Interactions of bovine serum albumin with cationic imidazolium and quaternary ammonium gemini surfactants: Effects of surfactant architecture

th ts: ha cati Bovine serum albumin (BSA) Micropolarity Secondary structure Quenching mechanism th a nary inv rois surfactants with shorter spacers or with longer hydrophobic chains have a larger effect on BSA unfolding, most imens al cha e nati added to a protein solution to prevent aggregation and unwanted adsorption during purification, filtration, freeze-drying, and stor- age [15]. Cationic surfactants can interact with proteins by electro- static and hydrophobic interactions. Attributing to the excellent deodorization and antimicrobial properties, quaternary ammo- nium compounds or cationic surfactants are usually used as bacte- ricide in various systems by protein denaturation [16–18]. physicochemical mechanism [5,8,31–34]. However, many investigations of BSA/surfactant are focused on the traditional surfactant [24,30,31,35–38], such as DTAB, SDS, and C12E8. Recent investigations on the interaction of BSA/gemini sur- factant [39–43] have revealed that such surfactants interact more efficiently with proteins as compared with conventional single chain surfactants. As a new generation of amphiphilic molecule [39–46], gemini surfactants are characterized by two hydrophobic chains and two polar headgroups covalently linked through a spacer group. Compared with conventional surfactants, they have a number of unique properties [45], such as lower critical micelle ⇑ Corresponding author. Fax: +86 531 88564750. Journal of Colloid and Interface Science 389 (2013) 175–181 Contents lists available at n r .co E-mail address: [email protected] (G. Xu). useful for their immense potential in applications but also provid- ing an intellectual challenge [1–4]. Studying of protein/surfactant interaction is an active field of interest over the past decades, ever since surfactants are known to interact with proteins thereby inducing polarity and stability changes of proteins [5–9], which is of great importance in biological, pharmaceutical, and cosmetic systems [10–13]. For example, anionic surfactants are widely used for the determination of molecular weights of proteins or size sep- aration by electrophoresis [14]. The nonionic surfactant is often strates in fundamental research [22–27], because its structure and physicochemical properties have been well characterized [28,29]. Consisting of 583 amino acid residues and 17 disulfide bonds, BSA has a net negative charge (the isoelectric point is about 4.7) in water medium at neutral pH [19,29]. Previous studies have shown that surfactants can protect the BSA structure against the thermal denaturation [30] or directly interact with BSA by binding which lead to substantial changes in BSA conformation [24,26]. Surfactants may interact with BSA through a variety of different 1. Introduction Proteins are considered to be the macromolecules due to their three d structures, large scale conformation denaturation. The investigations of th 0021-9797/$ - see front matter � 2012 Elsevier Inc. A http://dx.doi.org/10.1016/j.jcis.2012.08.067 and the imidazolium gemini surfactant interacts with BSA more strongly than its corresponding mono- mer and the quaternary ammonium gemini surfactant. These conclusions have been confirmed by the binding constants (Ka) and binding sites (n) for the BSA/surfactant system. Stern–Volmer quenching con- stants KSV of cationic surfactants binding to BSA are obtained, indicating that the probable quenching mechanism is initiated by ground-state complex formation rather than by dynamic collision. Moreover, the synchronous fluorescence spectra show that the surfactants mainly interact with tryptophan residues of BSA. � 2012 Elsevier Inc. All rights reserved. abundant and versatile ional and self-assembly nges, or the process of ve proteins are not only With the ability of binding agents and forming molecular aggre- gates, catalyzing enzyme reactions, and adsorbing to surfaces, globular proteins are frequently used in a variety of health care and pharmaceutical as functional ingredients [14,19–21]. Bovine serum albumin (BSA) is often selected as a model globular protein for studying the interaction between proteins and different sub- Keywords: Imidazolium gemini surfactant elucidate the effects of surfactant hydrophilic head group, spacer length, and hydrophobic chain length on the conformation of BSA. The results of fluorescence spectra and CD show that the imidazolium gemini Interactions of bovine serum albumin wi quaternary ammonium gemini surfactan Ting Zhou, Mingqi Ao, Guiying Xu ⇑, Teng Liu, Juan Z Key Laboratory of Colloid and Interface Chemistry, Shandong University, Ministry of Edu a r t i c l e i n f o Article history: Received 12 June 2012 Accepted 16 August 2012 Available online 19 September 2012 a b s t r a c t The interactions of BSA wi 12, 14, s = 2, 4, 6), quater ([C12mim]Br and DTAB) are