Acyl Chain Disorder and Azelaoyl Orientation in ... - ACS Publications

Article pubs.acs.org/Langmuir

Acyl Chain Disorder and Azelaoyl Orientation in Lipid Membranes Containing Oxidized Lipids Tiago Mendes Ferreira,*,† Rohit Sood,‡ Ruth Bar̈ enwald,† Göran Carlström,¶ Daniel Topgaard,§ Kay Saalwac̈ hter,† Paavo K. J. Kinnunen,‡ and O. H. Samuli Ollila*,‡ †

Institut für Physik - NMR, Martin-Luther-Universität Halle-Wittenberg, 06108 Halle, Germany Department of Neuroscience and Biomedical Engineering, Aalto University, 02150 Espoo, Finland ¶ Centre for Analysis and Synthesis, Lund University, SE-221 00 Lund, Sweden § Physical Chemistry, Lund University, SE-221 00 Lund, Sweden ‡

S Supporting Information *

ABSTRACT: Oxidized phospholipids occur naturally in conditions of oxidative stress and have been suggested to play an important role in a number of pathological conditions due to their effects on a lipid membrane acyl chain orientation, ordering, and permeability. Here we investigate the effect of the oxidized phospholipid 1-palmitoyl-2-azelaoyl-sn-glycero-3phosphocholine (PazePC) on a model membrane of 1palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) using a combination of 13C−1H dipolar-recoupling nuclear magnetic resonance (NMR) experiments and united-atom molecular dynamics (MD) simulations. The obtained experimental order parameter SCH profiles show that the presence of 30 mol % PazePC in the bilayer significantly increases the gauche content of the POPC acyl chains, therefore decreasing the thickness of the bilayer, although with no stable bilayer pore formation. The MD simulations reproduce the disordering effect and indicate that the orientation of the azelaoyl chain is highly dependent on its protonation state with acyl chain reversal for fully deprotonated states and a parallel orientation along the interfacial plane for fully protonated states, deprotonated and protonated azelaoyl chains having negative and positive SCH profiles, respectively. Only fully or nearly fully protonated azelaoyl chain are observed in the 13C−1H dipolar-recoupling NMR experiments. The experiments show positive SCH values for the azelaoyl segments confirming for the first time that oxidized chains with polar termini adopt a parallel orientation to the bilayer plane as predicted in MD simulations.

■

INTRODUCTION Under biochemical oxidative stress, mono- or polyunsaturated lipids oxidize and form a multitude of products that have potentially harmful effects on cellular membranes. The increase of lipid oxidation is correlated, for instance, with pathological conditions such as atherosclerosis,1,2 Parkinson’s disease,3,4 Alzheimer’s disease,5,6 preeclampsia,7,8 rheumatoid arthritis,9 multiple sclerosis,10,11 and ischemia−reperfusion.12 While it has been suggested that oxidized lipids play an important role in these and other pathological states, as well as in host defensive peptide functionality,13,14 the relevant molecular details involved are not clear.15 One major type of lipid oxidation product commonly found consists of oxidized phospholipids (OxPLs) that incorporate a hydrophilic group on the tip of their sn-2 lipid tail, for example, the fatty acid chain being replaced by dicarboxylic acids, hydroxy-aldehydes, keto-aldehydes, keto-acids, or other residues.16,17 From a thermodynamics and molecular point of view, the picture of how such OxPLs will fit in a membrane is particularly relevant. Because of the hydrophilic group in the © XXXX American Chemical Society

sn-2 chain terminus, the chain may no longer adopt the typical orientation toward the hydrophobic core of the bilayer but reorient instead in order to solvate its hydrophilic tip.14,15 The type of reorientation will depend on the chemical structure of the oxidized chain and is expected to affect the general properties of the bilayer. For instance, Wong-ekkabut et al.18 concluded from molecular dynamics (MD) simulations that the inclusion of OxPLs containing hydroperoxide or aldehyde groups increases the area per molecule of a phospholipid bilayer and therefore reduces its thickness and acyl chain SCH order parameters, giving rise to water defects at high concentration of OxPLs. The degree of disorder imposed was shown to depend on the hydrophilicity of the polar groups, the less hydrophilic aldehyde group imposing a slightly stronger disorder than the hydroperoxide group. A higher contrast of hydrophilicity may be simulated by using a neutral and a charged group in the Received: March 5, 2016 Revised: April 29, 2016

A

DOI: 10.1021/acs.langmuir.6b00788 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

the orientation of the azelaoyl chain will then be dependent on both the pKa of the carboxylic acid group and on the pH of the aqueous phase. The pKa will depend, in turn, on the dipolar environment of the carboxylic acid group. Such dependence may be important since dicarboxylic acid sn-2 residues are biologically relevant and pH variations may occur in pathological conditions, for example, atherosclerosis22 and cancer.23 PazePC specifically is a major product in oxidized low density lipoproteins (LDL) and has been shown to have high affinity toward the transcription factor PARγ,24 as well as to modulate the behavior of a pro-apoptotic protein.25 As it will be shown, the two distinct MD simulation models represented in Figure 1 have clearly distinct SCH order parameter profiles for the azelaoyl chain. SCH can be experimentally determined by different nuclear magnetic resonance (NMR) spectroscopy methods26−30 and is defined as

oxidized chain because the free energy cost of locating a charge in the hydrophobic region is higher compared with a polar moiety.19 This was done by Khandelia et al.20 by comparing the effects of 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine (PazePC) and 1-palmitoyl-2-(9′-oxo-nonanoyl)-sn-glycero-3phosphocholine (PoxnoPC) on a 1-palmitoyl-2-oleoyl-snglycero-3-phosphocholine (POPC) bilayer. The result was that the charged sn-2 azelaoyl chain of PazePC, with a terminal carboxylic group (modeled as −COO−) undergoes an acyl chain reversal, that is, flips toward the water region becoming perpendicular to the bilayer plane, while the neutral sn-2 nonanoyl chain of PoxnoPC with a terminal aldehyde group (−CHO) lies parallel to the bilayer interface. A similar comparative picture of having a polar and a charged group in the sn-2 tip is observed if one reproduces the MD simulation of PazePC in a POPC bilayer but uses the two alternative structures for the carboxylic acid group of the azelaoyl chain, that is, the deprotonated form (−COO−) and the protonated form (−COOH) as done in the framework of the present work. This is shown in Figure 1. It becomes readily clear from the snapshots of the distinct MD simulations that

SCH =

1 ⟨3 cos2 θ − 1⟩ 2

(1)

where θ is the angle between the C−H bond and the bilayer normal and the angular brackets denote ensemble average over a time-interval of approximately 1 μs or less. According to the MD simulations, a protonated chain on average lies parallel to the interfacial plane of the membrane and its C−H order parameters are mostly positive, while a deprotonated chain undergoes acyl chain reversal and has negative SCH values. It is therefore possible to experimentally discriminate between the two molecular models in Figure 1. Here, we study the orientation of the azelaoyl chain and its disordering effect on POPC concentrated multilamellar vesicles (40 wt % hydration) by performing two types of dipolar recoupling solid-state NMR experiments: R-type proton detected local field (R-PDLF) experiments31 for measuring the magnitude of the 13C−1H dipolar couplings and therefore determining |SCH| with C−H segmental resolution for both POPC and PazePC, and sign dipolar recoupling on-axis with scaling and shape preservation (S-DROSS) experiments27 for further determining the sign of the heteronuclear dipolar couplings. The sign determination is specially relevant for investigating the azelaoyl chain orientation as already emphasized. Aqueous solvents with three distinct pH values (5, 9, and 11) were used. We show that for the studied systems, the azelaoyl chain of PazePC is fully or nearly fully protonated and adopts a parallel orientation to the bilayer interfacial plane. Furthermore, we show that PazePC induces disordering of the POPC chains although without formation of stable bilayer pores. The combined methodology used gives a detailed molecular picture of the studied lipid bilayers and can be applied to any other oxidized phospholipids of interest.

