- Email: [email protected]
Monitoring the effects of strong cosolvent hexafluoroisopropanol in investigation of the tetrameric structure and stability of K+-channel KcsA
Archives of Biochemistry and Biophysics 498 (2010) 1–6
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Original paper
Monitoring the effects of strong cosolvent hexafluoroisopropanol in investigation of the tetrameric structure and stability of K+-channel KcsA Mobeen Raja * Department of Biochemistry of Membranes, Centre for Biomembranes and Lipid Enzymology, Institute of Biomembranes, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
a r t i c l e
i n f o
Article history: Received 16 March 2010 Available online 27 March 2010 Keywords: Potassium channel KcsA Tetrameric stability Hexafluoroisopropanol Tryptophan fluorescence Circular dichroism Membrane lateral pressure Protein–lipid interaction
a b s t r a c t Adsorption of small chain alcohols into lipid membranes significantly changes the conformational states of intrinsic membrane proteins. In this study, the effects of membrane-active strong cosolvent hexafluoroisopropanol (HFIP) on the intrinsic tetrameric stability of potassium channel KcsA were investigated. Presence of acidic phosphatidylglycerol (PG) in non-bilayer phosphatidylethanolamine (PE) or bilayer phosphatidylcholine (PC) significantly increased the tetrameric stability compared to zwitterionic pure PC bilayers. The stabilizing effect of PG in both lipid bilayers was completely abolished upon deletion of the membrane-anchored N-terminus. Tryptophan fluorescence and circular dichroism experiments indicated that HFIP destabilizes the tetramer possibly via drastic changes in the lateral pressure profile close to the membrane–water interface. The data suggest that HFIP disturbs the ionic, H-bonding and hydrophobic interactions among KcsA subunits where N-terminus presumably plays a crucial role in determining the channel proper folding and tetrameric structure via ionic/H-bond interactions between the helix dipole and the membrane lipids. Ó 2010 Elsevier Inc. All rights reserved.
Introduction The oligomerization of protein macromolecules on cell surfaces is believed to play a fundamental role in the regulation of cellular function, including signal transduction and the immune response [1,2]. Oligomerization can bring several functionally important advantages [3,4]: (i) Oligomerization can give shape to active sites or even lead to their compartment-alization. It has been estimated that about one sixth of oligomeric enzymes have their active sites located at oligomeric interfaces; (ii) Oligomerization can allow cooperativity between subunits, enabling allosteric regulation as an additional level of control; (iii) Oligomerization can serve as a tool to create multivalency in active or interactive sites, increasing affinity of the complexes for substrates or binding partners; (iv) Oligomerization can enhance protein stability; (v) For those proteins that have different activities in their oligomeric and monomeric states, oligomerization can provide an additional level of regulation; and (vi) For hetero-oligomeric complexes, oligomerization may allow the formation of enzymatic and signaling cascades. To remain soluble in the native state, oligomeric membrane proteins (MPs) need an amphiphilic environment, which can
* Present address: School of Systems Medicine, 6126 HRIF East, Alberta Diabetes Institute, University of Alberta, Edmonton, Canada AB T6G 2E1. Fax: +1 780 492 8836. E-mail address: [email protected] 0003-9861/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2010.03.014
be provided by detergents or lipids. a-Helical MPs typically contain long consecutive sequences of hydrophobic amino acids in the transmembrane segments that allow the helices to dock against each other, stabilized through van der Waals interactions [5–7]. The environment of the MP stabilizing the native state can play a role in the protein’s response to alcohols. Alcohols can also modulate the oligomerization of MPs in lipid bilayers. This can occur indirectly by redistributing lateral membrane pressure in a manner which correlates with alcohol hydrophobicity [8]. Among various alcohols, trifluoroethanol (TFE) is often used because of its high potential for stabilizing the a-helical structure [9,10]. The secondary structures stabilized by TFE are assumed to reflect conformations that prevail in the early stages of protein folding. However, hexafluoroisopropanol (HFIP), a compound with six F-atoms, is the most effective cosolvents for the structural stabilization of secondary structure forming peptides. In particular, it is one of the strongest helix-inducing and stabilizing cosolvents [11]. It has a pKa of 9.3; hence, it is more acidic than its hydrocarbon analogue TFE (pKa 12.4). The presence of two CF3 groups alters its properties to a great extent. It is a better H-bond donor and poorer H-bond acceptor than TFE. Hence, HFIP is potentially more powerful than TFE in terms of perturbing the ionic, H-bonding, and hydrophobic interactions in proteins. HFIP has been used to unfold aggregates of the Alzheimer’s amyloid peptide [12] or prion protein peptides [13].
2
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
The potassium channel KcsA is an oligomeric MP from Streptomyces lividans, which is a convenient model protein to study MP oligomerization [14]. The tetrameric structure of KcsA is highly stable in a wide range of detergents, even in SDS [15]. This high stability is caused not only by interactions between protein subunits but also by interactions between the protein and the surrounding lipid bilayer [16,17]. Previous studies on KcsA have shown that the stability of the tetramer in lipids is reduced by molar concentrations of alcohols, like TFE, in an indirect manner which can be related to their ability to alter the lateral membrane pressure and permeabilize the membrane [18,19]. More recently, TFE has been proven to be an authentic tool in investigation of channel assembly and stability of the tetrameric structure of KcsA [20]. The present study addresses the question that how does this tetrameric channel respond to HFIP-induced changes in membrane lateral pressure profile and how does any alteration in membrane profile affect the intrinsic stability, structure and conformation of KcsA in a lipid bilayer. Experimental procedures Reagents 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dioleoyl-sn-glycero-3-phospho-glycerol (DOPG) and 1,2-dioleoyl-snglycero-3-phospho-ethanolamine (DOPE) were purchased from Avanti Polar Lipids Inc. n-Dodecyl-b-D-maltoside (DDM)1 was from Anatrace Inc. 1,1,1,3,3,3-Hexafluoroisopropanol (HFIP) was obtained from Merck. Protein expression, purification and reconstitution Wild type (WT) and mutant-lacking the first 18 amino acids, named ‘‘(DN-KcsA)”-were expressed and purified according to the method described previously [17,20]. The purity of proteins was assessed by SDS–PAGE. Proteins were reconstituted in different lipid mixtures with a 1:1000 protein:lipid molar ratio, and the resultant proteoliposomes were obtained as described previously [17]. Stability assay of tetramer dissociation by HFIP Small aliquots of proteoliposomes were incubated with variable concentrations of HFIP for 1 h at room temperature. Samples were mixed with an electrophoresis sample buffer (50 mM Tris–HCl, pH 6.8, 50% glycerol, 0.01% bromophenol blue and 10% SDS) and run on 15% acrylamide gel in the presence of 0.1% SDS. Gels were stained by silver nitrate, scanned by a densitometer (Bio-Rad Laboratories) and quantified with the program Quantity One. The amount of tetramer (%) was plotted against HFIP (vol.%) for the stability assay, as described previously [17]. Tryptophan fluorescence spectroscopy All fluorescence experiments were performed in vesicle buffer at room temperature using a QuantaMaster QM-1/2005 spectrofluorometer (Photon Technology International, NJ) in a quartz cuvette. The samples were excited at 280 nm and emission spectra were collected between 300 and 400 nm. The bandwidths for both excitation and emission monochromators were 5 nm and the data were corrected as described previously [17]. All fluorescence 1 Abbreviations used: DDM, n-dodecyl-b-D-maltoside; HFIP, hexafluoroisopropanol; MP, membrane protein; PC, phosphatidylcholine; PG, phosphatidylglycerol; PE, phosphatidylethanolamine; SDS, sodium dodecyl sulphate; PAGE, polyacrylamide gel electrophoresis; TFE, trifluoroethanol; Trp, tryptophan; WT, wild type.
