Structural Transitions and Phase Diagram - ACS Publications

Jun 14, 2012 - Glyn O. Phillips Hydrocolloid Research Centre, Glyndwr University, Plas Coch, Mold Road, Wrexham LL11 2AW, United Kingdom...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Complexation of Bovine Serum Albumin and Sugar Beet Pectin: Structural Transitions and Phase Diagram Xiangyang Li,†,‡ Yapeng Fang,*,†,‡ Saphwan Al-Assaf,†,‡ Glyn O. Phillips,†,‡ Xiaolin Yao,† Yifeng Zhang,† Meng Zhao,† Ke Zhang,† and Fatang Jiang† †

Glyn O. Phillips Hydrocolloid Research Centre at HUT, Hubei University of Technology, Wuhan 430068, China Glyn O. Phillips Hydrocolloid Research Centre, Glyndwr University, Plas Coch, Mold Road, Wrexham LL11 2AW, United Kingdom

‡

S Supporting Information *

ABSTRACT: The complexation between bovine serum albumin (BSA) and sugar beet pectin (SBP) was studied in situ by coupling glucono-δ-lactone (GDL) induced acidification with dynamic light scattering and turbidity measurements. Individual measurements at specific pHs and mixing ratios were also carried out using zeta potentiometry, gel permeation chromatography−multiangle laser light scattering (GPC-MALLS), and isothermal titration calorimetry (ITC). These investigations together enabled the establishment of a phase diagram of BSA/SBP and the identification of the molecular events during protein/polysaccharide complexation in relation to the phase diagram, which showed five regions: (I) a stable region of mixed individual soluble polymers, (II) a stable region of intramolecular soluble complexes, (III) a quasi-stable region of intermolecular soluble complexes, (IV) an unstable region of intermolecular insoluble complexes, and (V) a second stable region of mixed individual soluble polymers, on lowering pH. We found for the first time that the complexation could take place well above the critical pHc, the value that most previous studies had regarded as the onset occurrence of complexation. A model of structural transitions between the regions was proposed. The borderline between region II and region III represents the BSA/SBP stoichiometry for intramolecular soluble complex at a specific pH, while that between region III and region IV identifies the composition of the intermolecular insoluble complex. Also studied was the effect of NaCl and CaCl2 on the phase diagram and structural transitions.

■

INTRODUCTION Numerous investigations have been undertaken on the complexation of protein/polysaccharide mixed systems due to a rapid increase in both fundamental and technological interest.1−3 Protein and polysaccharide are the two key ingredients of food systems, and their complexation and coacervation have been widely used to confer structure and stabilize food products and provide desired functionalities.1,4 Practical applications such as microencapsulation of active ingredients and purification of proteins in the pharmaceutical industry also utilize the technique of protein/polysaccharide complexation.5,6 Moreover they are implicated in many biological processes such as maintaining cell integrity and participating in cell division.1,7,8 Electrostatic interaction is the prevalent primary interaction controlling the complexation of proteins/ionic polysaccharides, more generally, proteins/polyelectrolytes including both synthetic and natural polyelectrolytes. Several parameters have been shown to exert influence, which include pH, ionic strength, ion type, protein/polyelectrolyte ratio, charge density, and shape and size of the polymers.6,9,10 Among these, pH and ionic strength are the most important since they both determine the number of charges carried by proteins and polysaccharides. There have been many attempts to understand protein/ polysaccharide complexation at the molecular level to elucidate © 2012 American Chemical Society

detailed structural transitions. It is generally envisaged that complexation occurs as a two-step process upon changing pH. First intramolecular soluble complexes are formed, which are followed by intermolecular complexes and insoluble complexes via aggregation leading to either a liquid−liquid phase separation (coacervation) or a precipitation depending on the charge density of the polysaccharide.11,12 However, detailed assignments of these steps from experimental observations remain inconsistent and controversial. This is partly due to the diversity of the systems under investigation and the techniques adopted. For examples, studies have covered both acid-stable proteins that remain soluble over a wide range of pH and acid-instable proteins that have a strong propensity to aggregate and precipitate.13,14 The complex natures of the molecular forces governing protein instability complicate the structural transitions of protein/ polysaccharide complexation. A variety of techniques have also been used to follow the structural and morphological transitions of protein/polysaccharide complexation upon varying pH. These include turbidimetry,6−8,11,13,15 static/dynamic light scattering,7,11,13,16 zeta potentiometry,7,13 circular dichroism,7 light Received: January 12, 2012 Revised: June 14, 2012 Published: June 14, 2012 10164

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

microscopy,7,8,16 etc. Different specific pHs, such as the onset point, the inflection point, the maximum point, the minimum point, or the end point of those monitored signals, have been determined to identify structural transitions of different length scales. Among those specific pHs, the most significant are pHc, pHφ, and pHd. The critical pH value (pHc) is traditionally defined as the onset increase of scattered light intensity and is generally considered to be associated with the formation of primary intramolecular soluble complexes.7,11 The definition of pHφ has two versions in the literature: one as the maximum of scattered light intensity8 and the other as the onset increase of turbidity.7,11,13,15 pHφ is associated with the formation of insoluble complexes causing phase separation. It is worthy of noting that the maximum of scattered light intensity often coincides with the onset increase of turbidity,8,11 but in some instances it lags well behind.7 The pHd is the end point of turbidity decrease when turbidity returns to its baseline value,11,13,17 and it is related to the dissociation of the protein/ polysaccharide complex due to the reduction of the charges carried by the polysaccharide around the pKa of its acidic groups. Realizing the need of an intermediate state to bridge the structural evolution from intramolecular soluble complexes to the insoluble complexes inducing phase separation, Mekhloufi et al. in the study of β-lactoglobulin/acacia gum proposed another critical pH (pHca) in the region of pHc−pHφ as the onset point for the aggregation of intramolecular soluble complexes forming intermolecular complexes.7 pHca was identified by a sharp increase in the hydrodynamic radius measured by 90° dynamic light scattering. In a similar pH region, Kaibara et al. proposed a pH of pHcrit′ as the point marking the completion of intramolecular complex formation because the scattered light intensity at pHcrit′ and afterward attained a constant level.8 They argued that the intramolecular complexes would remain stable and unaggregated throughout the plateau region until near pHφ. In the study of sodium caseinate/gum arabic, Ye et al. found the formation of stable nanoparticle complexes in the region of pHc− pHφ.13 They suggested schematically that the nanoparticles were formed by intermolecular aggregation of sodium caseinate and gum arabic.13 Such interpretations are often not mutually consistent and illustrate the need for a further exploration of the structural transitions of protein/polysaccharide complexation, in particular that between intra- and intermolecular complexes. Sugar beet pectin (SBP) is extracted from sugar beet pulp. Unlike conventional pectins from citrus peel and apple pomace, SBP is incapable of forming gels, but it has excellent emulsifying properties in stabilizing oil-in-water emulsions.18 Structurally, SBP differs from other pectins in that it has a higher proportion of branched regions (side chains), a higher content of acetyl groups, a higher content of phenolic esters (notably ferulic acid) in the side chains, and also a higher content of proteinaceous material covalently bound to the side chains.18,19 Kirby et al.20 by atomic force microscopy demonstrated for the first time the presence of protein−pectin complexes with protein attached to one end of the pectin chains. The disadvantage of SBP-stabilized emulsions is their unsatisfactory long-term stability, which can be attributed to insufficient steric stabilization provided by SBP at the oil− water interface.19 SBP complexing with proteins is seen as a potential approach to solve the problem. Two aspects of the study are (1) to further examine the structural transitions of protein/polysaccharide complexation by reference to SBP and a model protein bovine serum albumin (BSA) and to understand the resultant phase diagram in terms of

composition, pH, and charge density and (2) to investigate the emulsifying performance of SBP/BSA complexes in relation to structural transitions and phase diagram. This latter investigation will be presented in a subsequent publication.

