Characterization of interactions between β-lactoglobulin with surface active ionic liquids in aqueous medium

Journal of Molecular Liquids 259 (2018) 134–143

Contents lists available at ScienceDirect

Journal of Molecular Liquids journal homepage: www.elsevier.com/locate/molliq

Characterization of interactions between β-lactoglobulin with surface active ionic liquids in aqueous medium Shruti Chabba, Rajni Vashishat, Rakesh Kumar Mahajan ⁎ Department of Chemistry, UGC-Centre for Advanced Studies-I, Guru Nanak Dev University, Amritsar 143005, India

a r t i c l e

i n f o

Article history: Received 28 November 2017 Received in revised form 21 February 2018 Accepted 5 March 2018 Available online 06 March 2018 Keywords: β-Lactoglobulin Ionic liquids Aggregation Quenching Hydrophobic interactions Circular dichroism

a b s t r a c t The interaction of β-lactoglobulin (βLG) with surface active ionic liquids (ILs), 1-dodecyl-3-methylimidazolium bromide [C12mim][Br] and 1-hexyl-3-methylimidazolium dodecylsulfate [C6mim][C12OSO3] has been investigated employing tensiometry, conductivity, fluorescence, dynamic light scattering (DLS), zeta potential and circular dichroism (CD) methods. A significant difference was observed in the nature of interactions present in the βLG-[C12mim][Br] system as compared to βLG-[C6mim][C12OSO3]. The IL, [C12mim][Br] interacted strongly with βLG to form complexes at the interface as compared to weaker interactions observed in case of βLG-[C6mim] [C12OSO3] system. The results of intrinsic fluorescence measurements along with CD measurements have provided insights into the unfolding of βLG due to binding of ILs. Turbidity measurements have shown the coacervate formation in case of βLG-[C12mim][Br] which was also supported from DLS and zeta potential measurements. An excellent correlation of results obtained from multiple techniques has been used to characterize the interactional behavior of ILs and βLG in a wide concentration regime. The results of the present studies are expected to contribute in understanding the interactional behavior of βLG with surface active ILs and its effect on functionality of protein. © 2018 Elsevier B.V. All rights reserved.

1. Introduction β-Lactoglobulin (βLG), a typical lipocalin, is a globular protein abundantly found in whey of ruminant species. βLG has an isoelectric point (IEP) 5.1, molecular mass of 18.5 kDa and comprises 162 aa residues, with one free cysteine residue and two disulfide bridges [1–3]. βLG exists mainly as a beta sheet protein consisting of nine anti parallel beta strands and one α-helix [4]. The native protein is composed of 10–15% of α-helix structure and 48% of antiparallel β sheet rich structure [5]. βLG exhibits complex monomer-dimer equilibria depending upon the pH of the solution and the concentration of protein [6]. The protein in its three dimensional state comprises of eight anti parallel β sheets which forms a calyx acting as binding site for most of the hydrophobic compounds. βLG is of considerable interest to the food industry because it is the major protein present in the milk whey of ruminants [7]. Another remarkable property of βLG is its ability to bind hydrophobic compounds such as retinoids, fatty acids, vitamin D, lipids and aromatic compounds [8–13]. Owing to widespread applications of surfactants in variety of biological, industrial, pharmaceutical and food processing systems, their interaction with proteins is of paramount importance [14,15]. However, the ⁎ Corresponding author. E-mail address: [email protected] (R.K. Mahajan).

https://doi.org/10.1016/j.molliq.2018.03.020 0167-7322/© 2018 Elsevier B.V. All rights reserved.

interactions between surfactants and proteins are intricate as proteins have different levels of structures. Also these interactions are dependent upon the type of surfactant, type of the protein, pH as well as temperature [16,17]. The interactions between surfactants and proteins influence the structural and morphological changes in proteins which may affect their biological activity. Upon interaction with surfactants, proteins tend to denature forming partially or fully unfolded structures with little or no tertiary structure. The rearrangement of these unfolded structures may lead to formation of different intermediate structures through aggregation or other mechanisms which play a definite role in certain human diseases [18,19]. Hence, there is growing need to understand the interactions between surfactants and proteins to control the functionality of proteins with surfactants. The interactions between βLG and conventional surfactants have been reported by several researchers. Magdassi et al. studied the binding of sodium dodecylsulfate (SDS) and dodecyltrimethylammonium bromide DTAB to βLG using differential scanning calorimetry, surface tension and zeta potential measurements. They evidenced the formation of surfactant/βLG complexes as well as aggregates via various interactional forces depending upon the nature of surfactant and binding ratio of surfactant to protein [20]. Taheri-Kafrani et al. investigated the interactions of βLG with cationic surfactant dodecyltrimethylammonium bromide (DTAB) at different pH values using spectroscopic and calorimetric techniques and concluded that DTAB induces significant alterations in its tertiary

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

structure whereas modifies its secondary structure only slightly. However, no change in the retinol binding properties of βLG in the presence of various amounts of DTAB was observed signifying the stability of the retinol binding site [21]. In another report, the same authors reported the smaller denaturing effect of non-ionic surfactant, Triton X-100 as compared to anionic surfactant, sodium dodecylsulfate (SDS) on the tertiary structure of βLG [22]. Viseu and coworkers reported unfolding of βLG induced by cationic surfactant dodecyltrimethylammonium chloride (DTAC) and compared it with unfolding induced by chemical denaturant guanidine hydrochloride (GnHCl). By studying the kinetics of unfolding, they observed that the β to α transition in DTAC was much more cooperative as compared to the complete unfolding of protein β to D (denatured) in GnHCl [23]. The effect of various classes of surfactants (anionic, cationic and zwitterionic) on the aggregation behavior of βLG has been reported by Hansted et al. [24]. They concluded that the anionic and non-ionic surfactant micelles were found to inhibit aggregation of βLG by solubilizing the protein monomers thereby making them unavailable for protein-protein association. On the other hand, the cationic surfactants were found to promote the protein aggregation by combination of destabilization and charge neutralization. In recent times, ionic liquids (ILs) have received great technological and industrial importance because of their unique properties and diverse applications [25,26]. They have also emerged as novel class of surfactants as they demonstrate superior surface activity as compared to conventional surfactants. ILs are known for solubilizing biomolecules, altering the enzyme activity and stabilization of proteins in various biomedical and pharmaceutical applications [27–29]. Therefore, it is essential to explore the interactions of surfactant like ILs with the proteins in order to optimize the technologies for their applications. There are limited reports in literature pertaining to the studies of ILs with proteins. Yu et al. reported that the binding of 1-tetradecyl-3-methylimidazolium bromide ([C14mim][Br]) to bovine serum albumin (BSA) leads to the denaturation of protein at high [C14mim][Br] concentrations [30]. The stabilization of BSA by short chain ILs, 1-alkyl-3-methylimidazolium bromide [Cnmim][Br] (n = 4, 6, 8) has been reported by Yan et al. [31]. On the other hand, Pankaj et al. reported significant alterations in BSA structure induced by [C8mim][C12OSO3] in monomeric regimes, however, it was found to be stabilized in vesicular regimes [32]. The literature reports on studies of interaction of ILs with βLG are very few and moreover they focused on effect of short chain ILs on the native structure of βLG. The conformational transitions in βLG in the presence of ILs 1-ethyl-3-methylimidazolium ethylsulfate at different pH conditions (pH = 4 and 7.4) has been studied by Sankaranarayanan et al. [33]. The authors reported a sequential conversion of helical conformation to β turn and finally to β sheet structure at acidic pH due to an increase in local microviscosity. However, at neutral pH, the native β state transforms into helical state and reverts back to native β state in the presence of IL. Takeiyo and coworkers reported the IL induced structural modification of bovine milk βLG in aqueous 1-butyl-3-methylimidazolium nitrate and ethylammonium nitrate using FT-IR and circular dichroism spectroscopy [34]. They concluded that a high IL concentration leads to alterations in protein's tertiary structure by forming non-native αhelical structure of βLG. The novel aspect of the present work lies in the fact that there are no investigations in literature dealing with the interactions between βLG and surface active imidazolium based ILs in a wide concentration range. Also, the results from the present study are expected to broaden the potential applications for SAILs and βLG. Herein, the interactions of ILs, 1-dodecyl-3-methylimidazolium bromide [C12mim][Br] and 1hexyl-3-methylimidazolium dodecylsulfate [C6mim][C12OSO3] with βLG at pH 7 were investigated using various physicochemical and spectroscopic methods. The interactions of IL-βLG at the air-water interface were investigated using tensiometric measurements. A combination of techniques has been employed to resolve the interactional behavior of ILs with the βLG in broad concentration ranging from monomeric to micellar regime.