experiments were performed at 3 lM protein concentration present in proteoliposomes. Circular dichroism analysis Far-UV CD spectra were recorded between 200 and 250 nm (350 ll sample volume) on a Jasco J-810 spectropolarimeter (equipped with a temperature-controlled incubator) at 20 °C using 1 mm optical path length quartz cells. The step size was 0.5 nm with a 1.0-nm-bandwidth at a scan speed of 10 nm/min. All measurements were performed under nitrogen flow. As a blank, the CD spectra of lipid mixture in the absence and presence of HFIP (vol.%) were measured. Averages of 10 scans were obtained for blank and protein spectrum and all spectra were corrected for the buffer and vesicle scattering effects. Predicted percentages for different kinds of secondary structures were calculated using the K2d computer modeling program as described previously [21], and the results were expressed as mean residue ellipticity in units of degrees/cm2/dmol. The spectra were recorded at 0.1 mg/ml protein concentration in 50 mM Tris–HCl (pH 7.5). Results HFIP-induced tetramer dissociation in WT-KcsA Fig. 1A (upper panel) illustrates silver-stained gels of WT-KcsA dissociation in PC:PG (7:3 mol.%) and PE:PG (7:3 mol.%) lipid bilayers as a function of HFIP (vol.%). Dissociation of WT tetramer (T), which runs at 68 kDa, required 6 vol.% HFIP to completely dissociate the tetramer into its monomeric (M) subunits, which run at 18 kDa, as reported previously [17,20]. The amount of monomer in PC:PG bilayer almost remained the same throughout the whole range of HFIP; however, in PE:PG the monomer population was disappeared upon increasing HFIP concentration lipid bilayer, which reappeared upon complete tetramer dissociation. This effect can be explained by aggregation of KcsA in the presence of HFIP which could not be seen on SDS-gel. It was, however, interesting to note that the intensity of monomeric KcsA in PE:PG bilayer was relatively increased at higher HFIP concentration compared to PC:PG system. It suggests that non-bilayer lipid PE might be required for proper folding of monomeric KcsA as also reported previously [16]. Fig. 1B summarizes the data from the experiments illustrated in Fig. 1A. In addition, the summarized data also demonstrate KcsA tetrameric stability either in detergent micelles or in a pure PC bilayer (SDS-gels not shown) for comparison. DDM-solubilized WTKcsA tetramer (0.1 mg/ml) was dissociated at 2 vol.% HFIP. In pure PC, the stability of KcsA was increased such that 4 vol.% HFIP was required for complete tetramer dissociation. In PC:PG or PE:PG, the tetrameric stability was significantly increased as discussed above. However, the quantification of gels clearly indicated that tetrameric stability was unaffected in the range of 1–3 vol.% HFIP in PE:PG lipid bilayer compared to PC:PG system. These results agree well with the previous observation that non-bilayer lipid PE maximally stabilizes the tetramer [18,19]. HFIP-induced tetramer dissociation in DN-KcsA
DN-KcsA, which forms a stable tetramer, required significantly less amounts of HFIP to dissociate the tetramer such that 4 vol.% HFIP was able to completely dissociate the tetramer in both lipid bilayers (Fig. 1A lower panel). In addition, no monomeric DN-KcsA could be detected either in the absence or presence of HFIP. However, at 4 vol.% HFIP a monomeric band appeared upon tetramer dissociation indicating that HFIP solubilized the monomeric KcsA
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
3
Fig. 1. (A) Representative silver-stained gels of HFIP-induced tetramer dissociation in WT and DN-KcsA in PC:PG or PE:PG (7:3 mol.%) lipid bilayers. Tetrameric (T) and monomeric (M) KcsA are indicated. The concentration of HFIP (vol.%) is indicated at the top of each gel. (B) Quantification of silver-stained gels shown in panel A. In addition, the tetrameric stabilities analysed for DDM-solubilized protein or KcsA reconstituted in pure PC bilayers are also shown. The intensities of tetramer bands were assigned as a relative value of 100% observed for TFE untreated (0 vol.% TFE) sample. (C) Proteins were reconstituted in LUV’s and incubated with various concentrations of HFIP for 1 h at room temperature. All data points correspond to the average and ±SD of three experiments.
upon tetramer dissociation which was not soluble enough to be observed by SDS–PAGE. The quantification of the gels shown in Fig. 1A (lower panel) is summarized in Fig. 1C. Similar to WT-KscA, the stability of DN-KcsA in detergent micelles or pure PC was also investigated (gels not shown). DDM-solubilized DN-KcsA (0.1 mg/ ml) was also dissociated at 2 vol.% HFIP, however, a significant reduction in tetramer fraction was observed at 1 vol.% HFIP compared to WT-KcsA. In pure PC, the DN-KcsA tetramer was as stable as WT-KcsA. However, the stabilizing effect in PC:PG or PE:PG was completely abolished compared to WT-KcsA. These results indicated that N-terminus stabilized the tetramer via electrostatic interactions between the basic residues and the acidic PG. It was interesting to note that KcsA dissociation as a function of HFIP (vol.%) was quite similar to TFE-induced tetramer dissociation pattern reported previously [17,20]. However, the amount of HFIP required to dissociate the tetramer either in detergent micelles or in different lipid bilayers is 10 times less than TFE. Thus, HFIP, a strong cosolvent and more acidic than TFE, has stronger effects on KcsA tetrameric stability.
HFIP-induced changes in Trp fluorescence and protein unfolding Since TFE has been shown to unfold the tetrameric structure via increase in hydrophilicity or accessibility of Trp residues to a local environment [17,22], the effects of HFIP were also determined during HFIP-induced channel unfolding. Fluorescence emission spectra for the effects of 1 and 4 vol.% HFIP on WT and DN-KcsA are shown in Fig. 2A and B, respectively, and he effects of HFIP on Trp fluorescence properties are compiled in Table 1.
For comparison, similar amounts of HFIP were used regardless of dissociation concentration for both proteins. Incubation of WTKcsA with 1 and 4 vol.% HFIP led to fluorescence quenching. In addition, significant blue shifts in emission maxima were observed which is in contrast to TFE-induced unfolding behaviour of KcsA shown previously [17,22]. Blue shifts in emission maxima clearly indicated that Trp residues are located in a more hydrophobic environment upon HFIP insertion in between the lipid head groups. For DN-KcsA, relatively stronger effects of HFIP were monitored compared to WT-KcsA indicating that these changes are mainly elicited upon N-terminus deletion. In other words, N-terminus could be directly involved in major conformational or structural changes during tetramer dissociation. Effect of HFIP on the secondary structure of KcsA The effect of HFIP on the secondary and tertiary structure of both proteins were investigated in a PC:PG lipid bilayer. To obtain information about the secondary structure, circular dichroism (CD) spectroscopy was used. The resulting spectra of WT and DN-KcsA are shown in Fig. 2C and D, respectively, with the estimated secondary structure compiled in Table 2. Without HFIP, WT-KcsA protein contained slightly high content of random structure, moderate content of a-helical and low content of b-sheet structure. The structural information especially the a-helical content is in good agreement what has been published previously [15]. However, a slight reduction in a-helical content could be due to difference in protein reconstitution medium or presence of detergent micelles which seem to slightly increase the a-helical content compared to reconstituted samples as reported previously [15].