■

EXPERIMENTAL SECTION

Materials. SBP was supplied by San Ei Gen FFI Inc., Japan. It has the following molecular parameters: intrinsic viscosity [η], 0.33 mL/mg; weight-average molecular weight Mw, 5.3 × 105 Da; polydispersity Mw/ Mn, 1.9; radius of gyration Rg, 41 nm; as determined by Ubbelohde capillary viscometer and GPC-MALLS in 0.2 M NaCl at 25 °C. The critical overlap concentration of SBP was found to be 6.7 mg/mL in distilled water. SBP has a protein content of 1.6% as determined by the Bradford method. The number of free carboxylic groups per SBP molecule is estimated to be 470.21 BSA was purchased from SigmaAldrich (U.K.), which according to the supplier has a minimum purity of >96%.The reported Mw value of 6.7 × 104 Da was used for BSA for interpretation of results.22 SBP and BSA were used as received without further purification. All other chemicals used in the study were purchased from Fisher Scientific (U.K.) and were of analytical grade. Preparation of SBP and BSA Stock Solutions. BSA and SBP aqueous solutions at a concentration of 1.0 wt % were prepared by dispersing weighed amount of the powder samples into distilled water, respectively, followed by hydration at ambient temperature overnight on a roller mixer. The solutions were centrifuged at 4000 rpm for 30 min, and the supernatants were taken and passed through Nylon filters of 0.45 μm pore size to remove insoluble aggregates22 and/or insoluble matter if any present. The concentrations of the filtered solutions were determined by evaporating a known amount of the solutions at 60 °C until constant weights were achieved and weighing the dry residues. The solutions were diluted with distilled water (filtered through 0.20 μm Nylon filters) to give stock solutions of BSA and SBP at the same final concentration of 0.1 wt %. A level of 0.005% NaN3 was added to the stock solutions to suppress bacteria growth. Solutions used in the different experiments as described below were obtained by mixing the BSA and SBP stock solutions at appropriate mass ratios to give different protein/polysaccharide weight ratios (r = BSA/SBP). When the effect of salt was concerned, NaCl and CaCl2 were added. Zeta Potential Measurements. Zeta potential ζ measurements were made on a Zetasizer Nano-ZS apparatus (Malvern Instruments, U.K.) equipped with an MPT-2 pH autotitrator. The apparatus has a 4 mW He/Ne laser emitting at 633 nm. Electrophoretic mobility UE of charged particles was actually measured by means of laser Doppler velocimetry (LVD) at 17°, and ζ was calculated according to the Henry equation:23

ζ=

3ηUE 2εf (Ka)

(1)

where ε is the dielectric constant and η the viscosity of medium. f(Ka) is the Henry function which possesses a value of 1.5 under the Smoluchowski approximation. Since SBP are rather heterogeneous molecules with a broad distribution in size and surface charge, this prevented the resolution of zeta potentials of individual BSA and SBP components in their mixed solutions. The situation was even complicated when electrostatic complexes formed between BSA and SBP. Therefore, apparent mean zeta potential values ζapp are reported for BSA/SBP mixed solutions at different pHs, which are intensity-averaged values based on the intensity of light scattered by particles or species with particular electrophoretic mobility.24 Although it is not for a specific species, ζapp as an ensemble average could reflect the colloidal stability of BSA/SBP mixed solutions and could be linked to the structural transitions during complexation. Construction of Phase Diagram. The pH−composition phase diagram of BSA/SBP mixture was constructed from pH-induced structural transitions at various mixing ratios according to the experimental protocols established previously.7,13 Specifically, 10 mL of 0.1% BSA/SBP solutions was mixed vigorously with GDL powder at levels ranging from 0.25 to 1.0 wt % to initialize in situ acidification. 10165

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Light scattering and turbidity measurements were immediately taken for these samples. Light scattering measurements were performed on a Zetasizer 1000HS apparatus (Malvern Instruments, U.K.) equipped with a 10 mW He−Ne laser emitting at 633 nm. The average scattered light intensity at 90° (I90, counts s−1) and intensity autocorrelation function were recorded every 60 s for 200 min at 25 °C. The autocorrelation functions were analyzed by the second-order cumulant fit to determine z-averaged diffusion coefficients Dz.11,16 The diffusion coefficient was used to calculate z-averaged hydrodynamic radius Rh of particles through the Stokes−Einstein equation:7 Dz =

kBT 6πηR h

then dissociated by adjusting pH = 7.0 with NaOH and were analyzed on GPC-MALLS at 25 °C and pH 7.0. In order to quantify the loss of insoluble complexes during centrifugation, the BSA/SBP mixed solutions at pH 4.0 were directly adjusted to pH 7.0 without applying centrifugation, and these were analyzed by GPC-MALLS at pH 7.0 as controls. Characterization of Complexation by ITC. A CSC 4200 isothermal titration calorimeter (Calorimetry Sciences Corporation) was used to measure the complexation of BSA with SBP at pH 4.0 and 25 °C. 4% BSA and 0.5% SBP solutions were prepared in pH 4.0 phosphate buffer (2.5 mM) and were dialyzed against the buffer solvent to avoid possible pH and ionic strength mismatches. 10 μL aliquots of the 4% BSA solution was sequentially injected into a 1300 μL reaction cell initially containing either the buffer solvent (control) or the 0.5% SBP solution. A total of 24 injections were made for each measurement at an interval of 3000 s between successive injections. A stirring speed of 150 rpm was applied. The injection peaks were analyzed using the software BindWorks 3.1 to obtain a binding isotherm. Considering the nonnegligible steric effect of the ligand (BSA) on the subsequent binding and the nonspecific nature of the binding, the binding isotherm should be ideally and more accurately analyzed by the overlapping binding site model as proposed by Turgeon et al.25 However to simplify the fitting procedure, a model of independent binding sites integrated in the software was applied,26 which has been validated in the complexation of β-lactoglobulin/acacia gum and could quantify the binding process to a reasonable approximation:16

(2)

where η is the viscosity of medium and kBT is the thermal energy. Note that the effect of viscosity change during acidification on the calculation of Rh has been corrected (similarly for the calculation of ζ using eq 1). Viscosity change during acidification was assessed by steady shear flow measurement and one example is provided in Supporting Information Figure S1. The use of cumulant analysis is to obtain more robust data of mean hydrodynamic radius by minimizing the interference of air bubbles and large aggregates, etc. Additionally, CONTIN analysis of the autocorrelation functions was also applied to examine the change of distribution pattern of BSA/SBP mixtures during complexation.7 A Lambda UV/vis spectrophotometer (Perkin-Elmer) at a wavelength of 550 nm was used for turbidity measurements. The transmitted light intensity was recorded every 60 s for a period of 200 min and turbidity (τ, cm−1) was calculated as follows:

τ = (1/L) ln(I0/It)

{

Q = V ΔH [L] + (1 + [M]nK −

(3)

where L is the optical path length, I0 is the incident light intensity, and It is the transmitted light intensity. The change of pH with time during GDL-induced acidification was monitored by an Orion 4 Star multifunctional pH meter (Thermo Electro Corporation) at 25 °C for BSA/SBP at various mixing ratios. The pH−time curves were correlated with the time dependence of light scattering and turbidity measurements to obtain the information on structural transitions with pH. Since GDL-induced acidification could not reach pHs below 2.0, addition of 5 M HCl was used to decrease pH stepwise from 7.0 to 1.5. The change of turbidity at 550 nm with pH was measured on the same UV/vis spectrophotometer as used above. The pH adjustment by HCl was carried out in an external beaker maintained by water bath at 25 °C, and the solutions were circulated into the optical cell of the spectrophotometer using a Bio-Rad EP-1 Econo peristaltic pump. Characterization of Complexation by GPC-MALLS. A GPCMALLS system was used to probe the formation of complexes between BSA and SBP and to analyze the composition of the complexes formed. The system consisted of a ConstaMetric 3200 MS pump (Thermo Separation Products), an injection valve with 100 μL loop (Reodyne 7725i), a DAWN-DSP multiangle light scattering photometer (Wyatt Technology) equipped with He−Ne laser at wavelength of 633 nm, and a differential refractometer (RI 2000, Schambek, Germany). A set of two columns SB-803HQ and SB-806HQ in series with guard column material SUS 316 (Shodex OHpak, Japan) was used for separation. Depending if insoluble complexes were formed or not (judging from turbidity and according to the phase diagram constructed above), different GPC-MALLS experimental conditions were applied. When no clear turbidity was observed, such as at (pH = 6.0, r = 0.25, 0.5, 1.0, 2.0, 4.0), (pH = 5.0, r = 0.25, 0.5, 1.0, 2.0, 4.0), and (pH = 4.0, r = 0.125, 0.25), the BSA/SBP solutions were directly injected into GPCMALLS for analysis at 25 °C. GPC eluents used were 5 mM sodium acetate solutions buffered at respective pHs. These were filtered through 0.20 μm cellulose nitrate membrane and run at a flow rate of 0.45 mL/ min. When turbidity was observed and unstable complexes formed, such as at (pH = 4.0, r = 1.0, 2.0, 4.0, 8.0), the BSA/SBP solutions were first subjected to centrifugation at 4000 rpm for 30 min to remove insoluble complexes. The soluble complexes present in the supernatants were

(1 + MnK − [L]K )2 + 4K[L] )/2K

}

(4)

where Q is accumulative heat, and V, [L], and [M] are the volume of cell, ligand concentration (which here refers to BSA concentration), and macromolecule concentration (which here refers to SBP), respectively. Iterative curve fitting yielded thermodynamic parameters, which included the binding constant K, the binding enthalpy ΔH, and the stoichiometry n.

■

RESULTS Zeta Potential of BSA/SBP Mixtures. Figure 1a displays apparent mean zeta potential ζapp of BSA/SBP mixtures at varying protein/polysaccharide ratios (r) as a function of pH. The total biopolymer concentration is 0.1%. Pure BSA solution had an isoelectric point (IEP) of 4.73, which is in agreement with the literature value of 4.70.27,28 Pure SBP solutions attained a saturated ζapp value of ca. −40 mV at pHs above 5.0 and approached zero when the pH was lowered to 2.0. This can be attributed to the protonation of carboxylic groups in SBP chains around their pKa value. Decreasing the protein/polysaccharide ratio r resulted in shifts of ζapp profile to lower pHs and therefore lower IEPs for BSA/SBP mixtures. Figure 1b is the plot of IEP value against ratio r. Clearly there is a sharp decrease of IEP around r = 1.0 as r decreases, and afterward IEP remains nearly constant. The transition could mark the saturation of a SBP chain with BSA by electrostatic complexation. In other word, the maximum stoichiometry of BSA/SBP is about 1.0. When r > 1.0, the possible binding sites of SBP are fully occupied by excessive BSA molecules.28 When r < 1.0, SBP is in excess and the free carboxylic groups (unoccupied binding sites) dominate the IEP values via the mechanism of protonation. Similar observation and explanation have been made for BSA/gum arabic system and has been confirmed by ITC measurement of the binding stoichiometry.28 Phase Diagram of BSA/SBP Mixture. In order to establish the phase diagram for BSA/SBP mixtures within a pH− composition ordinate, light scattering and turbidimetry were 10166

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

whey protein/gum arabic system.11 The behavior however was different from that observed for the β-lactoglobulin/gum arabic system, which showed an increase and even a sharp rise in Rh between pHc and pHφ.7 The change of molecular size distribution of BSA/SBP during acidification is exemplified in the Supporting Information. The polydispersity in Rh considerably diminishes in the pH window between pHc and pHφ, resulting in a narrow monomodal distribution with a smaller size (Figure S2). The reduction in polydispersity and molecular size is in agreement with the Rh valley observed in this pH window. Since I90 measures the Rayleigh scattering of molecules or particles that are much smaller than the wavelength of the incident light, it can detect the change in size and shape at the length scale of macromolecules and macromolecular assemblies.7,29 The turbidity τ is caused by the loss of transmitted light due to the light scattering of particles of any diameter, being more sensitive to larger particles. τ detects the change in size at larger length scale, e.g., micrometers.7,29 Therefore, pHc has been traditionally considered to be an indication of the formation of soluble complex and pHφ associated with the formation of insoluble complexes inducing phase separation.7,11,13,15 To identify structural transition at low pHs, HCl was used to acidify BSA/SBP systems until below pH 2.0 meanwhile following turbidity change. Figure 3 shows the changes of turbidity at 550 nm (τ) during titration of HCl for different BSA/ SBP ratios r. These HCl-induced τ changes generally resemble those induced by GDL (Figure 2). However, τ disappears completely when pH is below pHd. The transitions at pHd can be attributed to the dissociation of protein/polysaccharide complex due to the reduction of the charges carried by SBP around the pKa of its carboxylic groups.13,17,28 pHd < 2 is consistent with the observation that SBP still carries a very small amount of negative charges at pH = 2.0 (Figure 1a), and a complete neutralization of SBP will only be achieved when pH < 2.0. The transition at pHd is unlikely to be caused by acid-induced degradation of BSA via aggregation or hydrolysis,30 since the transition was found reversible when titrated back to high pH (data not shown). The acid-induced degradation of BSA is a much slower process over a time scale of days,30 and therefore no remarkable degradation should occur during the titration. Additionally, complexing with SBP might also protect BSA from being degraded. Based on the pHc, pHφ, and pHd values derived from Figures 2 and 3, the phase diagram of BSA/SBP system was built in a pH− composition ordinate and is presented in Figure 4. Both pHc and pHφ display a sigmoidal dependence on r. On lowering r, pHc starts to decrease markedly at r = 1.0. It coincides with the transition observed in the IEP−r curve (Figure 1b). This similarly could be attributed to the saturation of SBP with BSA when all the possible binding sites have been occupied by electrostatic complexation. The weight ratio corresponds to a BSA/SBP number ratio of rn = 8, meaning that one SBP chain to a maximum can complex with 8 BSA molecules. Accordingly, the phase diagram can be separated into two domains: SBP-excessive domain (left) and BSA-excessive domain (right) as marked by the vertical dotted line at r = 1.0 in Figure 4. The value of pHd is nearly unchanged with the variation of r. This is readily understood because the dissociation of BSA/SBP complexes is controlled by the protonation of the carboxylic groups in SBP. Probing Complex Formation by GPC-MALLS. GPCMALLS was employed to probe the structure formation at different locations of the phase diagram (the positions marked by “+” in Figure 4). These included three sets of locations: those well above IEP (pH 7.0), close to IEP (pH 6.0), and markedly