135

2. Experimental section 2.1. Materials βLG was purchased from Sigma Aldrich and was used without further purification. 1-bromododecane (97%), 1-methylimidazole (≥99%), 1-hexyl-3-methylimidazolium chloride (98%), Sodium dodecylsulfate (98%) were purchased from Sigma Aldrich and used as received. ARgrade sodium dihydrogen phosphate and disodium hydrogen phosphate were purchased from Merck, India. AR grade dichloromethane and ethyl acetate were products of S.D. Fine chem. Ltd. India. The investigated ILs were synthesized using the procedure reported elsewhere [35,36]. The purity of the compounds was checked using 1H NMR. The solutions of ILs and βLG were prepared in phosphate buffer (10 mM pH = 7.4) using Sartorius analytical balance having an accuracy of ±0.0001 g. The structure of ILs and βLG has been depicted in Scheme 1.

2.2. Methods 2.2.1. Surface tension measurements The tensiometric measurements were made on a KRUSS (Hamburg, Germany) Easy Dyne tensiometer using the ring method at 298.15 ± 0.1 K. The surface tension of doubly distilled water (71.9 ± 0.1 mN m−1) was used for calibration purposes. The temperature was controlled using a thermostat and the accuracy was within ±0.1 K. All the measurements were done in triplicate with the accuracy 0.1 mN m−1.

2.2.2. Conductivity measurements The conductivity measurements were performed using Systronics 306 digital conductivity meter equipped with cell having unit cell constant. The measurements were carried out at 298.15 K and the samples were equilibrated using a thermostated glass vessel controlled by temperature controller. The conductivity meter was calibrated using freshly prepared 0.01 M AR grade potassium chloride (KCl) solution. The 0.01 M KCl solution has known conductance of 1411 μS/cm at 298.15 K. The measurements were performed with an uncertainty of b1%. 2.2.3. Fluorescence measurements The steady state fluorescence measurements were carried out on Hitachi-4500 spectrophotometer using quartz cell of path length 1 cm at 298.15 K. Pyrene having concentration 2 μM was used as probe. The excitation wavelength used for pyrene was 334 nm and the emission spectra were recorded in wavelength range 350–500 nm. The excitation and emission slit widths were fixed at 2.5 nm. The measurements were performed in triplicate. The intrinsic fluorescence of βLG was monitored in range of 300–450 nm by providing an excitation wavelength of 280 nm.

2.2.4. Far-UV circular dichroism (CD) measurements The circular dichroism measurements have been performed on Jasco-815 CD spectrometer at 298.15 K. The spectrum has been recorded using cuvette having path length 2 mm and the results have been expressed as average of 3 scans. The response time and the bandwidth were kept at 2 s and 0.2 nm, respectively while performing the experiments. The data is expressed as molar residue ellipticity [Ɵ] vs. wavelength (λ) given by [Ɵ] = 100Ɵobs/ncl, where Ɵobs is the observed ellipticity in mdeg, n is the number of amino acid residues, c is the concentration of protein and l is the path length in cm. At higher [C12mim] [Br] concentrations, the spectra could not be obtained with accuracy due to high HT voltage and hence the measurements were limited to lower IL concentrations only.

136

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

N

N

Br-

(a)

(b) 1-dodecyl-3-methylimidazolium bromide [C12mim][Br] N

N

-

O 3SO

(c) 1-hexyl-3-methylimidazolium dodecylsulfate [C6mim][C12OSO3] Scheme 1. (a) 3D structure of β-lactoglobulin (b) Molecular structures of [C12mim][Br] and (c) [C6mim][C12OSO3].

3. Results and discussion

cmc obtained from the tensiometric profiles have been listed in Table 1 and is found in agreement with the literature reports [35,37]. The cmc value for [C6mim][C12OSO3] is lower as compared to cmc of [C12mim][Br] which is attributed to the surface active nature of both the ions of former, hence displaying higher surface activity. The various surface parameters such as maximum surface excess concentration (Гmax), minimum headgroup area per unit molecule (Amin), standard Gibbs free energy of micellization (ΔG°m) and adsorption (ΔG°ads) have been calculated from tensiometric data using equations described in Annexure S1. The obtained parameters have been listed in Table 1. The conductivity profiles for ILs, [C12mim][Br] and [C6mim][C12OSO3] are shown in Fig. 2(a) and (b). The values of cmc and degree of counterion binding (β) obtained from conductivity plot are given in Table 1. Further, the cmc of ILs has also been obtained using steady state fluorescence measurements using pyrene as probe. The variation of I1/I3 as a function of concentration of IL has been presented in Fig. 3(a) and (b). The cmc values derived from steady state measurements are presented in Table 1.

3.1. Micellization of ILs in aqueous buffer solution

3.2. IL-βLG interactions at the interface

The surface tension (γ) vs. concentration plot for ILs, [C12mim][Br] and [C6mim][C12OSO3] are shown in Fig. 1(a) and (b) respectively. The

The qualitative information about the interaction between βLG and [C12mim][Br] at the air-water interface can be gained from tensiometric

2.2.5. Dynamic light scattering (DLS) measurements The Dynamic light scattering measurements were performed using zetasizer Nano ZS light scattering apparatus at 298.15 ± 0.1 K. The samples were thermostatted for 5 min before measurements. The measurements were carried out at scattering angle of 173°. Each experiment was repeated two times. The zeta potential measurements were performed with the same instrument. 2.2.6. Turbidity measurements A digital Nephelo-Turbidity meter 132 from Systronics was employed for recording the turbidity measurements at 298.15 K. For taking measurement, the sample cell was filled with 25 ml 5 μΜ βLG solution and concentrated stock solution of IL was titrated into the sample cell with stirring. Before recording measurement, the sample was equilibrated for 5 min.