4
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
Fig. 2. Effects of HFIP on Trp fluorescence emission spectra of WT (A) and DN-KcsA (B) in PC:PG (7:3 mol.%) bilayers. Trp fluorescence was monitored either in the absence (solid) or presence of 1 (dash) or 4 (dotted) vol.% HFIP. Far-UV CD spectra of WT (C) and DN-KcsA (D) in the absence (solid) or presence of 4 vol.% HFIP (dash) in PC:PG (7:3 mol.%) bilayers. Effect of 1 vol.% HFIP on both proteins elicited no significant change in the secondary structure content compared to HFIP-untreated samples; hence these spectra are not shown.
Table 1 Effect of HFIP on Trp fluorescence of WT and DN-KcsA in PC:PG (7:3 mol.%) lipid bilayer.a HFIP (vol.%)
0 1 4 a
Emission wavelength (nm)
Fluorescence quenching (%)
Blue shift (nm)
WT
DN
WT
DN
WT
DN
322 ± 0.8 319 ± 1.2 316 ± 1.6
321 ± 1.2 316 ± 1.4 308 ± 4
– 15 ± 6 47 ± 15
– 40 ± 11 82 ± 14
– 3 ± 0.4 6±2
– 5 ± 1.2 13 ± 3
Parameters of Trp fluorescence were derived from data shown in Fig. 2A and B.
Table 2 Effect of HFIP on secondary structure of WT and DN-KcsA in PC:PG (7:3 mol.%) lipid bilayer.a HFIP (vol.%)
0 4
WT
DN
a-Helix
b-Strand
Random
a-Helix
b-Strand
Random
42 ± 5 38 ± 5
15 ± 1.5 17 ± 4
43 ± 6 45 ± 8
32 ± 5 40 ± 3
32 ± 4 12 ± 2
36 ± 7 48 ± 5
a Parameters of secondary structure were derived from data shown in Fig. 2C and D. Structural analysis was performed for the average CD spectra corrected for the blank and for volume increase due to HFIP. The values are the average and ±SD of predicted percentage for secondary structures calculated by K2d computer modeling program [21].
Incubation of WT-KcsA with 1% (not shown) or 4 vol.% HFIP did not change the CD spectrum significantly. It was also interesting to note that 4 vol.% HFIP, which is quite close to the tetramer-tomonomer transition and at which 50% tetramer is stable in a PC:PG bilayer, still contains similar structural information as in HFIP-untreated KcsA protein. This suggests that HFIP has an indirect effect on tetrameric stability without changing the secondary structure.
For DN-KcsA, a decrease in a-helical content, increase in bsheet and slight reduction in random content was observed compared to WT when no HFIP was present in the medium. A reduction in a-helical content can be explained on the basis of a-helical Nterminus deletion. However, it seems that deletion of the N-terminus induced secondary structure in the protein. This could either be due to protein aggregation of monomeric KcsA as estimated by SDS–PAGE. Incubation of DN-KcsA with 4 vol.% HFIP significantly decreased the content of b-sheets and increased the a-helical content indicating unfolding or solubilization of aggregated monomers which could be seen by SDS–PAGE at 4 vol.% concentration of HFIP.
Discussion Short-chain alcohols as anesthetics molecules can bind specifically to proteins and this may influence the binding of ions to the protein acting as ion channels, shifting protein conformational equilibria [5,7]. The potassium channel KcsA, which acts as a representative ion channel, is a convenient model protein to study the effect of anesthetic agents on MP complexes. The extremely high stability of KcsA channel arises from interaction between hydrophobic part of the subunit and hydrogen bonding between polar residues in the selectivity filter [23]. However, TFE could disrupt these interactions resulting in dissociation of KcsA tetramer into its monomer suggesting that tetramer stabilizing interactions occur mainly in the transmembrane domain of the protein [18,19]. On the other hand, alcohols have significant effects on the physical properties of biological membranes. Such changes in the membrane properties, in turn, can lead to changes in the conformational states of intrinsic MPs, thus leading to an indirect (non-specific) mechanism for the modulation of protein behaviour by alcohol adsorption into lipid membranes [17–19,24]. The detailed molecular mechanism for these effects is not yet fully understood. Some
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
5
Fig. 3. Schematic model of the effect of HFIP on the membrane lateral pressure acting on KcsA tetrameric structure. (A) Each monomer is of KcsA shown in four distinct colors highlighting the positions of the N and C-termini. (B) The presence of low concentration of HFIP squeezes the tetramer with the help of N-terminus located at the membrane– water interface thereby protecting the tetramer against HFIP-induced drastic changes in the membrane lateral pressure profile. This results in stabilization of compact structure of KcsA tetramer. (C) Increasing HFIP concentration eventually destabilizes the tetramer into its monomeric subunits possibly via exposure of the pore domain and the N-terminus which are not inserted in the lipid bilayer anymore, because of the partitioning of HFIP in the lipid bilayer and the direct interactions of HFIP with the protein.
