Effect of Ring Size in Ï-Alicyclic Fatty Acids on the Structural and

Subscriber access provided by University of Sydney Library

Article

The Effect of Ring Size in #-Alicyclic Fatty Acids on the Structural and Dynamical Properties Associated with Fluidity in Lipid Bilayers David Poger, and Alan Edward Mark Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.5b02635 • Publication Date (Web): 06 Oct 2015 Downloaded from http://pubs.acs.org on October 10, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The Effect of Ring Size in ω-Alicyclic Fatty Acids on the Structural and Dynamical Properties Associated with Fluidity in Lipid Bilayers David Poger∗ and Alan E. Mark School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane QLD 4072, Australia E-mail: [email protected] Phone: +61 (0)7 3365 7562. Fax: +61 (0)7 3365 3872

1

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Fatty acids containing a terminal cyclic group such as cyclohexyl and cycloheptyl are commonly found in prokaryotic membranes, especially in those of thermo-acidophilic bacteria. These so-called ω-alicyclic fatty acids have been proposed to stabilise the membranes of bacteria by reducing the fluidity in membranes and increasing lipid packing and lipid chain order. In this article, molecular dynamics simulations are used to examine the effect of 3- to 7-membered cycloalkyl saturated and unsaturated (cyclopent-2-enyl and phenyl) rings in ω-alicyclic fatty acyl chains on the structure (lipid packing, lipid chain order and fraction of gauche defects in the chains) and dynamics (lateral lipid diffusion) of a model lipid bilayer. It was found that ω-alicyclic chains in which the ring was saturated reduced lipid condensation and lowered chain order which would be associated with enhanced fluidity. However, this effect was limited. The lateral diffusion of the lipids diminished as the ring size increased. In particular, ω-cyclohexyl and ω-cycloheptyl acyl tails led to a decrease in lipid diffusion. In contrast, ω-alicyclic acyl chains that contain an unsaturated ring promoted membrane fluidity both in terms of changes in membrane structure and lipid diffusion. This may indicate that saturated and unsaturated terminal rings in ω-alicyclic fatty acids fulfil alternative functions within membranes. Overall, the simulations suggest that ωalicyclic fatty acids in which the terminal ring is saturated might protect the membrane of thermo-acidophilic bacteria from high-temperature and low-pH conditions through a “dynamical barrier” that would limit lipid diffusion and transmembrane diffusion of undesired ions and molecules.

2


Page 2 of 37

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Introduction Lipid bilayers consisting of phospholipids in which the two fatty acyl chains are linear (saturated or unsaturated) have been thoroughly examined as models for biological membranes. However, while such fatty acids are typical of eukaryotic membranes, branched-chain fatty acids are common in bacterial and archaeal membranes where their concentration often exceeds that of linear-chain fatty acids 1 . The variations in the presence, nature and number of branching groups in fatty acids lead to distinct physico-chemical properties of the lipids that in turn, result in biological membranes with unique structural and functional properties. 2,3 Amongst branched-chain fatty acids are those that contain a terminal ring. They are often referred to as ω-alicyclic fatty acids in the literature. The structure of an ωalicyclic fatty acid in which the ring is cyclohexyl is shown in Figure 1. In some cell types, ω-alicyclic fatty acids can amount to 60–90% of all fatty acids in membrane lipids. 4–7 For example, ω-cyclopentyl, ω-cyclohexyl and ω-cycloheptyl fatty acids occur in bacteria 5–9 and plants, 10 and ω-cyclobutyl fatty acids are found in some bacteria in the form of fused cyclobutane moieties (ladderane fatty acids). 11 Unsaturated ω-cyclic fatty acids are found in plants. 12,13 Interestingly, a range of ω-cyclopent-2-enyl fatty acids (namely hydnocarpic acid, chaulmoogric acid and grolic acid) and their derivatives have also been shown to alter the properties of bacterial membranes, even to have antibacterial activity against Mycobacterium leprae and Mycobacterium vaccae. 12 The precise role of ω-alicyclic fatty acids in biological membranes is unclear. Lipids containing ω-alicyclic fatty acids are the major membrane components of members of the thermoacidophilic genera Alicyclobacillus and Sulfobacillus (they grow at temperatures of 40–70 °C in a pH range of 2–6) 4,6,7,14,15 . Based on a series of calorimetric and infrared, nuclear magnetic resonance (NMR) and fluorescence spectroscopic studies of monolayers and bilayers of 1,2di(ω-cyclohexyl)glycerophospholipids (mostly phosphatidylcholines) in a liquid-crystalline (i.e. fluid) phase, it has been claimed that ω-alicyclic fatty acids induce a reduction in trans–gauche isomerisation in lipid chains, 16 an increase in lipid chain order 16–20 and a de3


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

crease in the area per lipid 20,21 when compared to fluid-phase phospholipid bilayers composed of linear acyl tails. In addition, the permeability of liquid-crystalline lipid bilayers and vesicles containing ω-cyclohexyl phospholipids to water, small molecules (such as glucose and glycerol) and ions has been reported to be lower than that of the corresponding systems consisting of linear acyl chains. 16,17,20,22–24 It has thus been proposed that ω-alicyclic fatty acids diminish the degree of fluidity in membranes compared to linear-chain fatty acids and it is this that enables thermo-acidophiles to withstand acidic pH and high temperatures. However, other studies have suggested that ω-alicyclic fatty acids have little effect on, or indeed lead to enhanced, membrane fluidity. 25,26 Specifically, fluorescence anisotropy measurements of a fluorophore embedded in membranes containing ω-cyclopropyl, ω-cyclobutyl, ω-cyclopentyl and ω-cyclohexyl acyl chains, in which a moderate to large decrease in fluorescence anisotropy was observed, have been interpreted as indicating a gain in membrane fluidity. The aim of the present study is to shed light on the nature of the effect of ω-alicyclic fatty acyl chains in phospholipids on the structure and dynamics of membranes; in particular, to shed light on how these lipids affect the fluidity of membranes. Using molecular dynamic simulations, the effect of both ring size (from cyclopropane to cycloheptane) and saturation was examined. The degree of fluidity of phospholipid bilayers containing an sn-1 ω-alicyclic tail was characterised in terms of structural (area per lipid, bilayer thickness, lipid chain order and proportion of gauche conformers in the lipid tails) and dynamical (lateral mobility of lipids) properties of the bilayers. The results obtained were also compared with that of two phospholipids composed of two saturated linear fatty acids, namely DPPC (1,2dipalmitoyl-sn-glycero-3-phosphocholine) and UPPC (2-palmitoyl-1-undecanoyl-sn-glycero3-phosphocholine). Fluidity is a multifaceted collective property that is dependent on the balance between interactions between lipid headgroups, lipid chains and lipids and water. The degree of fluidity of a lipid bilayer can be characterised using a range of properties which, while they are inferred from experiment, are often intrinsically interrelated. 27–29 Molecular

