Accepted Manuscript

Recent advances in neutron reflectivity studies of biological membranes

Jeremy H. Lakey

PII: S1359-0294(19)30007-X

DOI: https://doi.org/10.1016/j.cocis.2019.02.012

Reference: COCIS 1274

To appear in: Current Opinion in Colloid & Interface Science

Received Date: 7 February 2019

Revised Date: 26 February 2019

Accepted Date: 26 February 2019

Please cite this article as: Lakey JH, Recent advances in neutron reflectivity studies of biologicalmembranes, Current Opinion in Colloid & Interface Science, https://doi.org/10.1016/j.cocis.2019.02.012.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Recent advances in neutron reflectivity studies of biological membranes

Jeremy H. Lakey

Institute for Cell and Molecular Bioscience, Newcastle University, Framlington Place,

Newcastle upon Tyne, NE2 4HH, UK

Abstract

Biological membranes are critical living interfaces which not only compartmentalise cells but

also provide specific surfaces upon which biochemical reactions take place. Their study is

essential for our understanding of life and disease, and also for the development of new

pharmaceuticals. Since they are dynamic structures reliant upon the surrounding aqueous

phase for their activity, techniques which can probe them at a molecular scale under

physiological conditions are crucial to understand their function. Here I describe some of the

recent developments in neutron reflectivity (NR) in this research area and suggest where

future developments in the technology might be usefully directed.

Keywords

Neutron reflectivity, model membranes, outer membrane, lipopolysaccharide, deuteration,

GISANS.

1. Introduction

It has been recounted many times that the geneticist J.B.S Haldane said that one of the

things you could learn from natural history about a Creator was that “he has an inordinate

fondness for beetles”. This conclusion was based upon the fact that there are over 300,000

beetle species compared to about 10,000 mammals. If a membrane biologist were asked the

same question then the answer might be “a great fondness for interfaces” since cells are

replete with boundaries and surfaces which govern how life runs at the molecular level.

However, if we look from the physics point of view, Wolfgang Pauli concluded that God

made the bulk and surfaces were the work of the devil. Due to the inherent difficulty in

studying cell membranes this conclusion would probably unite biologist and physicist alike.

Cells vary in the complexity of the membranes that they contain. The first cells were

presumably defined by the first cell or cytoplasmic membrane and therefore were likely to

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

have been surrounded by a single lipid bilayer. This pattern persists in the prokaryotes, the

simplest cells alive on the planet which include the bacteria and cyanobacteria. Their very

name is based upon their membrane structure since prokaryon indicates that they represent

a state before the existence of the nucleus, the membrane bound organelle in which nucleic

acids are packaged in eukaryotes such as fungi, animals and plants. As cells get more

complicated, it is their membrane architecture which is the most visible example of their

development and there are many techniques available to examine their structure and

dynamics ranging from atomic force microscopy to x-ray methods. To study them in relevant

situations they must be hydrated, at physiological temperatures and able to interact with

soluble components such as proteins or drugs. Neutron reflectivity (NR) is a method that can

probe the structure and dynamics of a single model membrane within an enclosed sample

cell, under close to physiological conditions, where it can be exposed to controlled changes

in environment or bathing solution. This review tries to highlight recent papers in the field

which show what is possible and where this science may go.

2. Membranes as biological interfaces.

Eukaryotes not only have nuclear membranes but also an array of membrane bound

compartments. In addition to this complexity there are two special membrane bound

compartments, the mitochondria and the chloroplasts, which are derived not from modified

cytoplasmic membrane but from free living cells that were taken up by the eukaryotes early

in their evolution from the archaea. The mitochondrion and chloroplast have two membranes

with a biochemically important intermembrane space in which crucial reactions take place.

These double membranes reflect the original structure of the free living forms and, as we

shall see later, these remain important in bacteria to this day. These membranes not only

create separate compartments within cells but also provide interfaces where reactions and

other activities occur.