Fig. 1. Variation of surface tension (γ) as a function of IL concentration in the absence and presence of 5 μM βLG in buffer for (a) [C12mim][Br] and (b) [C6mim][C12OSO3]. (The fitted lines to the pre-cmc data have been used to calculate the surface excess.) Various transitions discussed in the text are marked with vertical lines.

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143 Table 1 Critical micelle concentration (cmc) from surface tension (ST), conductometry (Cond) and steady state fluorescence (Flr) techniques, degree of counterion binding (β), surface tension at cmc (γcmc), surface excess concentration (Гmax), minimum area per molecule (Amin), Gibbs free energy of micellization (ΔG°m) and Gibbs free energy of adsorption (ΔG°ads) for ILs in buffer and in presence of 5 μM βLG solution at 298.15 K. cmc (mmol dm−3)

ILs in buffer [C12mim][Br] [C6mim][C12OSO3]

β

γcmc

Гmax

Amin

ΔG°m

ΔG°ads

ST

Flr

Cond

10.2 0.9

9.0 0.8

10.1 0.8

0.63 0.62

30.0 27.0

3.8 3.9

42.7 42.5

−34.8 −44.2

−45.5 −54.9

8.1 0.9

9.6 0.9

0.55 0.55

32.0 28.2

1.4 4.9

112.7 33.8

−34.2 −41.9

−51.3 −47.6

ILs in 5 μM βLG solution [C12mim][Br] 7.5 [C6mim][C12OSO3] 1.0

γcmc, Γmax, and Amin are expressed in m Nm−1, μmol m−2 and Å2, respectively. ΔG°m and ΔG°ad are expressed in kJ mol−1. (The error estimate in cmc is ±0.1 mmol dm−3.)

measurements. The variation of surface tension (γ) as a function of log of concentration (C) of [C12mim][Br] in the absence and presence of 5 μM βLG in buffer has been shown in Fig. 1(a). In the presence of βLG, a complex behavior has been observed owing to the presence of interactions between βLG and [C12mim][Br]. The γ value for 5 μM βLG solution was observed to be 58 mN m−1 indicating that the protein is feebly surface active and tends to get adsorb at air-water interface. Fig. 1 (a) reveals that with subsequent addition of [C12mim][Br], γ decreases sharply upto C1 after which it increases upto C1′ followed by a decrease in γ upto C2. As the [C12mim][Br] concentration is increased further, the γ decreases very slightly in region between C2 to C″2 followed by a decrease upto C3 to attain a plateau and does not changes with further addition of [C12mim][Br]. An initial decrease in γ of βLG solution upon addition of [C12mim][Br] upto C1 can be assigned to its adsorption on the protein surface forming highly surface active βLG-[C12mim][Br] complexes. The interaction between βLG and [C12mim][Br] in this dilute concentration regime is mainly electrostatic in nature as the positively charged [C12mim]+ ions of IL interacts with negatively charged amino acid residues present on the protein along with the hydrophobic interaction between the alkyl chain and hydrophobic patches present on the protein to some extent. The formation of these complexes between the IL and protein has been exemplified earlier as well [38,39]. Since the added [C12mim][Br] will be present in the form of monomers at this low concentration hence these complexes are named as monomer complexes. The concentration C1 is referred to as critical aggregation concentration (cac). Then a sudden rise in γ as a consequence of further addition of [C12mim][Br] is ascribed to the partial collapse of βLG[C12mim][Br] monomer complexes into the bulk. Consequently, the protein will come in contact with more number of IL molecules present in 2.4

with LG without LG

the bulk (as compared to interface) resulting in its unfolding (as observed from DLS measurements described in Section 3.6). Furthermore, the decrease in γ values up to C2 with less steep slope is attributed to the transformation of monomer complexes into aggregate complexes. These complexes being more hydrophobic tend to adsorb at the airwater interface and leads to decrease in γ upto C2. As the protein is unfolded, it provides more number of interaction sites restricting the [C12mim][Br] molecules present in bulk to migrate to the interface, thereby the γ decreases very slightly in the concentration regime from C2 to C2′. Further, the addition of [C12mim][Br] results in a linear decrease in γ upto C3. In the concentration regime between C2′ and C3, the unfolding process increases to much larger extent exposing more sites for interaction with IL, hence leading to growth of aggregate complexes in the bulk solution whereas [C12mim][Br] molecules travel to the airwater interface registering a decrease in γ. On reaching concentration C3, the saturation of βLG with [C12mim][Br] leads to the formation of free micelles in bulk along with the presence of βLG-[C12mim][Br] (aggregate) complexes. Using the surface tension data, various surface parameters such as maximum surface excess concentration (Гmax), minimum headgroup area per unit molecule (Amin), standard Gibbs free energy of micellization (ΔG°m) and adsorption (ΔG°ads) for [C12mim][Br] in the presence of βLG have been calculated and listed in Table 1. It can be observed from Table 1 that the cmc (C3) value of [C12mim][Br] decreases in the presence of βLG. Similar behavior has been reported in literature for interaction of cationic surfactants interactions with the protein [40]. From Table 1, it can be observed that the value of Гmax is lower for [C12mim][Br] in the presence of βLG as compared to that in buffer. As expected, a reverse behavior has been reflected in the Amin values. A lower Гmax and increase in Amin values for [C12mim][Br] in the presence of βLG suggests the reduced compactness of monolayer at the air-water interface in the presence of βLG which is because of the adsorption of surface active βLG-IL complexes at the interface. Also, the value of ΔG°ads for [C12mim][Br] in the presence of βLG is higher than ΔG°ads values in the absence of βLG indicating that the [C12mim][Br] shows greater feasibility of interfacial adsorption in the presence of βLG. The interaction of [C6mim][C12OSO3] with βLG takes place in entirely different fashion as compared to [C12mim][Br]. The variation of γ as a function of log of concentration (C) of [C6mim][C12OSO3] in the absence and presence of 5 μM βLG has been presented in Fig. 1(b). In the presence of βLG, γ is observed to decrease with increase in concentration of [C6mim][C12OSO3] upto C1 (critical aggregation concentration), beyond which it again decreases linearly until cmc is reached. Initially the interaction between the [C6mim][C12OSO3] and βLG is weak as indicated by slow decrease in γ and this decrease is mainly attributed to adsorption of constituent ions of [C6mim][C12OSO3] on the βLG forming βLG-[C6mim][C12OSO3] monomer complex at the air-interface. In this

without LG with LG

(a)

0.19

Turbidity (N.T.U)