short-chain alcohols are amphiphilic, they have been suggested to localize primarily in the headgroup region of the lipid bilayer [25]. This disrupts the packing in the lipid bilayer and leads to a variety of changes, among them observed increase in membrane fluidity [26], in membrane permeability [27], and lipid lateral mobility [28]. The present study was set out to investigate if strong cosolvent like HFIP affects tetramer stability via similar mechanism as TFE [18–20]. First, denaturation experiments were performed in which KcsA tetrameric stability was analysed. Incubation of detergent solubilized KcsA with HFIP resulted in tetramer dissociation indicating that HFIP directly affects the KcsA structure possibly via exposure of hydrophobic surface of the transmembrane helices. Second, an increased stability of KcsA in membranes compared to detergent micelles was observed indicating that lipid bilayer limited the exposure of the transmembrane helices of KcsA to HFIP. These effects were in total agreement with TFE-induced tetramer dissociation behaviour [18,19]. However, the advantage of using HFIP over TFE was that less amounts were required to investigate which interactions stabilize the oligomeric structure of KcsA in a particular environment. It was interesting to observe the maximum stability of KcsA tetramer in a PE:PG bilayer compared to PC:PG or pure PC since small headgroup size of PE allows larger insertion sites for alcohols which help maintaining the bilayer structure [18,19]. NMR-experiments have indicated that small alcohols insert mainly in the headgroup region of the lipid bilayer and that the insertion of small alcohols in the bilayer containing non-bilayer lipid PE results in an increased lateral pressure in the acyl chain region leading to a higher KcsA tetramer stability [19]. The stabilizing effects of PE and PG can also be justified by the fact that both lipids are important for efficient membrane association and tetramerization of KcsA [16] and that KcsA preferentially interacts with both these lipids [14,16]. The N-terminus, which is directly associated with the lipid headgroups, had a significant influence on tetramer stability. Increase in tetramer stability in the presence of acidic PG confirmed the stabilizing effect of N-terminus via electrostatic interactions between its basic residues and the acidic lipids as recently reported [20]. However, such effect was found to be more pronounced for HFIP compared to TFE [17–19] indicating that HFIP could promote even stronger ionic interactions between the charged PG and the basic residues thereby making electrostatic/H-bond switch much powerful. But, what happens to KcsA structure and conformation during HFIP-induced tetramer dissociation? Incubation of WT-KcsA with
HFIP elicited strong hydrophobicity in a local environment indicating that HFIP promoted tetramer squeezing or a more compact structure of KcsA (see model in Fig. 3). Even at or close to dissociating concentration of HFIP a strong hydrophobicity was monitored indicating that HFIP-induced changes in protein conformation might increase protein hydrophobicity by exposing hydrophobic side chains and shielding polar amide groups from the solvent [29]. This behaviour is completely opposite to TFE-induced KcsA unfolding mechanism [17,22]. Interestingly, no change in the secondary structure was observed for WT-KcsA which confirmed that HFIP exerted an indirect effect on tetrameric stability possibly via drastic changes in the membrane lateral pressure profile. Removal of the N-terminus elicited strong changes in Trp fluorescence properties and increase in b-sheet content compared to WT-KcsA indicating that tetramer might undergo significant structural and conformational changes upon N-terminus deletion. Furthermore, incubation of this protein with HFIP resulted in solubilization of b-sheet structure thereby increasing the a-helical content. Such propensity of HFIP to convert aggregation or b-sheet content of several peptides into the helical structure has also been illustrated in other studies [11–13]. Thus, N-terminus seems crucial for controlling overall protein structure and proper conformation. The application of these studies might be useful in understanding the role of membrane bound extramembranous parts of ion channels in oligomerization as well as the nature of small alcohols in affecting the membrane lateral pressure profile and consequent changes in the folding properties of integral MP complexes. Although ion channels and other MPs have been reported to harbor relatively specific sites for general anesthetic agents [30,31], the physicochemical and molecular features of these sites are not yet well-understood. In addition to the presence of physically circumscribed hydrophobic protein cavities, which constitute the alcohol and general anesthetic sites [32], N-terminus is possibly one of the major binding sites for anesthetic molecules. It seems to play an important role in maintaining the membrane lateral pressure profile which might be disturbed by a number of anesthetic agents. Acknowledgments The construct of N-terminal deleted (DN) KcsA was a precious gift from the late Jeanette G. Stam. The careful secretarial help of Irene van Duin and helpful suggestions of Antoinette Killian and Ben de Kruijff (Department of Biochemistry of Membranes, Utrecht
6
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
University) are gratefully acknowledged. This work was supported by funds of the Chemical Sciences Division (CW) of The Netherlands Organization for Scientific Research (NWO). References [1] A. Ullrich, J. Schlessinger, Cell 61 (1990) 203–212. [2] H. Metzger, J. Immunol. 149 (1992) 1477–1487. [3] D.S. Goodsell, A.J. Olson, Annu. Rev. Biophys. Biomol. Struct. 29 (2000) 105– 153. [4] M.H. Ali, B. Imperiali, Bioorg. Med. Chem. 13 (2005) 5013–5020. [5] J.L. Popot, D.M. Engelman, Biochemistry 29 (1990) 4031–4037. [6] J.L. Popot, D.M. Engelman, Annu. Rev. Biochem. 69 (2000) 881–922. [7] M.H.B. Stowell, D.C. Rees, Adv. Protein Chem. 46 (1995) 279–311. [8] D.E. Otzen, L. Nesgaard, Biochemistry 46 (2007) 4348–4359. [9] H.J. Dyson, P.E. Wright, Curr. Opin. Struct. Biol. 3 (1993) 60–65. [10] J.J. Yang, M. Buck, M. Pitkeathly, M. Kotik, D.T. Haynie, C.M. Dobson, S.E. Radford, J. Mol. Biol. 252 (1995) 483–491. [11] N. Hiroto, K. Mizuno, Y. Goto, Protein Sci. 6 (1997) 416–421. [12] S.J. Wood, B. Maleeff, T. Hart, R. Wetzel, J. Mol. Biol. 256 (1996) 870–877. [13] H. Zhang, K. Kaneko, J.T. Nguyen, T.L. Livshits, M.A. Baldwin, F.E. Cohen, T.L. James, S.B. Prusiner, J. Mol. Biol. 250 (1995) 514–526. [14] F.I. Valiyaveetil, Y. Zhou, R. MacKinnon, Biochemistry 41 (2002) 10771–10777. [15] D.M. Cortes, E. Perozo, Biochemistry 36 (1997) 10343–10352. [16] A. Van Dalen, S. Hegger, J.A. Killian, B. de Kruijff, FEBS Lett. 525 (2002) 33–38.
[17] M. Raja, R.E.J. Spelbrink, B. de Kruijff, J.A. Killian, FEBS Lett. 581 (2007) 5715– 5722. [18] E. van den Brink-van der Laan, V. Chupin, J.A. Killian, B. de Kruijff, Biochemistry 43 (2004) 5937–5942. [19] E. van den Brink-van der Laan, V. Chupin, J.A. Killian, B. de Kruijff, Biochemistry 43 (2004) 4240–4250. [20] M. Raja, J. Membr. Biol. 234 (2010) 235–240. [21] M.M. Raja, N.K. Tyagi, R.K.H. Kinne, J. Biol. Chem. 278 (2003) 49154–49163. [22] F.N. Barrera, M.L. Renart, M.L. Molina, J.A. Poveda, J.A. Encinar, A.M. Fernandez, J.L. Neira, J.M. Gonzalez-Ros, Biochemistry 44 (2005) 14344–14352. [23] D.A. Doyle, C.J. Morais, R.A. Pfuetzner, A. Kuo, J.M. Gulbis, S.L. Cohen, B.T. Chait, R. MacKinnon, Science 280 (1998) 69–77. [24] R.S. Cantor, Toxicol. Lett. 100 (1998) 451–458. [25] M. Patra, E. Salonen, E. Terama, I. Vattulainen, R. Faller, B. Lee, J. Holopainen, M. Karttunen, Biophys. J. 90 (2006) 1121–1135. [26] J.H. Chin, D.B. Goldstein, Mol. Pharmacol. 13 (1977) 435–441. [27] H. Komatsu, S. Okada, Chem. Phys. Lipids 85 (1997) 67–74. [28] S.Y. Chen, B. Yang, K. Jacobson, K.K. Sulik, Alcohol 13 (1996) 589–595. [29] Y.O. Kamatari, T. Konno, M. Kataoka, K. Akasaka, J. Mol. Biol. 259 (1996) 512– 523. [30] R.W. Peoples, C. Li, F.F. Weight, Annu. Rev. Pharmacol. Toxicol. 36 (1996) 185– 201. [31] K.W. Miller, Br. J. Anaesth. 89 (2002) 17–31. [32] M.P. Mascia, J.R. Trudell, R.A. Harris, Proc. Natl. Acad. Sci. USA 97 (2000) 9305– 9310.