The lipid bilayer forms the basis for all of these structures and has the important properties

of self-assembly, flexibility and fluidity which allow it to perform the many functions required

of membranes. In turn this means that the molecules which form these disparate structures

conform to a basic pattern of amphipathic structure, where a hydrophilic head group is

attached to one or more acyl chains composed mainly of –CH2– groups. This simple

architecture belies a wide array of molecular designs which confer changes in the physical

properties of the membranes, their surface interactions and even their metabolic function.

For example, membrane lipids normally quietly responsible for the structure of a cell

membrane can be converted after trauma, by enzymatic action, into powerful signalling

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

molecules involved in wound healing [1]. This process involves removing one acyl chain

which makes the lipid sufficiently water soluble to allow it to leave the bilayer and act upon

targets elsewhere in the cell.

Thus the individual parts of the membrane lipid play critical roles in the function of the

bilayer. The headgroups provide charge, molecular interactions, cell signalling and many

other roles. The lipid chains control the fluidity, stability and also contribute to the

biochemistry of the cell. Membranes are also characterised by their protein content which

defines the type and level of biochemical or physiological activity which that each membrane

undertakes. For example the Schwann cell membrane of the myelin sheath, which

electrically insulates nerve fibres, has a very low protein content (approx. 18% by weight),

whilst the mitochondrial inner membrane, which is highly metabolically active, has almost

80% protein. To understand these varied biological interfaces it is therefore essential to be

able to simultaneously study these various components within the assembled membrane

and observe how they work together. It is in this observation of dynamic mixtures of

biological molecules that NR provides unique information.

3. Neutron reflection studies of membranes

If one spreads a colourless oil on the surface of water and illuminates it with white light, one

can see rainbow patterns due to interfering reflection and refraction from the layers of

differing thickness and refractive index. In a similar way, specular neutron reflection

measures the differences in chemical composition across the membrane by the way

differences in neutron scattering length density (SLD, the neutron analogue of refractive

index) alter the angular and wavelength dependent intensity of reflected neutrons. The data

is fitted to models of the transbilayer structure composed of series of layers each defined by

a thickness, roughness and SLD. Neutron reflection can be applied to lipid monolayers at the

air-water interface or lipid bilayers assembled on solid substrates such as silicon. There

have been several recent reviews on applications of neutron reflection to membranes or thin

films where comparisons with diffraction and details of the experimental methods are clearly

described [2-4].

4. What can neutrons tell us about dynamic membrane biology?

Figure 1 provides a guide to the questions that neutron scattering methods can address. In

Fig1 A the simple lipid bilayer is shown with gaps which can occur due to incomplete layer

assembly or due to intrinsic dynamics, for example when using unsaturated phospholipids

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

[5]. As with all biological molecular assemblies, water is the common environment in which

all membrane studies take place. Here, the ability of neutrons to differentiate between the

distinct SLDs of hydrogen and deuterium is of great importance since the hydration state can

be measured more accurately than by other methods. In studying model bilayers, which

feature greatly in neutron reflection studies [3, 5-7], the use of different background solvent

contrasts, obtained by mixing H2O and D2O, allows the water content along the normal to the

membrane plane to be easily measured. In general the more well packed the membrane the

lower the water content of the hydrophobic core and so the quality of the membrane can be

readily measured. Subsequently the effects of membrane perturbing molecules such as

antibacterial peptides can be determined by the increasing water penetration into the

hydrophobic core of the bilayer even if the peptide itself cannot be visualised by neutron

reflection due to its small size, or insufficiently distinctive SLD, [8-11]. Finally, the right hand

side of Fig1A shows that lipid phase changes from solid or gel to liquid crystalline phase can

be revealed by thickness changes or water content. In Fig1B on the left, an asymmetric

membrane is shown where the lipid types are different on either side. This could be

determined by the different head group sizes which can be measured by NR. Such

asymmetry is found throughout nature but is most pronounced in bacterial outer membranes

OM discussed later in the review. However the measurement is made more accurate by

deuterating the lower leaflet which is then clearly resolved by neutrons [12]. Any subsequent

mixing of these two layers is quantifiable. In other cases different lipid types can be

selectively deuterated to examine their distribution [13] and the exchange between the

leaflets can be followed. Finally, on the right protein lipid interactions which lead to

alterations in lipid distribution can be followed or supported by small angle neutron scattering

measurements [14, 15]. In Fig 1C the possibilities of membrane protein studies are outlined.