/mS cm-1

C3 (cmc) 0.17

1.8

0.19 0.18 0.17

0.16

1.6

0.16 0.15

C2

0.0

1.4 0

2

4

6

8

10

12

C/mmol dm-3

14

16

18

20

0.15

0.0

0.4

0.8

0.5

1.2

C/mmol dm-3

C3 (cmc)

(c)

14

C2(cmc)

0.18

'' C 2

16

(b)

2.2

2.0

137

1.0 1.5 C/mmol dm-3

1.6

12 10 8 6 4 2 C2

0 0

5

10

15

C/ mmol dm-3

Fig. 2. Variation of specific conductance (κ) as a function of concentration for (a) [C12mim][Br] and (b) [C6mim][C12OSO3] in the absence and presence of 5 μM βLG in buffer. Various transitions discussed in the text are marked with vertical lines. For clarity purposes, the variation of κ vs. C for pure [C6mim][C12OSO3] has been shown as inset to Fig. 2(b). (c) Variation of turbidity of βLG as a function of concentration of [C12mim][Br].

138

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

Fig. 3. Variation of pyrene intensity ratio I1/I3 as a function of IL concentration in the absence and presence of 5 μM βLG in buffer for: (a) [C12mim][Br] and (b) [C6mim][C12OSO3]. Various transitions discussed in text are marked with vertical lines.

concentration regime, the interaction between the [C6mim][C12OSO3] and βLG is dominated by electrostatic forces. The constituent ions of [C6mim][C12OSO3] can interact with positively as well as negatively charged residues present on the βLG surface leading to slight unfolding of protein. As the concentration of[C6mim][C12OSO3] is increased further beyond C1, a linear decrease in γ is observed due to the formation of βLG-[C6mim][C12OSO3] aggregate complexes as a consequence of cooperative binding of [C6mim][C12OSO3] to the unfolded protein. These complexes are less surface active as compared to pure IL as indicated by the higher values of γ observed for [C6mim][C12OSO3] in the presence of βLG as compared to the γ values in buffer. This interaction process reaches saturation at C2, beyond which the added IL cannot bind to βLG any more, hence free micelles begin to form along with the presence of βLG-[C6mim][C12OSO3] aggregates. The tensiometric binding curve of βLG and SDS differs from the present results as the βLG-SDS complexes exhibited lower surface tension values below the cmc region as compared to SDS alone [20].This can be assigned to the presence of amphiphilic [C6mim]+ ion along with the [C12OSO3]− anion thereby modifying the structure of the complexes making it less surface active. The calculated parameters such as cmc, Гmax, Amin, ΔG°m and ΔG°ads for [C6mim][C12OSO3] in presence of βLG are presented in Table 1. It can be observed from Table 1 that the cmc (C2) of [C6mim][C12OSO3] is delayed in the presence of βLG due to the binding of IL monomers with βLG forming complexes which hinders the micellization process resulting in formation of micelles at higher concentration. The higher value of Гmax and lower values of Amin obtained for [C6mim][C12OSO3] in the presence of βLG suggests the increased compactness of airsolution interface due to the presence of βLG-[C6mim][C12OSO3] aggregates. The value of ΔG°ads for βLG-[C6mim][C12OSO3] system is lower as compared to its value in the buffer-[C6mim][C12OSO3] system suggesting the higher adsorption of [C6mim][C12OSO3] at the air-solution interface in the absence of βLG. 3.3. IL-βLG interactions in bulk (conductivity and turbidity measurements) The conductivity measurements have been used to probe the interactions between the protein and IL taking place in the bulk. The conductivity profiles depicting the variation of conductivity (κ) as a function of concentration of [C12mim][Br] and [C6mim][C12OSO3] in the absence and presence of 5 μM βLG solution have been presented in Fig. 2 (a) and (b), respectively. For [C12mim][Br]-βLG system, the κ values of [C12mim][Br] in the presence of βLG are found to be less as compared to κ values in its absence, which can be ascribed to the increased counterion association with the βLG. The conductivity profile for [C12mim] [Br] in the presence of βLG shows three lines with different slopes giving two break points which matches well with transitions C2 and C3 (cmc) obtained from tensiometry and steady state fluorescence techniques.

The transition corresponding to the C1 cannot be obtained due to the lesser sensitivity of the conductivity technique. An important parameter called degree of counterion binding (β) is related to degree of dissociation (α) as β = 1 − α where α is obtained by the ratio of slopes of post to pre-micellar region. The obtained values of cmc and β for [C12mim] [Br] in the absence and presence of βLG has been presented in Table 1. Initially, in the concentration regime from C0 to C2, an increase in κ with a lower slope has been observed. This increase in κ with lower slope occurs due to binding of added [C12mim][Br] to the βLG leading to formation of βLG-[C12mim][Br] (aggregate) complexes having lower mobility. Following this, further addition of IL leads to an increase in κ upto C3 but with a higher slope. The reason for obtaining higher slope can be ascribed to the formation of coacervates in this region which are confirmed from turbidimetric measurements (discussed later) [41]. At C3, the micellization of [C12mim][Br] takes place leading to the condensation of Br− into the stern layer of micelles from the solution, resulting in an increased κ with a reduced slope. It can be observed from Table 1 that the cmc (C3) of [C12mim][Br] is observed to be decreased in the presence of βLG indicating that the βLG-[C12mim] [Br] aggregate complexes did not collapse and assist in template driven micellization process. The degree of counterion binding, β, was found to decrease in the presence of βLG when compared to that in buffer. Similar findings have been reported earlier for the interactions of cationic surfactants with globular proteins [42]. The value of ΔG°m for [C12mim][Br] was found to be less in the presence of βLG as compared to βLG-free system indicating that coacervation is a favorable phenomenon over the free micelle formation. The conductivity profiles for [C6mim][C12OSO3] in the presence of βLG shows only one transition which corresponds well with the C2 (cmc) observed from the tensiometric measurements. Initially, the κ increases with a higher slope upto C2 and further increasing the concentration beyond C2 leads to increase in κ with smaller slope. The reason for increase in κ with a higher slope can be attributed to the binding of [C6mim][C12OSO3] to the protein forming aggregate complexes by virtue of electrostatic and hydrophobic interactions. With further addition of [C6mim][C12OSO3], the binding process reaches saturation and formation of free micelles commences which is indicated by a decrease in the slope. The obtained values of cmc and β have been listed in Table 1. In contrast to [C12mim][Br]-βLG system, the cmc value increases for [C6mim][C12OSO3] in the presence of βLG. A plausible explanation for this behavior can be assigned to the binding of [C6mim][C12OSO3] with βLG resulting in delay in micelle formation process. The β value for [C6mim][C12OSO3] has been observed to decrease in the presence of βLG as compared to its value in aqueous buffer solution Since the IL, [C12mim][Br] and βLG exhibited turbidity in the solution during their course of interaction, hence turbidity measurements have been performed for βLG-[C12mim][Br] system. The variation of turbidity