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi
Original paper
Monitoring the effects of strong cosolvent hexafluoroisopropanol in investigation of the tetrameric structure and stability of K+-channel KcsA Mobeen Raja * Department of Biochemistry of Membranes, Centre for Biomembranes and Lipid Enzymology, Institute of Biomembranes, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
a r t i c l e
i n f o
Article history: Received 16 March 2010 Available online 27 March 2010 Keywords: Potassium channel KcsA Tetrameric stability Hexafluoroisopropanol Tryptophan fluorescence Circular dichroism Membrane lateral pressure Protein–lipid interaction
a b s t r a c t Adsorption of small chain alcohols into lipid membranes significantly changes the conformational states of intrinsic membrane proteins. In this study, the effects of membrane-active strong cosolvent hexafluoroisopropanol (HFIP) on the intrinsic tetrameric stability of potassium channel KcsA were investigated. Presence of acidic phosphatidylglycerol (PG) in non-bilayer phosphatidylethanolamine (PE) or bilayer phosphatidylcholine (PC) significantly increased the tetrameric stability compared to zwitterionic pure PC bilayers. The stabilizing effect of PG in both lipid bilayers was completely abolished upon deletion of the membrane-anchored N-terminus. Tryptophan fluorescence and circular dichroism experiments indicated that HFIP destabilizes the tetramer possibly via drastic changes in the lateral pressure profile close to the membrane–water interface. The data suggest that HFIP disturbs the ionic, H-bonding and hydrophobic interactions among KcsA subunits where N-terminus presumably plays a crucial role in determining the channel proper folding and tetrameric structure via ionic/H-bond interactions between the helix dipole and the membrane lipids. Ó 2010 Elsevier Inc. All rights reserved.
Introduction The oligomerization of protein macromolecules on cell surfaces is believed to play a fundamental role in the regulation of cellular function, including signal transduction and the immune response [1,2]. Oligomerization can bring several functionally important advantages [3,4]: (i) Oligomerization can give shape to active sites or even lead to their compartment-alization. It has been estimated that about one sixth of oligomeric enzymes have their active sites located at oligomeric interfaces; (ii) Oligomerization can allow cooperativity between subunits, enabling allosteric regulation as an additional level of control; (iii) Oligomerization can serve as a tool to create multivalency in active or interactive sites, increasing affinity of the complexes for substrates or binding partners; (iv) Oligomerization can enhance protein stability; (v) For those proteins that have different activities in their oligomeric and monomeric states, oligomerization can provide an additional level of regulation; and (vi) For hetero-oligomeric complexes, oligomerization may allow the formation of enzymatic and signaling cascades. To remain soluble in the native state, oligomeric membrane proteins (MPs) need an amphiphilic environment, which can
* Present address: School of Systems Medicine, 6126 HRIF East, Alberta Diabetes Institute, University of Alberta, Edmonton, Canada AB T6G 2E1. Fax: +1 780 492 8836. E-mail address: [email protected] 0003-9861/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2010.03.014
be provided by detergents or lipids. a-Helical MPs typically contain long consecutive sequences of hydrophobic amino acids in the transmembrane segments that allow the helices to dock against each other, stabilized through van der Waals interactions [5–7]. The environment of the MP stabilizing the native state can play a role in the protein’s response to alcohols. Alcohols can also modulate the oligomerization of MPs in lipid bilayers. This can occur indirectly by redistributing lateral membrane pressure in a manner which correlates with alcohol hydrophobicity [8]. Among various alcohols, trifluoroethanol (TFE) is often used because of its high potential for stabilizing the a-helical structure [9,10]. The secondary structures stabilized by TFE are assumed to reflect conformations that prevail in the early stages of protein folding. However, hexafluoroisopropanol (HFIP), a compound with six F-atoms, is the most effective cosolvents for the structural stabilization of secondary structure forming peptides. In particular, it is one of the strongest helix-inducing and stabilizing cosolvents [11]. It has a pKa of 9.3; hence, it is more acidic than its hydrocarbon analogue TFE (pKa 12.4). The presence of two CF3 groups alters its properties to a great extent. It is a better H-bond donor and poorer H-bond acceptor than TFE. Hence, HFIP is potentially more powerful than TFE in terms of perturbing the ionic, H-bonding, and hydrophobic interactions in proteins. HFIP has been used to unfold aggregates of the Alzheimer’s amyloid peptide [12] or prion protein peptides [13].
2
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
The potassium channel KcsA is an oligomeric MP from Streptomyces lividans, which is a convenient model protein to study MP oligomerization [14]. The tetrameric structure of KcsA is highly stable in a wide range of detergents, even in SDS [15]. This high stability is caused not only by interactions between protein subunits but also by interactions between the protein and the surrounding lipid bilayer [16,17]. Previous studies on KcsA have shown that the stability of the tetramer in lipids is reduced by molar concentrations of alcohols, like TFE, in an indirect manner which can be related to their ability to alter the lateral membrane pressure and permeabilize the membrane [18,19]. More recently, TFE has been proven to be an authentic tool in investigation of channel assembly and stability of the tetrameric structure of KcsA [20]. The present study addresses the question that how does this tetrameric channel respond to HFIP-induced changes in membrane lateral pressure profile and how does any alteration in membrane profile affect the intrinsic stability, structure and conformation of KcsA in a lipid bilayer. Experimental procedures Reagents 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dioleoyl-sn-glycero-3-phospho-glycerol (DOPG) and 1,2-dioleoyl-snglycero-3-phospho-ethanolamine (DOPE) were purchased from Avanti Polar Lipids Inc. n-Dodecyl-b-D-maltoside (DDM)1 was from Anatrace Inc. 1,1,1,3,3,3-Hexafluoroisopropanol (HFIP) was obtained from Merck. Protein expression, purification and reconstitution Wild type (WT) and mutant-lacking the first 18 amino acids, named ‘‘(DN-KcsA)”-were expressed and purified according to the method described previously [17,20]. The purity of proteins was assessed by SDS–PAGE. Proteins were reconstituted in different lipid mixtures with a 1:1000 protein:lipid molar ratio, and the resultant proteoliposomes were obtained as described previously [17]. Stability assay of tetramer dissociation by HFIP Small aliquots of proteoliposomes were incubated with variable concentrations of HFIP for 1 h at room temperature. Samples were mixed with an electrophoresis sample buffer (50 mM Tris–HCl, pH 6.8, 50% glycerol, 0.01% bromophenol blue and 10% SDS) and run on 15% acrylamide gel in the presence of 0.1% SDS. Gels were stained by silver nitrate, scanned by a densitometer (Bio-Rad Laboratories) and quantified with the program Quantity One. The amount of tetramer (%) was plotted against HFIP (vol.%) for the stability assay, as described previously [17]. Tryptophan fluorescence spectroscopy All fluorescence experiments were performed in vesicle buffer at room temperature using a QuantaMaster QM-1/2005 spectrofluorometer (Photon Technology International, NJ) in a quartz cuvette. The samples were excited at 280 nm and emission spectra were collected between 300 and 400 nm. The bandwidths for both excitation and emission monochromators were 5 nm and the data were corrected as described previously [17]. All fluorescence 1 Abbreviations used: DDM, n-dodecyl-b-D-maltoside; HFIP, hexafluoroisopropanol; MP, membrane protein; PC, phosphatidylcholine; PG, phosphatidylglycerol; PE, phosphatidylethanolamine; SDS, sodium dodecyl sulphate; PAGE, polyacrylamide gel electrophoresis; TFE, trifluoroethanol; Trp, tryptophan; WT, wild type.