4


Page 4 of 37

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

packing within the plane of a lipid bilayer is estimated using the average interfacial area per lipid, which, in turn, is directly linked to the average bilayer thickness and the lipid molecular volume, both indicative of the transverse organisation of a lipid bilayer. The former depends on the conformation of the hydrocarbon chains and their relative orientation (that is order) within the bilayer. All other properties being equal, an increase in the area per lipid or a decrease in lipid chain order is interpreted as an increase in the fluidity of a bilayer. However, an increase in chain order can offset an increase in the area per lipid, resulting in the absence of or a small change in the level of fluidity of a bilayer. Finally, studies of the effect of cholesterol on the structure and dynamics of lipid bilayers have shown that the packing density and the level of chain order can affect the dynamical properties of lipids, such as their mobility in the plane of a bilayer. 30 The results of the simulations were compared with experimental data whenever possible. Note, UPPC was chosen as a reference because ω-alicyclic fatty acids are often derived from undecanoic and tridecanoic acids in bacteria. 4,6,8,9 The effect of the terminal unsaturated rings cyclopent-2-enyl and phenyl was also investigated. Overall, the simulations show that ω-alicyclic fatty acids enhanced some structural properties associated with membrane fluidity. The lateral mobility of lipids containing a terminal saturated ring was similar to, or lower than, DPPC. The presence of a terminal unsaturated ring in the tails tended to promote membrane fluidity in terms of both structure and dynamics.

Methods Simulation systems Nine systems were examined. Each consisted of a preassembled bilayer of 288 lipids and 12960 water molecules (corresponding to 45 water molecules per lipid) to ensure a fully hydrated state. The lipid molecules examined were DPPC (1,2-dipalmitoyl-sn-glycero-3phosphocholine; 16:0/16:0), UPPC (2-palmitoyl-1-undecanoyl-sn-glycero-3-phosphocholine; 5


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

11:0/16:0), and seven derivatives of UPPC (five saturated and two unsaturated) in which a terminal ring, saturated or unsaturated, was added to the sn-1 n-undecanoyl fatty acyl chain. The five (saturated) 11-cycloalkyl derivatives of UPPC included: cyclopropyl (c3 UPPC; 1(11-cyclopropylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine), cyclobutyl (c4 UPPC, 1-(11-cyclobutylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine), cyclopentyl (c5 UPPC, 1-(11-cyclopentylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine), cyclohexyl (c6 UPPC, 1-(11-cyclohexylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine), and cycloheptyl (c7 UPPC, 1-(11-cycloheptylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine).

The two unsatu-

rated cyclic groups studied were: (1R)-cyclopent-2-en-1-yl (HPPC, 1-[(R)-hydnocarpoyl]2-palmitoyl-sn-glycero-3-phosphocholine) and phenyl (pUPPC, 2-palmitoyl-1-(11-phenylundecanoyl)-sn-glycero-3-phosphocholine). The system consisting of a pure DPPC bilayer was taken from a previous study. 2 All other lipid bilayers were generated by modifying the sn-1 palmitoyl tails of an equilibrated DPPC bilayer obtained after 400 ns of simulation. 2 In all cases, the bilayers contained a single lipid species. The structures and the names of the lipids are given in Figure 2.

Simulation parameters All simulations were performed using Gromacs, version 3.3.3 31 in conjunction with the Gromos 54a7 united-atom force field. 29 This force field has been shown to reproduce a range of properties of unbranched and branched phosphatidylcholine bilayers in a liquidcrystalline phase. 2,3,29,32,33 Parameters for the ω-alicyclic groups were obtained from the Automated Topology Builder (ATB). 34 They were based on the corresponding propylcycloalkanes for saturated rings (that is cyclopropane, cyclobutane, cyclopentane, cyclohexane and cycloheptane in c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC, respectively), (3R)-3propylcyclopent-1-ene for HPPC and propylbenzene for pUPPC. Each system was simulated under periodic boundary conditions in a rectangular box. The temperature of the system was maintained by coupling to an external temperature bath at the reference temperature 6


Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

of 323 K with a coupling constant of 0.1 ps using a Berendsen thermostat. 35 This temperature is above the gel-to-fluid-phase-transition temperature (Tm ) of a pure DPPC bilayer (314.4 K). 36,37 To our knowledge, Tm has not been measured for a pure UPPC bilayer or any of the lipid bilayers shown in Figure 2 containing the ω-alicyclic fatty acyl chains described earlier. Nevertheless, studies on related phosphatidylcholine bilayers in which both the sn-1 and sn-2 chains were undecanoyl 38 or 11-cyclohexylundecanoyl have shown that shortening the length of the lipid tails or adding a terminal cyclohexyl group to the tails 18,22,39 lowers the Tm compared to a DPPC bilayer. It is thus reasonable to assume that all the lipid bilayers examined are in a liquid-crystalline phase at 323 K. The lipids and water were coupled independently to the heat bath. The pressure was kept at 1 bar in the lateral and normal directions with respect to the bilayer by weakly coupling to an anisotropic Berendsen barostat 35 using an isothermal compressibility of 4.6 ×10−5 bar−1 and a coupling constant of 1 ps. Using anisotropic pressure coupling ensured that the lipid bilayer could relax fully in all three box dimensions. The length of all covalent bonds within lipids were constrained using the Lincs algorithm. 40 The geometry of the Simple Point Charge (SPC) water molecules 41 was constrained using Settle. 42 A 2-fs timestep was used. Nonbonded interactions were evaluated using a twin-range cutoff scheme: interactions falling within the 0.8-nm short-range cutoff were calculated every step whereas interactions falling within the 1.4-nm long-range cutoff were updated every 10 fs, together with the pair list. A reaction-field correction 43 was applied to account for the truncation of the electrostatic interactions beyond the long-range cutoff using a relative dielectric permittivity constant of 62, as appropriate for SPC water. 44 Each system was first energy-minimised and then simulated at 50 K for 10 ps. The temperature was increased gradually over 180 ps until the final simulation temperature of 323 K was attained. Whether a given system had reached equilibrium was determined by examining the time evolution of the potential energy and the area per lipid. Once the systems were equilibrated (in general within 150–200 ns of simulation), data was collected for 200 ns. Each system was simulated three times with different initial velocities. An overview of the

7


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

simulations performed is given in Table 1.

