Effects of Lactose Permease on the Phospholipid Environment in

4642

Langmuir 2005, 21, 4642-4647

Effects of Lactose Permease on the Phospholipid Environment in Which It Is Reconstituted: A Fluorescence and Atomic Force Microscopy Study Sandra Merino,†,‡ O Å scar Dome`nech,‡,§ M.Vin˜as,| M. Teresa Montero,†,‡ and Jordi Herna´ndez-Borrell*,†,‡ Departament de Fisicoquı´mica and Quı´mica Fı´sica, Universitat de Barcelona, E-08028 Barcelona, Spain, Laboratori de Microbiologia, Campus de Bellvitge, Universitat de Barcelona, E-08907 L’Hospitalet de Llobregat, Spain, and Centre de Bioelectro` nica i Nanobiocie` ncia (CBEN), Parc Cientı´fic de Barcelona, Josep Samitier 1-5, E-08028 Barcelona, Spain Received November 25, 2004. In Final Form: February 24, 2005 The membrane transport protein lactose permease (LacY), a member of the major facilitator superfamily containing 12 membrane-spanning segments connected by hydrophilic loops, was reconstituted in liposomes whose composition was 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine and 1-palmitoyl-2-oleoylsn-glycero-3-phosphoglycerol in a 3:1 molar ratio. The structural order of the lipid membranes, in the presence and absence of LacY, was assessed using steady-state fluorescence anisotropy. The features of the anisotropy curves obtained with 1,6-phenyl-1,3,5-hexatriene and 1-(4-trimethylammoniumphenyl)6-phenyl-1,3,5-hexatriene p-toluenesulfonate suggest a surface effect of LacY on the membranes. Atomic force microscopy imaging of supported planar bilayers (SPBs) deposited onto mica was used to examine the effect of LacY on the nanostructure of the phospholipid matrix. Two separated domains were observed in SPBs formed from pure phospholipid mixture. Protein assemblies segregated from the rest of the matrix were observed after the extension of proteoliposomes. The effect of the protein on the electrostatic surface potential of the bilayer was also examined using a fluorescent pH indicator, 4-heptadecyl-7-hydroxycoumarin. Changes in surface potential were enhanced in the presence of the substrate (i.e., lactose). Taken together the results indicate that LacY is segregated into the phospholipid matrix and has moderate effects on the acyl chain order of the bilayers. The changes in surface electrical properties of the bilayers suggest a role for the phospholipid headgroups in proton transfer to the amino acids involved in substrate translocation.

Introduction Membrane proteins account for over 25% of total cell proteins.1 The cytoplasmic membrane of Escherichia coli, for instance, is believed to contain more than 200 protein types, of which 60 or more might be involved in transport functions. Among them, lactose permease (LacY),2 one of the best studied cytoplasmic membrane proteins, is often taken as a paradigm for secondary transport proteins that couple the energy stored in an electrochemical ion gradient to a concentration gradient (β-galactoside/H+ symport). LacY belongs to what is termed the major facilitator superfamily (MFS),3 most of whose members are predicted to contain 12 transmembrane segments. The secondary structure of LacY consists of 12 transmembrane R-helices, crossing the membrane in a zigzag fashion, that are connected by 11 relatively hydrophilic, periplasmic, and cytoplasmic loops, with both amino and carboxyl termini on the cytoplasmic surface (Figure 1). A three-dimensional (3D) model of a LacY mutant (Cys 154fGly) and a reaction * Author for correspondence: Departament de Fisicoquı´mica, Facultat de Farma`cia, Universitat de Barcelona, 08028-Barcelona, Spain. E-mail: [email protected]. † Departament de Fisicoquı´mica. ‡ Centre de Bioelectro ` nica i Nanobiocie`ncia. § Departament de Quı´mica Fı´sica. | Laboratori de Microbiologia. (1) Jones, D. T. FEBS Lett. 1998, 423, 281-285. (2) Kaback, H. R. In Transport Processes in Eukaryotic and Prokaryotic Organisms; Kaback, H. R., Konings, W., Eds.; Elsevier: Amsterdam, 1996; Vol. 2, pp 203-228. (3) Pao, S. S.; Paulsen, I. T.; Saier, M., Jr. Microbiol. Mol. Biol. Rev. 1998, 62, 1-34.

