ratio is 12 500 and the elastic modulus decreases at higher frequencies. However, when using eq 6 to fit the experimental data of E′′/ω, we found a relaxation time around 27 s. For this NDA concentration the R- helix secondary structure of BSA is lost (see Table 1). This could be one reason of the anomalous behavior of E′ for this NDA concentration. At long times, the elastic modulus remains almost constant and independent of the NDA concentration, except for the 1 500/1 NDA/BSA molar ratio, when the BSA R-helix content reaches a maximum and the E′ value shows a small decrement. Figure 4 shows the behavior of both E′ and E′′/ω for BSA and the 1 500/1 NDA/BSA molar ratio and different frequencies. As can be observed, in general the model fits better the experimental points at lower frequencies, probably due to the limitations of the instrument at higher frequencies. We also observe that, at low frequencies, the viscosity is very high and decreases steeply as the frequency increases up to ∼0.25 s-1 with little further change at higher frequencies. On the contrary, at higher frequencies E′ seems to reach a constant value larger than the one obtained for E′′/ω. This behavior is similar to the formation of gel-like network structures due to the interaction of some polysaccharides and proteins.42,43 Rheology at the Chloroform-Water Interface. Our result of the relaxation time of DPPC at the chloroform-water interface also shown in Table 3 is similar to the one found by Saulnier et al.28 by using a dichloromethane-water interface (10 s). However, they used a higher DPPC concentration (7.5 × 10-3 mg/mL), closer to the DPPC critical micelle concentration (cmc). An almost negligible increase of the relaxation time was observed due to the DPPC-NDA interaction for the highest NDA concentration used, varying from 11 s for DPPC alone to 13 s with 3.1 × 10 - 3M NDA (see Table 3). Notice in Table 3 that the presence of BSA in the water phase produces only a very small increase of τ. Meanwhile, the presence of NDA/BSA of varying molar ratios produces also negligible changes in the relaxation time, except for the 2 000/1 NDA/BSA molar ratio. The extrapolated values of the elastic modulus at higher frequencies (E′0 + ∆E′) show differences when BSA is incorporated to the aqueous phase. This value decreases from approximately 30 mN/n, obtained for DPPC at the chloroform-water interface, to an average value of 14 mN/m when including BSA, indicating that a more rigid film is formed when only molecules of DPPC are the adsorbed species at the interface. The variation of E′0, considered as the conservative elastic response of molecules at the interface,28 shows a slight
2732 J. Phys. Chem. B, Vol. 111, No. 10, 2007
Jua´rez et al.
TABLE 3: Relaxation Times in Seconds and Elastic Modulus (mN/m) at Long Times of Different NDA/BSA Molar Ratios at the Air-Water Interface and the Chloroform-Water (ch/water) Interface NDA/BSA
air/water τ(3s
DPPC ch/water τ(1s
DPPC/BSA ch/water τ(2s
air/water E0 + ∆E ( 2
DPPC ch/water E0 + ∆E ( 1
DPPC/BSA ch/water E0 + ∆E ( 0.6
0 300 900 150 0 200 0 600 0 125 00
19 19 18 19 20 20 0
11 12 12 13 10 11 13
15 15 16 16 10 12 15
80 81 81 76 81 80 0
29 30 30 29 30 28 31
14.0 13.3 13.2 12.0 12.0 13.0 16.1
decrement with the NDA concentration (around 2 mN/m for the 12 500/1 NDA/BSA molar ratio). On the contrary, ∆E′, characterized as the dissipative part of the rheological perturbation, shows the same slight average increment due to the NDA concentration. However, the sum of both terms remains almost constant. In agreement with Freer et al.44 and Saulnier et al.,28 we could interpret E′0 as the static response corresponding to a change in the surface pressure vs interfacial area change and ∆E′ as the dissipative contribution of protein-protein interactions (at the air-water interface) and of phospholipid-protein interactions (at the chloroform-water interface). In Figure 4 we also show the behavior of the elastic modulus and viscosity at the chloroform-water interface for different frequencies with and without BSA in the aqueous phase. Bestfit curves of the experimental data were obtained using the Maxwell model equations (eqs 5 and 6). The middle curves and data points correspond to fitted and experimental values of E′ of DPPC at the chloroform-water interface (with and without NDA molecules), and the curves at the bottom correspond to the behavior of E′ of the DPPC/BSA and DPPC/NDA/BSA interaction with a 1 500/1 NDA/BSA molar ratio. Notice that the viscosity values at the chloroform-water interface are also smaller than the ones at the air-water interface at low frequencies, probably due to the presence of DPPC and BSA at the interface, which could produce a more fluid film at the chloroform-water interface. The effect of interaction of NDA with DPPC at the chloroform-water interface does not seem to affect the elastic properties of the interface, as the DPPC concentration is near the cmc. In the presence of BSA, the elastic modulus of the DPPC at the chloroform-water interface
Figure 4. Behavior of E′ and the two-dimensional viscosity E′′/ω for different frequencies and different interfaces of aqueous BSA solutions. Filled symbols correspond to BSA solutions and open symbols to 3.7 × 10-4 NDA concentration solutions (1 500/1 NDA/BSA molar ratio). Circles correspond to BSA at the air-water interface, triangles to DPPC at the chloroform-water interface, and squares to DPPC/BSA at the chloroform-water interface. BSA concentration was the same (2.5 × 10-7 M), and the concentration of DPPC dissolved in chloroform was 1 × 10-5 M.