fluorescence, circular dich Journal of Colloid a www.elsevie ll rights reserved. cationic imidazolium and Effects of surfactant architecture ng on, Jinan 250100, PR China series of cationic imidazolium gemini surfactants ([Cn-s-Cnim]Br2, n = 10, ammonium surfactants (C12C2C12), and their corresponding monomers estigated by fluorescence using pyrene as a molecular probe, synchronous m (CD), and UV–visible absorption spectra. These surfactants are used to SciVerse ScienceDirect d Interface Science m/locate / jc is single chain counterparts and their efficiencies increase with de- crease in the length of spacer. We have investigated the interaction a-helicity and b-sheet were analyzed using the curve fitting meth- Int of cationic quaternary ammonium gemini surfactant C12C2C12 with BSA and gelatin by means of the surface tensiometer and fluores- cence spectroscopy and found that electrostatic and hydrophobic forces lead to the changes in the polarities of the microenviron- ments and in the conformation of protein [48]. And we have also investigated the properties of mixed gelatin/cationic imidazolium gemini surfactants solutions, which have special properties and potential application in many areas owing to the inherent nature of ionic liquids and the existence of imidazolium head groups [49–52]. The results indicate that the structure of imidazolium gemini surfactant has a great effect on the surface viscoelastic modulus of gelatin. The gemini surfactants with longer hydropho- bic chain or longer spacer can combine with gelatin to form a stronger film on the air/water surface, because of the formation of a cross-linked network with gelatin chains more easily via hydrophobic interaction [52]. Continuing of our previous work, that is, to explore the effect of the structure of cationic surfactants on the conformation of pro- tein, we carried out an investigation on the interactions of BSA with a series of imidazolium gemini surfactants and quaternary ammonium surfactants, by pyrene fluorescence probe and syn- chronous fluorescence spectroscopy, circular dichroism (CD), and UV–visible absorption spectra. The results show that the gemini surfactant with a shorter spacer or with a longer chain has a larger effect on BSA unfolding, and the interactions of BSA with imidazo- lium gemini surfactants are stronger than those with quaternary ammonium surfactants and their corresponding monomers. Stern–Volmer quenching constants KSV of different cationic surfac- tants binding to BSA are obtained by the Stern–Volmer equation, with the aim of comparing the electrostatic and hydrophobic inter- actions involved in protein/surfactant interaction. The results of the present study may be useful in optimizing the structure of sur- factants in protein/surfactant mixtures relevant for many formula- tions. Moreover, due to imidazolium gemini surfactants can result in much stronger denaturalization of protein than other surfac- tants in our work, they would have potential application as disin- fectants for manual processing lines. 2. Experimental 2.1. Materials The ionic liquid-type imidazolium gemini surfactants ([Cn-s-Cn- im]Br2) and 1,2-ethane bis(dimethyl dodecyl ammonium bromide) (C12C2C12) were synthesized and purified as described previously [48,49,51]. 1-methyl-2-dodecyl imidazole bromide ([C12mim]Br) was synthesized according to literature [53]. Their structures are illustrated in Fig. 1. Dodecyltrimethyl ammonium bromide (DTAB), bovine serum albumin (BSA), and pyrene probe were purchased concentration (cmc), higher adsorption efficiency, and better wet- ting property. Moreover, examining a wide range of surfactants with various head groups and different numbers of hydrophobic al- kyl chains, such as dicationic quaternary ammonium compounds [39,41,42,44] or the cationic imidazolium surfactant [47], could get a systematic understanding about the role of surfactants’ struc- tures in the interaction of protein/surfactant. Mir et al. [42] have studied the effect of spacer length and the hydrophobic counter- part of gemini homologues (C16CsC16Br2, s = 4, 5, 6) on the interac- tion of BSA, and the results showed that the gemini surfactants interact more efficiently with the proteins than their conventional 176 T. Zhou et al. / Journal of Colloid and from Sigma and were used as received without further purification. All solutions were prepared in sodium phosphate buffer with pH = 7.0, which was prepared by mixing 10 mmol L�1 Na2HPO4 od of the far-UV CD spectrum with the Jasco secondary structure manager. 2.2.3. UV–visible absorption measurement The UV–visible absorption spectra were obtained using a HP- 845X UV–visible spectrometer. The spectra were recorded in a range from 190 to 400 nm wavelengths, and 1.0 cm path length quartz cell was used for the absorbance measurement. 