■

MATERIALS AND METHODS

Sample Preparation. 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine (PazePC) were purchased from Avanti Polar Lipids. Chloroform with a purity higher than 99.8% was obtained from Sigma-Aldrich. The multilamellar vesicles (MLVs) of POPC/PazePC (70:30) were prepared in the following way. First a lipid film was prepared by blowing dry nitrogen over the surface of a solution of lipids in chloroform in order to evaporate the solvent. The resulting thin lipid film was further dried under reduced pressure for 24 h at room temperature. Three distinct samples were then prepared by hydrating the dried lipid films with solvents with distinct pH values; pH 5 (H2O/ HCl), pH 9 (100 mM boric acid/NaOH), and pH 11 (H2O/NaOH). After addition of solvent, giving the desired ratio of lipid and aqueous fraction (60/40 wt/wt), the mixture was left to equilibrate in a

Figure 1. Representative snapshots of two distinct MD simulations of the POPC/PazePC bilayer as described in the paper, with the carboxyl group of the azelaoyl chain fully protonated (left) and fully deprotonated (right). The atoms that are not from the azelaoyl chain were made transparent for a clear depiction of the orientation of the azelaoyl chain (solid yellow) in the bilayers. Visual Molecular Dynamics software (VMD)21 is used to create the figure. B


Article

Langmuir

any potential errors concerning the phase cycling set up or phase correction in the processing, we used the α (positive SCH) and β (negative SCH) carbons as reference.27 The parameters used are described according to Figure 1b,c of the original reference for the SDROSS experiment.27 The INEPT delay δ2 was 1.19 ms. The τ1 and τ2 in the S-DROSS recoupling blocks R were set as 39.4 and 89.4 μs, respectively. Radio frequency pulses were performed with the following nutation frequencies: 78.1 kHz (13C 90° and 180°), 50 kHz (1H SPINAL64 decoupling). The t1 increment (dipolar recoupling dimension) was 800 μs, and a total of 8 points along t1 were measured using 1024 scans for each with a recycle delay of 5 s and a spectral with of 149.5 ppm. NMR Numerical Simulations. The NMR numerical simulations of S-DROSS curves were performed with the SIMPSON simulation package 33 using the experimental 13 C−1H dipolar couplings determined by the R-PDLF experiments as input. CSA and homonuclear couplings were neglected in all the simulations and the input file rep2000 was used to simulate the random distribution of bilayer orientations in the samples studied. Determination of the Carboxyl Group Protonation State. Two independent NMR methods were used to estimate the protonation state of the carboxyl group in the azelaoyl chain. Carboxyl 13C Chemical Shift Detection. The 13C chemical shift of the terminal carboxyl carbon in PazePC was used for detecting the dependence of its protonation state on the pH of the distinct aqueous solvents used; previous studies on fatty acids in lipid bilayers have shown an increase of about 4−5 ppm of the carboxyl carbon chemical shift from a fully protonated (−COOH) to a fully deprotonated (−COO−) carboxyl group.34,35 Choline Potentiometer. According to this method, the interfacial charge in phosphatidylcholine bilayers can be deduced from the SCH order parameters of the α and β segments in the choline headgroup (see Figure 2c for labeling). Seelig and co-workers36,37 have shown that with the introduction of a cationic or anionic dialkyl surfactant at a molar fraction X± in a POPC bilayer, the ΔνQα and ΔνQβ quadrupole splittings from 2H NMR vary linearly (up to X± ≈ 0.3) as37

desiccator overnight. After a homogeneous mixture was obtained, approximately 25 mg of sample was inserted in a 4 mm HR-MAS rotor (Bruker) with subsequent centrifugation of the rotor to release possible air bubbles and to maximize the amount of sample inside the rotor. The sample with multilamellar vesicles (MLVs) of POPC was prepared with the same protocol and hydrated with water. All solvents were prepared with deionized water that was glass distilled and purified further by a Milli-Q water system (Millipore, Milford, MA). The POPC/PazePC (70:30) molar ratio was chosen since it gave clearly detectable signal also from PazePC. Preliminary experiments with lower PazePC concentrations were in line with reported results, but the resolution did not allow the determination of PazePC order parameters. Solid-State NMR Experiments. All experiments were performed under a magic angle spinning (MAS) frequency of 5 kHz. The 13C direct polarization (DP) and R-type proton detected local field (RPDLF) experiments were performed using an E-free 4 mm HCP CP− MAS probe with a Bruker Avance II-500 NMR spectrometer operating at a 1H Larmor frequency of 500.13 MHz (Lund University, Sweden). The S-DROSS experiment was performed using standard 4 mm CPMAS HX probes with a Bruker Avance III 400 spectrometer operating at a 1H Larmor frequency of 400.03 MHz (Martin-Luther University Halle-Wittenberg, Germany). The following experimental setups were used. 13 C DP Experiments. Radio frequency pulses were performed with the following nutation frequencies: 80.65 kHz (13C 90°) and 40 kHz (TPPM 1H decoupling pulses). A number of 128 (solvent pH 5 and 11) or 256 (solvent pH 9) scans were acquired using a recycle delay of 30 s and a spectral width of 200.8 ppm. R-PDLF Experiments To Measure |SCH| Values. The outcome of the R-PDLF sequence is a 2D spectrum with carbon chemical shift resolution in the direct dimension and dipolar splitting ΔνR‑PDLF in the indirect dimension. The 13C chemical shift-resolved dipolar splittings of the 13C−1H R-PDLF 2D spectra relate to the order parameters SCH of the resolved carbons as31 max Δν R ‐ PDLF = 0.315dCH |SCH|

(2)