decreases, which might be indicative of the selectivity of the position sites of BSA interacting with moderate concentrations of NDA molecules, which remain in the aqueous phase, while BSA hydrophobic residues interact at the interface. In general, the presence of NDA at the NDA/BSA molar ratio used (1 500/1), leads to a slight decrement of the E′ values of both, air-water and chloroform-water interfaces, which would be due to the small increment of the BSA R-helix content. Surface Tension Measurements. The air-water interfacial behavior of aqueous BSA solutions mixed with different NDA concentrations was also addressed. The results of surface tension (γ) for BSA at varying pH solutions are shown in Figure 5. Notice that, at short times (up to ∼18 s) and irrespective of pH, γ behaves approximately linear. At pH 4.6, however, there is an important slope variation at about 25 s, interpreted as an increase of the magnitude of dγ/dt1/2 and therefore, as an increment of the diffusion coefficient, according to the eq 7.31,45 The obtained results for systems at two different NDA concentrations and varying pH are shown in Table 4, where also the bulk diffusion coefficients (D0) and the average hydrodynamic radius, obtained from DLS measurements, are compared. The calculated short time diffusion constants are slightly dependent on the pH, and for pH 4.6 and 8.4, these are similar to the values obtained (0.5 × 10-8 m2/s) for native BSA at the air-water interface.46 Notice that, for pH 4.6, the coefficient at the air-water interface increases with the NDA concentration. This behavior is probably due to the extended conformation of the BSA molecule near the IP,12 and to the increase of the hydrophobic interactions a the interface.46 The D0 values obtained at pH 7.1 (6.6 × 10-11 m2/s) and 4.6 (5.2 × 10-11 m2/s) were close to previously reported data at pH 7.0 (6.52 × 10-11 m2/s) and at pH 5.0 (6.79 × 10-11
Figure 5. Surface tension of BSA aqueous solutions vs t1/2 in phosphated buffered solutions at different pH. BSA concentration was the same used (2.5 × 10-7 M). Experiments were carried on at 25 ( 0.1 °C.
Axisymmetric Drop Tensiometry of NDA/BSA
J. Phys. Chem. B, Vol. 111, No. 10, 2007 2733
TABLE 4: Diffusion Coefficient near the Air-Water Interface, Da-w, and in Bulk, D0, Hydrodynamic Radius, RH, and Surface Concentration, Γ, at 25 s at the Air-Water Interface for Different pH and NDA/BSA Molar Ratios pH 4.6 7.1 8.4
NDA/BSA
D0 m2/s (×10-11)
Da-w m2/s (×10-8)
RH (nm)
Γ ( 0.03 (mg/m2)
0 900 1500 0 900 1500 0 900 1500
5.2 4.5 4.4 6.6 6.4 6.3 6.4 6.1 6.0
0.5 0.8 7.4 1.5 0.5 0.9 0.5 0.7 0.5
4.7 5.4 5.5 3.7 3.8 3.9 3.8 4.0 4.1
0.67 0.62 0.61 0.76 0.74 0.73 0.74 0.73 0.72
m2/s)16 or pH 5.4 (5.6 × 10-11 m2/s)47, respectively. The results demonstrate the effect of pH and buffer conditions on the bulk diffusion constant of BSA as a consequence of the molecular conformation changes. The effect of the NDA/BSA interaction on the bulk diffusion constant is interpreted as an increase of the average hydrodynamic radius. At pH 4.6, the increase of the 1 500/1 NDA/BSA molar ratio complex average radius was around 17% higher. On the contrary, for pH 7.1 and 8.4 the change was around 6%. The concentrations of the BSA and NDA/BSA films at the air-water interface, Γ, around 25 s were calculated with eq 8 and are shown in Table 4. We notice that the surface concentration is slightly dependent on the pH and the NDA concentration. At pH 4.6 the protein adsorption is about 10% lower in comparison with the adsorption at pH 7.1 and 8.6. The influence of the NDA molecules on the BSA adsorption is more evident at pH 4.6, where the bulk diffusion coefficient is smaller and BSA molecules are more extended. Similar surface concentration was found for HSA (10-2 %), but it was reached at shorter times (