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

of βLG as a function of concentration of [C12mim][Br] has been presented in Fig. 2(c) and the transitions extracted from the plot (listed in Table 2) matches well with the transitions obtained from tensiometric and conductometric measurements. In the absence of [C12mim][Br], βLG shows negligible turbidity values. At pH = 7, βLG is above its IEP hence, it carries a net negative charge due to which its aggregation does not takes place. However, there is a marginal increase in turbidity upto certain concentration marked as C2 of [C12mim][Br], followed by a steep increase in turbidity upto C″2. The concentration, which marks the onset of coacervation process matches well with the C2 obtained from other techniques. The turbidity arises due to electrostatic interaction between negatively charged βLG residues and positively charged [C12mim]+ along with the hydrophobic interactions leading to the decrease in net charge of the protein. Ultimately, the addition of [C12mim][Br] leads to charge neutralization and formation of protein aggregates (coacervates) takes place leading to a maximum in observed turbidity values at C″2. Afterwards, the turbidity remains constant in concentration regime between C″2 and C3 (cmc). It is worthy to mention here that the transition corresponding to C″2 is not observed in tensiometric and conductometric measurements. With the further increase in [C12mim][Br] concentration from C″2 to C3, the turbidity remains more or less constant upto a certain concentration C3 which matches with the cmc obtained from other techniques. In this region, growth or reorganization of coacervates takes place as the added [C12mim][Br] molecules binds cooperatively on the surface of the aggregate complexes which also becomes evident from fluorescence and DLS measurements (discussed later). After C3 (cmc), the turbidity begins to decrease because the large aggregates break down into smaller aggregates due to the repulsions between the cationic IL monomers adsorbed on the protein. 3.4. Fluorimetry 3.4.1. Extrinsic fluorescence measurements Since pyrene is considered as an excellent probe to sense the changes in the micropolarity of the system, it has been exploited to study the IL-βLG interactions by capturing the polarity changes in the solution. Fig. 3(a) shows the variation of I1/I3 values of pyrene as a function of concentration of [C12mim][Br] in the absence and presence of 5 μM βLG. In the presence of βLG, a complex behavior has been observed. The concentrations where transitions have been observed have been listed in Table 2. As seen from Figure, the I1/I3 value in the presence of βLG is low (1.70) as compared to the buffer solution (1.9) indicating that the pyrene is present in the hydrophobic patches of the protein. However, with the addition of [C12mim][Br], the I1/I3 value was found to increase slightly. This behavior is attributed to some conformational changes taking place in the protein on interaction with [C12mim][Br] due to which pyrene is now experiencing relatively polar environment. These conformational changes have been validated from intrinsic fluorescence measurements as well (discussed later). As the concentration of [C12mim][Br] is increased further, the I1/I3 value decreases although Table 2 The various transition concentrations (mmol dm−3) observed from different techniques, surface tension (ST), conductivity (Cond), steady state fluorescence (Flr), intrinsic fluorescence (βLG Flr (λemission)) and turbidity measurements (Turb) in βLG-ILs systems at 298.15 K. Cond

Flr

βLG Flr (λemission)

Turb.

βLG-[C12mim][Br] C1 (cac) 0.2 C2 1.1 C3 (cmc) 7.5

– 1.1 9.6

0.5 1.1 8.1

0.4 1.8 7.4

– 1.2 10.2

βLG-[C6mim][C12OSO3] C1 (cac) 0.2 C2 (cmc) 1.0

– 0.9

0.1 0.9

0.1 1.3

– –

ST

(The error estimate in cmc is ±0.1 mmol dm−3.)

139

less sharply up to a concentration C1 (cac). The decreased polarity sensed by pyrene indicates its incorporation in hydrophobic domains formed by adsorption of [C12mim][Br] molecules on the βLG forming βLG-[C12mim][Br] monomer complexes. Further increase in [C12mim] [Br] concentration witnesses a steep decrease in I1/I3 value from C1 to C2 after which a plateau is observed. A steep decrease in micropolarity from C1 to C2 indicates considerable hydrophobic microenvironment sensed by the pyrene due to the formation of βLG-[C12mim][Br] aggregate complexes. The complexation of βLG with [C12mim][Br] is driven by electrostatic as well as hydrophobic forces leading to the unfolding of βLG. As the concentration of [C12mim][Br] is increased further, the I1/I3 value remains more or less constant indicating slight change in physiological environment of the βLG-[C12mim][Br] aggregate complexes. The obtained results are in agreement with the inferences made from tensiometry and turbidimetric measurements as well. This plateau region is followed by further decrease in micropolarity suggesting the formation of free micelle when concentration reaches C3 (cmc). The variation of I1/I3 values of pyrene as a function of concentration of [C6mim][C12OSO3] in the absence and presence of βLG is presented in Fig. 3(b). With the addition of [C6mim][C12OSO3], the I1/I3 value has been found to increase slightly similar to βLG-[C12mim][Br] system and is attributed to conformational changes occurring in the protein due to which pyrene is exposed to polar environment. Further increase in concentration of [C6mim][C12OSO3] leads to slight decrease in I1/I3 values upto C1 (cac) which is ascribed to the formation of βLG[C6mim][C12OSO3] monomer complexes. The formation of these complexes has been witnessed from tensiometric measurements also. Following this, I1/I3 values starts decreasing continuously indicating the decreased polarity sensed by pyrene as it is now embedded in βLG[C6mim][C12OSO3] aggregate complexes which have considerably higher hydrophobicity. This decrease continues until a plateau is formed where formation of free micelles takes place at C2 (cmc) and the microenvironment around the pyrene does not changes with further addition of IL. 3.4.2. Intrinsic fluorescence measurements Further, to understand the effect of ILs on the structure of βLG, the changes in fluorescence emission spectra has been monitored. The emission spectrum obtained by exciting the protein at 280 nm provides valuable information regarding the changes induced in the tertiary structure of the protein by the IL molecules. The variation in fluorescence intensity and shift in emission wavelength are helpful for providing insights into microenvironment of the fluorophore as well structural changes in protein. The intrinsic fluorescence offered by βLG is mainly due to the presence of two Trp residues, Trp 19 and Trp 61. The Trp19 is located in the A strand and is buried inside the hydrophobic cavity of the protein whereas Trp 61 is present on the C strand on the surface of the protein and has exposure to the solvent. The disulfide bridges, which act as fluorescence quenchers are present in close proximity to Trp 61, hence the intrinsic fluorescence of βLG is solely due to Trp 19 [43]. Upon exciting the protein at 280 nm, it shows an emission maxima centered at ≈330 nm. Fig. 4(a) illustrates the effect of increasing concentrations of [C12mim][Br] on the fluorescence emission intensity (Iflr) of 5 μM βLG. The various transitions obtained in different concentration regimes matches well with those obtained from tensiometry and other techniques. As observed from the Figure, initially, the addition of [C12mim] [Br] leads to an increase in Iflr upto a certain concentration, C1, with almost no change in λmax value. This increase in Iflr has been attributed to induced conformational change in protein structure on interacting with [C12mim][Br] which otherwise in native state quenches the fluorescence. Since the [C12mim][Br] is cationic in nature hence it will bind to the oppositely charged residues (glutamates and aspartates) present on the βLG leading to formation of βLG-[C12mim][Br] monomer complexes. The 3D structure of βLG reveals that the Glu and Asp residues are present in the vicinity of Trp 61 and hence, the site specific

