Subscriber access provided by University of Newcastle, Australia
Article
Reconciling differences between lipid transfer in free-standing and solid supported membranes: a time resolved small angle neutron scattering study Benny Wah, Jeffrey Michael Breidigan, Joseph Adams, Piotr Horbal, Sumit Garg, Lionel Porcar, and Ursula Perez-Salas Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b04013 • Publication Date (Web): 16 Mar 2017 Downloaded from http://pubs.acs.org on March 18, 2017
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 38
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
Reconciling differences between lipid transfer in free-standing and solid supported membranes: a time resolved small angle neutron scattering study. Benny Wah, a‡ Jeffrey M. Breidigan,a‡ Joseph Adams,a Piotr Horbal,a Sumit Garg, †d,a Lionel Porcar,b,c and Ursula Perez-Salas*a,d a
b
c
Physics Department, University of Illinois at Chicago, Chicago, IL 60607, USA. Large Scale Structure Group, Institut Laue-Langevin, Grenoble F-38042, France
Department of Chemical Engineering, Colburn Laboratory, University of Delaware, Newark, Delaware d
Materials Science Division, Argonne National Laboratory, Lemont, Illinois
KEYWORDS: Supported Lipid Bilayers, time resolved SANS, Calorimetry, Solvent-Exchange Deposition, SiO2 Nanoparticles, lipid vesicles, spontaneous lipid transport, lipid flip-flop, lipid exchange
ACS Paragon Plus Environment
1
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 2 of 38
ABSTRACT: Maintaining compositional lipid gradients across membranes in animal cells is essential to biological function, but what is the energetic cost to maintain these differences? It has long been recognized that studying the passive movement of lipids in membranes can provide insight into this toll. Unfortunately the reported values of inter, and particularly, intralipid transport rates of lipids in membranes show significant differences. To overcome this difficulty, biases introduced by experimental approaches have to be identified. The present study addresses
the
difference
in
the
reported
intra-membrane
transport
rates
of
dimyristoylphosphatidylcholine (DMPC) on flat solid supports (fast flipping) and in curved freestanding membranes (slow flipping). Two possible scenarios are potentially at play; one is the difference in curvature of the membranes studied and the other the presence (or not) of the support. Using DMPC vesicles and DMPC supported membranes on silica nanoparticles of different radii we found that an increase in curvature (from a diameter of 30nm to a diameter of 100nm) does not change the rates significantly, differing only by factors of order ~1. Additionally, we found that the exchange rates of DMPC in supported membranes are similar to the ones in vesicles. And as previously reported, we found that the activation energy for exchange on free-standing and supported membranes is similar (84 kJ/mol and 78 kJ/mol respectively). However, DMPC’s flip-flop rates increase significantly when in a supported membrane, surpassing the exchange rates and no longer limiting the exchange process. Although the presence of holes or cracks in supported membranes explains the occurrence of fast lipid flipflop in many studies, in defect-free supported membranes we find that fast flip-flop is driven by the surface’s induced disorder of the bilayer’s acyl chain packing as evidenced from their broad melting temperature behavior.
ACS Paragon Plus Environment
2
Page 3 of 38
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: Lipids are essential components of cellular membranes1. The membranes of eukaryotic cells bound compartments with specialized functions requiring unique protein and lipid compositions2. The quest to understand lipid trafficking in and between membranes as it relates to lipid homeostasis and metabolism, and how this lipid organization leads to proper membrane function, has led to numerous studies over the past four decades3. Evidence now suggests that vesicular transport plays a major part in protein and lipid transport along several energy-dependent metabolic pathways4. However, non-vesicular transport mechanisms also play critical roles in lipid homeostasis as demonstrated by the existence of lipid transport, even under conditions in which vesicular transport is blocked5. In this situation, the study of the passive movement of lipids between and within membranes can provide insight into the energetic toll to move lipids between and within membranes. Indeed, transbilayer flip-flop energetics can influence inter-organelle lipid transport by rearranging lipids from inner to outer leaflets or vice versa; this directly affects membrane curvature for example, and consequently vesicle budding, vesicle fission and vesicle fusion6. Thus, there is an interplay between vesicular and nonvesicular lipid transport that is crucial for the establishment and maintenance of intracellular lipid distribution. Although the study of the passive movement of lipids between and within membranes can provide insight into lipid transport regulation since it provides a way to gauge the energetic cost on active metabolic pathways, published work on the spontaneous transfer of lipids report a wide variation in the rates of transfer between and within membranes. Even in studies of model membrane systems, in which lipid composition is controlled, the transfer rates reported have been inconsistent. For example, the reported half-life for cholesterol’s transmembrane flipping varies by five to six orders of magnitude, ranging from several hours7 to a few minutes or