Results and Discussion To probe the effect of ω-alicyclic acyl chains on different phospholipid bilayers, a range of structural and dynamical properties was examined. These included the area per lipid, the bilayer thickness, the fatty acid chain order, the conformational preference of the tails and lipid diffusion. These properties were in turn used to assess changes in the degree of fluidity in the membrane. Note, fluidity is a collective property dependent on both the structural and dynamical characteristics of a membrane. The level of fluidity of a lipid bilayer results from a fine balance between lipid–lipid and lipid–water interactions. Therefore, a range of properties such as the area per lipid, the lipid chain order and conformation, and lipid diffusion that are indicative of the relative strength of lipid–lipid and lipid–water interactions have been analysed. Note, an increase in the attraction between lipid species or conversely a decrease in lipid–water interactions as observed in a liquid-ordered phase, translates into a tighter packing of lipids which, in turn, is associated with a lower area per lipid and a decrease in fluidity. 27,29

Overall Effect on the Bilayer Structure The area per lipid AL was calculated as the lateral area of the simulation box divided by the number of lipids in each leaflet over the last 200 ns of the individual simulations. The variation of AL with the size and the nature of the ring is shown in Figure 3A. In all cases, the addition of a ring increased AL when compared to that of DPPC (0.623 nm2 ). The ω-alicyclic chains containing a saturated ring (referred to as cn UPPC in Figure 3, with n = 3–7 for c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC) had only a limited effect on AL (AL increased by less than 1.5%). The unsaturated rings cyclopent-2-enyl in HPPC and phenyl in pUPPC caused larger variations in AL (+2.4 and +8%, respectively). Interestingly,

8


Page 8 of 37

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

the increase in AL for UPPC (c0 UPPC in Figure 3A) was similar to that observed for chains ending with a saturated ring or cyclopent-2-enyl. Therefore, the gain in fluidity in the bilayers associated with the looser packing of the lipids containing an ω-alicyclic tails with respect to a DPPC bilayer can be mostly ascribed to the shorter length of the sn-1 undecanoyl tail rather than to the presence of a cycle at the ω end. Note, some literature studies have suggested that ω-alicyclic fatty acids were associated with tighter packing of the lipid tails 21,24 and denser bilayers 20 than saturated linear-chain lipids in a fluid phase. However, these conclusions were based on either studies of supported monolayers of 1,2di(ω-cyclohexyl)phosphatidylcholines, 21 inferred from measurements of water permeability coefficients through large unilamellar vesicles of 1,2-di(ω-cyclohexyl)phosphatidylcholines, 24 or estimated from the mass of bilayers composed of lipid membrane extracts. 20 None of these approaches directly measure lipid packing or AL . The thickness DHH of each bilayer was derived from the total electron density profile of the system and corresponded to the distance between the two maxima of the electron density of the system along the bilayer normal (taken as the z-axis) over the last 200 ns of the individual simulations. The variation of DHH across the different lipid bilayers is displayed in Figure 3B. The lowest value for DHH was found for UPPC (3.17 nm) which, when compared to that for DPPC (3.59 nm), is consistent with X-ray and neutron diffraction studies showing that the bilayer thickness increased with the length of saturated acyl chains. 45,46 DHH increased with the size of the saturated ring reaching a value similar to that of a DPPC bilayer for c6 UPPC and c7 UPPC.

Order and Conformation of the Fatty Acid Chains The degree of order of the acyl chains was characterised in terms of the order parameter SCH that measures the relative orientation of the C–H bond with respect to the bilayer normal. It is defined as: 1 SCH = ⟨3 cos2 β − 1⟩ 2 9


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

where β is the angle between a C–H bond and the normal to the bilayer. The angular brackets denote an ensemble average over all of the lipids for each frame of the simulation. As Gromos 54a7 is an united-atom force field wherein aliphatic, nonpolar hydrogens are treated implicitly and incorporated into the carbon to which they are bound, the positions of the hydrogen atoms were constructed based on the positions of the neighbouring heavy atoms (carbons) assuming an ideal sp3 geometry for tetrahedral carbons in the linear part of the fatty acid chains, and a slightly distorted sp3 geometry determined for the carbon rings after geometry optimisation using density functional theory (B3LYP/6-31G*). 34 The |SCH | profiles of the sn-1 and sn-2 chains of all the lipids are displayed in Figure 4 together with those calculated for the palmitoyl chains in DPPC in a previous study using the same force field. 2 Note, for each methylene group, |SCH | was averaged over the two C–H bonds. In the terminal cycles, |SCH | was also averaged over equivalent CH or CH2 groups, that is CH or CH2 groups at positions 13 and 13′ in cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl and phenyl; 14 and 14′ in cyclopentyl, cyclohexyl, cycloheptyl and phenyl; and 15 and 15′ in cycloheptyl. From Figure 4, it is evident that while the shorter sn-1 undecanoyl chain in UPPC compared to the sn-1 palmitoyl chain in DPPC caused a decrease in order along both the sn-1 and sn-2 chains, the addition of a terminal saturated ring in the sn-1 tail in c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC, restored order along the two chains to a level similar to that in DPPC. In contrast, the rings were in general less ordered than the chains. For all of these lipids, the |SCH | values were lower than 0.05 for the terminal CH or CH2 groups (at position 13 in c3 UPPC, 14 in c4 UPPC and c5 UPPC, and 15 in c6 UPPC and c7 UPPC). However, the general level of order within the bilayers estimated as the average |SCH | value computed over the two tails for UPPC, c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC remained close to that of a DPPC bilayer (Figure 5). ⟨|SCH |⟩ values were around 0.156–0.164 (⟨|SCH |⟩ = 0.176 for DPPC). On the other hand, the presence of a terminal unsaturated ring in the sn-1 chain in HPPC and pUPPC led to an overall reduction in order. This is evident from both |SCH | in Figure 4 and ⟨|SCH |⟩ in Figure 5. Specifically,

10


Page 10 of 37

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

while the |SCH | values in the rings in HPPC and pUPPC were comparable to those in the saturated rings on the whole, the values along the non-cyclic region of the sn-1 and the sn-2 chains were lower than in c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC (Figure 4). In HPPC, the |SCH | values were intermediate between those of UPPC and DPPC whereas in the case of pUPPC, they were lower than UPPC for the sn-1 chain (by 8% on average) but similar for the sn-2 chain. This loss of order within HPPC and pUPPC bilayers resulted in ⟨|SCH |⟩ values (0.143 and 0.136, respectively; Figure 5) that were lower than those calculated for the lipid bilayers that contain acyl chains with a saturated ring. That saturated terminal rings were associated with some loss of order within the bilayers is consistent with previous studies that showed that di(ω-cyclohexyl)phosphatidylcholine bilayers exhibited lower gel-to-liquid-crystalline phase transition temperatures than the corresponding linear-chain lipids. 18,22 However, other experimental studies by Sunamoto et al. 17 and Mantsch et al. 19 concluded that ω-alicyclic fatty acids promoted order. As for the effect of ω-alicyclic fatty acids on lipid packing discussed earlier, the increase in order reported in these studies was estimated indirectly through changes in the fluorescence anisotropy of a membrane-embedded probe 17 and infrared spectroscopy. 19 In the fluorescence study of Sunamoto et al., 17 the intensity of fluorescence depolarisation of the fluorophore 1,6diphenyl-(1E,3E,5E )-hexa-1,3,5-triene (DPH) was used as a proxy for the level of chain order in di(ω-cyclohexyl)phosphatidylcholine bilayers. In this approach, it is assumed that the rotational motions of DPH result in fluorescence depolarisation which are associated with fluctuations in the orientation of acyl chains. 47 Despite its wide use, the precise localisation and orientation of DPH in the hydrophobic region of lipid bilayers in particular branched-lipid bilayers, is unclear. In fact, the behaviour of DPH has to date only been examined in model membranes containing linear-chain lipids. In these systems, it has been shown to lie either close to the carbonyl region, to spread broadly within the bilayer with its long axis aligned along the acyl chains, or to reside in the centre of bilayers parallel to the bilayer plane. 47,48 Therefore, as the localisation and orientation of DPH in lipid bilay-