140

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

Fig. 4. Variation of fluorescence emission intensity of βLG as a function of IL concentration for (a) [C12mim][Br] and (b) [C6mim][C12OSO3] when excited at λex = 280 nm. Various transitions discussed in the text are marked. The lines in figure are just guide to eye.

electrostatic binding of [C12mim][Br] to these residues either lead to slight unfolding of protein or conformational changes which will increase the distance between Trp 61 and Cys disulfide bridges leading to lesser quenching of Trp 61 by disulfide bridges and hence causing an increase in fluorescence intensity. This interpretation is consistent with earlier report by Taheri-Kafrani et al. where similar rise in tryptophan fluorescence intensity of βLG with the addition of surfactants at lower concentrations have been observed [21,22]. Also, since no change in λmax has been observed, this implies that the microenvironment of fluorophore does not changes in this dilute concentration regime as the [C12mim][Br] does not binds in its vicinity. However, with further addition of [C12mim][Br], the Iflr decreases sharply upto C2 with a slight blue shift of 3 nm. The blue shift suggests the shift of fluorophore to relatively more hydrophobic environment which can be explained by considering the binding of IL molecules to protein thereby screening the fluorophores from polar solvent. A sharp decrease in Iflr in this concentration regime can be assigned to partial unfolding of βLG induced by [C12mim][Br]. Along with the unfolding of βLG, the decrease in Iflr reflects the formation of βLG-[C12mim][Br] aggregate complexes due to enhanced hydrophobic interactions between the [C12mim][Br] tails and hydrophobic sites on the protein. With subsequent addition of [C12mim][Br] beyond C2, the Iflr decreases further although with a less steep slope upto C3 accompanied with a blue shift of 5 nm. The observance of slight blue shift is in contrast to the published reports where a red shift has been envisaged on interaction with cationic surfactants [21,23]. This diminution of Iflr with less steep slope has been ascribed to the growth of the aggregate complexes as the added [C12mim][Br] binds to unfolded protein via hydrophobic interactions. No observable change in λmax indicates that physiological environment around the fluorescence residues embedded inside the aggregate complexes does not changes. This has also been confirmed from steady state fluorescence measurements where micropolarity sensed by pyrene does not change much in the concentration regime between C2-C3. Above C3, the Iflr and λmax remains constant and does not change with further addition of [C12mim][Br] signifying no observable interactions between the βLG and [C12mim][Br] leading to the formation of free micelles in the solution. The effect of increasing concentrations of [C6mim][C12OSO3] on the Iflr of 5 μM βLG has been presented in Fig. 4(b). Initially, with addition of [C6mim][C12OSO3], Iflr has been found to increase after which it begins to decrease, remain constant between C1-C’1 and then again begins to decrease thereafter continuously upto C2. After this, an increasing addition of [C6mim][C12OSO3] leads to a further decrease in Iflr (5 nm blue shift) but with a less steep slope even when the concentration is twice as cmc of the IL. The initial increase in Iflr in monomeric region occurs

because [C12OSO3]− ion binds to the positively charged residues (Lysine and histidine) present on the protein thereby bringing the conformational change in the native structure. However, as observed from protein 3D structure, the positively charged residues are far away from Trp-61 therefore the increase in Iflr is less when compared to βLG-[C12mim][Br] system that signifies the quenching of fluorescence intensity. With further increasing the concentration of [C6mim][C12OSO3], Iflr decreases to little extent followed by a small concentration regime where it remains constant. The constancy in Iflr has been attributed to the cooperative binding character of IL to the protein causing its unfolding. Also, no change in λmax in this concentration regime signifies that the microenvironment around the fluorophore does not change during the course of binding. As the concentration of [C6mim][C12OSO3] is increased further, the Iflr begins to decrease as a consequence of IL binding to the unfolded βLG until 1.2 mmol dm−3 which coincides with cmc as observed from tensiometric and fluorescence measurements. This decrease in Iflr allied with slight blue shift of around 5 nm indicates that the microenvironment around the fluorophore is slightly changed. The reason for the changed microenvironment can be attributed to the amphiphilic nature of both the [C6mim]+ and [C12OSO3]− ions which can interact with the hydrophobic interior of the protein via hydrophobic interactions and also through electrostatic interaction with the oppositely charged amino acid residues present on the protein surface. To ascertain the binding effect of constituent ions of [C6mim][C12OSO3] on the native protein the fluorescence spectra of βLG with addition of [C6mim][Cl] and anionic surfactant, SDS has been recorded. The variation in the Iflr of βLG with increasing concentrations of [C6mim] and SDS has been shown in Fig. S1(a) and (b) respectively (Supporting Information). The Iflr of βLG has been observed to be quenched by [C6mim][Cl] in the studied concentration regime whereas it has been observed to increase in the presence of SDS. In the βLG-[C6mim][Cl] system, the λmax does not changes at all whereas it shows a blue shift of 8 nm in case of βLG-SDS system. Thus, these results clearly implies that in the monomeric region, the increase in the Iflr of βLG is mainly due to interaction with [C12OSO3]− ion. Also, the results demonstrate that the observed blue shift in the Iflr of βLG is due to presence of [C12OSO3]− ion near the vicinity of Trp, the [C6mim]+ ion though prefers to bind on the sites present on the surface of protein. It is interesting to note that addition of [C6mim][C12OSO3] leads to lesser decrease in Iflr of βLG as compared to addition of [C12mim][Br] which justifies their different interactional behavior with βLG. 3.5. Far-UV CD measurements Circular dichroism (CD) is considered as a powerful technique to monitor the changes in the secondary structure of the protein on