experiments were performed at 3 lM protein concentration present in proteoliposomes. Circular dichroism analysis Far-UV CD spectra were recorded between 200 and 250 nm (350 ll sample volume) on a Jasco J-810 spectropolarimeter (equipped with a temperature-controlled incubator) at 20 °C using 1 mm optical path length quartz cells. The step size was 0.5 nm with a 1.0-nm-bandwidth at a scan speed of 10 nm/min. All measurements were performed under nitrogen flow. As a blank, the CD spectra of lipid mixture in the absence and presence of HFIP (vol.%) were measured. Averages of 10 scans were obtained for blank and protein spectrum and all spectra were corrected for the buffer and vesicle scattering effects. Predicted percentages for different kinds of secondary structures were calculated using the K2d computer modeling program as described previously [21], and the results were expressed as mean residue ellipticity in units of degrees/cm2/dmol. The spectra were recorded at 0.1 mg/ml protein concentration in 50 mM Tris–HCl (pH 7.5). Results HFIP-induced tetramer dissociation in WT-KcsA Fig. 1A (upper panel) illustrates silver-stained gels of WT-KcsA dissociation in PC:PG (7:3 mol.%) and PE:PG (7:3 mol.%) lipid bilayers as a function of HFIP (vol.%). Dissociation of WT tetramer (T), which runs at 68 kDa, required 6 vol.% HFIP to completely dissociate the tetramer into its monomeric (M) subunits, which run at 18 kDa, as reported previously [17,20]. The amount of monomer in PC:PG bilayer almost remained the same throughout the whole range of HFIP; however, in PE:PG the monomer population was disappeared upon increasing HFIP concentration lipid bilayer, which reappeared upon complete tetramer dissociation. This effect can be explained by aggregation of KcsA in the presence of HFIP which could not be seen on SDS-gel. It was, however, interesting to note that the intensity of monomeric KcsA in PE:PG bilayer was relatively increased at higher HFIP concentration compared to PC:PG system. It suggests that non-bilayer lipid PE might be required for proper folding of monomeric KcsA as also reported previously [16]. Fig. 1B summarizes the data from the experiments illustrated in Fig. 1A. In addition, the summarized data also demonstrate KcsA tetrameric stability either in detergent micelles or in a pure PC bilayer (SDS-gels not shown) for comparison. DDM-solubilized WTKcsA tetramer (0.1 mg/ml) was dissociated at 2 vol.% HFIP. In pure PC, the stability of KcsA was increased such that 4 vol.% HFIP was required for complete tetramer dissociation. In PC:PG or PE:PG, the tetrameric stability was significantly increased as discussed above. However, the quantification of gels clearly indicated that tetrameric stability was unaffected in the range of 1–3 vol.% HFIP in PE:PG lipid bilayer compared to PC:PG system. These results agree well with the previous observation that non-bilayer lipid PE maximally stabilizes the tetramer [18,19]. HFIP-induced tetramer dissociation in DN-KcsA
DN-KcsA, which forms a stable tetramer, required significantly less amounts of HFIP to dissociate the tetramer such that 4 vol.% HFIP was able to completely dissociate the tetramer in both lipid bilayers (Fig. 1A lower panel). In addition, no monomeric DN-KcsA could be detected either in the absence or presence of HFIP. However, at 4 vol.% HFIP a monomeric band appeared upon tetramer dissociation indicating that HFIP solubilized the monomeric KcsA
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
3
Fig. 1. (A) Representative silver-stained gels of HFIP-induced tetramer dissociation in WT and DN-KcsA in PC:PG or PE:PG (7:3 mol.%) lipid bilayers. Tetrameric (T) and monomeric (M) KcsA are indicated. The concentration of HFIP (vol.%) is indicated at the top of each gel. (B) Quantification of silver-stained gels shown in panel A. In addition, the tetrameric stabilities analysed for DDM-solubilized protein or KcsA reconstituted in pure PC bilayers are also shown. The intensities of tetramer bands were assigned as a relative value of 100% observed for TFE untreated (0 vol.% TFE) sample. (C) Proteins were reconstituted in LUV’s and incubated with various concentrations of HFIP for 1 h at room temperature. All data points correspond to the average and ±SD of three experiments.
upon tetramer dissociation which was not soluble enough to be observed by SDS–PAGE. The quantification of the gels shown in Fig. 1A (lower panel) is summarized in Fig. 1C. Similar to WT-KscA, the stability of DN-KcsA in detergent micelles or pure PC was also investigated (gels not shown). DDM-solubilized DN-KcsA (0.1 mg/ ml) was also dissociated at 2 vol.% HFIP, however, a significant reduction in tetramer fraction was observed at 1 vol.% HFIP compared to WT-KcsA. In pure PC, the DN-KcsA tetramer was as stable as WT-KcsA. However, the stabilizing effect in PC:PG or PE:PG was completely abolished compared to WT-KcsA. These results indicated that N-terminus stabilized the tetramer via electrostatic interactions between the basic residues and the acidic PG. It was interesting to note that KcsA dissociation as a function of HFIP (vol.%) was quite similar to TFE-induced tetramer dissociation pattern reported previously [17,20]. However, the amount of HFIP required to dissociate the tetramer either in detergent micelles or in different lipid bilayers is 10 times less than TFE. Thus, HFIP, a strong cosolvent and more acidic than TFE, has stronger effects on KcsA tetrameric stability.
HFIP-induced changes in Trp fluorescence and protein unfolding Since TFE has been shown to unfold the tetrameric structure via increase in hydrophilicity or accessibility of Trp residues to a local environment [17,22], the effects of HFIP were also determined during HFIP-induced channel unfolding. Fluorescence emission spectra for the effects of 1 and 4 vol.% HFIP on WT and DN-KcsA are shown in Fig. 2A and B, respectively, and he effects of HFIP on Trp fluorescence properties are compiled in Table 1.
For comparison, similar amounts of HFIP were used regardless of dissociation concentration for both proteins. Incubation of WTKcsA with 1 and 4 vol.% HFIP led to fluorescence quenching. In addition, significant blue shifts in emission maxima were observed which is in contrast to TFE-induced unfolding behaviour of KcsA shown previously [17,22]. Blue shifts in emission maxima clearly indicated that Trp residues are located in a more hydrophobic environment upon HFIP insertion in between the lipid head groups. For DN-KcsA, relatively stronger effects of HFIP were monitored compared to WT-KcsA indicating that these changes are mainly elicited upon N-terminus deletion. In other words, N-terminus could be directly involved in major conformational or structural changes during tetramer dissociation. Effect of HFIP on the secondary structure of KcsA The effect of HFIP on the secondary and tertiary structure of both proteins were investigated in a PC:PG lipid bilayer. To obtain information about the secondary structure, circular dichroism (CD) spectroscopy was used. The resulting spectra of WT and DN-KcsA are shown in Fig. 2C and D, respectively, with the estimated secondary structure compiled in Table 2. Without HFIP, WT-KcsA protein contained slightly high content of random structure, moderate content of a-helical and low content of b-sheet structure. The structural information especially the a-helical content is in good agreement what has been published previously [15]. However, a slight reduction in a-helical content could be due to difference in protein reconstitution medium or presence of detergent micelles which seem to slightly increase the a-helical content compared to reconstituted samples as reported previously [15].