11


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ers containing ω-alicyclic fatty acyl chains is uncertain, great care needs to be taken when attributing the change in fluorescence depolarisation of DPH observed to a gain or a loss of chain order within di(ω-cyclohexyl)phospholipid bilayers. 17 In the infrared spectroscopy study of Mantsch et al., 19 it was claimed that the non-cyclic section of ω-cyclohexyl chains was less flexible than in saturated, linear-chain phosphatidylcholines. However, the length of these saturated, linear-chain phospholipids was not indicated. Figures 4 and 5 clearly show that in the simulations, ω-alicyclic tails containing a saturated ring have a level of order slightly lower that in a DPPC bilayer but higher than that in an UPPC bilayer. Potential loss in chain order described by the authors is therefore relative. The level of order within the bilayers was further examined by calculating the average fraction xg of gauche defects per carbon–carbon bond in the lipid tails. The occurrence of gauche rotamers in acyl chains indicates the level of conformational disorder within a lipid bilayer. For example, the gel-to-liquid-crystalline phase transition of phospholipid bilayers is associated with an increase in the number of gauche conformers in the acyl chains. 49 The torsion angles were calculated along the sn-1 and sn-2 chains. Note, for the the sn-1 chain in c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC, c7 UPPC, HPPC and pUPPC, only the non-cyclic part was considered, that is torsions about the bonds from the C2 –C3 bond to the C11 –C12 bond. A torsion angle φ was classified as gauche when 30° < |φ| ≤ 90°. 50 The variation of xg across the different bilayers investigated is shown in Figure 6. As can be seen, the fraction or concentration of gauche conformers in the lipid tails was virtually unaffected by the presence of a terminal ring in the sn-1 chain nor by the shorter sn-1 tail in UPPC when compared to DPPC. On average, there was 0.24–0.25 gauche rotamers per bond in the tails in all systems. This is in good agreement with previous estimates from theoretical calculations and infrared spectroscopy experiments for a liquid-crystalline DPPC bilayer. 49,51 Despite the absence of experimental values for UPPC, this is consistent with NMR spectroscopy estimates of the total number of gauche conformers per chain in the related saturated, linear 1,2-dilauroyland 1,2-dimyristoylphosphatidylcholine bilayers. 52 Interestingly, the fraction of gauche ro-

12


Page 12 of 37

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

tamers in HPPC and pUPPC was similar to that in ω-alicyclic fatty acyl chains that have a saturated ring.

Lipid Diffusion Finally, the normal (Brownian) diffusion of lipids within the plane of the bilayer was examined by calculating the lateral diffusion coefficient of the lipids DL using the Einstein relation: 1 DL = lim 4 t→∞

!

# 1" ∥rxy (t0 + t) − rxy (t0 )∥2 t0 t

$

where rxy (t) is the position of the centre of mass of the phosphate group of a lipid molecule " # in the bilayer plane at time t. ∥rxy (t + t0 ) − rxy (t0 )∥2 t0 is the mean squared displacement

(MSD) of the centre of mass of the phosphate of a lipid projected onto the plane of the bilayer over the time t and computed over all possible starting times t0 < t. The drift of

the bilayer leaflets during the simulations was corrected for by subtracting the motion of the centre of mass of the leaflet from the MSD of each lipid molecule. In practice, DL was calculated by averaging the MSD over all the lipids in the bilayer. DL was then derived from the slope of the linear regime of the time evolution of the average MSD displayed in Figure 7 for all the systems investigated. The MSD is linear in time over the time interval 20 ≤ t ≥ 120 ns in all cases (indicated by the gray dashed lines in Figure 7). The variation of DL with the size and the nature of the ring is shown in Figure 8. Overall, DL decreased with the size of the saturated ring, from about 4 µm2 ·s−1 for UPPC (considered to have a ring size of 0) down to 1.6–1.8 µm2 ·s−1 for c6 UPPC and c7 UPPC. Note, the values of DL for c3 UPPC, c4 UPPC and c5 UPPC were comparable to that of DPPC (2.4 µm2 ·s−1 ). The unsaturated rings led to different relative behaviour with respect to their corresponding saturated rings: DL for HPPC was similar to that of c5 UPPC (DL ≈ 2.2 µm2 ·s−1 ) whereas DL for pUPPC (3.4 µm2 ·s−1 ) was double that of c6 UPPC (1.6 µm2 ·s−1 ). To our knowledge, the lateral diffusion coefficient of lipids that include ω-alicyclic tails has not been reported in

13


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the literature. However, in the case of DPPC, using the Gromos 54a7 force field, there is a reasonable agreement between the calculated values of DL and experimental estimates. 3,33 The DL profile in Figure 8 suggests that as the size of the saturated ring increases, the mobility of the lipids within the plane of the bilayer diminishes. This is consistent with the fact that when compared to linear-chain lipids in fluid-phase bilayers, liposomes made from ω-cyclohexylphospholipids, were found to be less permeable to ions (such as H+ and OH− ) and small molecules (such as glucose, glycerol, glycol and erythritol). 16,17,22–24 In the framework of the free-volume theory, 53,54 the diffusing particle can move from its initial position if: i. it has sufficient energy to overcome its interactions with its current neighbours, and ii. there is a site of sufficient free volume that is available to move into. In the context of transverse diffusion of a particle through a membrane containing ω-alicyclic tails, if the first condition is met, then the limited mobility of the lipids within the bilayer, in particular for those containing ω-cyclohexyl and ω-cycloheptyl rings, reduces the probability of formation of an adjacent, transient volume accessible to the diffusing particle. Therefore, as proposed by the permeability studies cited above, 17,19,22–24 ω-alicyclic fatty acids containing a 6- or 7-membered saturated ring may create an effective barrier to permeation of ions and small molecules across cellular membranes. However, this barrier does not stem from a change in the structural properties of membranes that would lead to a lower degree of membrane fluidity than in linear-chain membranes (for example through denser lipid packing and increased lipid chain order), but from a dampening of lipid dynamics as highlighted by the variations of DL . In the context of cellular membranes, the lower diffusivity of lipids containing ω-cyclohexyl or ω-cycloheptyl fatty acids may constrain the mobility of lipids and proteins within the membranes, thereby contributing to the subdiffusion of those elements. 55 Subdiffusion, or anomalous diffusion, corresponds to the diffusion regime wherein the MSD grows nonlinearly with time, but instead scales as a power-law MSD ∝ tα with an anomalous scaling exponent α < 1. 56 Lipids containing ω-cyclohexyl or ω-cycloheptyl tails may thus facilitate the lateral segregation of lipids and membranes proteins through the formation of low-mobility domains

14


Page 14 of 37

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(islets) that would obstruct or trap lipids and proteins, as observed in theoretical 55,57 and experimental 58,59 studies.