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

interaction with various ligands or surfactant molecules. The far-UV CD spectra of βLG in the absence and presence of ILs [C12mim][Br] and [C6mim][C12OSO3] at different concentrations have been presented in Fig. 5(a) and (b), respectively. The far-UV CD spectrum of native βLG exhibits a broad negative minimum at 216 nm which corresponds to a typical β-sheet rich secondary structure of the protein consistent with the literature reports [44]. In case of βLG-[C12mim][Br] systems, we could measure CD spectra uptil concentration of 1.5 mmol dm−3 only. At higher [C12mim][Br] concentrations, the spectra could not be obtained with accuracy due to high HT voltage. It can be observed from Fig. 5(a) that with the addition of [C12mim][Br], the ellipticity values changes only slightly and slight red shift in band is observed at a concentration of 1.5 mmol dm−3. The slight changes in the CD spectrum suggests that the secondary structure of βLG is stabilized uptil this concentration of [C12mim][Br]. However, significant changes in tertiary structure have been revealed by the intrinsic fluorescence measurements. The stabilization of secondary structure of BSA at lower concentration of [C14mim][Br] has been reported earlier by Geng et al. [45]. Fig. 5(b) reveals that in the presence of 0.1 mmol dm−3 [C6mim] [C12OSO3], the CD band of βLG broadens with an increase in intensity indicating increase in α-helix content thus altering the βLG's secondary structure. As the concentration of [C6mim][C12OSO3] becomes 0.8 mmol dm−3, a minima started appearing at 208 nm with a shoulder at 222 nm suggesting further increase in α-helix content as a consequence of βLG unfolding. Finally, when the concentration of [C6mim] [C12OSO3] is 1.1 mmol dm−3 (near cmc), the value of ellipticity at 222 nm increases to much larger extent indicating enhancement in αhelicity. Further increase in concentration of [C6mim][C12OSO3] do not lead to any change in CD signal. Thus, the CD results indicate that the addition of [C6mim][C12OSO3] induces the formation of α-helical structures of βLG. It has been established earlier that the amino acid sequence of βLG and three N-terminal β-strands possess a high propensity to form α-helices [46]. Similar increase in α-helix content of βLG in presence of surfactants has been observed in previous reports [21,23]. 3.6. Dynamic light scattering measurements The change in size of βLG upon interacting with ILs has been examined using DLS measurements. The intensity weighted size distribution profiles of βLG in the presence of various concentrations of [C12mim] [Br] have been presented in Fig. 6(a). The hydrodynamic diameter (Dh) of native βLG has been observed to be 4.6 nm consistent with the literature reports [24]. Interestingly, the Dh increases sharply from 4.6 to 72 nm at a concentration corresponding to 0.5 mmol dm−3 (C1). This substantial increase in size may be accounted for the formation of βLG-[C12mim][Br]monomer complexes primarily through electrostatic

141

interactions between [C12mim][Br] and negatively charged residues present on the protein and consequently inducing the partial unfolding of βLG. With further increasing the concentration of [C12mim][Br] from C1 to C2, the Dh increases from 72 nm to 320 nm. The observance of such a large increase in size has been ascribed to the formation of βLG-IL aggregate structures via hydrophobic interactions consistent with the earlier reported literature [47,48]. As the [C12mim][Br] concentration is increased further, the Dh increases sharply from 320 nm to 1400 nm at concentration corresponding to 2.2 mmol dm−3. Since the unfolding process exposes more binding sites hence the further addition of [C12mim][Br] leads to the formation of aggregate complexes as a consequence of increased hydrophobic interactions between the alkyl chains of [C12mim][Br] and hydrophobic interior of the protein. As the concentration of [C12mim][Br] is added further, the size increases further from 1400 nm to 1700 nm uptil concentration 7.4 mmol dm−3. The increase in size in this concentration regime C2-C3 can be accounted for by the growth of aggregate complexes as added [C12mim][Br] molecules will be adsorbed on the protein backbone via hydrophobic interactions with already existing ILs present on the surface of the protein. This increase in size is in accordance with the inferences made from turbidity measurements which indicated the presence of coacervates in this concentration regime due to charge neutralization (between C2 to C3). Finally, when concentration reaches 8.2 mmol dm−3, we observed the presence of bimodal distributions, one having smaller size and other having higher Dh. The presence of bimodal distributions point towards the saturation of [C12mim][Br] binding to protein and added [C12mim] [Br] monomers, therefore, begins to form micelle like aggregates. After the cmc is reached, the large aggregates begin to break down slowly into smaller aggregates due to the repulsions as cationic IL monomers get adsorbed on the protein forming an IL double layer around the protein molecule. This behavior is also consistent with the decreased turbidity values after cmc. The variation in size of βLG upon interaction with [C6mim][C12OSO3] has been presented in Fig. 6(b). The Dh has been observed to increase slowly with increasing concentrations of [C6mim][C12OSO3] indicating the formation of βLG-[C6mim][C12OSO3] complexes and slight unfolding of protein. The Dh of native βLG increases from 4.6 nm to 6.5 nm when concentration of [C6mim][C12OSO3] is 0.43 mmol dm−3. As the concentration of [C6mim][C12OSO3] is increased further to 0.98 mmol dm−3, the Dh becomes 10.1 nm. However, with further increase in concentration a small decrease in size of the complexes is observed. 4. Conclusions Keeping in mind the promising role of ionic liquids (ILs) in various biomedical and pharmaceutical applications [27–29], the present work

Fig. 5. Far-UV CD spectra of 5 μM βLG in the presence of ILs (a) [C12mim][Br] and (b) [C6mim][C12OSO3] at various concentrations.

142

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143

Fig. 6. Size distribution profiles of βLG-IL systems at different concentrations of ILs for (a) [C12mim][Br] and (b) [C6mim][C12OSO3] using dynamic light scattering measurements.

is aimed at investigating the interactions of surfactant like ILs (1-dodecyl-3-methylimidazolium bromide [C12mim][Br] and 1-hexyl-3methylimidazolium dodecylsulfate [C6mim][C12OSO3]) with globular protein, β-lactoglobulin (βLG). Both the ILs interacted differently with βLG in different concentration regimes as indicated by variation in measured physicochemical properties such as extent of unfolding, changes in zeta potential (supporting information) and hydrodynamic diameter. The tensiometric measurements revealed that the [C12mim][Br] showed increased adsorption efficacy in the presence of βLG as compared to [C6mim][C12OSO3], whose adsorption efficacy decreases in the presence of βLG. The intrinsic fluorescence experiments revealed that in βLG[C6mim][C12OSO3] system, the increase in the Iflr of βLG is mainly due to interaction with [C12OSO3]− ion and the observed blue shift in the Iflr of βLG is due to presence of [C12OSO3]− ion near the vicinity of Trp whereas the [C6mim]+ ion prefers to bind on the sites present on the surface of protein. The addition of [C12mim][Br] to βLG leads to formation of coacervates which is evidenced from appearance of turbidity and large aggregates observed in DLS measurements in concentration regime C2 to C3. These results also validate the cooperative role of electrostatic and hydrophobic interactions in case of βLG-[C12mim][Br] interactions. The circular dichroism (CD) results depicted that addition of [C12mim][Br] leads to stabilization of secondary structure in monomeric regime whereas on interaction with [C6mim][C12OSO3], an increase in α-helical content of βLG has been observed. Casting light on the binding of ILs to βLG may prove useful in their application as additives to modify the structure of whey in food industry. It has been reported that increasing the temperature of βLG solution results in aggregation of protein and eventually leads to formation of gels [49,50]. Since, the milk processing requires heat treatments; therefore, the present study will be extended to investigate the effect of these ILs on heat induced aggregation of βLG.

Acknowledgements Financial support by the Department of Science and Technology (DST) New Delhi, India (Project no. SR/S1/PC-02/2011) is strongly acknowledged. One of the authors SC is thankful to University Potential for Excellence (UPE) scheme, Guru Nanak Dev University, Amritsar for the award of research fellowship (RF). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.molliq.2018.03.020.