mechanism derived from X-ray diffraction studies is now available,4 culminating years of effort. This mechanism confirms the basic structural features elucidated by noncrystallographic approaches5 based on a combination of molecular biology and various types of spectroscopy. The physiological activity of the transmembrane proteins may be influenced by or dependent on the physical properties of neighboring phospholipids.6 Such dependence has been demonstrated in the case of, among others, β-hydroxybutyrate dehydrogenase, an enzyme integrated into the mitochondrial inner membrane, Ca2+-ATPase,7 melibiose permease8 from E coli, another member of the MFS, and LacY itself,9 for which the requirement of phosphatidylethanolamine in functions in vivo has been clearly demonstrated.10 This phospholipid is also required for the correct folding of the protein.11 Hence, an understanding of the physicochemical properties and the influence of the phospholipids that constitute the matrix in which the transmembrane proteins are embedded is of critical importance. (4) Abramson, J.; Smirnova, I.; Kasho, V.; Verner, G.; Kaback, H. R.; Iwata, S. Science 2003, 301, 610-615. (5) Kaback, H. R.; Sa`hin-To´th, M.; Weinglass, A. B. Nat. Rev. Mol. Cell Biol. 2001, 2, 610-620. (6) Lee, A. G. Biochim. Biophys. Acta 2003, 1612, 1-40. (7) Warren, G. B.; Toon, N. J. M.; Birdsall, A. G.; Lee, J. C.; Metcalfe, J. C. Biochemistry 1974, 13, 5501-5507. (8) Dumas, F.; Tocanne, J. F.; Leblanc, G.; Lebrun, M. Ch. Biochemistry 2000, 39, 4846-4854. (9) Le Coutre, J.; Narasimhan, L. R.; Patel, C. K. N.; Kaback, H. R. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 10167-10171. (10) Bogdanov, M.; Dowhan, W. J. Biol. Chem. 1995, 270, 732-739. (11) Bogdanov, M.; Sun. J.; Kaback H. R.; Dowhan W. J. Biol. Chem. 1996, 271, 11615-11618.

10.1021/la047102d CCC: $30.25 © 2005 American Chemical Society Published on Web 04/08/2005

Effect of Transport Protein on a Lipid Bilayer

Langmuir, Vol. 21, No. 10, 2005 4643

Figure 1. Secondary structure of lactose permease from Escherichia coli showing the charged residues (boldface). The one-letter amino acid code is used. The putative transmembrane helices are shown in boxes that are connected by hydrophilic loops.