ACS Paragon Plus Environment
3
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 4 of 38
seconds 8, and to even tens of nanoseconds to a few milliseconds 9. Non-invasive approaches like time resolved small angle neutron scattering (TR-SANS) or sum -frequency vibrational spectroscopy (SFVP) have shown that the movement of lipids is extremely sensitive to chemical structure finding that the transfer rates of unaltered lipid molecules are dramatically different from their labeled counterparts10. However, even when studying a single chemical structure, issues persist. Studies using the same lipids and apparently similar time resolved non-invasive approaches to investigate the movement of lipids across the bilayer (flip-flop) find drastically different results: in single flat supported membranes flip-flop of lipids is found to be fast (less than a minute) 11 while in vesicles it has been reported to take hours12. The source of these apparently contradicting results could be the presence of the supporting surface or differences in curvature. In order to determine the most likely source of the inconsistency we compared the movement of DMPC in free-standing membranes (vesicles) and supported membranes on silica nanoparticles using TR-SANS. Differences in lipid transport between free vesicles and lipid-coated nanoparticles is also of particular interest for nanotechnology and nano-medicine applications13. Indeed, the various nanoparticle constructs hope to improve targeted (cellular or tissue) delivery of drugs, improve bioavailability, sustain release of drugs or solubilize drugs for systemic delivery. Nanoparticles, because of their small size, can move through tissue in inflammatory sites, allowing for efficient uptake by a variety of cell types and be selectively accumulated at target sites. Yet each construct may hold advantages over others. For example, mesoporous silica nanoparticles can host highly hydrophobic molecules such as many anticancer drugs and a lipid coating increase its bioavailability by protecting it from monocytes and macrophages which readily absorb
ACS Paragon Plus Environment
4
Page 5 of 38
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
circulating nano-materials13c. Hence studying how these constructs are alike or how they differ can only improve future designs. MATERIALS AND METHODS: Materials. 1,2-Dimyristoyl-sn-glycero-3-phosphocholine
(hDMPC),
1,2-dimyristoyl-sn-glycero-3-
phospho-(1'-rac-glycerol) (sodium salt) (DMPG), and 1,2-dimyristoyl-d54-sn-glycero-3phosphocholine (dDMPC) were obtained from Avanti Polar Lipids (Alabaster, AL) and used without further purification. LUDOX® AS-40 colloidal silica (SiO2) solution of nanoparticles with a diameter of 30 nm and at a 40.8 weight% SiO2 concentration, lot 200703, pH 10.1, specific gravity 1.295; were obtained from Sigma Aldrich and used as received. MP-1040 colloidal SiO2 solution with nanoparticles with a diameter of 100 nm and at a 40.6 weight% SiO2 concentration, lot 240724, pH 9.2, specific gravity 1.297; were obtained from Nissan Chemical and used as received. The SiO2 beads were prepared by the water glass process and had densities of 2.2-2.6 g/cm3 (reported by the manufacturer). HPLC grade ethanol was purchased from Fisher Chemicals. Deionized water was further purified with a Millipore Simplicity UV purifier and was used for all solutions. D2O and deuterated (D6) ethanol (d-ethanol) was obtained from Cambridge Isotope Laboratories, Inc. A mini-extruder from Avanti Polar Lipids was used with 1 mL Hamilton syringes for extrusion of unilamellar vesicles (SUV). Preparation of unilamellar lipid vesicles. The lipids, in powder form, were used as received. When making vesicles with DMPG, precise amounts of either hDMPC or dDMPC and DMPG were weighed out in a vial. Chloroform was
ACS Paragon Plus Environment
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 38
then added to the vials, and the vials were stirred sufficiently to dissolve all lipids. Dry lipid films were obtained by applying a constant stream of nitrogen to the liquid in the vials. The vials were then placed in a vacuum oven overnight at 60°C to assure the complete removal of chloroform. The dried d/h DMPC/DMPG mixtures were then re-dispersed in solvents made of appropriate ratios of D2O and H2O to achieve the correct neutron contrast match point for silica (see SANS Contrast Matching of Silica Nanoparticles section below). We prepared 100nm and 30nm unilamellar vesicles in diameter via extrusion. ‡ The size of the SUVs was determined by the mesh size of the polycarbonate membranes used during the extrusion process which were 30nm and 100nm. Lipid solutions were extruded in 1 mL syringes by passing through 100nm polycarbonate filters 51 times at 40°C, which is well above the melting temperature, Tm of DMPC. For 30nm vesicles an additional 51 passes were made using the appropriate filter. Supported Lipid Bilayer (SLB) Preparation. Assembly of single SLBs on 100nm and 30nm in diameter silica nanoparticles was done via a solvent exchange method14. An initial motivation to use the solvent exchange method was to be able to self assemble a lipid bilayer one leaflet at a time, and therefore potentially create asymmetric bilayers. However, so far, this procedure failed to do so. Figure 1 shows a water/ethanol ratio study of 30nm silica nanoparticles with lipid amounts consistent with the coating of the silica nanoparticles with one and two complete leaflets. A clear suspension was achieved in 30 % to 50% by volume ethanol aqueous solutions as long as there was enough lipid to form a full bilayer around the silica nanoparticles. The lipid required to form a full bilayer was calculated using a 0.6nm2 area per lipid15, the size of the silica particles obtained from their