Comparison with the Fluidity of Lipid Bilayers in Experimental Studies The effect of ω-alicyclic fatty acyl chains on the structural and dynamical properties of a lipid bilayer was examined and compared systematically with that of saturated, linear-chain lipids (DPPC and UPPC) in order to shed light on how the presence of terminal saturated rings modify the fluidity of bilayers. The level of fluidity was assessed in terms of both the structure (area per lipid, bilayer thickness, lipid chain order and fraction of gauche defects within the chains) and dynamics (lipid lateral diffusion) of each bilayer. It has been claimed in the literature that ω-alicyclic chains, especially ω-cyclohexyl and ω-cycloheptyl chains, protected thermo-acidophilic bacteria from low-pH and high-temperature conditions by reducing the fluidity of their membrane through greater lipid chain order, increased lipid packing 17–21 leading to lower permeability to water, ions and small molecules. 16,17,20,22–24 However, changes in the fluidity have not been measured directly. In this work, it was found that in fact, ω-alicyclic fatty acids containing a saturated ring tended to affect structural properties associated with a gain in fluidity, namely an increase in the area per lipid and a decrease in lipid chain order. The effect was, however, in general limited and comparable to the degree of fluidity achieved in a DPPC bilayer in a liquidcrystalline state for rings ranging from cyclobutane to cycloheptane. In terms of the bilayer thickness, while it increased with the size of the saturated ring, it plateaued at a value comparable to that of a DPPC bilayer for the ω-cyclohexyl and ω-cycloheptyl tails. In contrast, the lateral mobility of lipids containing fatty acyl chains with a terminal saturated ring was either similar to or lower than that of DPPC. The simulations suggest that contrary to previous assumptions, these ω-alicyclic fatty acids, in particular ω-cyclohexyl and ω-cycloheptyl fatty acids, stabilise the membrane of thermo-acidophilic bacteria through a 15


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

“dynamical barrier” that restricts the diffusion of lipids within the plane of the membrane and the transmembrane diffusion of undesired osmolytes (such as protons), rather than a structural barrier associated with low area per lipid and high lipid chain order. At the same time, their effect on the structure of membranes, in terms of area per lipid, membrane thickness, lipid chain order and conformational preference, is overall comparable to membranes composed of mainly linear-chain lipids to maintain the biological functions of the membranes and membrane proteins. The effect of ω-alicyclic fatty acyl chains in which the ring is unsaturated (cyclopent-2-enyl or phenyl), was also investigated. They tended to systematically enhance the fluidity of lipid bilayers, both structurally and dynamically, to a greater extent than DPPC and their corresponding chains with a saturated ring (c5 UPPC and c6 UPPC). The effect was greater for the ω-phenyl fatty acids. This indicates that the increase in fluidity observed for the unsaturated rings characterised by a larger AL (Figure 3A), reduced thickness (Figure 3B) and lower order parameters |SCH | (Figures 4 and 5) than in their corresponding saturated bilayers, was not associated with an enhanced flexibility of the chains as might be reflected in greater xg values (Figure 6). Instead, it was a result of steric hindrance due to the rings. As the cyclopent-2-en-1-yl and phenyl groups are bulkier than the cyclopentyl and cyclohexyl groups, this may promote a looser packing of the lipids and a decrease in order within the tail region. ω-Alicyclic chains wherein the ring is unsaturated may play a role in membranes that is distinct from that of ω-alicyclic chains with a saturated ring. Based on the results from the simulations, the antibacterial activity of ω-cyclopent-2-enyl fatty acids (hydnocarpic acid, chaulmoogric acid and grolic acid) and derivatives towards Mycobacterium leprae and Mycobacterium vaccae 12 may arise from alterations of the fluidity of bacterial membranes (“hyper-fluidification”) that could disrupt bacterial membranes or impair their functions or those of membrane proteins.

16


Page 16 of 37

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Conclusion In summary, the simulations show that ω-alicyclic fatty acids change structural and dynamical properties associated with the fluidity of lipid bilayers. However, the effect on fluidity of terminal saturated and unsaturated rings in the acyl tails was different. ω-Alicyclic fatty acyl chains containing a saturated ring gave rise to a generally moderate increase in structural aspects of fluidity through looser lipid packing (greater area per lipid) and decreased order of the lipid tails, but their dynamical effect involved a decrease in lateral lipid diffusion, in particular for ω-cyclohexyl and ω-cycloheptyl acyl tails. In contrast, ω-alicyclic acyl chains in which the ring was unsaturated enhanced the fluidity of lipid bilayers both structurally and dynamically, indicating that terminal saturated and unsaturated rings have markedly different effects on membranes. In the case of acyl chains containing a terminal saturated ring, the simulations suggest that they might thus fulfil a dual function in the membranes of thermo-acidophilic bacteria: on the one hand, contributing to the formation of a “dynamical barrier” that would impede the permeation by ions, in particular protons at low pH, and small molecules across membranes, while, at the same time, stabilising membranes by restricting the lateral mobility of lipids despite the high temperature. Work aimed at verifying the effect of ω-alicyclic acyl chains on membrane permeability by simulating systems that include osmolytes (ions and small molecules) is currently being undertaken.

Acknowledgement This work was funded from the National Health and Medical Research Council (NHMRC project grant APP1044327) with the assistance of high-performance computing resources provided by iVEC through the National Computational Merit Allocation Scheme supported by the Australian Government (projects m39 and m72).