Figure 1. (a) Apparent mean zeta potential ζapp as a function of pH for BSA/SBP mixtures with varying protein/polysaccharide ratios r. The total biopolymer concentration was 0.1 wt %. (b) Plot of isoelectric point (IEP) against protein/polysaccharide ratio r. The logarithmic xaxis was broken for inclusion of the data points of pure BSA and SBP (indicated in yellow).

employed to follow the structural transitions during GDLinduced acidification. Figures 2a−i show the evolution of turbidity at 550 nm (τ), scattered light intensity at 90° (I90), and hydrodynamic radius (Rh) with reduction of pH for various protein/polysaccharide ratios r. Note that progressively smaller pH window of measurements for r < 1 is due to the increasing content of SBP that has weaker pH buffering ability compared with BSA. The addition of GDL more rapidly reduced the pH of these systems below 5.0, rendering a smaller pH window for measurements. The evolving profiles of τ and I90 are in general similar to those reported for other protein/polysaccharide systems.7 Two characteristic pHs can be determined: pHc as the onset increase of I90 and pHφ as the maximum of I90. Specifically (refer to Figure 2e as an example), above pHc, τ and I90 are constant. Once pH < pHc, I90 starts to increase and deviate from its baseline value. Thereafter, it rises sharply to a maximum at pHφ. Between pHc and pHφ, τ remains nearly unchanged and only starts to increase strongly when I90 reaches its maximum. Accompanied with the sharp increase in τ, white precipitates appear and grow and sediment with further decreasing pH. Averaged hydrodynamic size Rh starts with hundreds of nanometers at pHs above pHc and exhibits a continuous decline before it increases sharply at pHφ. It should be noted that Rh values >1500 nm were not measured properly as they are beyond the sizing limit of the instrument and therefore do not necessarily represent true values of particle size. A valley in Rh can be observed in the pH window between pHc and pHφ, which is more pronounced at high protein/polysaccharide ratios such as r = 20, 10, 4, and 2. A similar Rh valley was also observed in the 10167

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Figure 2. Evolution of the turbidity at 550 nm (τ), scattered light intensity at 90° (I90), and hydrodynamic radius (Rh) as a function of pH during GDLinduced acidification for BSA/SBP mixtures of varying protein/polysaccharide ratios r: (a) r = 20; (b) r = 10; (c) r = 4; (d) r = 2; (e) r = 1; (f) r = 0.5; (g) r = 0.25; (h) r = 0.1; (i) r = 0.05. The total biopolymer concentration was fixed at 0.1 wt %. Example of determining characteristic pHs is illustrated in (e), where pHc is the onset increase of I90 and pHφ is the maximum of I90.

below IEP (pH 4.0), but all above pHc. Figures 5a−c show I90 signals of the GPC-MALLS elution profiles of BSA/SBP mixtures measured at the different locations. Pure SBP is eluted

with a peak centered around 5.2 mL, and pure BSA is at 6−7 mL. At pH 6.0 and 5.0, the SBP peak shows increased light scattering intensity with increasing proportion of SBP. This is different 10168

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Figure 3. Change of the turbidity at 550 nm (τ) with pH for BSA/SBP systems acidified by HCl. The protein/polysaccharide ratio r is indicated beside the curves. The total biopolymer concentration was fixed at 0.1%. pHd corresponds to the pH value where turbidity returns to its baseline value.

Figure 4. Phase diagram of BSA/SBP system in a pH−composition ordinate: ○ (pHc), □ (pHφ), and △ (pHd). r is the protein/ polysaccharide mixing ratio by weight. rn is the protein/polysaccharide mixing ratio by number that was calculated based on Mw of BSA and SBP reported in the Experimental Section. The dashed line marks the IEP of BSA and merges with the pHc curve at high r end. The vertical dotted line at r = 1.0 represents the maximum stoichiometry of BSA/SBP, which to the left is SBP-excessive domain and to the right is BSAexcessive domain. The curves together divide the phase diagram into five regions: I, II, III, IV, and V. The assignment of the five regions is given in the Discussion. Symbols (+) mark the positions where GPC-MALLS measurements were conducted to probe the formation of soluble complexes. Asterisk (∗) marks the positions where GPC-MALLS measurements were conducted to analyze the composition of insoluble complexes.

from the results at pH 4.0 where the mixed systems (r = 0.125 and 0.25) produce greatly higher light scattering intensity in comparison with the pure SBP and BSA. Figure 5d plots I90 at 5.2 mL as a function of SBP proportion (1/r + 1). A linear relationship was observed at pHs 6.0 and 5.0, which deviated strongly at pH 4.0. This is a clear indication that complexes were absent at pHs 6.0 and 5.0 while they did already exist at pH 4.0. The complexation of BSA on to SBP changed the light scattering of SBP which otherwise would be proportional to the SBP content in the mixture. It is worthy of noting that the complexes formed could be slightly perturbed under the GPC-MALLS conditions, i.e., by dilution effect in the GPC column and by the presence of a very small amount of salt (5 mM sodium acetate). However, the crucial point is that the pH has been maintained by buffering at

Figure 5. GPC-MALLS elution profiles (I90) of BSA/SBP systems with various protein/polysaccharide ratios r measured at three sets of pHs: (a) pH 6.0, (b) pH 5.0, and (c) pH 4.0. The total biopolymer concentration was 0.1 wt %, and the experimental conditions refer to the locations marked by “+” in the phase diagram of Figure 4. (d) Plot of I90 at the elution volume of 5.2 mL as a function of SBP proportion (1/r + 1). The left and right y-axes represent pure BSA and SBP, respectively.