Here we examine the effects of LacY on the structural order of lipid membranes by measuring, as a function of temperature, the steady-state fluorescence anisotropy (rs) of 1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene p-toluenesulfonate (TMA-DPH) and 1,6-phenyl1,3,5-hexatriene (DPH) incorporated into liposomes and proteoliposomes of LacY with 1-palmitoyl-2-oleoyl-snglycero-3-phosphoethanolamine (POPE)/1-palmitoyl-2oleoyl-sn-glycero-3-phosphoglycerol (POPG) as the phospholipid matrix. Recent studies12 have shown that the insertion of LacY into proteoliposomes depends on the presence of phospholipids in the fluid phase at the temperature of reconstitution. Therefore, to both visualize and elucidate lipid phase behavior, atomic force microscopy (AFM) was used to study the supported planar bilayers formed from POPE/POPG liposomes and proteoliposomes of LacY. Although the protein becomes randomly oriented when it is reconstituted in liposomes, its presence might affect the electrostatic surface potential of the membranes. The protein contribution to the membrane electrostatic surface potential can be assessed by using a fluorescent pH indicator, 4-heptadecyl-7-hydroxycoumarin (HHC),13 which has been extensively used in determining the electrostatic surface potential of monolayers14 and liposomes.15 Its fluorometric titration, when incorporated in liposomes, facilitates determination of the interfacial pK of the probe and calculation of variation in the electrostatic surface potential (∆ψ)16 after reconstitution of LacY in the same phospholipid matrix. This information helps us to relate the electrical properties of the phospholipid bilayer surface and LacY activity. (12) Merino, S.; Dome`nech, O Å .; Montero, M. T.; Herna´ndez-Borrell, J. Biosens. Bioelectron. 2005, 20, 1843-1846. (13) Ferna´ndez, M. S.; Fromherz, P. J. Phys. Chem. 1977, 81, 17551761. (14) Petrov, J. G.; Mo¨bius, D. Langmuir 1989, 5, 523-528. (15) Ferna´ndez, M. S. Biochim. Biophys. Acta. 1981, 646, 23-26. (16) Va´zquez, J. L.; Berlanga, M.; Merino, S.; Dome`nech, O Å .; Vin˜as, M.; Montero, M. T.; Herna´ndez-Borrell, J. Photochem. Photobiol. 2001, 73, 14-19.

Materials and Methods Chemicals. 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG) were purchased from Avanti Polar Lipids (Alabaster, AL). The probes 1,6-phenyl-1,3,5-hexatriene (DPH), 1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene p-toluenesulfonate (TMA-DPH) and 4-heptadecyl-7-hydroxycoumarin (HHC) were obtained from Molecular Probes (Eugene, OR). N-Dodecyl-β-D-maltoside (DDM) was purchased from Sigma Chemical Co. (St. Louis, MO). β-D-Galactopyranosyl-1-thio-βD-galactopyranoside (TDG) and isopropyl-1-thio-β-D-galactopyranoside (IPTG) were purchased from Ecogen (Barcelona, Spain), and Bio-Beads SM-2 was purchased from Bio-Rad (Hercules, CA). All other common chemicals were ACS grade. Bacterial Strains and Protein Purification. E. coli T-184 cells [lacI+O+Z-(A), rpsL, met, thr, recA, hsdN, hsdR/F, laclqO+ZD118 (Y+A+)] were kindly provided by Dr. H. Ronald Kaback from the HHMI-UCLA. LacY was extracted and purified from the overproducing strain E. coli T-184, according to previous studies.17 Briefly, cells were grown aerobically at 37 °C in LuriaBertani broth containing ampicillin (100 µg mL-1) and streptomycin (10 µg mL-1). Dense cultures were diluted in a 40 L fermentor and grown for 1 h at 37 °C before induction with 0.3 mM IPTG. After growing for another 3 h at constant temperature, the cells were harvested and disrupted by passage through a French Press. The membrane fraction was isolated by centrifugation and extracted with 2% (w/v) DDM, and LacY was purified by affinity chromatography on immobilized monomeric avidin beads, as described previously.18 The column was washed with approximately 200 mL of 50 mM NaPi, 150 mM NaCl, pH 7.50, 0.02% DDM, and the permease was eluted with 2 mM biotin in the same buffer, concentrated, and dialyzed using a MicroProDiCon membrane (Spectrum). The protein was quantified using the spectrophotometric assay Micro-BSA (Pierce). Vesicle Preparation and Protein Reconstitution. Chloroform/methanol (50:50, v/v) solutions containing the appropriate amounts of phospholipids were dried under a stream of oxygen(17) Voss, J.; Hubbell, W.; Herna´ndez-Borrell, J.; Kaback, H. R. Biochemistry 1997, 49, 15055-15061. Zhao, M.; Zen, K. C.; Herna´ndezBorrell, J.; Altenbach, C.; Hubbell, W. L.; Kaback, H. R. Biochemistry 1999, 48, 15970-15977. (18) Frillingos, S.; Sahin-To´th, M.; Persson, B.; Kaback., H. R. Biochemistry 1994, 35, 12915-12918.