ΔνiQ (X ±) = ΔνiQ (0) ± miX ±

where 0.315 is a scaling factor specific of the dipolar recoupling sequence used in the R-PDLF experiment and dmax CH is the rigid dipolar coupling constant equal to 22.1 kHz. The pulse sequence parameters are described according to Figures 1c and 2c of the original reference for the R-PDLF experiment.31 The refocused-INEPT (rINEPT) delays τ1 and τ2 were 1.79 and 1.20 ms, respectively. Radio frequency pulses with the nutation frequencies: 45.00 kHz (R1871 pulses), 80.65 kHz (13C 90° and 180° pulses), 40 kHz (TPPM 1H decoupling pulses). The t1 increment was equal to 11.11 μs × 18 × 2, and 32 points in the indirect dimension were recorded using 128 (solvent pH 5 and 11) or 256 (solvent pH 9) scans, with recycle delay of 5 s and a spectral width of 149.5 ppm. S-DROSS Experiments to Measure the Sign of SCH. The outcome of the S-DROSS sequence is a 2D spectrum with carbon chemical shift resolution in the direct dimension and sine modulation of the different 13 C peaks in the indirect dimension, t1. According to average Hamiltonian theory32 and the effective Hamiltonian of the SDROSS sequence and phase cycling reported by Gross et al.,27 the sine modulation of the intensity for an arbitrary 13C peak will be

I(t1) =

∫ψ P(ψ ) sin(ξ⟨ωD⟩t1) dψ

where the constant mi is positive and depends on the ionic surfactant charge and the average position of the ionic group in the bilayer. In their original publication, Seelig et al. suggested mα and mβ to have opposite signs. At the time, the sign of quadrupolar splittings was not accessible, and it was assumed that both Δν0i ’s were positive. Now knowing that Δν0α and Δν0β are negative and positive,38 respectively, we rewrote the original equation as eq 5 considering now mi to be always positive. The real value of the quadrupolar splittings is39 3 ΔνiQ = − χSCD 4

(6)

with the quadrupole coupling constant χ approximately equal to 167 kHz for methylenes. Since it is known from experiments that SCD ≈ SCH,38 the interfacial charge relation to the quadrupolar splittings in eq 5 can be rewritten as (i) (i) SCH (X ±) = SCH (0) ∓

4 −1 χ miX ± 3

(7)

where the order parameters of the choline segments under neutral (α) (β) (0) and SCH (0), are equal to +0.06 and −0.03, charge, SCH respectively.27 According to the potentiometer concept, the negative charge induced by deprotonation of azelaoyl chain carboxylic acid should then be detectable by measuring the changes of the S(α) CH and S(β) CH order parameters. Molecular Dynamics Simulations. The lipid bilayer molecular dynamics (MD) simulations were run having all the PazePC molecules either in the protonated or in the deprotonated form. Both simulations contained 90 POPC molecules and 38 PazePC molecules. The system with protonated PazePC was solvated with 7632 water molecules while the system with deprotonated PazePC had 7250 water molecules and 38 K+ ions to neutralize the total charge in the simulation box. The

(3)

where ξ is a scaling factor specific of the S-DROSS, P(ψ) is the probability of finding a bilayer with its bilayer normal at an angle ψ with respect to the external magnetic field, and the time averaged dipolar coupling angular frequency is 1 max ⟨ωD⟩ = 2πdCH SCH (3 cos2 ψ − 1) 2

(5)

(4)

For samples with randomly oriented bilayers, if SCH is positive, the intensity will go initially from zero to negative values, while when SCH is negative there will be a positive initial intensity buildup. To avoid C


Article

Langmuir

Figure 2. 13C NMR spectral features used for investigating the molecular structure of the POPC/PazePC (70:30) multilamellar vesicles in pH 5 solvent. (a) 13C rINEPT spectra from multilamellar vesicles of POPC and POPC/PazePC at 25 °C. The asterisks label the exclusive peaks of PazePC. (b) Sections of the R-PDLF spectrum of the POPC/PazePC multilamellar vesicles. Each highly resolved 13C−1H dipolar splitting is directly proportional to the order parameter magnitude |SCH| of its assigned carbon according to eq 2 in the text. The α and β 13C−1H dipolar splittings were used as sensors of the electrostatic interfacial charge of the bilayer. (c) Label scheme used for identifying the distinct carbons in POPC and PazePC. Note that all labels in the R-PDLF spectra, except for *, refer exclusively to POPC.

■

potassium ions were used to avoid the artifacts that arise from overestimated sodium binding in the used force-field.40 The system with protonated PazePC was run for 167 ns, the last 67 ns being used for analysis, while the system with deprotonated PazePC was run for 120 ns using the last 60 ns for analysis. For the pure POPC bilayer, we used simulation data from a previous publication,30 available at ref 41. The Berger-based force field42 with double bond correction by Bachar et al.43 introduced in ref 44 was used for describing POPC molecules in all simulations. The force field for the deprotonated PazePC was taken from ref 20. For describing protonated PazePC molecules, the hydrogen was added together with ad hoc partial charges giving a total charge equal to zero. All the force-field parameters and simulation files, including the generated trajectories, are available at refs 45 and 46. The starting structures for the simulations with PazePC were made by replacing the appropriate amount of POPC molecules by PazePC molecules from the end structure of a POPC bilayer simulation. A time step of 2 fs was used with leapfrog integrator. Covalent bond lengths were constrained with the LINCS algorithm.47,48 PME49,50 with real space cutoff 1.0 nm was used for electrostatics. Plain cutoff was used for the Lennard-Jones interactions with a 1.0 nm cutoff. The neighbor list was updated every fifth step with cutoff 1.0 nm. Temperature was coupled separately for lipids, water, and ions to 298 K with the velocity-rescale method51 with coupling constant 0.1 ps−1. Pressure was semi-isotropically coupled to the atmospheric pressure with the Beredsen barostat.52 The coordinates were written every 20 ps.

RESULTS AND DISCUSSION The R-PDLF spectrum of the sample with POPC/PazePC multilamellar vesicles is shown in Figure 2. The spectrum displays highly resolved 13C−1H dipolar splittings from which the SCH magnitude for all the distinct C−H bonds in POPC, as well as for most of the azelaoyl C−H bonds in PazePC, can be calculated. POPC peaks/splittings are assigned according to a previous determination.30 In order to make an interpretation of the large set of experimental SCH magnitudes measured, it is convenient to make use of the simulated MD models. Before doing so, it is crucial however to determine the protonation state of the PazePC azelaoyl chain in the MLVs studied, since significant structural differences are expected between protonated and deprotonated azelaoyl chains according to the MD simulations performed as illustrated in Figure 1. The Protonation State of the Azelaoyl Chain. The pKa values of carboxylic groups in fatty acids vary between ∼4 and 9 in water depending on acyl chain length,53,54 while values around ∼7 in vesicles35 and above 12 in apolar environments55 are measured. Thus, the protonation state of the carboxyl group at the end of the azelaoyl chain is expected to depend on its environment, that is, on its penetration depth into the bilayer. On the other hand, the penetration depth is expected to depend on the charge and polarity of the azelaoyl chain terminal end, that is, on the protonation state,14,15,18,20 as demonstrated in Figure 1. The protonation state of the azelaoyl D