10169

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

respective pHs. The perturbing effects exerted by dilution and salt are considered negligible and if any will both tend to dissociate the complexes formed. Despite this effect of dissociation, complex formation is still detectable at pH 4.0. Therefore, the perturbing effects do not contradict with the qualitative explanation of the GPC-MALLS data. Quantification of the Composition of BSA/SBP Complexes. The composition and/or stoichiometry of the BSA/SBP complexes formed in different regions of the phase diagram at pH = 4.0 were characterized by ITC and GPC-MALLS measurements. Figure 6a show the ITC thermogram recorded on the titration of 4.0% BSA into 0.5% SBP at pH 4.0. This titration process is equivalent to increasing r from zero at the fixed pH 4.0. Figure 6b is the control ITC thermogram where BSA is titrated into the buffer solvent. The titration of BSA into SBP produces distinct

injection peaks whereas the control titration only gives injection peaks with minimal amounts of heat. The comparison suggests that the complexation between BSA and SBP did take place at pH 4.0 and was accompanied by a release of heat (exothermic event). It further confirms the conclusion from the above GPC-MALLS measurements that complexation could occur well above pHc. The derived binding isotherm fits well to the model of independent binding sites (Figure 6c).26 The fitting yielded a set of thermodynamic parameters with n = 4.0, K = 42300 M−1, and ΔH = −190 kJ/mol. The stoichiometry n = 4.0 suggests that 4 BSA molecules are bound to one SBP molecule at pH 4.0. ITC experiments at pH 4.0 using other concentrations of BSA and SBP, e.g., titration of 2.0% BSA into 0.5% SBP, gave almost the same stoichiometry n. For unstable complexes formed at pH = 4.0 and r = 1.0, 2.0, 4.0, 8.0 (the positions marked by “∗” in Figure 4), GPC-MALLS was used to quantify their composition/stoichiometry. Because of the presence of insoluble complexes, the samples could not be directly subjected to GPC-MALLS measurements. Centrifugation was used to remove insoluble complexes or large aggregates, and the remaining soluble complexes were dissociated by adjusting to pH 7.0 before loading on to GPC-MALLS for analysis at pH 7.0. In parallel control experiments, the samples were directly adjusted to pH 7.0 without removal of insoluble complexes. Figure 7 shows the refractive index (RI) signals of GPC-MALLS results for control samples (solid lines) and those after removal of insoluble complexes (dashed lines). The difference between the pair of the lines thereby quantifies the insoluble complexes or large aggregates lost. Since the GPCMALLS measurements were carried out at pH 7.0 where no interaction between BSA and SBP is present, the two biopolymers would be eluted as separated individual molecules. The first peak is attributed to SBP and the second to BSA. With increasing ratio r, more and more SBP are lost as insoluble complexes. At r = 1.0, significant amount of SBP is found to be present as soluble complexes, which decreases to a very low level at r = 2.0. At r = 4.0 and 8.0, no SBP exists as soluble complexes, and all were precipitated as insoluble complexes. The excessive BSA is present as individual molecules in the solution. According to the difference in RI signal, it is possible to calculate the BSA/SBP composition of the insoluble complexes formed at pH 4.0. The calculated BSA/SBP composition of the insoluble complexes is 2.5, 2.1, 2.6, and 2.2 respectively at r = 1.0, 2.0, 4.0, and 8.0. This corresponds to an average BSA/SBP composition of 2.35. Effect of Salts. Figures 8a and 8b show respectively the evolutions of I90 and τ with pH during GDL-induced acidification of BSA/SBP systems in the presence of different concentrations of NaCl. Figure 8c plots pHc and pHφ against the salt concentration. As can be seen from Figure 8a, the initial increase in NaCl concentration leads to an overall increase in light scattering, and I90 reaches a maximum at NaCl = 20 mM. Correspondingly, pHc also increases with increasing the salt concentration and reaches a maximum at NaCl = 20 mM (Figure 8c). This indicates that addition of small amounts of NaCl are favorable for the formation of complexes between BSA and SBP. However, a further increase in NaCl concentration leads to decrease in light scattering. This is accompanied by a decrease in pHc. When NaCl ≥ 500 mM, no increase in I90 and τ was observed during acidification, and their values are constantly low, which suggests that the complexation between BSA and SBP is completely suppressed. Interestingly, for the systems with 20 ≤

Figure 6. ITC thermograms recorded for the titration of 4.0% BSA into (a) 0.5% SBP solution and (b) 2.5 mM phosphate buffer solvent at pH 4.0. The corresponding binding isotherm is displayed in (c). The solid line represents the curve fitting using the model of independent binding sites (eq 4). The obtained thermodynamic parameters are listed beside. 10170

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Figure 8. Evolution of (a) the scattered light intensity at 90° (I90) and (b) the turbidity at 550 nm (τ) as a function of pH during GDL-induced acidification of 0.1 wt % BSA/SBP mixed system in the presence of different NaCl concentrations. The protein/polysaccharide ratio r = 1.0. (c) Plots of pHc (circles) and pHφ (squares) against the concentrations of NaCl (black) and CaCl2 (red).

NaCl ≤ 200 mM, I90 and τ exhibit pronounced shoulders preceding their maxima. The addition of divalent salt CaCl2 has similar effects as NaCl (data for I90 and τ not shown). The difference is that the maximum in I90 and pHc occur at CaCl2 ≈ 3 mM (Figure 8c). The CaCl2 concentration required to completely suppress complex formation is ≥50 mM.

■

Figure 7. Refractive index signals of the GPC-MALLS elution profiles for BSA/SBP formed at pH 4.0 with various protein/polysaccharide ratios r: (a) r = 1.0, (b) r = 2.0, (c) r = 4.0, and (d) r = 8.0. The total biopolymer concentration was 0.1 wt %. The solid lines are controls in which the samples were adjusted to pH 7.0 to dissociate all complexes and analyzed by GPC-MALLS at pH 7.0. The dashed lines are for the samples that were subjected to centrifugation to remove insoluble complexes first, followed by dissociation of soluble complexes via pH adjustment to pH 7.0.

DISCUSSION Structural Transitions in Relation to Phase Diagram. Current conceptual understanding about the pH-induced structural transitions during protein/anionic polysaccharide complexation can be summarized as follows: (i) The complexation between protein and anionic polysaccharide can occur slightly above the IEP of protein despite the protein being overall negatively charged in this region.6,9,31,32 This is due to uneven distribution of charged groups on the protein surface producing 10171

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Figure 9. Proposed model of the structural transitions of BSA/SBP during complexation. The coils stand for SBP and the spheres for BSA. Note that the small amount of intrinsic proteins present in SBP was not detailed in the cartoon, as they are not dominating contributor to the complex formation in comparison with the highly charged polysaccharide parts of SBP.

positive patches which interact with the anionic polysaccharide. (ii) Above pHc, protein and anionic polysaccharide exist as individual soluble polymers without being associated each other to form any new entities.11,15 (iii) The pH range of pHc−pHφ is considered to be the formation of soluble complex, which

includes both intramolecular and intermolecular complexes of different stability.7,8,12,13,15,16 (iv) Below pHφ, the intermolecular complexes become insoluble, resulting in an unstable system undergoing either liquid−liquid phase separation or precipitation.7,13 (v) At pHd, the complexes dissociate due to the 10172

