Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Article
An X-ray and Neutron Reflectivity Study Shows CLIC1 undergoes Cholesterol Dependant Structural Re-Organisation in Lipid Monolayers Khondker Rufaka Hossain, Stephen Andrew Holt, Anton P. Le Brun, Heba Al Khamici, and Stella M. Valenzuela Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02872 • Publication Date (Web): 10 Oct 2017 Downloaded from http://pubs.acs.org on October 12, 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 21
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
An X-ray and Neutron Reflectivity Study Shows CLIC1 undergoes Cholesterol Dependant Structural Re-Organisation in Lipid Monolayers. Khondker R. Hossain 1, 3, Stephen A. Holt 3, Anton P. Le Brun 3, Heba Al Khamici 1, 2, a and Stella M. Valenzuela 1, 2* 1 School of Life Sciences, University of Technology Sydney, Sydney, New South Wales 2007, Australia; EMails:
[email protected] (K.R.H);
[email protected] (H.A)
2 Centre for Health Technologies, University of Technology Sydney, Sydney, New South Wales 2007, Australia;
3 Australian Centre for Neutron Scattering, Australian Nuclear Science and Technology Organisation (ANSTO), Lucas Heights, New South Wales 2234, Australia; E-Mails:
[email protected] (S.A.H);
[email protected] (A.L.B)
*Author to whom correspondence should be addressed; E-Mail:
[email protected] (SMV); Tel.: +612-95141917; Fax: +61-2-9514 8206.
Abstract CLIC1 belongs to the ubiquitous family of Chloride Intracellular Ion Channel proteins that are evolutionarily conserved across species. The CLICs are unusual in that they exist mainly as soluble proteins but possess the intriguing property of spontaneous conversion from the soluble to an integral membrane-bound form. This conversion is regulated by the membrane lipid composition especially cholesterol together with external factors such as oxidation and pH. However, the precise physiological mechanism regulating CLIC1 membrane insertion is currently unknown. In this study, X-ray and neutron reflectivity experiments were performed to study the interaction of CLIC1 with different phospholipid monolayers prepared using POPC, POPE or POPS with and without cholesterol in order to better understand the regulatory role of cholesterol on CLIC1 membrane insertion. Our findings demonstrate for the first time two different structural orientations of CLIC1 within phospholipid monolayers, dependent upon the absence or presence of cholesterol. In phospholipid monolayers devoid of cholesterol, CLIC1 was unable to insert into the lipid acyl chain region. However, in the presence of cholesterol, CLIC1 showed significant insertion within the phospholipid acyl chains occupying an area per protein molecule between 6 ~ 7 nm2 with a total CLIC1 thickness ranging from ~ 50 Å to 56 Å across the entire monolayer. Our data strongly suggests that cholesterol not only facilitates the initial docking or binding of CLIC1 to the membrane but also promotes deeper penetration of CLIC1 into the hydrophobic tails of the lipid monolayer. Key words: CLIC1 protein; cholesterol; phospholipids: POPC; POPE; POPS; X-ray reflectivity; neutron reflectivity; membrane insertion.
a Present Address: Laboratory of Cancer Biology and Genetics, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, 20817, United States of America
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
Page 2 of 21
Introduction The Chloride Intracellular Ion Channel (CLIC) protein family is a unique class of eukaryotic chloride ion channels that are highly conserved across species, with vertebrates expressing six different CLIC proteins (CLIC1-CLIC6) 1-4. They exist in two forms, namely, a cytosolic soluble form with a glutaredoxin-fold and a membrane-bound form with an undefined topology 5-10. In addition, CLIC1 has been shown to adopt more than one stable conformation in its soluble form and is thus categorised as a metamorphic protein 11-12. The crystal structure of the soluble CLIC1 protein resolved by X-ray crystallography at 1.4 Å resolution, shown in Figure 1A, demonstrates that CLIC1 is monomeric and adopts a fold similar to the GST-superfamily. CLIC1 consists of two distinct domains, the N-terminal domain which contains a single putative transmembrane domain (PTMD) within the thioredoxin-fold and a completely α-helical C-terminal domain (see Figure 1A) 6. It is speculated that the reversible spontaneous conversion of CLIC1 from the soluble to the membrane-bound form involves the structural rearrangement of its N-terminal domain (helix 1 and β-sheet 2 of PTMD) such that the exposed hydrophobic surfaces of the PTMD crosses the membrane as a single α-helix 8, 11, 13-15. It is then believed to undergo redox sensitive oligomerisation to form functional ion channels that are likely to consist of a tetrameric to hexameric assembly of subunits within a phospholipid bilayer 5, 8. Despite considerable progress in understanding CLIC1 membrane interactions, the mechanism and structural details of the CLIC1-membrane assembly are not well understood. However, membrane insertion is believed to be facilitated by a redox environment 8-9, 11, 13-14, low pH 16-18 and the membrane lipid composition 5-6, 8, 19-20. However, conflicting data were also published showing that the structural integrity