17


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 37

References (1) Kaneda, T. Iso- and anteiso-fatty acids in bacteria: biosynthesis, function, and taxonomic significance. Microbiol. Rev. 1991, 55, 288–302 (2) Poger, D.; Caron, B.; Mark, A. E. Effect of methyl-branched fatty acids on the structure of lipid bilayers. J. Phys. Chem. B 2014, 118, 13838–13848 (3) Poger, D.; Mark, A. E. A ring to rule them all: the effect of cyclopropane fatty acids on the fluidity of lipid bilayers. J. Phys. Chem. B 2015, 119, 5487–5495 (4) Oshima, M.; Ariga, T. ω-Cyclohexyl fatty acids in acidophilic thermophilic bacteria. Studies on their presence, structure, and biosynthesis using precursors labeled with stable isotopes and radioisotopes. J. Biol. Chem. 1975, 250, 6963–6968 (5) Suzuki, K.-I.; Saito, K.; Kawaguchi, A.; Okuda, S.; Komagata, K. Occurrence of ωcyclohexyl fatty acids in Curtobacterium pusillum strains. J. Gen. Appl. Microbiol. 1981, 27, 261–266 (6) Poralla, K.; König, W. A. The occurrence of ω-cycloheptane fatty acids in a thermoacidophilic bacillus. FEMS Microbiol. Lett. 1983, 16, 303–306 (7) Tsaplina, I. A.; Osipov, G. A.; Bogdanova, T. I.; Nedorezova, T. P.; Karavaiko, G. I. Fatty acid composition of lipids in thermoacidophilic bacteria of the genus Sulfobacillus. Microbiology 1994, 63, 821–830 (8) Kaneda, T. Stereoselective synthesis of chaulmoogric acid and related fatty acid from 2(±)-cyclopentenecarboxylic acid by Bacillus subtilis (ATCC 7059). Biochem. Biophys. Res. Commun. 1981, 99, 1226–1229 ˇ (9) Rezanka, T.; Siristova, L.; Melzoch, K.; Sigler, K. Identification of (S )-11-cycloheptyl4-methylundecanoic acid in acylphosphatidylglycerol from Alicyclobacillus acidoterrestris. Chem. Phys. Lipids 2009, 159, 104–113 18


Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(10) Misra, T. N.; Singh, R. S.; Pandey, H. S.; Prasad, C.; Singh, B. P. Two long chain compounds from Achyranthes aspera. Phytochemistry 1993, 33, 221–223 (11) Sinninghe Damsté, J. S.; Strous, M.; Rijpstra, W. I. C.; Hopmans, E. C.; Geenevasen, J. A. J.; van Duin, A. C. T.; van Niftrik, L. A.; Jetten, M. S. M. Linearly concatenated cyclobutane lipids form a dense bacterial membrane. Nature 2002, 419, 708–712 (12) Cabot, M. C.; Goucher, C. R. Chaulmoogric acid: assimilation into the complex lipids of mycobacteria. Lipids 1981, 16, 146–148 (13) Meija, J.; Soukup, V. G. Phenyl-terminated fatty acids in seeds of various aroids. Phytochemistry 2004, 65, 2229–2237 (14) De Rosa, M.; Gambacorta, A.; Minale, L.; Bu’Lock, J. D. Cyclohexane fatty acids from a thermophilic bacterium. J. Chem. Soc. D, Chem. Commun. 1971, 1334 (15) Wisotzkey, J. D.; Jurtshuk, Jr., P.; Fox, G. E.; Deinhard, G.; Poralla, K. Comparative sequence analyses on the 16S rRNA (rDNA) of Bacillus acidocaldarius, Bacillus acidoterrestris, and Bacillus cycloheptanicus and proposal for creation of a new genus, Alicyclobacillus gen. nov. Int. J. Syst. Bacteriol. 1992, 42, 263–269 (16) Kannenberg, E.; Blume, A.; Poralla, K. Properties of ω-cyclohexane fatty acids in membranes. FEBS Lett. 1984, 172, 331–334 (17) Sunamoto, J.; Iwamoto, K.; Inoue, K.; Endo, T.; Nojima, S. Liposomal membranes XI. A suggestion to structural characteristics of acido-thermophilic bacterial membranes. Biochim. Biophys. Acta 1982, 685, 283–288 (18) Lewis, R. N. A. H.; McElhaney, R. N. Thermotropic phase behavior of model membranes composed of phosphatidylcholines containing ω-cyclohexyl fatty acids. Differential scanning calorimetric and

31

P NMR spectroscopic studies. Biochemistry 1985, 24,

4903–4911 19


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 37

(19) Mantsch, H. H.; Madec, C.; Lewis, R. N. A. H.; McElhaney, R. N. An infrared spectroscopic study of the thermotropic phase behavior of phosphatidylcholines containing ω-cyclohexyl fatty acyl chains. Biochim. Biophys. Acta 1989, 980, 42–49 (20) Krischke, W.; Poralla, K. Properties of Bacillus acidocaldarius mutants deficient in ω-cyclohexyl fatty acid biosynthesis. Arch. Microbiol. 1990, 153, 463–469 (21) Asgharian, B.; Rice, D. K.; Cadenhead, D. A.; Lewis, R. N. A. H.; McElhaney, R. N. Monomolecular film behavior of a homologous series of 1,2-bis(ωcyclohexylacyl)phosphatidylcholines at the air/water interface. Langmuir 1989, 5, 30– 34 (22) Endo, T.; Inoue, K.; Nojima, S.; Terashima, S.; Oshima, T. Synthesis and physical properties of phosphatidylcholines containing ω-cyclohexyl fatty acyl chains. Chem. Phys. Lipids 1982, 31, 61–74 (23) Benz, R.; Hallmann, D.; Poralla, K.; Eibl, H. Interaction of hopanoids with phosphatidylcholines containing oleic and ω-cyclohexyldodecanoic acid in lipid bilayer membranes. Chem. Phys. Lipids 1983, 34, 7–24 (24) Jansen, M.; Blume, A. A comparative study of diffusive and osmotic water permeation across bilayers composed of phospholipids with different head groups and fatty acyl chains. Biophys. J. 1995, 68, 997–1008 (25) Blume, A.; Dreher, R.; Poralla, K. The influence of branched-chain and ω-alicyclic fatty acids on the transition temperature of Bacillus subtilis lipids. Biochim. Biophys. Acta 1978, 512, 489–494 (26) Siristova, L.; Luhovy, R.; Sigler, K.; Rezanka, T. Biosynthesis of ω-alicyclic fatty acids induced by cyclic precursors and change of membrane fluidity in thermophilic bacteria Geobacillus stearothermophilus and Meiothermus ruber. Extremophiles 2011, 15, 423– 429 20


Page 21 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(27) Nagle, J. F.; Tristram-Nagle, S. Structure of lipid bilayers. Biochim. Biophys. Acta 2000, 1469, 159–195 (28) Petrache, H. I.; Dodd, S. W.; Brown, M. F. Area per lipid and acyl length distributions in fluid phosphatidylcholines determined by 2 H-NMR spectroscopy. Biophys. J. 2000, 79, 3172–3192 (29) Poger, D.; van Gunsteren, W. F.; Mark, A. E. A new force field for simulating phosphatidylcholine bilayers. J. Comput. Chem. 2010, 30, 1117–1125 (30) Hofsäß, C.; Lindahl, E.; Edholm, O. Molecular dynamics simulations of phospholipid bilayers with cholesterol. Biophys. J. 2003, 84, 2192–2206 (31) van der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. Gromacs: fast, flexible, and free. J. Comput. Chem. 2005, 26, 1701–1718 (32) Poger, D.; Mark, A. E. On the validation of molecular dynamics simulations of saturated and cis-monounsaturated phosphatidylcholine lipid bilayers: a comparison with experiment. J. Chem. Theory Comput. 2010, 6, 325–336 (33) Poger, D.; Mark, A. E. Lipid bilayers: the effect of force field on ordering and dynamics. J. Chem. Theory Comput. 2012, 8, 4807–4817 (34) Malde, A. K.; Zuo, L.; Breeze, M.; Stroet, M.; Poger, D.; Nair, P. C.; Oostenbrink, C.; Mark, A. E. An Automated force Field Topology Builder (ATB) and repository: version 1.0. J. Chem. Theory Comput. 2011, 7, 4026–4037 (35) 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 (36) Mabrey, S.; Sturtevant, J. M. Investigation of phase transitions of lipids and lipid