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

unscreened intramolecular repulsion at low ionic strength might also contribute to the observation of the large Rh.11 Region II: This is the region below the IEP of pure BSA and above pHc. Since BSA in the region is overall positively charged, it interacts electrostatically with SBP and forms intramolecular complexes (Figure 9). When r < 1.0 (namely in the SBPexcessive domain), SBP molecules have enough binding sites to accommodate BSA molecules. The complexes formed have a negative zeta potential ζapp of < −25 mV (Figure 1a). The potential is high enough to stabilize the complexes and keep them separated from forming intermolecular complexes. This region, therefore, can be described as a “stable region of intramolecular soluble complexes”. When r > 1.0 (namely in the BSA-excessive domain), BSA in excess would increase its possibility of finding suitable binding sites in SBP, and therefore the interaction could start at slightly high pH above IEP where some positive batches are responsible for binding (Figure 4).6,31,32 On the other hand, BSA in excess also would cause an immediate saturation of SBP, resulting in falling of zeta potential into ζapp > −25 mV. This potential is not high enough to keep intramolecular complexes stable, leading to a direct entry into region III, namely, the region for intermolecular complexes. The region II therefore has a very narrow pH window and is indiscernible from region III in the protein-excessive domain (Figures 4 and 9). The change in molecular size during intramolecular complexation would be negligible, in spite that a slight shrinkage could be possible due to reduced intramolecular repulsion. The interspersing of BSA molecules into SBP upon complexation also would not cause a significant change in molecular size and shape. This would explain why I90 and τ measured in this region are constant with pH and Rh slightly declines (Figure 2). Region III: Region III is encompassed by the pHc and pHφ curves. A further reduction in pH in this region reduces the zeta potential of the intramolecular complexes formed. The zeta potential in the range of −25 mV < ζapp < −10 mV (Figure 1a) is not enough to provide a full stabilization and incipient instability appears.33 Moreover, with the protein being more positively charged, the short-range attractive interaction between BSA and SBP could overcome the longer-range repulsive interaction between intramolecular complexes bridging them together to form intermolecular complexes. It is worthy of note that the intermolecular complex formation in this region is still limited to small scale bridging or aggregation, and hence the complexes are still soluble (Figure 9). These soluble complexes are however partially removable when subjected to centrifugation as shown in Figures 7a,b. Because of this incipient instability, the complexes formed in this region can be described as “quasi-stable”, and the region is therefore coined as a “quasi-stable region of intermolecular soluble complexes”. The formation of intermolecular complexes would alter the molecular shape, and particularly as discussed below the complexes shrank in size and became more compact. This explains the increase in light scattering I90 in the region (Figure 2). Turbidity τ however still does not appear at this stage because no large particle comparable to the wavelength of incident light was developed (Figure 2). Furthermore, since the bridging or aggregation occurs only to a low extent and in a limited number, the overall average hydrodynamic radius Rh does not increase. On the contrary, as a result of reduction in intramolecular repulsion, the size of the complexes reduces slightly, leading to a decrease in Rh. For r > 1.0 and in the BSA-excessive region, a conspicuous Rh valley was observed (Figure 2). This can be attributed to a more

neutralization of anionic polysaccharide via protonation. Protein and polysaccharide then exist as individual soluble polymers.11,13 Some aspects of the structural transitions remain questionable, particularly regarding (ii) and (iii). It has often been observed that at low protein/polysaccharide ratios pHc is considerably lower than IEP of the protein, as in the sodium caseinate/gum arabic system.13 This behavior was observed in the present system (region II in Figure 4). In the pH region of IEP−pHc, protein is positively charged and therefore should have interacted with anionic polysaccharide resulting in complex formation. This contradicts the arguments (ii) and (iii) which consider that electrostatic complex is only formed below pHc. Moreover, the transition from primary intramolecular complex to intermolecular complexes, and their identification can be regarded as somewhat ambiguous. On the basis of the change in Rh, Mekhloufi et al. proposed pHca to be the onset pH for the formation of intermolecular complexes.7 Kaibara et al. proposed a pH of pHcrit′ as the point marking the completion of intramolecular complex, which was followed by a stable region before aggregating intermolecularly.8 By means of ITC, Girard et al. revealed a transition from intramolecular complex to intermolecular complex in the binding isotherm of βlactoglobulin to pectin.25 Despite these proposals, the identification of these two states is still not clarified. The existence of electrostatic complex below IEP and above pHc, as revealed by the GPC-MALLS measurements (Figure 6), demonstrates that the reported structural transitions of protein/ anionic polysaccharide complexation need to be reconsidered. Specifically, the view that electrostatic complexes of protein/ anionic polysaccharide are formed only below pHc must be considered doubtful.7,28 From Figure 6c, it can be seen that the complexes eluted at about the same volume as for the pure SBP, i.e., at 5.2 mL. This indicates that the complexation of BSA on to SBP did not alter the size of SBP significantly, and hence the complexation at this stage (IEP > pH > pHc) was limited only within individual SBPs, that is, intramolecularly. In view of these experimental findings, we redefined the phase diagram of protein/anionic polysaccharide systems as illustrated in Figure 4. The corresponding structural transitions in relation to the phase diagram are proposed in Figure 9. Together with the IEP of pure BSA (the dashed line in Figure 4), pHc, pHφ, and pHd curves divide the phase diagram into five regions: I, II, III, IV, and V. The phase diagram can further be separated into SBPexcessive domain and BSA-excessive domain as marked by the maximum stoichiometry of BSA/SBP at r = 1.0. Region I: This is the region above the IEP of the protein where BSA and SBP are both negatively charged. The polysaccharide SBP in this region has a zeta potential value ζapp < −35 mV (Figure 1a). The high absolute value prevents the approaching of similarly charged BSA molecules. Consequently no complexation happens, and the region is a stable region of mixed individual soluble biopolymers (Figure 9). I90 and τ measured in this region are nearly independent of pH (Figure 2). Rh in principle should be constant as well. The slight decrease in Rh with acidification (Figure 2) could be a result of the diminishing of air bubbles introduced during rapid mixing at the beginning of the measurements. Another possibility could be the dissociation of SBP aggregates with pH. Rh measured in the region is in the magnitude of several hundreds of nanometers, which is far larger than the size of individual BSA or SBP. This discrepancy should come from air bubble or from polysaccharide aggregates as proposed previously.7 SBP indeed has limited solubility as insoluble matters could be found during centrifugation. The 10173

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Effect of Salts. Salts and ionic strength influence protein/ polysaccharide complexation by exerting an electrostatic screening effect.6 A general phenomenon observed in most previous studies is the existence of an optimal salt concentration for the formation of complexes.3,6,15,32 The present system showed an optimal concentration of 20 and ∼3 mM for NaCl and CaCl2, respectively (Figure 8c). The phenomenon has been attributed to short-range attractive interaction coupled with longer-range repulsive interaction dominating the complexation of protein to anionic polysaccharide.32 At high salt concentration, the Debye length k−1 becomes so small that all the electrostatic interactions are “weakened” or completely screened, resulting in the suppression of complex formation. However, if the salt concentration is too low (or when k−1 is large), the long-range repulsion between like-charged patches of protein and polysaccharide (Fl as illustrated in Figure 10) cannot be screened