Proteins including toxins [16, 17] can interact simply with membrane surfaces or with specific

protein (or lipid) receptors [18-20]. Furthermore the interaction of drugs or nanoparticles can

be readily measured [21]. Integral membrane proteins usually span the membrane and

naturally have a different SLD to the lipid component [22, 23], a difference that can be

enhanced by protein deuteration to reveal their individual structures within the membrane

[24]. In some cases these proteins are channel forming and they thus introduce water into

the hydrophobic core of the membrane, as shown by the blue channel. This can be

measured by solvent contrast methods but is difficult to differentiate from membrane defects

shown in Fig. 1A. Finally the dynamics of membrane proteins including multimerisation or

conformational change can be addressed. The limitation of specular reflection data to only

one dimension limits our ability to see such events and in the final section of this review I

discuss the potential developments which may make this essential next step achievable

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

5. Deuteration of biological molecules.

Through the use of deuterated lipids and proteins, their positions within the membrane can

be selectively determined against a background of different molecules [25-27]. The different

chemical composition and hydration of phospholipid head groups compared to their acyl tails

is often enough to allow us to model their thickness and position but by selectively

deuterating these regions one can significantly enhance the contrast between one region

and another giving one confidence in the derived models. Selective deuteration of all or parts

of lipids is most easily achievable by purchasing commercially available materials but such

studies can be extended by bespoke deuteration of lipids and proteins relevant to the

membrane study in question [28].

Useful levels of deuteration can be achieved in any laboratory which purifies the native

material from bacteria and higher levels up to complete (per-) deuteration can be obtained

by working with dedicated deuteration laboratories often associated with major neutron

research centres [26]. Specific lipids, such as lipopolysaccharide, can be isolated from the

natural producer organism simply grown in media where water has been replaced by D2O or

also with deuterated carbon sources [26]. Where the lipids are found in mammals which

cannot be fed high concentrations of D2O modified yeast cells have been found to be good

sources of deuterated lipids such as cholesterol [29]. Deuterated proteins can be produced

recombinantly by having their genes expressed in bacteria or yeast which can be grown in

deuterated media. This can be made more specific by deuterating part of the protein as

elegantly shown for the interaction of the Parkinson’s disease related protein α‑Synuclein

with membranes [18]. Thus the different components of the membrane are purified as

separate samples and then recombined to reproduce the model of the natural membrane.

An important study has been carried out into the properties of deuterated lipids produced in

a natural host, in this case the yeast Pichia pastoris. It was shown that growth in deuterated

media altered the molecular composition of the natural membrane so that the model

membranes formed from them had different dimensions [30]. In another study the ratio of

phosphatidylethanolamine, phosphatidylglycerol and cardiolipin in Escherichia coli lipid

extracts was found to be similar in normal and deuterated samples [7]. In the future it will be

interesting to see if deuterated lipids show specific isotope effects.

In an interesting alternative to combining purified proteins with lipids to construct the models,

it has been shown that proteins can be inserted into supported lipid bilayers using cell free

methods, where the proteins are inserted directly into the membrane from isolated

ribosomes using a cell free synthesis method [24] [31]. The proteins are made from amino

acids added to the solution, and it is possible to use deuterated versions so that the proteins

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

have bespoke SLD values. In one example, the insertion of the hepatitis C p7 viroporin into

model bilayers was not clear when hydrogenated protein was used, but when deuterated

protein was synthesised the insertion became clearly measurable [24]. This insertion could

also be followed by electrochemical techniques since the protein increases the conductance

of lipid bilayers in which it is inserted.