4644

Langmuir, Vol. 21, No. 10, 2005

Merino et al.

free N2 in a conical tube. The thin film obtained was kept under high vacuum for approximately 3 h to remove organic solvent traces. Multilamellar liposomes were obtained after redispersion of the film in 50 mM KPi buffer, pH 7.50, and sonication for 30 min in a bath sonicator. The liposomes were then dissolved in 0.5% (w/v) of DDM and mixed with the solubilized protein and incubated at 4 °C for 30 min with gentle agitation. Extraction of DDM was achieved by addition of polystyrene beads (BioBeads SM-2, Bio-Rad).19 The first two extractions were performed at room temperature for 2 h each, and the last extraction was performed at 4 °C overnight. The vesicles were collected by centrifugation for 4 h at 35000 rpm and resuspended in a small volume of 50 mM KPi pH 7.50, followed by three cycles of freezing and thawing. The samples were frozen and stored in liquid nitrogen at a final protein concentration of 1 mg/mL. Immediately before use, the samples were subjected to a brief period of sonification in a bath sonicator at a temperature below 37 °C. Steady-State Anisotropy Experiments. The experimental procedures to measure the steady-state anisotropy of TMA-DPH and DPH were adapted from previous studies.20,21 Essentially, 3 µL of concentrated stock solution of either probe in methanol was added to 1500 µL of vesicle (liposome or proteoliposome) suspension for 30 min at 37 °C. The final lipid/fluorescent probe ratio was 1000:1; mol/mol. The excitation wavelength was 381 nm, and emission was measured at 426 nm. The anisotropy was recorded at three degree intervals in the range between 0 and 35 °C. The vertically and horizontally polarized emission intensities were corrected for background scattering by subtraction of the corresponding polarized intensities of a blank containing the unlabeled suspension. Steady-state anisotropy (rs) values were calculated according to

rs )

IVV - GIVH IVV + 2GIVH

(I)

Results

where IVV and IVH are the intensities measured in directions parallel and perpendicular to the exciting beam, respectively, and G is the grating correction factor equal to IHV/IHH. The following equation was fitted to the anisotropy versus temperature data

rs ) rs2 +

rs1 - rs2 1 + 10B′(T/(Tm)-1)

(II)

where T is the absolute temperature, Tm is the midpoint phase transition, and rs1 and rs2 are the upper and lower values of rs; B′ is the slope factor which is correlated with the extent of cooperativity (B) by B ) [1 - 1/(1 + B′)]; the introduction of B yields a convenient scale of cooperativity ranging from 0 to 1. Measurement of the Absolute Surface Potential of Charged Liposomes. The variation in the electrostatic membrane potential (∆ψ) was calculated in triplicate, adapting a method used elsewhere.15 Briefly, in these experiments the molar ratio of total lipid to fluorescent probe was 1000:1. Similar molar ratios and ionic strength were maintained in the experiments carried out in the presence of LacY in order to calculate ∆pK in the presence of liposomes. Fluorescent probe was incorporated into the bilayer by bath incubation at 37 °C for 30 min. Samples were kept at room temperature for another 30 min to achieve thermal stabilization before measurements were performed. The electrostatic surface potential was then calculated using

∆ψ )