3. Results and discussion 3.1. Fluorescence property for the BSA/surfactant systems BSA contains three aromatic amino acids, namely, tyrosine (Tyr), tryptophan (Trp) and phenylalanine (Phe), and fluorescence is usually dominated by the contribution of these residues. Phe is not excited in most cases, and its quantum yield in protein is rather low, so the emission from this residue can be ignored [54]. Syn- chronous fluorescence spectra store good information while ana- lyzing protein/agent systems by applying various Dk [55]. When Dk values between excitation and emission wavelength are stabi- lized at 20 or 60 nm, synchronous fluorescence can provide the characteristic information about Tyr and Trp residues of BSA, respectively [48,56]. As seen in Fig. 2, the intrinsic fluorescence of BSA is almost completely contributed by Trp alone because the fluorescence intensity of BSA for Dk = 60 nm is much higher than that for Dk = 20 nm. The result is in consistent with the fact that fluorescence intensity of protein is mainly dominated by Trp, whose absorbance at the wavelength of excitation and quan- tum yield of emission are considerably greater than the respective values for Tyr and Phe [24]. The fluorescence intensity decreases from 743 to 233 for Dk = 60 nm and from 125 to 107 for Dk = 20 nm, respectively, at 0.05 mmol L�1 [C12-4-C12im]Br2. The decreasing extent of fluorescence intensities for Dk = 60 is much stronger than that of Dk = 20 when [C12-4-C12im]Br2 is added. This and NaH2PO4 (A.R. Shanghai Reagent Co.) in appropriate amounts. Water used in the experiments was triply distilled by a quartz water purification system. 2.2. Methods 2.2.1. Fluorescence measurement Fluorescence spectra were performed with a Perkin Elmer LS55 Luminescence Spectrometer using 1.0 cm quartz cells. Synchro- nous fluorescence spectra were acquired by fixing the difference of excitation and emission wavelength (Dk = kem � kex). The inte- rior fluorescence spectra of BSA were scanned over the wavelength range of 290–450 nm with a fixed excitation wavelength at 280 nm. Pyrene spectra were recorded with a fixed excitation at 335 nm. Intensities, I1 and I3, were taken from the emission inten- sities at 373 and 384 nm, respectively. Then, the ratio of I1/I3 was used to estimate the polarity of microenvironment. The concentra- tion of pyrene was 1.0 � 10�6 mol L�1. The excitation and emission slits for all the fluorescence measurements were fixed at 10.0 and 3.5 nm, respectively. 2.2.2. Circular dichroism measurement (CD) Far-UV CD spectra were measured on a Jasco J-810 spectropo- larimeter, using a bandwidth of 2.0 nm and a cell of 0.1 cm path length over the wavelength range from 190 to 260 nm. Spectra were collected with a scan speed of 50 nm/min. The contents of erface Science 389 (2013) 175–181 indicates [C12-4-C12im]Br2 molecules mainly interact with Trp res- idues compared to Tyr residues. The effects of the other surfactants ([Cn-s-Cnim]Br2, n = 10, 12, 14, s = 2, 4, 6; [C12mim]Br; C12C2C12; N - C ) [C zoli Int N N Br- N N Br- CH2 s CnH2n+1 CnH2n+1 Br (1) [Cn-s-Cnim]Br2 (n=10,12,14; s=2,4,6) (2 Fig. 1. Chemical structures of cationic imida 400 600 800 nc e In te ns ity 0 mmol/L 0.005 mmol/L 0.05 mmol/L 0.5 mmol/L Δλ = 60 nmA T. Zhou et al. / Journal of Colloid and DTAB) on synchronous fluorescence intensity of BSA exhibit the similar behavior, and therefore, they are shown in Supporting Information, Fig. S1. The fluorescence intensity and the wavelength of emission maximum (kmax) are sensitive to protein conformation, which can be effectively used to probe protein folding/unfolding [21]. The effect of [C12-4-C12im]Br2 on BSA intrinsic fluorescence inten- sity at 298 K is shown in Fig. 3. The fluorescence intensity of BSA is quenched by binding of [C12-4-C12im]Br2, and a significant blue shift is observed in the kmax (350 nm? 331 nm). According to the literature [24], kmax occurs at shorter wavelength, indicating Trp residues are exposed to a more hydrophobic environment, which is in consistent with the binding of [C12-4-C12im]Br2 near the Trp sites of BSA. It is well known that BSA contains two Trp res- idues with intrinsic fluorescence [57]: Trp-134 and Trp-213, the former is located on the surface of the molecule and the later with- in a hydrophobic pocket of the protein. The kmax shifting (to shorter wavelength) of Trp residues toward a more nonpolar environment might be in two cases: (1) the Trp residues toward the core of pro- tein, which is possible when protein gets stabilized and (2) hydro- phobic interaction between the nonpolar moieties of surfactants 200 250 300 350 400 0 200 Fl uo re sc e Wavelength / nm 200 250 300 350 400 0 30 60 90 120 150 Δλ = 20 nm Fl uo re sc en ce In te ns ity Wavelength / nm 0 mmol/L 0.005 mmol/L 0.05 mmol/L 0.5 mmol/L B Fig. 2. Synchronous spectra of BSA (0.1 g L�1) in the presence of [C12-4-C12im]Br2. The concentrations of [C12-4-C12im]Br2 are 0, 0.005, 0.05 and 0.5 mmol L�1, respectively. and exposed Trp residues, which is possible during protein unfold- ing. The studies on BSA secondary structure change (see CD results in 3.2) show the unfolding of protein in the presence of surfactants. N CH3 12H25 N+ CH3 CH3 CH2 2 C12H25 N+ CH3 C12H25 CH3 12mim]Br (3) C12C2C12 um and quaternary ammonium surfactants. 