Article

Langmuir terminal carboxylic group is therefore determined by a complex interplay between the carboxyl proton dissociation in different environments and the change in free energy due to acyl chain reorientation. In addition, the PazePC concentration is quite high (∼0.6 M) in the MLV samples compared with the acid or base concentration of the solvent (10−1 M), thus the system pH is not fully determined by the solvent. Also the local proton concentration at the bilayer interface and interior may not be equal to the concentration in solution. Due to these complications, the system pH cannot be straightforwardly deduced from the system composition. Thus, we decided to carefully determine the PazePC protonation state in PazePC/ POPC MLV samples with solutions having different pH values of 5, 7, and 9. To estimate the PazePC protonation state, we first used the dipolar slices of the R-PDLF spectra of the headgroup α and β segments of POPC as a potentiometer, based on the work of Seelig and co-workers36,37 and as described in the Materials and Methods section. The α and β dipolar slices, at 59.4 and 65.9 ppm of the R-PDLF spectra, respectively, are shown in Figure 3

Figure 4. 13C direct polarization spectra at the carbonyl chemical shift region acquired from the various POPC/PazePC samples hydrated with solvents with different pH. 13

C chemical shift of the COOH group in PazePC (less than 0.5 ppm) are observed as a function of solvent pH. The significant 4−5 ppm increase in carboxyl chemical shift related to deprotonation35 is therefore not detected (see Materials and Methods) supporting the conclusion that PazePC molecules adopted the same protonation state in all distinct samples. Moreover, the dipolar slices from R-PDLF spectra for the azeoyl segments in Figure 5 display the same dipolar splittings

Figure 3. Solvent pH dependence of the dipolar slices of the R-PDLF spectra at the chemical shift positions of the α and β carbons of POPC.

for all studied samples. Only small (∼20 Hz) and nonsystematic changes are observed in these dipolar splittings, close to the frequency resolution of 10 Hz used in the experiments, and the corresponding order parameter change calculated from eq 2 is ∼0.003, which indicates an almost negligible conformational change of the choline headgroup in comparison to pure POPC bilayers. To estimate an upper bound for the amount of negatively charged PazePC molecules in the bilayer using the potentiometer concept,36,37 we assumed that the mα and mβ should be in the range of the values found by Scherer et al.37 (no larger than 10 kHz), which implies no more than 11% of deprotonated PazePC molecules according to eq 7. On the other hand, we cannot exclude that the very small variations observed in the headgroup R-PDLF dipolar slices in Figure 3 may arise from interactions between the POPC molecules with fully protonated PazePC molecules. The carboxyl carbon chemical shifts from different samples are shown in Figure 4. Only small nonsystematic changes of the

Figure 5. Solvent pH dependence of dipolar slices of the R-PDLF spectra at the chemical shifts indicated corresponding to four distinct azelaoyl carbons.

in the different solvents showing that the azelaoyl chain order parameters are not affected by the solvent pH. Since different orientations and SCH order parameters of the protonated and deprotonated azelaoyl chains are expected (see Figure 1), this result also indicates that the oxidized acyl chain has the same protonation state in all the different samples. In summary, three independent sets of observables indicate that the protonation state of the azelaoyl carboxyl terminus is the same in the MLVs samples with solvent pH values of 5, 9, and 11. Minute changes in the headgroup order parameters and E


Article

Langmuir

Figure 6. Disordering effect of PazePC on POPC bilayers hydrated with 40 wt % of H2O/HCl (pH equal to 5) and at a temperature of 25 °C. (left) Contour maps for the crowded spectral region from 29 to 30.5 ppm of the 13C−1H R-PDLF spectra taken from pure POPC MLVs (top) and those with 30 mol % of PazePC (bottom) and selected dipolar slices showing the decrease of dipolar splittings with the presence of PazePC in the bilayers. (right) The SCH magnitude segmental profiles for the sn-1 and sn-2 acyl chains of POPC determined from the full R-PDLF spectra and comparison with the corresponding MD simulation profiles. The full 2D spectra with assignments are given as Supporting Information.

deprotonated and fully protonated PazePC molecules. Disordering of acyl chains was also observed when probes with 7nitrobenz-2-oxa-1,3-diazol-4-yl (NBD) group attached to an acyl chain of phosphatidylcholine were embedded in a POPC bilayer.62 Interestingly, similar acyl chain reorientation is suggested for these probes and oxidized lipids.62−64 Together the results indicate that the relocalization of the acyl chain ends to the water interface leads to the general disordering effect in the acyl chain region. Stable bilayer pores have been suggested to arise at high concentrations of oxidized lipids65−67 in a lipid bilayer. Such pore formation does not occur in the studied systems since the C−H bond order parameters in the first segments of acyl chains, glycerol backbone, and choline segments of POPC are not affected by the presence of PazePC in experiments, as seen in Figure 6 and in Supporting Information. This means that the decrease of order parameters observed must arise from an increase of acyl chain gauche content rather than from formation of bilayer pores or increased molecular wobbling, which would affect all C−H bond order parameters in POPC.68 Also the small decrease in POPC ppm values due to the addition of PazePC can be related to the increase of gauche content.69 However, the presence of PazePC in the bilayers could increase the water permeation also in the studied conditions by increasing the polarity of the lipid bilayer core.18,70 The R-PDLF spectra does not show any sign of phase separation since only single dipolar splittings are observed for each acyl chain segment; however, we cannot exclude the existence of nanoscopic domains between which molecular exchange takes place on time scales faster than milliseconds.30 Based on typical R-PDLF spectra and order parameter values, we conclude that the system is in the lamellar liquid crystalline

the results in a later section indicate a fully or nearly fully protonated state with all the used solvents. In micellar and small unilamellar vesicle samples where the lipid concentration is lower and the system pH is determined by the buffer solution, the deprotonated form is observed also at pH values around ∼7.56−59 We conclude that the current data allow us to address the properties of protonated PazePC and that a solvent with pH higher than 11 is needed to detect the deprotonated form of PazePC in POPC MLVs. Since all the samples gave essentially the same results, we restrict the following discussion to the experimental observations of POPC/PazePC MLVs at pH 5. PazePC Disorders the Hydrophobic Region of the Bilayer. While bilayer structural effects induced by acyl chain oxidation are considered physiologically relevant,13−15 unambigous experimental data on such effects is sparse.60,61 The issue is largely complicated by the unequal effects of different oxidation products and the scarcity of spectroscopic experiments with well-defined oxidized products.60,61 Here we directly measure the 13C−1H dipolar couplings and corresponding order parameter magnitudes |SCH| for all the POPC C−H bonds, giving highly resolved details on the molecular structure of the lipid bilayer with atomistic resolution. Figure 6 displays the 13C−1H dipolar splittings from the crowded methylene region of the R-PDLF spectrum at 29−30.5 ppm, showing a clear disordering of the acyl chains, that is, a decrease of acyl chain order parameters, due to the presence of oxidized phospholipids in the POPC bilayer. The result confirms the predicted decrease of the acyl chain order parameters in previous MD simulations of oxidized phospholipids in lipid bilayers18,20 and agrees with the interpretation of EPR experiments for PazePC.60 The disordering of acyl chains is also observed in both of our MD simulations, with fully F