-2.3RT (∆pK - ∆pK0) F

(POPC)sand the pK of the probe in solution and ∆pK is the difference between the pK determined in POPE:POPG (3:1, mol/ mol), of either the liposome or proteoliposome system, and the pK of the probe in solution. R is the universal gas constant, T is the temperature (310 K), and F is Faraday’s constant. Supported Planar Bilayer Formation and Atomic Force Microscopy Observations. The spread of the supported planar bilayers was obtained by using the vesicle fusion technique described elsewhere.12 Briefly, 80 µL of either liposomes or proteoliposomes in 10 mM Hepes pH 7.40, 150 mM NaCl, 20 mM CaCl2; I ) 0.19 m buffer, was pipetted onto freshly cleaved mica and incubated for 10 min at room temperature, before being washed with 10 mM Hepes pH 7.40, 150 mM NaCl; I ) 0.15 m (image buffer). The tip was immediately immersed into the liquid cell. To perform all these experiments it was necessary to drift equilibrate and thermally stabilize the cantilever for 30 min in the presence of buffer. All images were recorded in tapping mode with a commercial Digital Instruments (Santa Barbara, CA) Nanoscope III AFM fitted with a 15 µm scanner (d-scanner). Standard Si3N4 tips, with a nominal force constant of 0.1 N/m (Digital Instruments), were used, and the forces were minimized during the scans. Before every sample, the AFM liquid cell was washed with ethanol and ultrapure water (Milli Q reverse osmosis system) and allowed to dry in a N2 stream. Mica (green muscovite mica) were cleaved with Scotch tape and glued onto a Teflon disk by a waterinsoluble epoxy. These Teflon disks were glued onto a steel disk and then mounted on to the piezoelectric scanner. All images were scanned under aqueous solution. Prior to imaging the sample, the tip was stabilized in buffer by scanning of the mica. Force plots were recorded for every sample to control the repulsion of the tip, and images were flattened with the Nanoscope III program.

(III)

where ∆pK0 is the difference between the pK determined in a electrically neutral liposomesthat we take as the reference (19) Rigaud, J. L.; Levy, D.; Mosser, G.; Lambert, O. Eur. Biophys. J. 1998, 27, 305-319. (20) Va´zquez, J. L.; Montero, M. T.; Merino, S.; Dome`nech, O Å .; Berlanga, M.; Vin˜as, M.; Herna´ndez-Borrell, J. Langmuir 2001, 17, 1009-1014. (21) Merino, S.; Va´zquez, J. L.; Dome`nech, O Å .; Berlanga, M.; Vin˜as, M.; Montero, M. T.; Herna´ndez-Borrell, J. Langmuir 2002, 18, 32883292.

The temperature dependence of TMA-DPH and DPH liposomes and proteoliposomes is shown in Figure 2. The features of the phospholipid phase transition, for TMADPH (Figure 2A) and DPH (Figure 2B) liposomes, were almost unaffected by the presence of LacY at a proteinto-lipid molar ratio of 1/3000. However, the anisotropy of DPH dropped to lower values than those observed with TMA-DPH for the same temperature increases. Thus, the anisotropy of TMA-DPH dropped from 0.34 to 0.24, approximately, when the temperature increased from 0 to 35 °C, while that of DPH dropped from 0.32 to 0.13 for the same temperature range. Parameters Tm and B for the TMA-DPH liposomes, calculated according to described methods, are listed in Table 1. As can be seen, LacY caused an approximate 0.3 °C increase in the temperature of the phospholipid phase transition (Tm) and a small change in B. Conversely, the presence of LacY induces a decrease of 1.6 °C in the Tm and a lowering of B when DPH was the probe used. The topography of the supported planar bilayer (SPB) of the POPE/POPG phospholipid matrix is shown in Figure 3A and reveals the existence of two different domains (see differences in contrast). Line profile analysis, such as the one shown in Figure 3B, allows us to establish that (i) the height of the bilayer is 4.4 ( 0.2 nm (n ) 10), by measuring it at the edge of the SPB and (ii) there is a difference of 0.64 ( 0.07 nm (n ) 10) between the upper and the lower domain (see Figure 3B). Figure 3C shows the AFM topographic image of a proteolipid sheet of POPE/POPG and LacY at a protein-to-lipid molar ratio of 1/30. Here, two regions or domains are observed underlying supporting mica (white asterisk in Figure 3C). The upper domain regions (white star) constitute the largest region while the lower domain forms small patches (black asterisk). This can be corroborated by the line profile analysis shown in Figure 3D, which corresponds to the line drawn in Figure 3C. When a part of this image is magnified