4
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
Fig. 2. Effects of HFIP on Trp fluorescence emission spectra of WT (A) and DN-KcsA (B) in PC:PG (7:3 mol.%) bilayers. Trp fluorescence was monitored either in the absence (solid) or presence of 1 (dash) or 4 (dotted) vol.% HFIP. Far-UV CD spectra of WT (C) and DN-KcsA (D) in the absence (solid) or presence of 4 vol.% HFIP (dash) in PC:PG (7:3 mol.%) bilayers. Effect of 1 vol.% HFIP on both proteins elicited no significant change in the secondary structure content compared to HFIP-untreated samples; hence these spectra are not shown.
Table 1 Effect of HFIP on Trp fluorescence of WT and DN-KcsA in PC:PG (7:3 mol.%) lipid bilayer.a HFIP (vol.%)
0 1 4 a
Emission wavelength (nm)
Fluorescence quenching (%)
Blue shift (nm)
WT
DN
WT
DN
WT
DN
322 ± 0.8 319 ± 1.2 316 ± 1.6
321 ± 1.2 316 ± 1.4 308 ± 4
– 15 ± 6 47 ± 15
– 40 ± 11 82 ± 14
– 3 ± 0.4 6±2
– 5 ± 1.2 13 ± 3
Parameters of Trp fluorescence were derived from data shown in Fig. 2A and B.
Table 2 Effect of HFIP on secondary structure of WT and DN-KcsA in PC:PG (7:3 mol.%) lipid bilayer.a HFIP (vol.%)
0 4
WT
DN
a-Helix
b-Strand
Random
a-Helix
b-Strand
Random
42 ± 5 38 ± 5
15 ± 1.5 17 ± 4
43 ± 6 45 ± 8
32 ± 5 40 ± 3
32 ± 4 12 ± 2
36 ± 7 48 ± 5
a Parameters of secondary structure were derived from data shown in Fig. 2C and D. Structural analysis was performed for the average CD spectra corrected for the blank and for volume increase due to HFIP. The values are the average and ±SD of predicted percentage for secondary structures calculated by K2d computer modeling program [21].
Incubation of WT-KcsA with 1% (not shown) or 4 vol.% HFIP did not change the CD spectrum significantly. It was also interesting to note that 4 vol.% HFIP, which is quite close to the tetramer-tomonomer transition and at which 50% tetramer is stable in a PC:PG bilayer, still contains similar structural information as in HFIP-untreated KcsA protein. This suggests that HFIP has an indirect effect on tetrameric stability without changing the secondary structure.
For DN-KcsA, a decrease in a-helical content, increase in bsheet and slight reduction in random content was observed compared to WT when no HFIP was present in the medium. A reduction in a-helical content can be explained on the basis of a-helical Nterminus deletion. However, it seems that deletion of the N-terminus induced secondary structure in the protein. This could either be due to protein aggregation of monomeric KcsA as estimated by SDS–PAGE. Incubation of DN-KcsA with 4 vol.% HFIP significantly decreased the content of b-sheets and increased the a-helical content indicating unfolding or solubilization of aggregated monomers which could be seen by SDS–PAGE at 4 vol.% concentration of HFIP.
Discussion Short-chain alcohols as anesthetics molecules can bind specifically to proteins and this may influence the binding of ions to the protein acting as ion channels, shifting protein conformational equilibria [5,7]. The potassium channel KcsA, which acts as a representative ion channel, is a convenient model protein to study the effect of anesthetic agents on MP complexes. The extremely high stability of KcsA channel arises from interaction between hydrophobic part of the subunit and hydrogen bonding between polar residues in the selectivity filter [23]. However, TFE could disrupt these interactions resulting in dissociation of KcsA tetramer into its monomer suggesting that tetramer stabilizing interactions occur mainly in the transmembrane domain of the protein [18,19]. On the other hand, alcohols have significant effects on the physical properties of biological membranes. Such changes in the membrane properties, in turn, can lead to changes in the conformational states of intrinsic MPs, thus leading to an indirect (non-specific) mechanism for the modulation of protein behaviour by alcohol adsorption into lipid membranes [17–19,24]. The detailed molecular mechanism for these effects is not yet fully understood. Some
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
5
Fig. 3. Schematic model of the effect of HFIP on the membrane lateral pressure acting on KcsA tetrameric structure. (A) Each monomer is of KcsA shown in four distinct colors highlighting the positions of the N and C-termini. (B) The presence of low concentration of HFIP squeezes the tetramer with the help of N-terminus located at the membrane– water interface thereby protecting the tetramer against HFIP-induced drastic changes in the membrane lateral pressure profile. This results in stabilization of compact structure of KcsA tetramer. (C) Increasing HFIP concentration eventually destabilizes the tetramer into its monomeric subunits possibly via exposure of the pore domain and the N-terminus which are not inserted in the lipid bilayer anymore, because of the partitioning of HFIP in the lipid bilayer and the direct interactions of HFIP with the protein.