300 350 400 450 0 200 400 600 800 Fl uo re sc en ce In te ns ity Wavelength / nm without [C12-4-C12im]Br2 0.001 mmol/L 0.003 mmol/L 0.005 mmol/L 0.007 mmol/L 0.01 mmol/L 0.015 mmol/L 0.02 mmol/L 0.05 mmol/L 0.07 mmol/L 0.15 mmol/L 0.5 mmol/L A L Fig. 3. Emission spectra of BSA (0.1 g L�1) in the presence of [C12-4-C12im]Br2. c ([C12-4-C12im]Br2): A? L, from 0 to 0.5 mmol L�1 as seen in the right of the figure. erface Science 389 (2013) 175–181 177 Therefore, the case (2) seems to be more feasible. Fig. 4 illustrates kmax of BSA as a function of the surfactant con- centrations. Initial binding of surfactant has a slightly effect on the kmax of BSA. With increasing contents of surfactants, all systems are accompanied by the decrease in kmax. At higher surfactant concen- trations, kmax values keep constant, and further addition of surfac- tants does not alter the microenvironment of Trp residues. As seen in Fig. 4A and B, the kmax decreases from 350 nm (pure BSA) to 327, 331, 333, 334, and 329 nm for [C12-2-C12im]Br2, [C12-4-C12im]Br2, [C12-6-C12im]Br2, [C10-4-C10im]Br2 and [C14-4-C14im]Br2, respectively. The decrease extents of kmax follow the order: [C12-6-C12im]Br2 < [C12-4-C12im]Br2 < [C12-2-C12im]Br2, and [C10-4-C10im]Br2 < [C12-4-C12im]Br2 < [C14-4-C14im]Br2. We can obtain the conclusion that [Cn-s-Cnim]Br2 (n = 10, 12, 14; s = 2, 4, 6) with a shorter spacer or with a longer chain has a larger effect on BSA conformation. At low [Cn-s-Cnim]Br2 concentration, the sur- factant molecules primarily bind with the Trp-134 on surface of BSA [57] through electrostatic and p–p interactions, causing the protein to expand somewhat. The gemini surfactant with shorter spacer has a more localized electric field, then enhancing the electrostatic attraction between Trp-134 and the surfactant’s head- groups. This causes the BSA to expand more easily. With increasing [Cn-s-Cnim]Br2 concentration, more and more surfactant molecules are adsorbed on the surface of protein. Except for electrostatic interaction, the hydrophobic interaction is gradually enhancing. The hydrophobic tails of the surfactants will insert the hydropho- bic pocket of BSA. Trp-213 is gradually exposed to the surface of BSA and interacts with the surfactants. Surfactants with shorter spacer own the higher charge density [41,42], which have a larger ability to interact with Trp-213 via electrostatic attraction, whereas the surfactants with longer hydrophobic chains would in- sert the hydrophobic pocket of BSA more easily due to the stronger hydrophobic interaction [34]. These factors of the gemini is defined as ‘‘hydrophobic index’’, is sensitive to the polarity of the microenvironment at the site of pyrene molecules [31,33]. Pyr- ene is a hydrophobic probe, so the I1/I3 value increases with the in- crease in the polarity [19]. Fig. 5 shows the dependence of the I1/I3 values on the surfactants concentrations in the absence and pres- ence of BSA. As can be seen, for all BSA/surfactant mixed systems, the I1/I3 values decrease considerably at low surfactant concentra- tion compared with those of pure surfactants and pure BSA (I1/ I3 = 1.38 for pure BSA, pyrene is solubilized in the hydrophobic re- gions of BSA), indicating that the hydrophobicity of the microenvi- ronment at the site of pyrene molecules is strengthened. With increasing surfactant concentrations, the I1/I3 values of the BSA/ surfactant systems continue to decrease. This may be that the pro- tein is gradually unfolding when it is interacting with the surfac- tant, inducing more and more hydrophobic residues to be exposed. Then, these hydrophobic residues bind with the hydro- phobic chains of surfactant, which can enhance the hydrophobicity of environment for the protein/surfactant complexes. Besides, mi- celles form at high surfactant concentration, and the pyrene mole- cules also exist in the hydrophobic core of the surfactant micelles. These factors all induce the pyrene molecules to sense more hydro- phobic microenvironment and decrease the I1/I3 values. From Fig. 5A and B, we can observe that the binding of surfactants on BSA reaches saturation at high surfactant concentrations, and the I1/I3 values of BSA/[Cn-s-Cnim]Br2 are 1.34, 1.27, 1.23, 1.32, and 1.25 for BSA/[C12-6-C12im]Br2, BSA/[C12-4-C12im]Br2, BSA/[C12-2- C12im]Br2, BSA/[C10-4-C10im]Br2, and BSA/[C14-4-C14im]Br2, respectively. These I1/I3 values follow the order: [C12-6-C12im]- Br2 > [C12-4-C12im]Br2 > [C12-2-C12im]Br2, and [C10-4-C10im]Br2 > Interface Science 389 (2013) 175–181 10-3 10-2 10-1 330 340 350 360 [C12-2-C12im]Br2+0.1g/L BSA [C12-4-C12im]Br2+0.1g/L BSA [C12-6-C12im]Br2+0.1g/L BSA λ m ax / n m C / mmol⋅L-1 A 10-3 10-2 10-1 100 101 330 340 350 360 [C10-4-C10im]Br2+0.1g/L BSA [C12-4-C12im]Br2+0.1g/L BSA [C14-4-C14im]Br2+0.1g/L BSA B 10-3 10-2 10-1 100 101 