pronounced shrinking of molecular size and formation of compact structure arising from extensive reduction in intramolecular repulsion upon a saturation binding.11 This also explains the diminishing of polydispersity in Rh (Figure S2) in the region. Region IV: This is region between the pHφ and pHd curves. In the region BSA becomes even more positively charged, and the zeta potential of the complexes is further reduced to ζapp > −10 mV (Figure 1a). The low potential leads to an extremely unstable system that undergoes large scale and extensive bridging or aggregation (Figure 9). It gives rise to insoluble intermolecular complexes that show liquid−solid phase separation, namely precipitation. This region can be regarded as an “unstable region of intermolecular insoluble complex”. The phase separations are companied with the development of precipitates. Consequently, τ and Rh increases rapidly (Figure 2). The formation of insoluble complexes reduces the number of soluble molecules scattering light, hence leading to a drop in I90 (Figure 2). Region V: This is the region below pHd curve. In the region, the negatively charged carboxylic groups of SBP are protonated and neutralized. The loss of the charges on SBP results in the disappearance of electrostatic interaction with BSA, and therefore the complexes are dissociated. Both SBP and BSA exist as individual soluble molecules (Figure 9). The region is a stable region of mixed individual soluble polymers. As a result of the dissociation of insoluble complexes, turbidity τ disappears. Physical Interpretation of Phase Diagram. It is clear above that the different regions in the phase diagram correspond to different stages of structural evolution during the complexation of BSA/SBP. The quantitative physical meaning of the phase borderlines such as pHc and pHφ is yet to be established. The composition/stoichiometry of BSA/SBP complexes formed at pH 4.0 and in different regions of the phase diagram was analyzed by ITC (Figure 6) and GPC-MALLS (Figure 7). The titration of BSA into SBP in ITC experiment corresponds to an increasing r ratio from zero at fixed pH of 4.0 and therefore spans largely over the intramolecular complex region (II). The analysis of binding of BSA onto SBP yielded a stoichiometry of n = 4.0 (Figure 6c). Significantly, this agrees exactly with the coordinate (rn = 4.0, pH = 4.0) read from the pHc borderline of the phase diagram as shown in Figure 4. The physical meaning of the pHc borderline therefore is presumably the BSA/SBP stoichiometry for intramolecular soluble complex at a specific pH. The pHc borderline in region II moves to higher r with increasing pH until the inflection point at r = 1.0 (Figure 4). This means that BSA/SBP stoichiometry increases with decreasing pH until the maximum possible stoichiometry r = 1.0 is reached. This could be interpreted by the number of positive charges carried by BSA being reduced with increasing pH, allowing more proteins to be bound per SBP molecule. The maximum stoichiometry of r = 1.0 might be determined by the charge and steric capacity of SBP receiving BSA molecules and agrees with that suggested by the IEP curve (Figure 1b). GPC-MALLS analysis of the complexes formed at pH = 4.0 and r = 1.0, 2.0, 4.0, and 8.0 gave an average BSA/SBP composition of insoluble complexes of 2.35. This value is in reasonable agreement with the coordinate (r = 2.0, pH = 4.0) read from the pHφ borderline of the phase diagram as shown in Figure 4. The physical meaning of the pHφ borderline therefore could possibly represent the composition of the intermolecular insoluble complex formed at a specific pH.

Figure 10. Schematic representation of short- and longer-range electrostatic interactions governing the complexation of protein (BSA) and anionic polysaccharide (SBP).

effectively, making the complexation formation less favorable. The optimal salt concentration is estimated to appear when the Debye length k−1 is in the order of protein radius Rpro:32 k−1 ≈ R pro

(5)

k−1 (nm) = 0.304/ |NaCl (M)|

(6)

k−1 (nm) = 0.176/ |CaCl 2 (M)|

(7)

Estimation using the above equations yielded Rpro ≈ 2.1 and 3.2 nm for BSA, based on the optimal concentration of NaCl and CaCl2, respectively. The hydrodynamic structure of BSA was reported to resemble a prolate ellipsoid with a = b = 2.1 nm; c = 7.0 nm.35 The estimated value of Rpro agrees reasonably well with the equatorial radius of the protein, i.e., a and b. It is therefore implied that BSA in the complex might take an orientation with its long c-axis parallel to the segments of SBP chain. Burgess argued in another way in that the rigid chain conformation of polysaccharide at extremely low salt concentration due to intramolecular repulsion is unfavorable for the alignment of 10174

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

■

protein during binding.36 These all can be unified in the framework of the Debye length effect. As pointed out earlier, in a salt concentration range following immediately the optimal values, i.e., 20 ≤ [NaCl] ≤ 200 mM; I90 and τ displayed pronounced plateau shoulders before their maxima (Figures 8a,b). The shoulders fall in the region III, namely the quasi-stable region of intermolecular soluble complexes as defined in the phase diagram (Figure 4). It indicates that in this range of salt concentration the formation of intermolecular soluble complexes is promoted. This could be due to screening of the medium-range repulsive interaction between intramolecular complexes (Fm as illustrated in Figure 10), allowing effective bridging to form intermolecular complexes. However, beyond this particular range, too high salt concentration would screen short-range attraction (Fs as illustrated in Figure 10) as well leading to an overall suppression of complex formation.

■

CONCLUSIONS

■

ASSOCIATED CONTENT

REFERENCES

(1) Schmitt, C.; Sanchez, C.; Desobry-Banon, S.; Hardy, J. Structure and Technofunctional Properties of Protein-Polysaccharide Complexes: A Review. Crit. Rev. Food Sci. Nutr. 1998, 38, 689−753. (2) Doublier, J. L.; Garnier, C.; Renard, D.; Sanchez, C. ProteinPolysaccharide Interactions. Curr. Opin. Colloid Interface Sci. 2000, 5, 202−214. (3) de Kruif, C. G.; Weinbreck, F.; de Vries, R. Complex Coacervation of Proteins and Anionic Polysaccharides. Curr. Opin. Colloid Interface Sci. 2004, 9, 340−349. (4) Burova, T. V.; Grinberg, N. V.; Golubeva, I. A.; Mashkevich, A. Y.; Grinberg, V. Y.; Tolstoguzov, V. B. Flavour Release in Model Bovine Serum Albumin/Pectin/2-Octanone Systems. Food Hydrocolloids 1999, 13, 7−14. (5) Wang, Y. F.; Gao, J. Y.; Dubin, P. L. Protein Separation via Polyelectrolyte Coacervation: Selectivity and Efficiency. Biotechnol. Prog. 1996, 12, 356−362. (6) Fang, Y. P.; Li, L. B.; Inoue, C.; Lundin, L.; Appelqvist, I. Associative and Segregative Phase Separations of Gelatin/Kappacarrageenan Aqueous Mixtures. Langmuir 2006, 22, 9532−9537. (7) Mekhloufi, G.; Sanchez, C.; Renard, D.; Guillemin, S.; Hardy, J. pH-induced Structural Transitions during Complexation and Coacervation of Beta-lactoglobulin and Acacia Gum. Langmuir 2005, 21, 386−394. (8) Kaibara, K.; Okazaki, T.; Bohidar, H. B.; Dubin, P. L. pH-induced Coacervation in Complexes of Bovine Serum Albumin and Cationic Polyelectrolytes. Biomacromolecules 2000, 1, 100−107. (9) Park, J. M.; Muhoberac, B. B.; Dubin, P. L.; Xia, J. L. Effects of Protein Charge Hetergeneity in Protein-Polyelectrolyte Complexation. Macromolecules 1992, 25, 290−295. (10) Sanchez, C.; Mekhloufi, G.; Schmitt, C.; Renard, D.; Robert, P.; Lehr, C. M.; Lamprecht, A.; Hardy, J. Self-Assembly of Betalactoglobulin and Acacia Gum in Aqueous Solvent: Structure and Phase-ordering Kinetics. Langmuir 2002, 18, 10323−10333. (11) Weinbreck, F.; de Vries, R.; Schrooyen, P.; de Kruif, C. G. Complex Coacervation of Whey Proteins and Gum Arabic. Biomacromolecules 2003, 4, 293−303. (12) Weinbreck, F.; Rollema, H. S.; Tromp, R. H.; de Kruif, C. G. Diffusivity of Whey Protein and Gum Arabic in Their Coacervates. Langmuir 2004, 20, 6389−6395. (13) Ye, A. Q.; Flanagan, J.; Singh, H. Formation of Stable Nanoparticles via Electrostatic Complexation between Sodium Caseinate and Gum Arabic. Biopolymers 2006, 82, 121−133. (14) Kobori, T.; Matsumoto, A.; Sugiyama, S. pH-Dependent Interaction between Sodium Caseinate and Xanthan Gum. Carbohydr. Polym. 2009, 75, 719−723. (15) Weinbreck, F.; Nieuwenhuijse, H.; Robijn, G. W.; de Kruif, C. G. Complexation of Whey Proteins with Carrageenan. J. Agric. Food Chem. 2004, 52, 3550−3555. (16) Schmitt, C.; da Silva, T. P.; Bovay, C.; Rami-Shojaei, S.; Frossard, P.; Kolodziejczyk, E.; Leser, M. E. Effect of Time on the Interfacial and Foaming Properties of Beta-lactoglobulin/Acacia Gum Electrostatic Complexes and Coacervates at pH 4.2. Langmuir 2005, 21, 7786−7795. (17) Ru, Q. M.; Wang, Y. W.; Lee, J.; Ding, Y. T.; Huang, Q. R. Turbidity and Rheological Properties of Bovine Serum Albumin/Pectin Coacervates: Effect of Salt Concentration and Initial Protein/ Polysaccharide Ratio. Carbohydr. Polym. 2012, 88, 838−846. (18) Siew, C. K.; Williams, P. A. Role of Protein and Ferulic Acid in the Emulsification Properties of Sugar Beet Pectin. J. Agric. Food Chem. 2008, 56, 4164−4171. (19) Funami, T.; Nakauma, M.; Ishihara, S.; Tanaka, R.; Inoue, T.; Phillips, G. O. Structural Modifications of Sugar Beet Pectin and the Relationship of Structure to Functionality. Food Hydrocolloids 2011, 25, 221−229. (20) Kirby, A. R.; MacDougall, A. J.; Morris, V. J. Sugar Beet PectinProtein Complexes. Food Biophys. 2006, 1, 51−56. (21) Funami, T.; Zhang, G. Y.; Hiroe, M.; Noda, S.; Nakauma, M.; Asai, I.; Cowman, M. K.; Al-Assaf, S.; Phillips, G. O. Effects of the