Article

Langmuir

(including signs) for the azelaoyl chain C−H bonds, using the MD simulation models for the structural interpretation. Figure 7 shows the rINEPT and R-PDLF spectra at the chemical shift regions of the azelaoyl peaks marked with * in Figure 2. Six azelaoyl peaks with distinct chemical shifts and

Lα phase without macroscopic phase separation. The phase separation for a DMPC/PazePC mixture observed with 31P NMR experiments in the same temperature probably follows from the higher phase transition temperature for DMPC compared with POPC.71 The experimentally measured order parameter decrease is reproduced by MD simulations with both protonated and deprotonated PazePC molecules. Therefore, it is reasonable to use the MD models to analyze the changes also in other bilayer properties. Table 1 shows the area per molecule and bilayer Table 1. Area per Molecule and Hydrophobic Thickness Calculated from Simulations system

Apl (nm2)a

d (nm)b

POPC POPC/PazePC (protonated) POPC/PazePC (deprotonated)

66.6 ± 0.1 65.9 ± 0.2 64.4 ± 0.2

2.9 2.6 2.7

a

Area per lipid. bHydrophobic thickness from crossing of water and acyl chain number densities.

hydrophobic thickness calculated from the simulations. The presence of both deprotonated and protonated PazePC in a bilayer slightly decreases the area per molecule and the bilayer thickness, the effect being slightly stronger for deprotonated PazePC molecules. The decrease in thickness is in agreement with previous scattering studies,59,72 while the decrease of the area per molecule is in contrast with Langmuir balance experiments with different PazePC mixtures.73,74 An increase of area per molecule was also observed in most MD simulations with different oxidized species.18,20,67 A decrease of area per molecule was only observed for simulations with deprotonated oxidized acyl chains, and was attributed to the sodium ion binding,20,75 which is, however, too strong in the used model.40 To minimize this artifact, we instead used potassium ions as counterions, but still observed a decrease of area per molecule for both, the deprotonated and the protonated (no ions present) systems. The protonated form of PazePC has not been previously simulated. The simultaneous decrease of order parameters, area per molecule, and bilayer thickness does not follow the traditionally expected relations between these structural parameters.76 This probably follows from the atypical oxidized acyl chain structures where only one of the PazePC acyl chains enter the bilayer interior. Our results together with previous studies suggest that lipid oxidation decreases the membrane thickness about 2−3 Å, but the relation to the lateral packing density is not straightforward. Form factors from scattering experiments could be used to judge whether the predicted area per molecule decrease due to PazePC in POPC bilayer is a simulation artifact or a real effect. Orientation of the Oxidized Chain in the Bilayer. The reorientation adopted by an oxidized acyl chain in order to locate their polar or charged terminal ends close to the hydrophilic/hydrophobic interface has been extensively discussed in the literature15,18,20,73,77,78 as has its potential physiological relevance.13−15 Several experiments73,77,78 and MD simulations,18,20,75 including the simulations presented here, support the existence of an acyl chain reorientation toward a parallel orientation to the bilayer interface or acyl chain reversal, but direct experimental measurements of such structures had been until now missing. Here, we present experimental data for the protonated form of the PazePC azelaoyl acyl chain in POPC MLVs by measuring SCH values

Figure 7. 13C−1H dipole coupling magnitudes and signs for six azelaoyl carbons. (I) Region of the 13C rINEPT spectra of the POPC/ PazePC MLVs (solvent pH 5) containing peaks from the azelaoyl chain. (II) Sections of the R-PDLF spectrum for the azelaoyl peaks showing six distinct carbons with different 13C−1H dipolar couplings. (III) Slices of the R-PDLF spectra at the chemical shifts labeled a−f with the corresponding experimental S-DROSS slices at the same chemical shifts (black) and numerically simulated S-DROSS curves using positive order parameters (gray). The bottom R-PDLF and SDROSS profiles refer to the α (positive SCH) and β (negative SCH) segments and were used as reference. In order to compare the experimental and simulated S-DROSS curves, the curves were normalized such that their maximum absolute value is unity. Error bars depict the maximum amplitude in noise-only regions of the experimental spectrum. G


Article

Langmuir

studied and that the measured order parameters would be an average over both conformations. Despite of the incomplete model, the MD simulations predict that the two different chain conformations of the protonated and deprotonated forms produce very different order parameters with different sign. Thus, these conformations could be experimentally distinguished with dipolar recoupling NMR experiments. This is to be confirmed by, for example, performing the R-PDLF and S-DROSS experiments for a system with a significant fraction of deprotonated PazePC molecules.

dipolar couplings are clearly observed in the R-PDLF spectrum. We assume the peaks at 24.2−24.4 ppm and 28.1−28.4 ppm to correspond to the carbons 3−7 of the azelaoyl chain (Figure 2 for labeling) since carbons 2 and 8 should be at the typical 13C chemical shift region of covalently bound carbons to carbonyls from 34 to 48 ppm.79 The peak labeled with f below 33.7 ppm is assigned to carbon 8 of the azelaoyl chain since we expect the remaining carbon 2 to have a similar chemical shift (close to 33.9) to the carbons 2 and 2′ of the POPC sn-1 and sn-2 chains, respectively. The high spectral resolution achieved allowed us to accurately determine the order parameters for all azelaoyl peaks except for the ones labeled a and f with order parameters too small for an accurate determination. However, we can estimate that these smaller order parameters should be in the range of −0.03 to +0.03. The S-DROSS intensity modulation curves measured for the different azelaoyl carbons and shown in Figure 7 prove that all of these carbons have C−H bonds with positive SCH values. As shown in Figure 8, this is also the case for MD simulation with

■

CONCLUSION The POPC/PazePC (70:30%mol) MLVs hydrated with 40 wt % of solvents with pH values of 5, 9, and 11 were all found to be in an homogeneous lamellar liquid crystalline Lα phase at a temperature of 25 °C. The set of experiments and simulations performed indicates that in all the studied samples, the azelaoyl carboxyl group is fully or almost fully protonated and that the azelaoyl chain adopts an orientation parallel to the interfacial plane of the membrane. This conformation can be differentiated from acyl chain reversal by measuring the order parameter signs. At the concentration studied, PazePC significantly decreases the acyl chain ordering without formation of bilayer stable pores. The protonation state of PazePC, as well as the disorder of POPC acyl chains and the orientation of the choline group, are not affected by the used solvent pH. This indicates that solvent pH above 11 is needed to detect deprotonated PazePC form in PazePC/POPC MLVs.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b00788. Contour plots for selected regions of the 13C−1H RPDLF spectra of POPC/PazePC and pure POPC MLVs, complete assignment of the POPC R-PDLF dipolar splittings, and all POPC order parameter magnitudes determined experimentally with R-PDLF spectroscopy and calculated from the MD simulation trajectories (PDF)

Figure 8. Simulated and experimental SCH profiles of the azelaoyl chain. The assignment of the carbons 3−7 of the azelaoyl chain was done in order to get a best fit.