ACS Paragon Plus Environment
6
Page 7 of 38
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
characterization using SANS (Figure 2), and the silica concentration used (typically a five to ten fold dilution from the stock solution provided by the manufacturer). To remove all ethanol from the suspension and obtain fully aqueous solutions, the samples were dialyzed at least twice, each at a volume ratio of 1:50 using Spectra/Por® Biotech Grade Pre-wetted Dialysis Tubing with a pore size of 100-500 Daltons (or g/mol) over a span of at least 4 hours. Initially it was found that during the dialysis process, pure DMPC lipid coated silica nanoparticles would precipitate out. As a result, it was necessary to add small amounts (6 mole %) of charged lipids (DMPG) to promote inter-particle repulsion and therefore stability to the colloidal suspensions. 6 mole% DMPG was the least amount of DMPG necessary for colloidal stability. Additional z-potential measurements were done on a temperature controlled Malvern Zetasizer Nano Z instrument which showed that the z-potential for pure silica solutions were equivalent to DMPC vesicles with 6 mole % DMPG as well as lipid coated silica nanoparticles with DMPG as shown in Table 1. SLBs in this study were made with dDMPC or hDMPC in addition to 6% DMPG.
ACS Paragon Plus Environment
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
Page 8 of 38
Figure 1. Visual study of colloidal stability by varying ethanol volume% in water/ethanol mixtures. (A) shows the phase aggregation state of DMPC following Hohner et al.14. (B) Shows the behavior of 30nm silica particles in varying water /ethanol mixtures. (C) Colloidal stability of 30nm silica after adding lipids to form a single lipid leaflet sheath. Colloidal stability was expected at high ethanol concentrations but none was found. (D) between 30% and 50% ethanol it was possible
to
form
stable
silica
coated
nanoparticles.
Zeta Potential (mV) DMPC vesicles with DMPG
-51.6 ± 13.6
Table 1. Z-potential values for vesicles, bare
DMPC vesicles without DMPG
-10.2 ± 8.2
silica particles and lipid coated silica particles
100nm silica nanoparticles
-49.1 ± 9.9
30nm silica nanoparticles
-41.0 ± 17.4
in water. The addition of 6 mole % of DMPG stabilizes the lipid coated silica particles, coinciding with the recovery of the potential
100nm silica w/ dDMPC +DMPG
-47.2 ± 7.2
100nm silica w/ hDMPC+DMPG
-50.6 ± 6.6
30nm silica w/ dDMPC+DMPG
-42.8 ± 5.1
measured on bare silica.
ACS Paragon Plus Environment
8
Page 9 of 38
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
Small Angle Neutron Scattering (SANS) SANS is a powerful technique to obtain structure information of particles that are a few tens to hundreds of nanometers in size because the scattered intensity, , is directly related to their shape, size and composition as follows:
= c −
(1)
Indeed, a critical part in obtaining this information is to have a scattering length density (SLD) difference between the particles and the solvent. The SLD is directly obtained from the chemical and isotopic make-up of the components involved in the scattering. Particularly isotopic differences, like those between H2O and D2O, can be of great value. Indeed, the SLD of an aqueous solvent can be precisely tuned with varying ratios of H2O and D2O without altering its chemical identity, which is water. Similarly, lipid chemical identity is preserved with hydrogen to deuteron substitutions, yet their SLD is drastically different; for example the SLD of hDMPC is 0.3 x 10-6 Å-2 while for dDMPC (with 54 deuteron substitutions) it is 5.4 x 10-6 Å-2 12. And as with H2O and D2O, mixtures of “d” and “h” lipids will produce varying membrane SLDs according to the volume fraction of each lipid type. In contrast to deuteron substitutions, the chemical difference between DMPG and hDMPC is insignificant, both giving essentially identical SLD values. P(Q), the form factor of the particles, contains all the information of the shape and size of the particles and the constant c is related to the concentration of particles. P(Q), depends on the magnitude of the scattering vector, Q, which is related to the scattering angle, θ, by Q=4π sin(θ/2)/λ, where λ is the neutron wavelength.
ACS Paragon Plus Environment
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
Page 10 of 38
The SANS data presented in this study was collected on the NG3 30 m SANS instruments at the National Institute of Standard and Technology Center for Neutron Research (NIST-CNR), Gaithersburg, MD and on D22 at the Institut Laue Langevin Grenoble. Scattering intensity measurements were taken over a broad Q-range: 0.003