Effect of Transport Protein on a Lipid Bilayer

Langmuir, Vol. 21, No. 10, 2005 4645

into the bilayer leads to a negative value of ∆ψ being obtained. As can be seen in Table 2 the values obtained for the proteoliposomes were positive. It should be noted that when lactose was added (10 mM) the surface potential shifted to even more positive values, +103.35 mV. Similar changes were obtained by addition of the analogue substrate TDG. The addition of sucrose did not significantly affect the surface potential values. Discussion

Figure 2. Steady-state polarization of TMA-DPH (A) and DPH (B) for vesicles formed with POPE/POPG (3:1, mol/mol). Liposomes (filled circles) and proteoliposomes of LacY (open diamonds) are shown. Each point indicates the mean value of three experiments. Data were fitted to sigmoid curves as described in the methods section. Table 1. Transition Temperature (Tm) and Cooperativity (B) Values Obtained by the Nonlinear Adjustment of Steady-State Polarization Data from Figure 2a proteoliposomes of LacY

liposomes probe

Tm (°C)

B

Tm (°C)

B

TMA-DPH DPH

15.4 ( 0.4 18.9 ( 0.4

0.69 0.74

15.7 ( 0.4 17.3 ( 0.2

0.67 0.68

a

Results represent data of three independent experiments.

(Figure 4), we can observe how the upper domain is covered by round protrusions, segregated from the lower domain, in a quasi-crystalline arrangement. It can be seen that the lower domain region shows a featureless surface. The surface pK values and electrical surface potentials (∆ψ ) calculated according to eq III are shown in Table 2. The reference surface potential (0 mV) was assigned to the neutral phospholipid (POPC). As expected, the introduction of the negatively charged phospholipid POPG

The experiments presented in this paper were designed to investigate the effects of LacY insertion into lipid bilayers (liposomes and SPBs) where it is reconstituted. It has been demonstrated that LacY is fully functional when reconstituted in native E. coli extract membranes22,23 and binary mixtures of phosphatidylglycerol and phosphatidylethanolamine.9,24 For this reason we selected POPE/POPG (3:1, mol/mol) as a suitable phospholipid matrix. We measured the fluorescence anisotropy of TMA-DPH and DPH in liposomes and proteoliposomes. The changes in the features of the phospholipid phase transition in the presence of LacY were far from dramatic. To explore the effects of LacY on distinct regions of the bilayer, we employ the fluorescent dyes TMA-DPH and DPH. TMA-DPH remains anchored at the aqueous interface of the phospholipids bilayer and as can be seen there was little to no effect of LacY on TMA-DPH anisotropy. On the other hand, significant changes with the more deeply embedded DPH dye were observed. It is important to mention here that for these experiments transmembrane proteins are normally reconstituted into proteoliposomes at low proteinto-lipid ratio.25 Thus, most of DPH and TMA-DPH molecules would be in lipid a long way from the protein in the membrane. Hence, the measurable and repetitive changes in Tm and B that we have detected would become significant if higher protein concentrations, as those used to form SPBs, could be used in this method. Actually, if we assume that Tm might respond linearly with increasing concentrations of the protein,25 the decrease of Tm calculated from DPH anisotropy measurements would reflect that there is a moderate change in the phospholipid acyl chain order in the presence of LacY. In addition, the slight increase of Tm in TMA-DPH liposomes in the presence of LacY suggest that hydrogen bonds might be formed between the headgroups of the phospholipids and some residues in the putative loops, most of them positive26 at physiological pH (see Figure 1). A note of caution, however, should be added with regard to this interpretation. The characteristics of the SPBs formed from liposomes should be different from those formed from proteoliposomes. Despite the fact that separate domains cannot be observed in POPE/POPG bilayers using solid-state NMR at 30 °C,27 we unambiguously observed two different domains in the SPBs formed with the phospholipid matrix (22) Viitanen, P.; Newman, M. J.; Foster, D. L.; Wilson, T. H.; Kaback, H. R. Methods Enzymol. 1986, 125, 429-452. Jung, K.; Jung, H.; Wu, J.; Prive´, G. G.; Kaback, H. R. Biochemistry 1993, 32, 12273-12278. Jung, K.; Jung, H.; Kaback, H. R. Biochemistry 1994, 33, 3980-3985. (23) Offenbacher, H.; Wolfbeis, O. S.; Fu¨rliger. Sens. Actuators 1986, 9, 73-84.; Kiefer, H.; Klee, B.; John, J.; Stierhof, Y. D.; Ja¨hnig, F. Biosens. Bioelectron. 1991, 6, 233-237; Klee, B.; John, E.; Ja¨hnig, F. Sens. Actuators, B 1992, 7, 376-379. (24) Zhao, M.; Zen, K. C. Hubbell, W. L. Biochemistry 1999, 38, 74077412.; Va´zquez-Ibar, J. L.; Weinglass, A. B.; Kaback, H. R. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 3487-3492. (25) Dumas, F.; Tocanne, J. F.; Leblanc, G.; Lebrun, M. Ch. Biochemistry 2000, 39, 4846-4854. (26) Heijne, G. J. Mol. Biol. 1992, 225, 487-494. (27) Hallock, K. J.; Lee, D. K.; Omnaas, J.; Mosberg, H. I.; Ramamoorthy, A. Biophys. J. 2002, 83, 1004-1013.