Structural transitions of protein/anionic polysaccharide during complexation were revisited using BSA/SBP mixtures. A phase diagram was established in a pH−composition ordinate using a range of complementary techniques. It was found that complexation could take place well above the critical pHc, the value that most previous studies had regarded as the onset formation of complexation. A new model of structural transition was proposed in relation to the phase diagram, which included the evolutions from individual soluble polymers to intramolecular soluble complexes, further to intermolecular soluble and insoluble complexes, and finally to another state of individual soluble polymers. The BSA/SBP stoichiometry of the intramolecular soluble complexes and the composition of the intermolecular insoluble complex are indicated by respective borderlines in the phase diagram. The effect of salts could be explained in terms of combined electrostatic interactions of different ranges. It would be interesting in future studies to correlate the different structures existing in protein/polysaccharide complexes with their functionalities, e.g., emulsifying and foaming performance.

S Supporting Information *

Changes of viscosity and molecular size distribution during complexation. This material is available free of charge via the Internet at http://pubs.acs.org.

■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] or [email protected]; Tel +86-(0)-2788015996. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge the financial support from Phillips Hydrocolloids Research Ltd. (U.K.) and the National Natural Science Foundation of China (No. 31171751, No. 31101260, and No. 31071520) and from the Planned Key Science and Technology Project, Wuhan Science and Technology Bureau (No. 201120822280). 10175

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176

Langmuir

Article

Proteinaceous Moiety on the Emulsifying Properties of Sugar Beet Pectin. Food Hydrocolloids 2007, 21, 1319−1329. (22) Yohannes, G.; Wiedmer, S. K.; Elomaa, M.; Jussila, M.; Aseyev, V.; Riekkola, M.-L. Thermal Aggregation of Bovine Serum Albumin Studied by Asymmetrical Flow Field-Flow Fractionation. Anal. Chim. Acta 2010, 675, 191−198. (23) Cited from Zetasizer Nano Series User Manual, Malvern Instruments Ltd., United Kingdom. (24) Connah, M. T.; Kaszuba, M.; Morfesis, A. High Resolution Zeta Potential Measurements: Analysis of Multi-component Mixtures. J. Dispersion Sci. Technol. 2002, 23, 663−669. (25) Girard, M.; Turgeon, S. L.; Gauthier, S. F. Thermodynamic Parameters of Beta-lactoglobulin-Pectin Complexes Assessed by Isothermal Titration Calorimetry. J. Agric. Food Chem. 2003, 51, 4450−4455. (26) Cited from the User’s Manual for Isothermal Titration Calorimeter Model CSC 4200 (Calorimetry Science Corp.). (27) Rangsansarid, J.; Fukada, K. Factors Affecting the Stability of O/ W Emulsion in BSA Solution: Stabilization by Electrically Neutral Protein at High Ionic Strength. J. Colloid Interface Sci. 2007, 316, 779− 786. (28) Vinayahan, T.; Williams, P. A.; Phillips, G. O. Electrostatic Interaction and Complex Formation between Gum Arabic and Bovine Serum Albumin. Biomacromolecules 2010, 11, 3367−3374. (29) Aymard, P.; Williams, M. A. K.; Clark, A. H.; Norton, I. T. A Turbidimetric Study of Phase Separating Biopolymer Mixtures during Thermal Ramping. Langmuir 2000, 16, 7383−7391. (30) Estey, T.; Kang, J.; Schwendeman, S. P.; Carpenter, J. F. BSA Degradation under Acidic Conditions: A Model for Protein Instability during Release from PLGA Delivery Systems. J. Pharm. Sci. 2006, 95, 1626−1639. (31) Bowman, W. A.; Rubinstein, M.; Tan, J. S. Polyelectrolyte-Gelatin Complexation: Light-Scattering Study. Macromolecules 1997, 30, 3262− 3270. (32) Seyrek, E.; Dubin, P. L.; Tribet, C.; Gamble, E. A. Ionic Strength Dependence of Protein-Polyelectrolyte Interactions. Biomacromolecules 2003, 4, 273−282. (33) Siposova, K.; Kubovcikova, M.; Bednarikova, Z.; Koneracka, M.; Zavisova, V.; Antosova, A.; Kopcansky, P.; Daxnerova, Z.; Gazova, Z. Depolymerization of Insulin Amyloid Fibrils by Albumin-Modified Magnetic Fluid. Nanotechnology 2012, 23, 055101. (34) Israelachvili, J. N. Intermolecular and Surface Forces, 3rd ed.; Academic Press: San Diego, CA, 2011; p 312. (35) Wright, A. K.; Thompson, M. R. Hydrodynamic Structure of Bovine Serum Albumin Determined by Transient Electric Birefringence. Biophys. J. 1975, 15, 137−141. (36) Burgess, D. J. Practical Analysis of Complex Coacervate Systems. J. Colloid Interface Sci. 1990, 140, 227−238.

10176

dx.doi.org/10.1021/la302063u | Langmuir 2012, 28, 10164−10176