of CLIC1 was unchanged under varying pH environments 17, 21 and that CLIC1 was able to form equally effective functional channels under reducing conditions 8-9. These studies therefore suggest the involvement of other factors that may be deemed necessary for the spontaneous membrane insertion of the CLIC proteins. We have previously proposed that specific lipid-protein interaction facilitates the spontaneous membrane insertion of CLIC1 and have demonstrated through Langmuir monolayer studies that cholesterol plays a crucial role in the autonomous membrane insertion of the protein 19-20, 22. Furthermore, we demonstrated that the optimal membrane composition for CLIC1 insertion is a combination of POPE, POPS and cholesterol in a mole ratio of 4:1:1 19, which is consistent with the observations recorded by Singh et al. 2006 8. In addition to the Langmuir study 19, our previous impedance spectroscopy studies on CLIC1 ion channel activity within tethered bilayer lipid membranes (tBLMs), confirmed the regulatory role of membrane cholesterol in the spontaneous membrane insertion and therefore the ion channel activity of CLIC1 within these artificial model membranes 20, 22. These studies also strongly support a mechanism involving CLIC1’s ability to bind to, and or form, a relatively stable CLIC1-cholesterol pre-complex when pre-incubated with free cholesterol or membrane bound cholesterol, and therefore speculate the presence of a cholesterol-binding domain within CLIC1 19-20, 22. Although cholesterol has been implicated to play a crucial role in CLIC1 membrane insertion, the mechanism by which it facilitates this insertion remains largely unresolved. Here-in, we provide insights into both the structural conformation of the membrane form of CLIC1 and a mechanism for CLIC1’s interaction with membrane cholesterol, studied for the first time using specular X-ray (XR) and neutron reflectometry (NR). Specular reflection allows the depth (thickness, t) of proteins to be determined in artificial monolayer membranes at the nanometre scale and also helps determine protein structural changes once in contact with the model membrane; thereby enabling us to probe the effect of cholesterol on the structural form adopted by CLIC1 and its localisation within different phospholipid monolayers.
2
ACS Paragon Plus Environment
Page 3 of 21
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
Earlier results have demonstrated that CLIC1 interactions with a phospholipid monolayer correlate closely to those observed with bilayer membranes 19-20, 22. Therefore, we predict that the structural features of CLIC1 within a lipid monolayer will resemble (perhaps not completely) the model structure previously proposed for CLIC1 within a bilayer 5, 8-9, 14. Based on these previous studies, we postulate a structural model as shown in Figure 1B, to schematically represent a number of interactions that may arise between CLIC1 and the phospholipid monolayer. The crystal structure of the soluble CLIC1 protein resolved by X-ray crystallography showed it was comprised of two distinct domains, with approximate overall dimensions of 50 Å x 50 Å x 20 Å as illustrated in Figure 1A 6. Studies have shown that the N-terminal domain of CLIC1, upon interaction with membranes, changes conformation such that the N and C-domains are located on opposite sides of the membrane, with the cysteine residue (Cys24) localised at the trans face of the membrane while the tryptophan (Trp35) residue is localised within the phospholipid bilayer 5, 8, 11. Subsequently, we have also presumed that the C-domain of CLIC1 (shown in green in Figure 1B) remains in the buffer subphase, while the Nterminal domain is available for interaction with the lipid monolayer. A recent study performed using FRET spectroscopy and EPR demonstrated that the hydrophobic transmembrane region (TMR) (helices ß1-a1-ß2) at the N-terminal domain of CLIC1 undergoes a structural reorganisation such that all the β-strands present in the reduced monomeric form become disordered and completely α-helical, exposing the TMR which then inserts into the artificial phospholipid bilayer as a single α-helix 5. Since, in this study a lipid monolayer is used instead of a bilayer, it is likely that the α-helices (shown in purple in Figure 1B), comprising the N-domain, are folded remaining buried within the hydrophobic phospholipid tail region of the monolayer, rather than extending out beyond the monolayer. Alternatively, one of the α-helices may be in close contact with the hydrophobic tails as shown in Figure 1B.
Figure 1. Schematic representation of CLIC1 protein. (A) Schematic ribbon diagram of the structure of reduced CLIC1. The N-and C-domains are shown in purple and green, respectively. The putative transmembrane region (PTMD) comprised of α1 helix and the β2 strand, located in the N-terminal domain, is shown in blue (Figure generated with DNA star software using PBD file 1K0m) 6. (B) Schematic cartoon representation of putative structural models of CLIC1 interacting with a phospholipid monolayer in the absence of cholesterol. The lipid monolayer is shown in orange and the N-terminal and C-terminal domains of CLIC1 are shown in purple and green respectively (Figure generated using Microsoft PowerPoint and not drawn to scale).