21


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 37

mixtures by high sensitivity differential scanning calorimetry. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 3862–3866 (37) Huang, C. H.; Lapides, J. R.; Levin, I. W. Phase-transition behavior of saturated, symmetric chain phospholipid bilayer dispersions determined by Raman spectroscopy: correlation between spectral and thermodynamic parameters. J. Am. Chem. Soc. 1982, 104, 5926–5930 (38) Lewis, R. N. A. H.; Mak, N.; McElhaney, R. N. A differential scanning calorimetric study of the thermotropic phase behavior of model membranes composed of phosphatidylcholines containing linear saturated fatty acyl chains. Biochemistry 1987, 26, 6118–6126 (39) Koynova, R.; Caffrey, M. Phases and phase transitions of the phosphatidylcholines. Biochim. Biophys. Acta 1998, 1376, 91–145 (40) Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. Lincs: A linear constraint solver for molecular simulations. J. Comput. Chem. 1997, 18, 1463–1472 (41) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; Hermans, J. In Intermolecular Forces; Pullman, B., Ed.; Reidel: Dordrecht, The Netherlands, 1981; pp 331–342 (42) Miyamoto, S.; Kollman, P. A. Settle: An analytical version of the Shake and Rattle algorithm for rigid water models. J. Comput. Chem. 1992, 13, 952–962 (43) Tironi, I. G.; Sperb, R.; Smith, P. E.; van Gunsteren, W. F. A generalized reaction field method for molecular dynamics simulations. J. Chem. Phys. 1995, 102, 5451–5459 (44) Heinz, T. N.; van Gunsteren, W. F.; H¨ unenberger, P. H. Comparison of four methods to compute the dielectric permittivity of liquids from molecular dynamics simulations. J. Chem. Phys. 2001, 115, 1125–1136 22


Page 23 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(45) Rawicz, W.; Olbrich, K. C.; McIntosh, T.; Needham, D.; Evans, E. Effect of chain length and unsaturation on elasticity of lipid bilayers. Biophys. J. 2000, 79, 328–339 (46) Kuˇcerka, N.; Nieh, M.-P.; Katsaras, J. Fluid phase lipid areas and bilayer thicknesses of commonly used phosphatidylcholines as a function of temperature. Biochim. Biophys. Acta 2011, 1808, 2761–2771 (47) Lentz, B. R. Use of fluorescent probes to monitor molecular order and motions within liposome bilayers. Chem. Phys. Lipids 1993, 64, 99–116 (48) Hurjui, I.; Neamtu, A.; Dorohoi, D. O. The interaction of fluorescent DPH probes with unsaturated phospholipid membranes: a molecular dynamics study. J. Mol. Struct. 2013, 1044, 134–139 (49) Mendelsohn, R.; Senak, L. In Biomolecular Spectroscopy; Clark, R. J. H., Hester, R. E., Eds.; Wiley, New York, 1993; pp 339–380 (50) Moss, G. P. Basic terminology of stereochemistry (IUPAC Recommendations 1996). Pure Appl. Chem. 1996, 68, 2193–2222 (51) Snyder, R. G.; Tu, K.; Klein, M. L.; Mendelssohn, R.; Strauss, H. L.; Sun, W. Acyl chain conformation and packing in dipalmitoylphosphatidylcholine bilayers from MD simulation and IR spectroscopy. J. Phys. Chem. B 2002, 106, 6273–6288 (52) Douliez, J.-P.; Léonard, A.; Dufourc, E. J. Restatement of order parameters in biomembranes: calculation of C–C bond order parameters from C–D quadrupolar splittings. Biophys. J. 1995, 68, 1727–1739 (53) Cohen, M. H.; Turnbull, D. Molecular transport in liquids and glasses. J. Chem. Phys. 1959, 31, 1164–1169 (54) Macedo, P. B.; Litovitz, T. A. On the relative roles of free volume and activation energy in the viscosity of liquids. J. Chem. Phys. 1965, 42, 245–256 23


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(55) Nicolau, Jr., D. V.; Hancock, J. F.; Burrage, K. Sources of anomalous diffusion on cell membranes: a monte carlo study. Biophys. J. 2007, 92, 1975–1987 (56) Metzler, R.; Klafter, J. The random walk’s guide to anomalous diffusion: a fractional dynamics approach. Phys. Rep. 2000, 339, 1–77 (57) Jeon, J.-H.; Martinez-Seara Monne, H.; Javanainen, M.; Metzler, R. Anomalous diffusion of phospholipids and cholesterols in a lipid bilayer and its origins. Phys. Rev. Lett. 2012, 109, 188103 (58) Schwille, P.; Korlach, J.; Webb, W. W. Fluorescence correlation spectroscopy with single-molecule sensitivity on cell and model membranes. Cytometry 1999, 36, 176– 182 (59) Chiantia, S.; Ries, J.; Schwille, P. Fluorescence correlation spectroscopy in membrane structure elucidation. Biochim. Biophys. Acta 2009, 1788, 225–233

24


Page 24 of 37

Page 25 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Tables Table 1: Overview of the Bilayer Systems Simulated lipida DPPCd UPPC c3 UPPC c4 UPPC c5 UPPC HPPC c6 UPPC pUPPC c7 UPPC a b

c

d

nature of the sn-1 tailb nature of the sn-2 tail 16:0 16:0 11:0 16:0 11:1cpr(11) 16:0 11:1cbu(11) 16:0 11:1cpe(11) 16:0 11:1∆cpe(11) 16:0 11:1chx(11) 16:0 11:1ph(11) 16:0 11:1chp(11) 16:0

total time (ns)c 400 450 350 350 350 350 400 350 400

The names of the lipids are given in Figure 2. cbu, cyclobutyl; chp; cycloheptyl; chx, cyclohexyl; cpe, cyclopentyl; ∆cpe, (1R)-cyclopent-2-en-1-yl; cpr, cyclopropyl; ph, phenyl. Each system was simulated three times and analysed over the last 200 ns of each individual simulation. From reference 2.