320 330 340 350 360 [C12-2-C12im]Br2+0.1g/L BSA C12C2C12+0.1g/L BSA [C12mim]Br+0.1g/L BSA DTAB+0.1g/L BSA λ m ax / n m C C / mmol⋅L-1 C / mmol⋅L-1 178 T. Zhou et al. / Journal of Colloid and surfactants all result in the chain of the protein unfolding more easily, accompanying with exposed more hydrophobic residues. Then, these hydrophobic residues further interact with the hydro- phobic chains of surfactants, which can enhance the hydrophobic microenvironment for protein/surfactant complexes and cause the quenching of fluorescence intensity of BSA. The intrinsic fluo- rescence of BSA in the presence of [C12-2-C12im]Br2, C12C2C12, [C12mim]Br, and DTAB is similar to that of [C12-4-C12im]Br2 (Supporting Information, Fig. S2). Fig. 4C shows the kmax of BSA as a function of the [C12-2-C12im]Br2, C12C2C12, [C12mim]Br, and DTAB concentrations, respectively. For the surfactants with different polar headgroups, it is observed that the decrease extents of kmax of BSA follow the order: DTAB < [C12mim]Br < C12C2C12 < [C12-2-C12im]Br. This confirms the conclusion that imidazolium surfactants interact with BSA more strongly than quaternary ammonium surfactants with the same hydrophobic chain. The explanations may be that imidazolium surfactants display the strong p–p interaction, namely, the aromatic ring stacking be- tween the imidazolium rings and the residues (such as Trp, Tyr, and Phe) in BSA, besides electrostatic and hydrophobic interactions [39,40,47], attributing to the existence of aromatic functional groups imidazolium ions for [Cn-4-Cnim]Br2 or [C12mim]Br. This aromatic ring stacking leads to the more expanding of polypeptide and the larger unfolding of BSA. Meanwhile, as shown in Fig. 4C, we can also obtain the conclusion that BSA has a stronger affinity for the gemini surfactants than for their corresponding monomers. As is well known, gemini surfactants have double hydrophobic tails and two polar headgroups [45,46], which may have stronger hydrophobic interactions than that of their corresponding mono- mers with BSA. The ratio of the first (373 nm) to the third(384 nm) vibrational peaks of monomeric pyrene in the emission spectra, (I1/I3), which [C12-4-C12im]Br2 > [C14-4-C14im]Br2. This suggests the environment is more hydrophobic when BSA is mixed withFig. 4. The maximum emission wavelengths of BSA (0.1 g L�1) as a function of surfactant concentrations. (A) [C12-s-C12im]Br2, s = 2, 4, 6; (B) [Cn-4-Cnim]Br2, n = 10, 12, 14; (C) [C12-2-C12im]Br2, C12C2C12, [C12mim]Br and DTAB. [C12-2-C12im]Br2 or [C14-4-C14im]Br2. This also reflects the gemini 10-3 10-2 10-1 100 101 1.2 1.4 1.6 1.8 2.0 I 1 /I 3 C / mmol⋅L-1 [C 12 -2-C 12 im]Br 2 [C12-2-C12im]Br2+0.1g/L BSA [C12-4-C12im]Br2 [C12-4-C12im]Br2+0.1g/L BSA [C12-6-C12im]Br2 [C12-6-C12im]Br2+0.1g/L BSA A 10-4 10-3 10-2 10-1 100 101 102 103 1.2 1.4 1.6 1.8 2.0 I 1 /I 3 C / mmol⋅L-1 [C10-4-C10im]Br2 [C10-4-C10im]Br2+0.1g/L BSA [C12-4-C12im]Br2 [C12-4-C12im]Br2+0.1g/L BSA [C14-4-C14im]Br2 [C14-4-C14im]Br2+0.1g/L BSA B Fig. 5. I1/I3 values of pyrene as a function of surfactants concentrations without and with of BSA (0.1 g L�1). surfactant with shorter spacer or with longer hydrophobic chain may interact with BSA more strongly, which is in well agreement with the intrinsic fluorescence results. 3.2. The change of BSA secondary structure in the presence of surfactants The secondary structure of BSA is composed of a-helicity, b- sheet, random coil, and b-turn. The decrease in a-helical content in the secondary structure can reflect the unfolding extent of BSA [32]. To investigate the effect of surfactants on the secondary struc- ture of BSA, circular dichroism (CD) measurements were under- taken in the wavelength region of 190–260 nm. As seen in Fig. 6, the CD spectra of BSA exhibit two negative bands in the UV region at 208 and 222 nm, which is the characteristic of the a-helical structure of protein [22,40,48]. At low [C12-4-C12im]Br2 concentra- aromatic ring stacking (p–p interaction), the imidazolium surfac- (quaternary ammonium or imidazolium compounds) and Br . The intrinsic fluorescence of BSA in the absence and presence of NaBr is measured (Supporting Information, Fig. S5), and the intensity has slightly decrease when NaBr is added. Therefore, it can be con- firmed that the quenching effect is mainly induced by the cationic parts of the surfactants. Fluorescence quenching is described by the well-known Stern– Volmer equation [58]: F0 F ¼ 1þ KSV ½Q � ð1Þ where F0 and F represent the steady-state fluorescence intensities in the absence and presence of quencher (surfactant), respectively. KSV is the Stern–Volmer quenching constant, and [Q] is the concentra- tion of quencher. Hence, Eq. (1) is applied to determine KSV by linear regression of a plot of F0/F against [Q]. Taking [C12-4-C12im]Br2 as an example, the KSV values of [C12-4-C12im]Br2 from Stern–Volmer plots at different temperatures are listed in Table 1. The data show that the Stern–Volmer quenching