protonated azelaoyl chains but not for simulation with deprotonated chains. The peaks a−f are assigned to the azelaoyl chain carbons 3−7 in Figure 8 to give the best agreement between protonated simulation and experimental order parameters. In addition to the correct signs, the order parameter magnitudes of protonated azelaoyl chain are close to the experimental values. However, the agreement is not within experimental accuracy and two distinct order parameters for the distinct C−H bonds in a methylene group are observed for some segments in simulations ( forking38) but not in experiments, indicating that the atomistic resolution structures in simulations do not fully correspond to the experiments. This may be due to the insufficient length of the simulation to allow full conformational sampling or from inaccuracies in the force field description.38 The typical simulation time scales are sufficient to sample conformational space for regular acyl chains,38 but the dynamics of azelaoyl might be significantly slower. On the other hand, the conformations of the glycerol backbone and the beginning of sn-2 acyl chain are suggested to be relevant for atypical conformations of oxidized acyl chains14 but are not correctly captured by the used simulation model.30,80 In addition, we cannot fully exclude the presence of a small fraction of deprotonated PazePC in the samples

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: samuli.ollila@aalto.fi. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS O.H.S.O. acknowledges Emil Aaltonen foundation for financial support and the computational resources provided by the Aalto Science-IT project. Himanshu Khandelia and Jirasak Wongekkabut are acknowledged for sharing their simulation files. Roman Volinsky is acknowledged for the help in sample preparation. The authors acknowledge financial support from the European Commission under the Seventh Framework Program by means of the grant agreement for the Integrated Infrastructure Initiative No. 262348 European Soft Matter Infrastructure (ESMI). H


Article

Langmuir

■

(24) Davies, S. S.; Pontsler, A. V.; Marathe, G. K.; Harrison, K. A.; Murphy, R. C.; Hinshaw, J. C.; Prestwich, G. D.; Hilaire, A. S.; Prescott, S. M.; Zimmerman, G. A.; McIntyre, T. M. Oxidized Alkyl Phospholipids Are Specific, High Affinity Peroxisome Proliferatoractivated Receptor γ Ligands and Agonists. J. Biol. Chem. 2001, 276, 16015−16023. (25) Wallgren, M.; Lidman, M.; Pham, Q. D.; Cyprych, K.; Gröbner, G. The oxidized phospholipid PazePC modulates interactions between Bax and mitochondrial membranes. Biochim. Biophys. Acta, Biomembr. 2012, 1818, 2718−2724. (26) Davis, J. H. The description of membrane lipid conformation, order and dynamics by 2H NMR. Biochim. Biophys. Acta, Rev. Biomembr. 1983, 737, 117−171. (27) Gross, J. D.; Warschawski, D. E.; Griffin, R. G. Dipolar Recoupling in MAS NMR: A Probe for Segmental Order in Lipid Bilayers. J. Am. Chem. Soc. 1997, 119, 796. (28) Dvinskikh, S. V.; Castro, V.; Sandström, D. Efficient solid-state NMR methods for measuring heteronuclear dipolar couplings in unoriented lipid membrane systems. Phys. Chem. Chem. Phys. 2005, 7, 607. (29) Leftin, A.; Brown, M. F. An NMR database for simulations of membrane dynamics. Biochim. Biophys. Acta, Biomembr. 2011, 1808, 818−839. (30) Ferreira, T. M.; Coreta-Gomes, F.; Ollila, O. H. S.; Moreno, M. J.; Vaz, W. L. C.; Topgaard, D. Cholesterol and POPC segmental order parameters in lipid membranes: solid state 1H-13C NMR and MD simulation studies. Phys. Chem. Chem. Phys. 2013, 15, 1976−1989. (31) Dvinskikh, S. V.; Zimmermann, H.; Maliniak, A.; Sandstrom, D. Measurements of motionally averaged heteronuclear dipolar couplings in MAS NMR using R-type recoupling. J. Magn. Reson. 2004, 168, 194−201. (32) Levitt, M. H. Spin Dynamics; John Wiley & Sons Ltd.: Chichester, U.K., 2008. (33) Bak, M.; Rasmussen, J. T.; Nielsen, N. C. SIMPSON: A general simulation program for solid-state NMR spectroscopy. J. Magn. Reson. 2011, 213, 366−400. Part of special issue Magnetic Moments Groundbreaking papers from the pages of the Journal Magnetic Resonance - and recollections from the scientists behind them. (34) Cistola, D. P.; Small, D. M.; Hamilton, J. A. Ionization behavior of aqueous short-chain carboxylic acids: a carbon-13 NMR study. J. Lipid Res. 1982, 23, 795−9. (35) Small, D. M.; Cabral, D. J.; Cistola, D. P.; Parks, J. S.; Hamilton, J. A. The Ionization Behavior of Fatty Acids and Bile Acids in Micelles and Membranes. Hepatology 1984, 4, 77S−79S. (36) Seelig, J.; MacDonald, P. M.; Scherer, P. G. Phospholipid head groups as sensors of electric charge in membranes. Biochemistry 1987, 26, 7535−7541. (37) Scherer, P. G.; Seelig, J. Electric Charge Effects on Phospholipid Headgroups - Phosphatidylcholine in Mixtures with Cationic and Anionic Amphiphiles. Biochemistry 1989, 28, 7720−7728. (38) Ollila, O. S.; Pabst, G. Atomistic resolution structure and dynamics of lipid bilayers in simulations and experiments. Biochim. Biophys. Acta, Biomembr. 2016, DOI: 10.1016/j.bbamem.2016.01.019. (39) Ferreira, T. M.; Bernin, D.; Topgaard, D. NMR Studies of Nonionic Surfactants. Annu. Rep. NMR Spectrosc. 2013, 79, 73−127. (40) Catte, A.; Girych, M.; Javanainen, M.; Miettinen, M. S.; Monticelli, L.; Mäaẗ tä, J.; Oganesyan, V. S.; Ollila, O. H. S. The electrometer concept and binding of cations to phospholipid bilayers, 2015, DOI: 10.5281/zenodo.32175 (41) Ollila, O. H. S.; Ferreira, T.; Topgaard, D. MD simulation trajectory and related files for POPC bilayer (Berger model delivered by Tieleman, Gromacs 4.5), 2014, DOI: 10.5281/zenodo.13279 (42) Berger, O.; Edholm, O.; Jähnig, F. Molecular dynamics simulations of a fluid bilayer of dipalmitoylphosphatidylcholine at full hydration, constant pressure, and constant temperature. Biophys. J. 1997, 72, 2002−2013. (43) Bachar, M.; Brunelle, P.; Tieleman, D. P.; Rauk, A. Molecular Dynamics Simulation of a Polyunsaturated Lipid Bilayer Susceptible to Lipid Peroxidation. J. Phys. Chem. B 2004, 108, 7170−7179.