3
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
Page 4 of 21
Materials and Method Phospholipids: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylcholine (POPC), 1-palmitoyl-2-oleoylsn-glycero-3-phosphatidylserine (POPS), 1-palmitoyl-2-oleoyl-sn-glycero-3 phosphatidylethanolamine (POPE) and chain deuterated 1-palmitoyl-(d31)-2-oleoyl-sn-glycero-3phosphatidylcholine (d31-POPC) were purchased from Avanti Polar Lipids (Alabaster, USA) and used as received. Cholesterol (Chol) was purchased from Sigma Aldrich (Australia) and used as received. Lipid stock solutions were prepared in spectroscopic grade chloroform (Sigma) at a concentration of 1 mg/ml and stored at -20°C. Recombinant wild-type deuterated CLIC1 (d-CLIC1) protein was produced by the National Deuteration Facility at the Australian Nuclear Science and Technology Organization (ANSTO), Australia.
Sample preparation for X-ray and Neutron Reflectivity experiments The XR and NR profiles were recorded for phospholipid and phospholipid-cholesterol (5:1 mole ratio) monolayers prepared using a Nima Technologies 601 Langmuir trough (Nima Technologies, Coventry, UK) by methods previously described in Hossain.KR et.al 2016 19. Briefly, the phospholipid or phospholipid: cholesterol (5:1 mole ratio) monolayer was compressed to a surface pressure (π) of 20mN m-1 at a rate of 20cm2 min-1 and then held at that constant pressure to ensure the formation of a stable monolayer. Once XR and NR data of the phospholipid (± cholesterol) monolayer was collected, recombinant CLIC1-wt (in XR experiments) and deuterated or hydrogenated (d/h)CLIC1-wt (in NR experiments) (50µg) was injected underneath the preformed monolayer into the KCL/Hepes buffer (0.1 M KCl, 0.1mM Hepes and 0.01 mM CaCl2) subphase (pH 6.5). The XR and NR reflectivity profiles of CLIC1 interaction with the monolayer was recorded with the exception of holding the monolayer at a constant area (instead of constant pressure) in order to prevent structural changes in the monolayer while collecting the data. For NR experiments, the reflectivity data was recorded after a 20% increase in the initial monolayer area was observed following d/h-CLIC1 insertion. X-ray Reflectometry measurements at the Air-Water interface The Panalytical X’Pert Pro reflectometer (Cu-Ka X-rays, λ = 1.542 Å) at ANSTO (Lucas Heights, Australia) was used to perform XR measurements. Calibration of the reflectometer was performed on a KCl/Hepes buffer subphase (pH 6.5) in a Langmuir trough. The background scattering from the subphase was measured at every data point with the detector offset by 2θ = ± 0.7°. For each experiment, the lipid monolayer was aligned to the incident X-ray beam and measurements were performed by scanning from an angle of incidence of 0.04° degrees to 4.00° degrees in 0.01° degree steps for a total time period of 40 min. After XR data collection from the lipid monolayer, CLIC1-wt protein was injected underneath the same lipid monolayer and 3 scans were performed for each phospholipid (±cholesterol) monolayer thereby obtaining three XR profiles of lipid-CLIC1 interactions over a total period of three hours. However, we only present the XR profiles from the third scan for each phospholipid (± cholesterol) monolayer collected after 3 hours following CLIC1 addition, in order to complement our previously published Langmuir results 19. For XR experiments, the absolute reflectivity, R was derived by subtracting the background scattering followed by normalisation of the data to the incident beam flux. The error bars on the data represent the counting statistics at each data point. Specular Neutron Reflectometry measurements at the Air-Water interface The Platypus, time-of-flight reflectometer at the 20MW OPAL Research Reactor at ANSTO (Lucas Heights, Australia) was used to perform NR measurements probing interfacial thickness and density 4
ACS Paragon Plus Environment
Page 5 of 21
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 the lipid-CLIC1 monolayer in the direction normal to the solution/air interface. The measurements used neutrons of wavelengths ranging between 2.8 and 18Å incident on the interface at two angles, namely 1.1° and 4.8°, giving a Q-range from ∼0.013 to ∼0.32 Å-1. Instrument calibration was confirmed using a D2O KCL/Hepes buffer subphase (pH 6.5). Additional data were also collected on the CRISP reflectometer at the ISIS pulsed neutron source, Rutherford Appleton Laboratory, UK. The neutron wavelength range used for the CRISP reflectometer was between 0.5-6.5Å and data was collected at incident angles of 0.3°, 0.7° and 1.5° producing an overall similar Q range for both instruments. Data reduction followed the same procedure for both the NR instruments. The data was reduced by dividing the collected data through by direct beam runs for each angle, acquired under the same slit settings, the different angles were then ‘stitched together using the overlap regions, a background subtracted and normalized to unit reflectivity below the critical edge. For datasets where there was no critical edge a scale factor from the D2O buffer run collected under identical instrument settings was used. For NR experiments, two different contrast KCL/Hepes buffer (pH 6.5) subphase: D2O and air contrast matched water, (ACMW) which is a mixture of composition 92% H2O and 8% D2O by volume, were used to generate contrast variation. In addition, both hydrogenated and deuterated POPC and CLIC1, abbreviated as h/d31-POPC and h/d-CLIC1 respectively, were used to create contrast variation such that particular components of the monolayer can be highlighted or made invisible to the neutrons.