constants KSV decrease with T. Zhou et al. / Journal of Colloid and Int tions (60.002 mmol L�1), there is little change in the negative ellip- ticity at 208 and 222 nm, but at high concentrations (>0.01 mmol L�1), the negative ellipticity decreases. The content of a-helicity decreases and that of b-sheet increases when the sur- factant concentrations increase from 0.002 to 0.15 mmol L�1 (Sup- porting Information, Fig. S3), indicating that the surfactants disrupt the secondary structure and lead to the unfolding of BSA molecule. With more and more surfactant molecules binding on BSA, the BSA may swell and the main chain of the protein is unfolded with exposing the hydrophobic residues, and then, the hydrogen-bond- ing networks [26] are destroyed and a-helices are broken to give a more open disordered structure, which will increase the percent- age of b-sheet. The more decrease in a-helical content, the larger extent of BSA unfolding [38]. The a-helical content of BSA (0.1 g L�1) in the pres- ence of [C12-s-C12im]Br2 or [Cn-4-Cnim]Br2 at 0.05 mmol L�1 is shown in Fig. 7. The decrease extent of a-helical content of BSA molecule in the presence of imidazolium gemini surfactant follows the order [C12-6-C12im]Br2 < [C12-4-C12im]Br2 < [C12-2-C12im]Br2 and [C10-4-C10im]Br2 < [C12-4-C12im]Br2 < [C14-4-C14im]Br2, sug- gesting that the imidazolium gemini surfactants with shorter spacer or with longer hydrophobic chain interact with BSA more strongly. This can result in a greater degree of protein unfolding, which is in consistent with our conclusions obtained from fluorescence. The secondary structure of BSA also can be greatly affected by the surfactants with different polar headgroups. For comparison, the secondary structures of BSA in the presence of imidazolium or quaternary ammonium cationic surfactants are shown in Supporting Information (Fig. S4). The a-helicity contents of BSA/ surfactants decrease to 29.6%, 33.8%, 38.3%, and 39.1% for 200 220 240 260 -20 -10 0 10 Wavelength / nm CD / m de g without [C12-4-C12im]Br2 0.002 mmol/L 0.01 mmol/L 0.05 mmol/L 0.07 mmol/L 0.15 mmol/L Fig. 6. CD spectra of BSA (0.1 g L�1) at different concentrations of [C12-4-C12im]Br2. The concentrations of [C12-4-C12im]Br2 are 0, 0.002, 0.01, 0.05, 0.07 and 0.15 mmol L�1, respectively. tant has a greater effect on the protein unfolding. 3.3. Mechanism of interaction between surfactants and BSA Quenching can be caused by a variety of factors [25], including molecular rearrangements, collision, excited-state reactions, and ground-state complex formation. Usually, the different mecha- nisms of quenching are classified as dynamic quenching and static quenching, which can be distinguished by their differing depen- dence on temperature and viscosity [25]. Higher temperature re- sults in faster diffusion and hence larger amounts of dynamic quenching. In our work, cationic surfactants contain cationic parts � [C12-2-C12im]Br2, C12C2C12, [C12mim]Br, and DTAB (in all cases, the surfactant concentration is 0.05 mmol L�1), respectively, namely, the decrease extent of a-helical content follows the order: DTAB < [C12mim]Br < C12C2C12 < [C12-2-C12im]Br2. It is seen that the decrease extent of a-helicity in the presence of gemini surfac- tant is bigger than its corresponding monomer. In addition, the de- crease extent of a-helicity in the presence of imidazolium surfactant is higher than that of the quaternary ammonium surfac- tant with the same hydrophobic chain. This reconfirms the conclu- sion obtained from the fluorescence results that, owing to the 29 30 31 32 33 34 α - he lic ity % Spacer Length (s) A 2 4 6 10 12 14 30 32 34 36 38 Chain Length (n) B Fig. 7. Contents of a-helicity of BSA (0.1 g L�1) in the presence of [C12-s-C12im]Br2. (A) [C12-s-C12im]Br2, s = 2, 4, 6; (B) [Cn-4-Cnim]Br2, n = 10, 12, 14. erface Science 389 (2013) 175–181 179 increasing the temperature, indicating that the probable quenching mechanism is initiated by ground-state complex formation rather than by dynamic collision. To distinguish static and dynamic quenching, one additional method is by careful examination of the absorption spectra of the fluorophores. Collisional quenching only affects the excited states of the fluorophores, and thus, no changes in the absorption spectra are expected. In contrast, ground-state complex formation will frequently result in perturbation of the absorption spectrum of Table 1 Stern–Volmer quenching constants, binding constants, and binding sites for the interaction of [C12-4-C12im]Br2 with BSA at different temperatures. T (K) 293 298 303 308 Ksv (104 L mol�1) 7.66 7.41 7.23 7.07 Ka (104 L mol�1) 6.21 5.84 5.40 5.02 n 1.2 ± 0.1 1.2 ± 0.1 1.1 ± 0.1 1.1 ± 0.1 180 T. Zhou et al. / Journal of Colloid and Int the fluorophore [32,43,58]. The UV–visible absorption spectra of BSA without and with [C12-4-C12im]Br2 are shown in Fig. 8. As can be seen, the addition of [C12-4-C12im]Br2 to BSA solution pro- duces a decrease in the intensity of