REFERENCES

(1) Stocker, R.; Keaney, J. Role of oxidative modifications in atherosclerosis. Physiol. Rev. 2004, 84, 1381−1478. (2) Esterbauer, H.; Gebicki, J.; Puhl, H.; Jurgens, G. The role of lipidperoxidation and antioxidants in oxidative modification of LDL. Free Radical Biol. Med. 1992, 13, 341−390. (3) Simonian, N.; Coyle, J. Oxidative stress in neurodegenerative diseases. Annu. Rev. Pharmacol. Toxicol. 1996, 36, 83−106. (4) Jenner, P. Oxidative stress in Parkinson’s disease. Ann. Neurol. 2003, 53, S26−S36. (5) Behl, C.; Davis, J.; Lesley, R.; Schubert, D. Hydrogen peroxide mediates amyloid β protein toxicity. Cell 1994, 77, 817−827. (6) Markesbery, W. Oxidative stress hypothesis in Alzheimer’s disease. Free Radical Biol. Med. 1997, 23, 134−147. (7) Morris, J.; Gopaul, N.; Endresen, M.; Knight, M.; Linton, E.; Dhir, S.; Angard, E.; Redman, C. Circulating markers of oxidative stress are raised in normal pregnancy and pre-eclampsia. BJOG 1998, 105, 1195−1199. (8) Walsh, S. Lipid peroxidation in pregnancy. Hypertens. Pregnancy 1994, 13, 1−32. (9) Firuzi, O.; Fuksa, L.; Spadaro, C.; Bousova, I.; Riccieri, V.; Spadaro, A.; Petrucci, R.; Marrosu, G.; Saso, L. Oxidative stress parameters in different systemic rheumatic diseases. J. Pharm. Pharmacol. 2006, 58, 951−957. (10) Mazza, M.; Pomponi, M.; Janiri, L.; Bria, P.; Mazza, S. Omega-3 fatty acids and antioxidants in neurological and psychiatric diseases: An overview. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 2007, 31, 12− 26. (11) Haider, L.; Fischer, M. T.; Frischer, J. M.; Bauer, J.; Hoeftberger, R.; Botond, G.; Esterbauer, H.; Binder, C. J.; Witztum, J. L.; Lassmann, H. Oxidative damage in multiple sclerosis lesions. Brain 2011, 134, 1914−1924. (12) Oliver, C.; Starkereed, P.; Stadtman, E.; Liu, G.; Carney, J.; Floyd, R. Oxidative damage to brain proteins, loss of glutaminesynthetase activity, and production of free-radicals during schema reperfusion-induced injury to gerbil brain. Proc. Natl. Acad. Sci. U. S. A. 1990, 87, 5144−5147. (13) Kinnunen, P. K. J. Amyloid Formation on Lipid Membrane Surfaces. Open Biol. J. 2010, 2, 163−175. (14) Kinnunen, P. K.; Kaarniranta, K.; Mahalka, A. K. Proteinoxidized phospholipid interactions in cellular signaling for cell death: From biophysics to clinical correlations. Biochim. Biophys. Acta, Biomembr. 2012, 1818, 2446−2455. (15) Volinsky, R.; Kinnunen, P. K. J. Oxidized phosphatidylcholines in membrane-level cellular signaling: from biophysics to physiology and molecular pathology. FEBS J. 2013, 280, 2806−2816. (16) Reis, A.; Spickett, C. M. Chemistry of phospholipid oxidation. Biochim. Biophys. Acta, Biomembr. 2012, 1818, 2374−2387. Part of special issue Oxidized phospholipids - their properties and interactions with proteins. (17) Domingues, M. R. M.; Reis, A.; Domingues, P. Mass spectrometry analysis of oxidized phospholipids. Chem. Phys. Lipids 2008, 156, 1−12. (18) Wong-ekkabut, J.; Xu, Z.; Triampo, W.; Tang, I.-M.; Tieleman, D. P.; Monticelli, L. Effect of Lipid Peroxidation on the Properties of Lipid Bilayers: A Molecular Dynamics Study. Biophys. J. 2007, 93, 4225−4236. (19) Israelachvili, J. Intermolecular and Surface Forces; Academic Press: London, 1991. (20) Khandelia, H.; Mouritsen, O. G. Lipid Gymnastics: Evidence of Complete Acyl Chain Reversal in Oxidized Phospholipids from Molecular Simulations. Biophys. J. 2009, 96, 2734−2743. (21) Humphrey, W.; Dalke, A.; Schulten, K. VMD − Visual Molecular Dynamics. J. Mol. Graphics 1996, 14, 33−38. (22) Ö örni, K.; Rajamäki, K.; Nguyen, S. D.; Lähdesmäki, K.; Plihtari, R.; Lee-Rueckert, M.; Kovanen, P. T. Acidification of the intimal fluid: the perfect storm for atherogenesis. J. Lipid Res. 2015, 56, 203−214. (23) Lee, E. S.; Gao, Z.; Bae, Y. H. Recent progress in tumor pH targeting nanotechnology. J. Controlled Release 2008, 132, 164−170. I


Article

Langmuir (44) Ollila, S.; Hyvönen, M. T.; Vattulainen, I. Polyunsaturation in Lipid Membranes: Dynamic Properties and Lateral Pressure Profiles. J. Phys. Chem. B 2007, 111, 3139−3150. (45) Ollila, S. MD simulation trajectory and related files for POPC bilayer with 30 mol % protonated pazePC (Berger, Gromacs 4.5.), 2016; http://dx.doi.org/10.5281/zenodo.44660. (46) Ollila, S., Khandelia, H. MD simulation trajectory and related files for POPC bilayer with 30 mol% of deprotonated pazePC (Berger, Gromacs 4.5.), 2016; http://dx.doi.org/10.5281/zenodo.44622. (47) Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. LINCS: a linear constraint solver for molecular dynamics simulations. J. Comput. Chem. 1997, 18, 1463−1472. (48) Hess, B. P-LINCS: A Parallel Linear Constraint Solver for Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 116−122. (49) Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: An N· log(N) method for Ewald sums in large systems. J. Chem. Phys. 1993, 98, 10089. (50) Essman, U. L.; Perera, M. L.; Berkowitz, M. L.; Larden, T.; Lee, H.; Pedersen, L. G. A smooth particle mesh ewald potential. J. Chem. Phys. 1995, 103, 8577−8592. (51) Bussi, G.; Donadio, D.; Parrinello, M. Canonical sampling through velocity rescaling. J. Chem. Phys. 2007, 126, 014101. (52) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; DiNola, A.; Haak, J. R. Molecular dynamics with coupling to an external bath. J. Chem. Phys. 1984, 81, 3684−3690. (53) Kanicky, J. R.; Poniatowski, A. F.; Mehta, N. R.; Shah, D. O. Cooperativity among Molecules at Interfaces in Relation to Various Technological Processes: Effect of Chain Length on the pKa of Fatty Acid Salt Solutions. Langmuir 2000, 16, 172−177. (54) Kanicky, J. R.; Shah, D. O. Effect of Premicellar Aggregation on the pKa of Fatty Acid Soap Solutions. Langmuir 2003, 19, 2034−2038. (55) Kolthoff, I. M.; Chantooni, M. K., Jr.; Bhowmik, S. Dissociation constants of uncharged and monovalent cation acids in dimethyl sulfoxide. J. Am. Chem. Soc. 1968, 90, 23−28. (56) Mattila, J.-P.; Sabatini, K.; Kinnunen, P. K. Oxidized Phospholipids as Potential Novel Drug Targets. Biophys. J. 2007, 93, 3105−3112. (57) Mattila, J.-P.; Sabatini, K.; Kinnunen, P. K. J. Interaction of Cytochrome c with 1-Palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine: Evidence for Acyl Chain Reversal. Langmuir 2008, 24, 4157− 4160. (58) Pande, A. H.; Kar, S.; Tripathy, R. K. Oxidatively modified fatty acyl chain determines physicochemical properties of aggregates of oxidized phospholipids. Biochim. Biophys. Acta, Biomembr. 2010, 1798, 442−452. (59) Makky, A.; Tanaka, M. Impact of Lipid Oxidization on Biophysical Properties of Model Cell Membranes. J. Phys. Chem. B 2015, 119, 5857−5863. (60) Megli, F. M.; Russo, L. Different oxidized phospholipid molecules unequally affect bilayer packing. Biochim. Biophys. Acta, Biomembr. 2008, 1778, 143−152. (61) Jurkiewicz, P.; Olżyńska, A.; Cwiklik, L.; Conte, E.; Jungwirth, P.; Megli, F. M.; Hof, M. Biophysics of lipid bilayers containing oxidatively modified phospholipids: Insights from fluorescence and EPR experiments and from MD simulations. Biochim. Biophys. Acta, Biomembr. 2012, 1818, 2388−2402. (62) Huster, D.; Müller, P.; Arnold, K.; Herrmann, A. Dynamics of Membrane Penetration of the Fluorescent 7-Nitrobenz-2-Oxa-1,3Diazol-4-yl (NBD) Group Attached to an Acyl Chain of Phosphatidylcholine. Biophys. J. 2001, 80, 822−831. (63) Chattopadhyay, A.; London, E. Parallax method for direct measurement of membrane penetration depth utilizing fluorescence quenching by spin-labeled phospholipids. Biochemistry 1987, 26, 39− 45. (64) Mazéres, S.; Schram, V.; Tocanne, J.; Lopez, A. 7-nitrobenz-2oxa-1,3-diazole-4-yl-labeled phospholipids in lipid membranes: differences in fluorescence behavior. Biophys. J. 1996, 71, 327−335. (65) Smith, H. L.; Howland, M. C.; Szmodis, A. W.; Li, Q.; Daemen, L. L.; Parikh, A. N.; Majewski, J. Early Stages of Oxidative Stress-