Data Analysis The final reflectivity profile is presented as a function of momentum transfer (Q), as defined by Q = 4 , where θ is the angle of incidence and λ is the wavelength. The procedure to obtain and fit lipid-protein profiles was as described 23 and is briefly outlined here. X-ray reflectivity data and NR data (from both Platypus and CRISP) were modelled using the Motofit analysis program in the Igor Pro environment (Wavemetrics) 24. The interface was described as a series of slabs (layers), with each layer characterised by a certain thickness (t, Å), a Gaussian interfacial roughness (σ, Å) and scattering length density, ( , Å-2) in the case of XR or ( , Å-2) in the case of NR. The difference between the data and the fit, χ2, was minimised by a Genetic Algorithm least squares fitting routine which proceeds by varying t, / , and σ. Since we have a system in which a protein is inserting through the ‘stack’ of layers then a disruption to the layering might well be expected, it was therefore decided to treat roughness as a fitting parameter in all experiments instead of fixing the interfacial roughness as capillary wave contribution. In all cases, the simplest possible physically reasonable model with the least number of parameters that adequately described the data was selected. For XR experiments, the errors given are the standard deviations reported by MOTOFIT. For NR experiments, multiple reflectivity profiles were collected for each phospholipid and CLIC1 interactions using different contrast variation. Since isotopic variation has no apparent impact on the structure, data from multiple H/D contrasts for each sample were simultaneously constrained to fit to the same layer where the thickness and roughness was kept constant and only the was allowed to vary between datasets as required. In order to have a precise measure of parameter uncertainties in NR experiments, a Monte-Carlo (MC) resampling technique was applied by methods described by Holt et.al (2009) 25. This technique quantifies confidence limits on the fitted parameters and provides correlations between the fit parameters. The MC resampling synthesises ‘new’ datasets by applying random perturbations (linked to the counting statistics) to individual data points. This new dataset is then fitted with the same model parameters. At least N = 5000 synthetic datasets were produced and analysed in this manner to produce N variations of each parameter. The fits to the synthetic data were analysed producing a frequency plot of fitted values. These parameter distributions were statistically analysed with the parameter value reported as the midpoint of the 95% confidence interval. The error 5
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
Page 6 of 21
is reported as twice the standard deviation of the distributions. A histogram is produced for the frequency distribution obtained from MC analysis. Comparison of the fitted / for each layer to the theoretical / values of each component within the layer enables the volume fraction to be calculated. Since CLIC1 and phospholipid head-groups have exchangeable protons in their structure, the theoretical of the hydrogenated and deuterated lipids and proteins were calculated for two different contrasts: D2O and ACMW. The percentage deuteration of CLIC1 was also taken into consideration when calculating the theoretical scattering length density of the protein in the different solvent contrasts. Table 1 presents the molecular volumes and the theoretically calculated and values of the lipids and CLIC1 used in this study. In monolayers containing lipids (L), proteins (P) and water (W), the electron density/ scattering length density of each layer ( ) is the sum of the theoretical weighted by the volume fractions φi of each component and water as given by the equation:
= (! x #! ) + (% x #% ) + (& x #& ) Therefore, due to significant differences in the molecular electron densities and scattering length densities of the lipid, protein and water (see Table 1), their individual volume fractions were computed from the fitted / profile of the layer and subsequently used to determine the average area per molecule (A, Å2) of a component (i) using the equation, ' = 2
() *
. Once the average area per
molecule is known, the surface excess of a component (Г, mol/m ) can be calculated by the given -
-
equation, Г = , . / -0123 x 4 6 # , where, 7. is the Avogadro’s number and # is the calculated 5
volume fraction of the component (i) in the monolayer.
Table 1. Summary of the molecular Volumes (89 ), theoretical electron densities (:;