25


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 37

Figure Captions Figure 1 Structure of an ω-alicyclic fatty acid in which the terminal ring is a cyclohexyl group. Figure 2 Structures of the lipids investigated. DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine; c3 UPPC, 1-(11-cyclopropylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine; c4 UPPC, 1-(11-cyclobutylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine; c5 UPPC, 1-(11-cyclopentylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine; c6 UPPC, 1-(11-cyclohexylundecanoyl)-2-palmitoyl-sn-glycero-3-phosphocholine; c7 UPPC, 1-(11-cycloheptylundecanoyl)-2palmitoyl-sn-glycero-3-phosphocholine; HPPC, 1-[(R)-hydnocarpoyl]-2-palmitoyl-sn-glycero3-phosphocholine ((R)-hydnocarpoyl: 11-[(1R)-cyclopent-2-en-1-yl]undecanoyl); pUPPC, 2palmitoyl-1-(11-phenylundecanoyl)-sn-glycero-3-phosphocholine; UPPC, 2-palmitoyl-1-undecanoyl-sn-glycero-3-phosphocholine. Figure 3 Average (A) area per lipid AL and (B) bilayer thickness DHH calculated from the simulations as a function of the size of the ω ring in the sn-1 chain. cn UPPC (n = 3–7) corresponds to the lipids c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC that contain a saturated ω-alicyclic ring. n = 0 corresponds to UPPC. The dotted lines in the panels indicate the average AL and DHH calculated for a DPPC bilayer. 2,3 Figure 4 Carbon–hydrogen order parameter |SCH | profiles of the sn-1 (

and

) and sn-2 ( ) fatty

acyl chains of hydrated UPPC, c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC, c7 UPPC, HPPC and pUPPC bilayers calculated from the simulations. (

26

) |SCH | in the non-cyclic part of the


Page 27 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

sn-1 chain (carbons 2–11); ( ) |SCH | in the ring. The values of |SCH | in the ring are averaged over equivalent CH or CH2 groups within the ring (that is at positions 13 and 13′ , 14 and 14′ and 15 and 15′ where relevant). In HPPC, the |SCH | values corresponding to the CH groups at positions 13′ and 14′ are indicated by orange asterisks. The position of the first atom in the ring (carbon 12) is shown by a gray dashed line. The |SCH | profiles of the sn-1 ( sn-2 (

) and

) palmitoyl chains calculated from the simulations of a hydrated DPPC bilayer 2 are

plotted in each panel as a reference. The |SCH | values are averaged over all the sn-1 and sn-2 acyl chains in the bilayers and over the three simulations. Error bars are mostly within the size of the symbols. Figure 5 Average carbon–hydrogen order parameter ⟨|SCH |⟩ calculated over the two fatty acyl chains from the simulations as a function of the size of the ω ring in the sn-1 chain. cn UPPC (n = 3–7) corresponds to the lipids c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC that contain a saturated ω-alicyclic ring. n = 0 corresponds to UPPC. The dotted lines in the panels indicate the value of ⟨|SCH |⟩ calculated for a DPPC bilayer. 2 Error bars are mostly within the size of the symbols. Figure 6 Average fraction xg of gauche defects per bond in the fatty acyl chains of the bilayers (excluding the ring in the sn-1 chain, if any) as a function of the size of the ω ring in the sn1 chain. cn UPPC (n = 3–7) corresponds to the lipids c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC that contain a saturated ω-alicyclic ring. n = 0 corresponds to UPPC. The dotted lines in the panels indicate the average xg calculated for a DPPC bilayer. 2 Error bars are mostly within the size of the symbols.

27


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7 Time evolution of the mean squared displacement (MSD) of the centre of mass of the phosphate group of the lipids as a function of the averaging time in the simulations of hydrated hydrated UPPC, c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC, c7 UPPC, HPPC and pUPPC. The gray dashed lines at 20 ns and 120 ns indicate the time interval over which the lateral diffusion coefficient was computed. Figure 8 Lateral diffusion coefficient DL of the lipids calculated from the simulations of hydrated UPPC, c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC, c7 UPPC, HPPC and pUPPC bilayers as a function of the size of the ω ring in the sn-1 chain. cn UPPC (n = 3–7) corresponds to the lipids c3 UPPC, c4 UPPC, c5 UPPC, c6 UPPC and c7 UPPC that contain a saturated ω-alicyclic ring. n = 0 corresponds to UPPC. The dotted line indicates the average DL calculated for a DPPC bilayer. 3

28


Page 28 of 37

Page 29 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

O HO

Figure 1

29


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

N

Page 30 of 37

R 1= O O P O O

H UPPC 13 13

12

O

13’

c3UPPC

O

13

O

12

DPPC 13’

c4UPPC 13 14

14

12

14’

15 13’ 14’

c6UPPC

c5UPPC O O P O O

14

13’

O

N

12

13 14 15

12

O

15’ 13’ 14’

O

c7UPPC

R1

O

13

11

O

12 13’

HPPC

Figure 2

30


13 14 14 14’

12

15 13’ 14’

pUPPC

Page 31 of 37

0.70

A

cnUPPC HPPC pUPPC

AL (nm2)

0.68 0.66 0.64 0.62 0.60

0

3

4 5 ring size

6

7

B 3.6 DHH (nm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3.4 cnUPPC HPPC pUPPC

3.2 3.0

0

3

4 5 ring size

6

7

Figure 3

31


Langmuir

0.3 UPPC

c3UPPC

|SCH|

0.2 0.1 sn-1 sn-2

sn-1 sn-2

0 c5UPPC

c4UPPC

|SCH|

0.2 0.1 sn-1 sn-2

sn-1 sn-2

0 c7UPPC

c6UPPC

|SCH|

0.2 0.1 sn-1 sn-2

sn-1 sn-2

0 pUPPC

HPPC 0.2

|SCH|

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 37

0.1 0

sn-1 sn-2 2

4

sn-1 sn-2

* *

6 8 10 12 14 Carbon position

2

4

6 8 10 12 14 Carbon position

Figure 4

32


Page 33 of 37

0.20 0.18 〈|S CH|〉

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

0.16 0.14

cnUPPC HPPC pUPPC

0.12 0.10

0

3

4 5 ring size

6

7

Figure 5

33


Langmuir

0.26 0.25 xg

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 37

0.24 0.23

cnUPPC 0

3

4 5 ring size

HPPC pUPPC 6

7

Figure 6

34


Page 35 of 37

1.2 UPPC c3UPPC c4UPPC c5UPPC

MSD (nm2)

1.0 0.8 0.6 0.4 0.2 0

c6UPPC c7UPPC HPPC pUPPC

1.0 MSD (nm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

0.8 0.6 0.4 0.2 0

0

20 40 60 80 100 120 140 time (ns)

Figure 7

35


Langmuir

5.0 DL (µm2·s–1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 37

cnUPPC

HPPC pUPPC

4 5 ring size

6

4.0 3.0 2.0 1.0

0

3

7

Figure 8

36


Page 37 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Graphical TOC Entry

LINEAR-CHAIN LIPID BILAYER

ω-ALICYCLIC LIPID BILAYER

VS

37


Effect of Ring Size in Ï-Alicyclic Fatty Acids on the Structural and

Recommend Documents

Effect of Ring Size in Ï-Alicyclic Fatty Acids on the Structural and