Induced Membrane Permeabilization: A Neutron Reflectometry Study. J. Am. Chem. Soc. 2009, 131, 3631−3638. (66) Cwiklik, L.; Jungwirth, P. Massive oxidation of phospholipid membranes leads to pore creation and bilayer disintegration. Chem. Phys. Lett. 2010, 486, 99−103. (67) Lis, M.; Wizert, A.; Przybylo, M.; Langner, M.; Swiatek, J.; Jungwirth, P.; Cwiklik, L. The effect of lipid oxidation on the water permeability of phospholipids bilayers. Phys. Chem. Chem. Phys. 2011, 13, 17555−17563. (68) Ferreira, T. M.; Topgaard, D.; Ollila, O. H. S. Molecular Conformation and Bilayer Pores in a Nonionic Surfactant Lamellar Phase Studied with 1H-13C Solid-State NMR and Molecular Dynamics Simulations. Langmuir 2014, 30, 461−469. (69) Ferreira, T. M.; Medronho, B.; Martin, R. W.; Topgaard, D. Segmental order parameters in a nonionic surfactant lamellar phase studied with 1H-13C solid-state NMR. Phys. Chem. Chem. Phys. 2008, 10, 6033−6038. (70) Volinsky, R.; Cwiklik, L.; Jurkiewicz, P.; Hof, M.; Jungwirth, P.; Kinnunen, P. Oxidized Phosphatidylcholines Facilitate Phospholipid Flip-Flop in Liposomes. Biophys. J. 2011, 101, 1376−1384. (71) Wallgren, M.; Beranova, L.; Pham, Q. D.; Linh, K.; Lidman, M.; Procek, J.; Cyprych, K.; Kinnunen, P. K. J.; Hof, M.; Gröbner, G. Impact of oxidized phospholipids on the structural and dynamic organization of phospholipid membranes: a combined DSC and solid state NMR study. Faraday Discuss. 2013, 161, 499. (72) Mason, R.; Walter, M. F.; Mason, P. E. Effect of Oxidative Stress on Membrane Structure: Small-Angle X-Ray Diffraction Analysis. Free Radical Biol. Med. 1997, 23, 419−425. (73) Sabatini, K.; Mattila, J.-P.; Megli, F. M.; Kinnunen, P. K. Characterization of Two Oxidatively Modified Phospholipids in Mixed Monolayers with DPPC. Biophys. J. 2006, 90, 4488−4499. (74) Volinsky, R.; Paananen, R.; Kinnunen, P. Oxidized Phosphatidylcholines Promote Phase Separation of Cholesterol-Sphingomyelin Domains. Biophys. J. 2012, 103, 247−254. (75) Beranova, L.; Cwiklik, L.; Jurkiewicz, P.; Hof, M.; Jungwirth, P. Oxidation Changes Physical Properties of Phospholipid Bilayers: Fluorescence Spectroscopy and Molecular Simulations. Langmuir 2010, 26, 6140−6144. (76) Kinnun, J. J.; Mallikarjunaiah, K.; Petrache, H. I.; Brown, M. F. Elastic deformation and area per lipid of membranes: Atomistic view from solid-state deuterium NMR spectroscopy. Biochim. Biophys. Acta, Biomembr. 2015, 1848, 246−259. (77) van Kuijk, F. J.; Sevanian, A.; Handelman, G. J.; Dratz, E. A. A new role for phospholipase A2: protection of membranes from lipid peroxidation damage. Trends Biochem. Sci. 1987, 12, 31−34. (78) Greenberg, M. E.; Li, X.-M.; Gugiu, B. G.; Gu, X.; Qin, J.; Salomon, R. G.; Hazen, S. L. The Lipid Whisker Model of the Structure of Oxidized Cell Membranes. J. Biol. Chem. 2008, 283, 2385−2396. (79) Couperus, P. A.; Clague, A. D. H.; van Dongen, J. P. C. M. Carbon-13 chemical shifts of some model carboxylic acids and esters. Org. Magn. Reson. 1978, 11, 590−597. (80) Botan, A.; Favela-Rosales, F.; Fuchs, P. F. J.; Javanainen, M.; Kanduč, M.; Kulig, W.; Lamberg, A.; Loison, C.; Lyubartsev, A.; Miettinen, M. S.; Monticelli, M.; Mäaẗ tä, J.; Ollila, O. H. S.; Retegan, M.; Róg, T.; Santuz, H.; Tynkkynen, J. Toward Atomistic Resolution Structure of Phosphatidylcholine Headgroup and Glycerol Backbone at Different Ambient Conditions. J. Phys. Chem. B 2015, 119, 15075− 15088.

J


Acyl Chain Disorder and Azelaoyl Orientation in ... - ACS Publications

Recommend Documents