short-chain alcohols are amphiphilic, they have been suggested to localize primarily in the headgroup region of the lipid bilayer [25]. This disrupts the packing in the lipid bilayer and leads to a variety of changes, among them observed increase in membrane fluidity [26], in membrane permeability [27], and lipid lateral mobility [28]. The present study was set out to investigate if strong cosolvent like HFIP affects tetramer stability via similar mechanism as TFE [18–20]. First, denaturation experiments were performed in which KcsA tetrameric stability was analysed. Incubation of detergent solubilized KcsA with HFIP resulted in tetramer dissociation indicating that HFIP directly affects the KcsA structure possibly via exposure of hydrophobic surface of the transmembrane helices. Second, an increased stability of KcsA in membranes compared to detergent micelles was observed indicating that lipid bilayer limited the exposure of the transmembrane helices of KcsA to HFIP. These effects were in total agreement with TFE-induced tetramer dissociation behaviour [18,19]. However, the advantage of using HFIP over TFE was that less amounts were required to investigate which interactions stabilize the oligomeric structure of KcsA in a particular environment. It was interesting to observe the maximum stability of KcsA tetramer in a PE:PG bilayer compared to PC:PG or pure PC since small headgroup size of PE allows larger insertion sites for alcohols which help maintaining the bilayer structure [18,19]. NMR-experiments have indicated that small alcohols insert mainly in the headgroup region of the lipid bilayer and that the insertion of small alcohols in the bilayer containing non-bilayer lipid PE results in an increased lateral pressure in the acyl chain region leading to a higher KcsA tetramer stability [19]. The stabilizing effects of PE and PG can also be justified by the fact that both lipids are important for efficient membrane association and tetramerization of KcsA [16] and that KcsA preferentially interacts with both these lipids [14,16]. The N-terminus, which is directly associated with the lipid headgroups, had a significant influence on tetramer stability. Increase in tetramer stability in the presence of acidic PG confirmed the stabilizing effect of N-terminus via electrostatic interactions between its basic residues and the acidic lipids as recently reported [20]. However, such effect was found to be more pronounced for HFIP compared to TFE [17–19] indicating that HFIP could promote even stronger ionic interactions between the charged PG and the basic residues thereby making electrostatic/H-bond switch much powerful. But, what happens to KcsA structure and conformation during HFIP-induced tetramer dissociation? Incubation of WT-KcsA with
HFIP elicited strong hydrophobicity in a local environment indicating that HFIP promoted tetramer squeezing or a more compact structure of KcsA (see model in Fig. 3). Even at or close to dissociating concentration of HFIP a strong hydrophobicity was monitored indicating that HFIP-induced changes in protein conformation might increase protein hydrophobicity by exposing hydrophobic side chains and shielding polar amide groups from the solvent [29]. This behaviour is completely opposite to TFE-induced KcsA unfolding mechanism [17,22]. Interestingly, no change in the secondary structure was observed for WT-KcsA which confirmed that HFIP exerted an indirect effect on tetrameric stability possibly via drastic changes in the membrane lateral pressure profile. Removal of the N-terminus elicited strong changes in Trp fluorescence properties and increase in b-sheet content compared to WT-KcsA indicating that tetramer might undergo significant structural and conformational changes upon N-terminus deletion. Furthermore, incubation of this protein with HFIP resulted in solubilization of b-sheet structure thereby increasing the a-helical content. Such propensity of HFIP to convert aggregation or b-sheet content of several peptides into the helical structure has also been illustrated in other studies [11–13]. Thus, N-terminus seems crucial for controlling overall protein structure and proper conformation. The application of these studies might be useful in understanding the role of membrane bound extramembranous parts of ion channels in oligomerization as well as the nature of small alcohols in affecting the membrane lateral pressure profile and consequent changes in the folding properties of integral MP complexes. Although ion channels and other MPs have been reported to harbor relatively specific sites for general anesthetic agents [30,31], the physicochemical and molecular features of these sites are not yet well-understood. In addition to the presence of physically circumscribed hydrophobic protein cavities, which constitute the alcohol and general anesthetic sites [32], N-terminus is possibly one of the major binding sites for anesthetic molecules. It seems to play an important role in maintaining the membrane lateral pressure profile which might be disturbed by a number of anesthetic agents. Acknowledgments The construct of N-terminal deleted (DN) KcsA was a precious gift from the late Jeanette G. Stam. The careful secretarial help of Irene van Duin and helpful suggestions of Antoinette Killian and Ben de Kruijff (Department of Biochemistry of Membranes, Utrecht
6
M. Raja / Archives of Biochemistry and Biophysics 498 (2010) 1–6
University) are gratefully acknowledged. This work was supported by funds of the Chemical Sciences Division (CW) of The Netherlands Organization for Scientific Research (NWO). References [1] A. Ullrich, J. Schlessinger, Cell 61 (1990) 203–212. [2] H. Metzger, J. Immunol. 149 (1992) 1477–1487. [3] D.S. Goodsell, A.J. Olson, Annu. Rev. Biophys. Biomol. Struct. 29 (2000) 105– 153. [4] M.H. Ali, B. Imperiali, Bioorg. Med. Chem. 13 (2005) 5013–5020. [5] J.L. Popot, D.M. Engelman, Biochemistry 29 (1990) 4031–4037. [6] J.L. Popot, D.M. Engelman, Annu. Rev. Biochem. 69 (2000) 881–922. [7] M.H.B. Stowell, D.C. Rees, Adv. Protein Chem. 46 (1995) 279–311. [8] D.E. Otzen, L. Nesgaard, Biochemistry 46 (2007) 4348–4359. [9] H.J. Dyson, P.E. Wright, Curr. Opin. Struct. Biol. 3 (1993) 60–65. [10] J.J. Yang, M. Buck, M. Pitkeathly, M. Kotik, D.T. Haynie, C.M. Dobson, S.E. Radford, J. Mol. Biol. 252 (1995) 483–491. [11] N. Hiroto, K. Mizuno, Y. Goto, Protein Sci. 6 (1997) 416–421. [12] S.J. Wood, B. Maleeff, T. Hart, R. Wetzel, J. Mol. Biol. 256 (1996) 870–877. [13] H. Zhang, K. Kaneko, J.T. Nguyen, T.L. Livshits, M.A. Baldwin, F.E. Cohen, T.L. James, S.B. Prusiner, J. Mol. Biol. 250 (1995) 514–526. [14] F.I. Valiyaveetil, Y. Zhou, R. MacKinnon, Biochemistry 41 (2002) 10771–10777. [15] D.M. Cortes, E. Perozo, Biochemistry 36 (1997) 10343–10352. [16] A. Van Dalen, S. Hegger, J.A. Killian, B. de Kruijff, FEBS Lett. 525 (2002) 33–38.
[17] M. Raja, R.E.J. Spelbrink, B. de Kruijff, J.A. Killian, FEBS Lett. 581 (2007) 5715– 5722. [18] E. van den Brink-van der Laan, V. Chupin, J.A. Killian, B. de Kruijff, Biochemistry 43 (2004) 5937–5942. [19] E. van den Brink-van der Laan, V. Chupin, J.A. Killian, B. de Kruijff, Biochemistry 43 (2004) 4240–4250. [20] M. Raja, J. Membr. Biol. 234 (2010) 235–240. [21] M.M. Raja, N.K. Tyagi, R.K.H. Kinne, J. Biol. Chem. 278 (2003) 49154–49163. [22] F.N. Barrera, M.L. Renart, M.L. Molina, J.A. Poveda, J.A. Encinar, A.M. Fernandez, J.L. Neira, J.M. Gonzalez-Ros, Biochemistry 44 (2005) 14344–14352. [23] D.A. Doyle, C.J. Morais, R.A. Pfuetzner, A. Kuo, J.M. Gulbis, S.L. Cohen, B.T. Chait, R. MacKinnon, Science 280 (1998) 69–77. [24] R.S. Cantor, Toxicol. Lett. 100 (1998) 451–458. [25] M. Patra, E. Salonen, E. Terama, I. Vattulainen, R. Faller, B. Lee, J. Holopainen, M. Karttunen, Biophys. J. 90 (2006) 1121–1135. [26] J.H. Chin, D.B. Goldstein, Mol. Pharmacol. 13 (1977) 435–441. [27] H. Komatsu, S. Okada, Chem. Phys. Lipids 85 (1997) 67–74. [28] S.Y. Chen, B. Yang, K. Jacobson, K.K. Sulik, Alcohol 13 (1996) 589–595. [29] Y.O. Kamatari, T. Konno, M. Kataoka, K. Akasaka, J. Mol. Biol. 259 (1996) 512– 523. [30] R.W. Peoples, C. Li, F.F. Weight, Annu. Rev. Pharmacol. Toxicol. 36 (1996) 185– 201. [31] K.W. Miller, Br. J. Anaesth. 89 (2002) 17–31. [32] M.P. Mascia, J.R. Trudell, R.A. Harris, Proc. Natl. Acad. Sci. USA 97 (2000) 9305– 9310.