References [1] L. Sawyer, G. Kontopidis, Biochim. Biophys. Acta 1482 (2000) 136–148. [2] L. Liang, M. Subirad, J. Phys. Chem. B 114 (2010) 6707–6712. [3] S.G. Hambling, A.S. McAlpine, L. Sawyer, Advanced Dairy Chemistry, Elsevier Applied Science, London, 1992. [4] H.L. Monaco, G. Zanotti, P. Spadon, M. Bolognesi, L. Sawyer, E.E. Eliopoulos, J. Mol. Biol. 197 (1987) 695–706. [5] X.L. Ql, C. Holt, D. Mcnulty, D.T. Clarke, S. Brownlow, G.R. Jones, Biochem. J. 324 (1997) 341–346. [6] M. Verheul, J.S. Pedersen, S.P. Roefss, K.G. DeKruif, Biopolymers 49 (1999) 11–20. [7] D.W.S. Wong, W.M. Camirand, A.E. Pavlath, Crit. Rev. Food Sci. Nutr. 36 (1996) 807–844. [8] G. Kontopidis, C. Holt, L. Sawyer, J. Dairy Sci. 87 (2004) 785–796. [9] Y. Zhang, Q. Zhong, J. Agric. Food Chem. 60 (2012) 1880–1886. [10] L. Liu, K. Michelsen, E.N. Kitova, P.D. Schnier, J.S. Klassen, J. Am. Chem. Soc. 134 (2012) 3054–3060. [11] J.M. Berg, J.L. Tymoczko, L. Stryer, Biochemistry, W. H. Freeman and Co., New York, 2002. [12] D.C. Harris, B.R. Saks, J. Jayawickramarajah, J. Am. Chem. Soc. 133 (2011) 7676–7679. [13] L. Liang, M. Subirad, J. Phys. Chem. B 114 (2010) 6707. [14] K.P. Ananthapadmanabhan, Interactions of Surfactants with Polymers and Proteins, CRC Press, London, 1993. [15] M.N. Jones, Chem. Soc. Rev. 21 (1992) 127–136. [16] N. Ballery, M. Desmadril, P. Minard, J.M. Yon, Biochemistry 32 (1993) 708–714. [17] K. Kuwajima, Proteins 6 (1989) 87–103. [18] E.K. Kumar, N.P. Prabhu, Phys. Chem. Chem. Phys. 16 (2014) 24076–24088. [19] S. Yamamoto, K. Hasegawa, I. Yamaguchi, S. Tsutsumi, J. Kardos, Y. Goto, F. Gejyo, H. Naiki, Biochemistry 43 (2004) 11075–11082. [20] S. Magdassi, Y. Vinetsky, P. Relkin, Colloids Surf. B 6 (1996) 353–362. [21] A. Taheri-Kafrani, E. Asgari-Mobarakeh, A.K. Bordbar, T. Haertlé, Colloids Surf. B 75 (2010) 268–274. [22] A. Taheri-Kafrani, A.K. Bordbar, H. Mousavi, T. Haertlé, J. Agric. Food Chem. 56 (2008) 7528–7534. [23] M.A. Viseu, E.P. Melo, T.I. Carvalho, R.F. Correia, S.M.B. Costa, Biophys. J. 93 (2007) 3601–3612. [24] J.G. Hansted, P.L. Wejse, H. Bertelsen, D.E. Otzen, Biochim. Biophys. Acta 1814 (2011) 713–723. [25] T. Welton, Chem. Rev. 108 (1999) 2071–2084. [26] R.D. Rogers, K.R. Seddon, Science 302 (2003) 792–793. [27] F. Rantwjik, R.M. Lau, R.A. Sheldon, Trends Biotechnol. 21 (2003) 131–138. [28] H. Weingartner, C. Cabrele, C. Herrman, Phys. Chem. Chem. Phys. 14 (2012) 415–426. [29] T.L. Greaves, C.J. Deummond, Chem. Rev. 108 (2008) 206–237. [30] F. Geng, L. Zheng, J. Liu, L. Yu, C. Tung, Colloid Polym. Sci. 287 (2009) 1253–1259. [31] H. Yan, J. Wu, G. Dai, A. Zhong, H. Chen, J. Yang, D. Han, J. Lumin. 132 (2012) 622–628. [32] P. Bharmoria, K.S. Rao, T.J. Trivedi, A. Kumar, J. Phys. Chem. B 118 (2014) 115–124. [33] K. Sankaranarayanan, B. Shreedhar, B.U. Nair, A. Dharbathayan, J. Phys. Chem. B 117 (2013) 1234–1240. [34] T. Takekiyo, Y. Koyama, K. Yamazaki, H. Abe, Y. Yoshimura, J. Phys. Chem. B 117 (2013) 10142–10148. [35] K.S. Rao, T.J. Trivedi, A. Kumar, J. Phys. Chem. B 116 (2012) 14363–14374. [36] J. Dupont, C.S. Consort, P.A. Suarez, R.F. Souza, Org. Synth. 79 (2002) 236–241. [37] B. Dong, N. Li, L. Zheng, L. Yu, T. Inoue, Langmuir 23 (2007) 4178–4182. [38] T. Singh, P. Bharmoria, M. Morikawa, N. Kimizuka, A. Kumar, J. Phys. Chem. B 116 (2012) 11924–11935.

S. Chabba et al. / Journal of Molecular Liquids 259 (2018) 134–143 [39] M. Kumari, J.M. Maurya, M. Tasleem, P. Singh, R. Patel, J. Photochem. Photobiol. B 138 (2014) 27–35. [40] T. Chakraborty, I. Chakraborty, S.P. Moulik, S. Ghosh, Langmuir 25 (2009) 3062–3074. [41] D. Mitra, D. Bhattacharya, S.P. Moulik, J. Phys. Chem. B 112 (2008) 6609–6619. [42] S. Ghosh, A. Banerjee, Biomacromolecules 3 (2002) 9–16. [43] G. Kontopidis, C. Holt, L. Sawyer, J. Mol. Biol. 318 (2002) 1043–1055. [44] X. Zhang, T.A. Keiderling, Biochemistry 45 (2006) 8444–8452.

143

[45] F. Geng, L. Zheng, L. Yu, G. Li, C. Tung, Process Biochem. 45 (2010) 306–311. [46] S.M. Andrade, T.I. Carvalho, M.I. Viseu, S.M.B. Costa, Eur. J. Biochem. 271 (2004) 734–744. [47] G. Singh, T.S. Kang, J. Phys. Chem. B 119 (2015) 10573–10585. [48] P.P. Mishra, N. Kishore, J. Colloid Interface Sci. 354 (2011) 234–247. [49] M. Pouzot, D. Durand, T. Nicolai, Macromolecules 37 (2004) 8703–8708. [50] D. Durand, J.C. Gimel, T. Nicolai, Physica A 304 (2002) 253–265.