spectra, accompanied by a red shift of the maximum absorption peak, which is caused by the n-P� transition of the peptide bond in BSA [59]. This confirms the probable quenching mechanism is initiated by ground-state complex formation, namely, a static quenching procedure. There- fore, the quenching data can be analyzed according to the follow- ing equation [25,32,58]: lg½ðF0 � FÞ=F� ¼ lgKa þ n lg½Q � ð2Þ where n is the binding sites. Ka is the binding constant of the surfac- tant with BSA, which can be determined by the slope of double log- arithm regression curve of lg[(F0 � F)/F] versus lg[Q] based on the Eq. (2). The corresponding values of Ka and n for BSA/[C12-4- C12im]Br2 at different temperatures are shown in Table 1. The num- ber of binding sites is n = 1.1 ± 0.1 under different temperatures, which suggests one molecule of the protein combines with one molecule of the surfactant. Both Ka and n are decreasing with increasing temperature, indicating the lower stability of ground- state complex at higher temperature due to the more drastic motil- ity between molecules. Similar decreasing trend of Ka is observed for the other surfactants and BSA mixed systems, which are not shown here, suggesting the quenching mechanisms of fluorescence of BSA by the surfactants are all the static quenching procedures. The values of Ka and n for the different BSA/surfactant ([C12-s- C12im]Br2, [Cn-4-Cnim]Br2, C12C2C12, [C12mim]Br, DTAB) systems at 298 K are measured (Supporting Information, Table S1). These results also reconfirm the conclusions obtained from CD and fluorescence spectra. 2.0 2.5 3.0 without [C12-4-C12im]Br2 0.005 mmol/L 0.05 mmol/L A 150 200 250 300 350 400 0.0 0.5 1.0 1.5 A bs or ba nc e Wavelength / nm 0.5 mmol/L D Fig. 8. UV–visible spectra of BSA (0.1 g L�1) in the presence of [C12-4-C12im]Br2. The concentrations of [C12-4-C12im]Br2 are 0, 0.005, 0.05 and 0.5 mmol L�1, respectively. 4. Conclusions The interaction between protein and surfactant is an active field of research interest, which is of great importance in biological, pharmaceutical, and cosmetic industries [10–13]. Based upon our earlier studies on the imidazolium gemini surfactants [Cn-4- Cnim]Br2 [49–52], in this work, we have investigated compara- tively that the BSA interacts with cationic gemini surfactants with imidazolium and quaternary ammonium groups, and their corre- sponding monomers via fluorescence probe technique of pyrene, circular dichroism, synchronous fluorescence, and UV–visible absorption spectra. The choices of the surfactants with different structure (imidazolium or quaternary ammonium; gemini or single chain) allow us to characterize the BSA/surfactant interactions as a function of the polar head group, the hydrophobic tail length, and spacer length of surfactants. The results show that [Cn-s-Cnim]Br2 may interact with BSA strongly, and [Cn-s-Cnim]Br2 with shorter spacer or longer hydrophobic tail have a larger effect on BSA unfolding. Additionally, the synchronous fluorescence spectra show that the surfactants mainly interact with the Trp residues of BSA, and fluorescence quenching of BSA goes probably through the static quenching procedure. The interactions between BSA and gemini surfactants with qua- ternary ammonium group have been studied [39,40,42], which have revealed that the interaction between them is stronger than that between BSA and conventional single chain surfactants. Significantly, our results confirm that the interactions between [Cn-s-Cnim]Br2 and BSA are not only much stronger in comparison with those of their corresponding monomers, but also they are stronger than those of between the quaternary ammonium ones with the same hydrophobic chain and BSA. This is attributed to the stronger p–p interaction between the imidazolium rings and the residues (such as Trp, Tyr, and Phe) in BSA, besides electrostatic and hydrophobic effects. Therefore, it may contribute to a funda- mental understanding on the effect of surfactant architecture in the interaction between surfactant and protein, and it may also provide guidance in selecting appropriate surfactants in pharma- ceutical, cosmetic, and biological systems. Acknowledgments We are grateful for financial support from the Natural Science Foundation of China (20833010) and the National Natural Science Foundation of Shandong Province, China (ZR2010BM002). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2012.08.067. References [1] R.G. Eckenhoff, J.S. Johansson, Pharmacol. Rev. 49 (1997) 343. [2] E. Dickinson, Colloids Surf. B 15 (1999) 161. [3] D.J. McClements, Curr. Opin. Colloid Interface Sci. 9 (2004) 305. [4] H.C. Mahler, W. Friess, U. Grauschopf, S. Kiese, J. Pharm. Sci. 98 (2009) 2909. [5] J.A. Reynolds, S. Herbert, H. Polet, J. Steinhardt, Biochemistry 6 (1967) 937. [6] I.M. Kolthoff, K. Yamashita, T.B. Hie, Proc. Natl. Acad. Sci. USA 72 (1975) 2044. [7] C.L. Mesa, J. Colloid Interface Sci. 286 (2005) 148. [8] J.M. Valderrama, J.M.R. Patino, Curr. Opin. Colloid Interface Sci. 15 (2010) 271. [9] D. Otzen, Biochim. Biophys. Acta 1814 (2011) 562. [10] Y.H. Miao, L.E. Helseth, Colloids Surf. B 66 (2008) 299. [11] T.T. Le, P. Sabatino, B. Heyman, M. Kasinos, H.H. Dinh, K. Dewettinck, J. Martins, P.V. Meeren, Food Hydrocolloids 25 (2011) 594. [12] H.J. Lee, A. McAuley, K.F. Schilke, J. McGuire, Advan. Drug Delivery Rev. 63 (2011) 1160. [13] H.M. Kwaambwa, A.R. Rennie, Biopolymers 97 (2011) 209. erface Science 389 (2013) 175–181 [14] E. Blanco, J.M. Ruso, G. Prieto, F. Sarmiento, J. Colloid Interface Sci. 316 (2007) 37. [15] H. Tani, T. Kamidate, H. Watanabe, J. Chromatogr. 780 (1997) 229. [16] G. Viscardi, P. Quagliotto, C. Barolo, P. Savarino, E. Barni, E. Fisicaro, J. Org. Chem. 65 (2000) 8197. [17] M. Simões, M.O. Pereira, I. Machado, L.C. Simões, M.J. Vieira, J. Ind. Microbiol. Biotechnol. 33 (2006) 741. [18] L. Caillier, E.T. Givenchy, R. Levy, Y. Vandenberghe, S. Geribaldi, F. Guittard, J. Colloid Interface Sci. 332 (2009) 201. [19] M. Vasilescu, D. Angelescu, Langmuir 15 (1999) 2635. [20] J.W. Brauner, C.R. Flach, R. Mendelsohn, J. Am. Chem. Soc. 127 (2005) 100. [21] S. Ghosh, Colloids Surf. B 41 (2005) 209. [22] R.G. Reed, R.C. Feldhoff, O.L. Clute, T.P. Je, Biochemistry 14 (1975) 4578. [23] J.A. Hamilton, D.P. Cistola, J.D. Morrisett, J.T. Sparrow, D.M. Small, Proc. Natl. Acad. Sci. USA 81 (1984) 3718. [24] S. Deep, J.C. Ahluwalia, Phys. Chem. Chem. Phys. 3 (2001) 4583. [25] U.S. Mote, S.H. Han, S.R. Patil, G.B. Kolekar, J. Luminesc. 130 (2010) 2059. [26] R. Kumaran, P. Ramamurthy, J. Fluoresc. 21 (2011) 1499. [27] S. Jana, S. Ghosh, S. Dalapati, N. Guchhait, Photochem. Photobiol. Sci. 11 (2012) 323. [28] F. Karush, J. Am. Chem. Soc. 72 (1950) 2705. [29] A. Valstar, M. Vasilescu, C. Vigouroux, P. Stilbs, M. Almgren, Langmuir 17 (2001) 3208. [30] Y. Moriyama, Y. Kawasaka, K. Takeda, J. Colloid Interface Sci. 257 (2003) 41. [31] N.J. Turro, X.G. Lei, K.P. Ananthapadmanabhan, M. Aronson, Langmuir 11 (1995) 2525–2533. [32] C.X. Sun, J.H. Yang, X. Wu, X.R. Huang, F. Wang, S.F. Liu, Biophys. J. 88 (2005) 3518. [33] C.C. Ruiz, J.M. Hierrezuelo, J. Aguiar, J.M.P. Garcia, Biomacromolecules 8 (2007) 2497. [34] T. Chakraborty, I. Chakraborty, S.P. Moulik, S. Ghosh, Langmuir 25 (2009) 3062. [35] A. Valstar, M. Almgren, W. Brown, Langmuir 16 (2000) 922. [36] D. Kelley, D.J. McClements, Food Hydrocolloids 17 (2003) 73. [37] A.A. Mikhailovskaya, B.A. Noskov, S.Y. Lin, G. Loglio, R. Miller, J. Phys. Chem. B 115 (2011) 9971. [38] H.K. Mandal, A. Kundu, S. Balti, A. Mahapatra, J. Colloid Interface Sci. 378 (2012) 110. [39] Y.J. Li, X.Y. Wang, Y.L. Wang, J. Phys. Chem. B 110 (2006) 8499. [40] N. Gull, P. Sen, R.H. Khan, K. Din, Langmuir 25 (2009) 11686. [41] L.F. Chen, Y.Z. Shang, H.L. Liu, Y. Hu, J. Colloid Interface Sci. 301 (2006) 644. [42] M.A. Mir, J.M. Khan, R.H. Khan, G.M. Rather, A.A. Dar, Colloids Surf. B 77 (2010) 54. [43] S. Tu, X.H. Jiang, L.M. Zhou, W.M. Yin, H.C. Wang, M. Duan, P.L. Liu, X.M. Jiang, J. Luminesc. 132 (2012) 381. [44] M.J. Rosen, J.H. Mathias, L. Davenport, Langmuir 15 (1999) 7340. [45] R. Zana, Adv. Colloid Interface Sci. 97 (2002) 205. [46] X.D. Zhai, M.H. Liu, J. Colloid Interface Sci. 295 (2006) 181. [47] H. Yan, J.Y. Wu, G.L. Dai, A.G. Zhong, H. Chen, J.G. Yang, D.M. Han, J. Luminesc. 132 (2012) 622. [48] D. Wu, G.Y. Xu, Y.H. Sun, H.X. Zhang, H.Z. Mao, Y.J. Feng, Biomacromolecules 8 (2007) 708. [49] M.Q. Ao, G.Y. Xu, Y.Y. Zhu, Y. Bai, J. Colloid Interface Sci. 326 (2008) 490. [50] M.Q. Ao, G.Y. Xu, J.Y. Pang, T.T. Zhao, Langmuir 25 (2009) 9721. [51] M.Q. Ao, P.P. Huang, G.Y. Xu, X.D. Yang, Y.J. Wang, Colloid Polym. Sci. 287 (2009) 395. [52] M.Q. Ao, G.Y. Xu, W.L. Kang, L.W. Meng, H.J. Gong, T. Zhou, Soft. Matter. 7 (2011) 1199. [53] Q.Q. Baltazar, J. Chandawalla, K. Sawyer, J.L. Anderson, Colloids Surf. A: Physicochem. Eng. Aspects 302 (2007) 150. [54] W.G. Liu, K.D. Yao, G.C. Wang, H.X. Li, Polymer 41 (2000) 7589. [55] D. Patra, C. Barakat, R.M. Tafech, Colloids Surf. B 94 (2012) 354. [56] F.M. Winnik, S.T.A. Regismond, Colloids Surf., A 118 (1996) 1. [57] Y. Moriyama, D. Ohta, K. Hachiya, Y. Mitsui, K. Takeda, J. Protein Chem. 15 (1996) 265. [58] B. Ojha, G. Das, Chem. Phys. Lipids 164 (2011) 144. [59] H. Polet, J. Steinhardt, Biochemistry 7 (1968) 1348. T. Zhou et al. / Journal of Colloid and Interface Science 389 (2013) 175–181 181 Interactions of bovine serum albumin with cationic imidazolium and quaternary ammonium gemini surfactants: Effects of surfactant architecture 1 Introduction 2 Experimental 2.1 Materials 2.2 Methods 2.2.1 Fluorescence measurement 2.2.2 Circular dichroism measurement (CD) 2.2.3 UV–visible absorption measurement 3 Results and discussion 3.1 Fluorescence property for the BSA/surfactant systems 3.2 The change of BSA secondary structure in the presence of surfactants 3.3 Mechanism of interaction between surfactants and BSA 4 Conclusions Acknowledgments Appendix A Supplementary material References