4646

Langmuir, Vol. 21, No. 10, 2005

Merino et al.

Figure 3. Topography images obtained from the extension of POPE/POPG (3:1, mol/mol) by AFM (A) liposomes and (B) LacY proteliposomes: upper domain (white star), lower domain (black asterisk), bare mica (white asterisk). Images were obtained in tapping mode using imaging buffer. Cross sections measured along the black lines are presented for each image (B) and (D), respectively. Table 2. Experimental Electrostatic Surface Potentials Calculated Using Equation IIIa liposomes

POPC POPE/POPG (3:1)

proteoliposomes of LacY

pK

∆ψ (mV)

pK

∆ψ (mV)

8.13 ( 0.07 8.59 ( 0.19

0 -28 ( 12

6.92 ( 0.18

+74 ( 12

Data are means of ( standard deviation of three independent experiments. a

Figure 4. Higher magnification image of the inset in Figure 3.

headgroup.28 Such domains could be interpreted at the molecular level by taking into account, first, that POPE readily forms hydrogen bonds29 and, second, that POPG may form clusters.30 Importantly, AFM provided evidence of protein assemblies which are unambiguously segregated from the rest of the phospholipid matrix. This observation is in agreement both with other studies31 that suggest the formation of phospholipid domains in the presence of LacY and with the segregation of LacY in SPBs of POPC.12 For other proteins and peptides such an observation could be

under study. Two findings allow us to assign the higher level to POPE and the lower level to POPG: (i) the area fraction covered by the lower domain is in agreement with the nominal composition of the liposome preparation and (ii) the POPE molecule is slightly longer than that of POPG due to differences in the size and orientation of the

(28) Tocanne, J. F. and Teissie´, J. Biochim. Biophys. Acta 1990, 1031, 111-142. (29) Hauser, H.; Pascher, I.; Pearson, H. R.; Sundell, S. Biochim. Biophys. Acta 1981, 650, 21-51. (30) So¨derlund, T.; Jutila, A.; Kinnunen, P. K. J. Biophys. J. 1999, 76, 896-907. (31) Lehtonen, J. Y. A.; Kinnunen, P. K. J. Biophys. J. 1997, 72, 1247-1257.