6. A biological case study - the Gram-negative bacterial outer membrane (OM).

Here, partly due to my own research but also highlighting that of several other groups, I

discuss a series of recent papers on bacterial outer membranes (OM) which also highlight

general trends in reflectometry studies of membranes. As stated in the introduction, Gram-

negative bacteria possess a double membrane structure which provides extra protection

from the outside environment since the outer membrane acts as a molecular sieve. Not only

is this of biological importance but these bacteria are increasingly important causes of

antibiotic resistant infections from species such as Escherichia coli, Klebsiella pneumoniae,

Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter sp. This has meant

that there has been great interest in studying this structure in vitro since, due to the very

small size of bacterial cells, many of the methods successfully used to study membranes of

eukaryotes in vivo are not applicable. The particular challenge of the OM is that it is highly

asymmetric with “normal” phospholipids present on the inner face and complex glycolipids

called lipopolysaccharides (LPS) on the outer face. This asymmetry is an essential feature of

the membrane enabling the robust outer layer to resist environmental challenges in the gut,

such as bile salts and proteases. LPS consists of a lipid A section with 5-7 acyl chains to

which is attached the core oligosaccharide of the head group with about ten sugar units. This

can be extended to include many hundreds of units in so called smooth strains of bacteria. It

has been shown that we can gain an insight into OM function using symmetric bilayers

containing mixtures of LPS and phospholipids [32, 33] but the full capabilities of NR are

exploited on fully asymmetric models.

One way to create asymmetric surfaces is to use hybrid bilayer membranes where the solid

surface is covered with a hydrophobic layer upon which a lipid monolayer is created. This is

shown in Figure 2 reproduced from [8]. An example of long chain “smooth” LPS being

studied in such a system has been published by Schneck’s group [34]. Building upon

previous work [35, 36] they were able to create layers containing both short and long LPS

chains and thus use neutrons to measure how these extend into the aqueous environment.

Furthermore they were also able to measure the interactions of two layers by creating

another opposing one, across a small water layer, at the air-water interface. This, together

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

with floating bilayer studies [5, 6], further demonstrates the possibilities of investigating

several interacting membranes using NR.

We have created complete asymmetric models by combining two monolayers, one of

phospholipid and one of LPS, formed sequentially at the air-water interface and co -

assembled using Langmuir-Blodgett and Langmuir-Schaefer techniques onto silicon or gold

surfaces. In Figure 3 a complete asymmetric outer membrane model assembled on a silicon

surface is presented with a layer of bound antibacterial protein (colicin N). An alternative

model in which the outer membrane floats on a ≈ 2nm water layer has also been developed

[6].

As outlined in Figure 1 the great strength of neutron reflectivity is the ability to resolve, in Fig

3, the hydrogenous LPS (either purified from bacteria or commercially available) and the

deuterated phospholipid (commercially available), allowing the asymmetry to be not only

observed but accurately quantified, along with a measure of the % surface coverage from

solvent contrast experiments (Figure 3). The hydrophobic tails in either deuterated or

hydrogenated form show SLDs which do not change much when the background solvent is

exchanged from H2O to D2O. However, the solvated headgroups of the core

oligosaccharide, and to a lesser extent the phospholipid, show SLD values which partially

track the bulk solvent and reveal the level of hydration. In this study the layer of bound

protein occupied about 40% of the surface by volume and this is shown by the high degree

of SLD change upon solvent exchange. In the same paper we were also able to measure the

thickness of core oligosaccharide from two versions of LPS and show that this acts as an

insulator to reduce the binding of toxic proteins to the bacterial surface [17]. In a subsequent

study, it was shown that the lipopeptide antibiotic polymyxin only penetrates into membranes

when the LPS is in the liquid crystalline form [8]. The phase change could be measured both

by NR and FTIR, and, provided biophysical support to biological observations that bacteria

maintain their LPS in a liquid crystalline rather than gel state. The combined use of NR and

FTIR is a promising development in the field of membrane research [37].

These models have also been used to support the development of new polymyxin type

antibiotics which are needed to overcome resistance to this important class of antibiotics.