Effect of Transport Protein on a Lipid Bilayer

the result of the favored insertion into fluid-phase domains.32,33 However, this cannot be the case here because POPE and POPG are both in the fluid phase at the temperature of reconstitution. The high hydrophobicity of LacY (12-R-helices embedded in the phospholipid matrix) is likely to be the main force that drives the final integration of the protein within the bilayer. In agreement with such hydrophobicity the protein tends to self-aggregate,34 resulting in the formation of enriched domains such as those observed here. This would explain why the features of the phospholipid phase change are almost unaffected as can be observed in our anisotropy measurements. However, a marginal but measurable electrostatic interaction indicates that hydrogen bonding might exist between LacY and the phospholipid headgroups. This is reasonable considering both the high hydrophobicity of LacY and the interfacial properties of the phospholipids studied.35 Moreover, it is quite conceivable that due to the positive charge exposed to the interface by the protein, the lipid ring immediately surrounding the protein should be enriched in the negatively charged phospholipid (POPG). It is interesting to recall that this phospholipid, as well as the POPE,25 may establish intermolecular hydrogen bonding.36 Thus, in a bioenergetic context, hydrogen bonding should be related to the socalled “microlocalized” chemiosmotic scheme, which basically assumes the existence of a protonic network at the membrane interface.37 Thus, our observation that hydrogen bonding may occur at the protein-membrane interface (32) Rinia, H. A.; Kik, R. A.; Demel, R. A.; Snel, M. E.; Killian, J. A.; van der Eerden, J. P. J. M.; de Kruijff, B. Biochemistry 2000, 39, 58525858. (33) Mall, S.; Broadbridge, R.; Sharma, R. P.; Lee, A. G.; East, J. M. Biochemistry 2000, 39, 2071-2078. (34) Patzlaff, J. S.; Moeller, J. A.; Barry, B. A.; Brooker, R. J. Biochemistry 1998, 37, 15363-15375. Engel, C. K.; Chen, L.; Prive´, G. G. Biochim. Biophys. Acta 2002, 1464, 47-56. (35) Langner, M.; Kubica, K. Chem. Phys. Lipids 1999, 101, 3-35. (36) Boggs, J. M. Biochim. Biophys. Acta 1987, 906, 353-404.

Langmuir, Vol. 21, No. 10, 2005 4647

supports the idea that either PE or PG might transfer the proton via a hydrogen bond to the amino acid or amino acids38 involved in the translocation (probably Glu 269 and His 322). Finally, from the changes in the electrostatic surface potential observed when lactose and TDG are added to charged proteoliposomes, it can be assumed that LacY undergoes a conformational change. This change fits with the proposed mechanism of action3 and is assumed to arise from the binding of the substrate to the protein. Although indirectly, this demonstrates that LacY is only active in the presence of POPG and POPE because those changes were not observed with neutral phospholipid matrixes (data not shown). The effects of LacY on other phospholipid matrixes, neutral and charged, are currently in progress in our laboratory, and the results will be shown in a subsequent paper. Acknowledgment. S.M. and O.D. are recipients of “Recerca i Doce`ncia” fellowships from the Universitat de Barcelona. We thank Professor Ronald Kaback for the material provided and for hosting J.H.B. during a sabbatical period. Financial support by the “Accio´ Gaspar de Portola`-02” from DURSI (Generalitat de Catalunya) is gratefully acknowledged. This work was supported by Grants TIC 2002-04280-C03-01 and SAF2002-00698/ FEDER from the Ministerio de Ciencia y Tecnologı´a, Spain. LA047102D (37) Kell, D. R. Biochim. Biophys. Acta 1979, 505, 1-44. (38) There is a controversy over whether the H+ or the hydronium ion (H3O+) is the transported species and whether a single amino acid or more constitute the binding site. For a detailed discussion see: Johnson, J. L.; Lockheart, M. S. K.; Brooker R. J. J. Membr. Biol. 2001, 181, 215-224, and ref 4.