The resistance mechanism can involve modifications of the LPS and the effect that these

changes have on the structure and dynamics of the membrane can be observed by X-ray

[11] and neutron reflectivity [9]. It is also possible to observe the behaviour of novel

antibiotics [10] or anti-microbial peptides in the membrane model [38]. These models lack

integral outer membrane proteins [14] and their role in the penetration of this class of

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

antibiotics has not been measured. The significant challenge is to introduce the proteins in

an oriented form so that they conform to the overall asymmetry of the OM. One way is to fix

the proteins to the solid substrate, which then acts as a symmetry breaker but it may

preclude the subsequent use of successive monolayer assembly to achieve asymmetric lipid

membranes [39, 40]. It has been shown that molecularly complete outer membranes can be

assembled on surfaces using outer membrane vesicles and that they show asymmetry [41].

It would be interesting to apply NR to these models, since they should mimic the OM more

closely. However, since NR averages the contributions it may be necessary to start with

bacterial preparations with simplified OM such as reduced outer membrane protein variety

and one short form of LPS. Use of deuterated antibiotics etc. may allow these to be

observed against the natural background or the OM may be made using deuterated bacteria

[41]. Furthermore native eukaryotic biological membranes are also being used to create

planar membrane models which can be investigated by many methods including NR [42]

7. Conclusions.

For the membrane biochemist, knowledge of the lateral motions of lipids and membrane

proteins in the plane of the bilayer is of great interest. The fluid mosaic model proposed in

the 1960’s has been modified over the years to include possible phase separations, rafts

and protein–lipid complexes but understanding the mobility and dynamics of the membrane

is all four dimensions, including time [43], remains a serious challenge. The time dimension

is being addressed by faster beamlines so that changes can be followed with resolution of

minutes or shorter [43, 44]. Resolution in the plane of the bilayer is currently less than

afforded by X-ray grazing incidence diffraction [45] or small angle scattering at the air-water

interface but we need to use neutrons since they are capable of penetrating the

experimental cells where we can construct complete bilayer models under controlled

conditions (Figure 4). Off specular NR signals have been recorded for LPS membranes [46]

whilst experiments with grazing incidence small angle neutron scattering (GISANS), which

can potentially resolve nm sized structures have shown promise with self-organising colloidal

materials [47] and block copolymers [48, 49]. Furthermore grazing incidence neutron spin

echo spectroscopy (GINSES) has been applied to measure the depth dependent dynamics

of extended polymer brushes on silicon substrates [50]. Neutron reflection has contributed

significantly to our understanding of biological membranes but we have only scratched the

surface of what could be measured. Sample preparation is developing fast e.g. [42] and

since three or four dimensional [43] experiments are clearly possible we shall be able to

probe areas of membrane behaviour inaccessible to all other methods. If we can develop

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

reliable applications in these areas we will certainly solve some crucial basic science

questions and probably save lives too.

Acknowledgements.

I wish to thank my patient collaborators in neutron scattering for their help and support over

the years in developing realistic biological models that have had influence beyond the world

of neutron science. Particular thanks go to Nico Paracini and Dan Peters for helpful

comments on the manuscript. All opinions and mistakes in this review are my own.

References .

[1] Jans R, Mottram L, Johnson DL, Brown AM, Sikkink S, Ross K, et al. Lysophosphatidic

Acid Promotes Cell Migration through STIM1-and Orai1-Mediated Ca-i(2+) Mobilization and

NFAT2 Activation. J Invest Dermatol. 2013;133:793-802.

[2] Foglia F, Lawrence MJ, Barlow DJ. Studies of model biological and bio-mimetic

membrane structure: Reflectivity vs diffraction, a critical comparison. Current Opinion in

Colloid & Interface Science. 2015;20:235-43.

[3] Fragneto G, Delhom R, Joly L, Scoppola E. Neutrons and model membranes: Moving

towards complexity. Current Opinion in Colloid & Interface Science. 2018;38:108-21.

[4] Heinrich F, Losche M. Zooming in on disordered systems: Neutron reflection studies of

proteins associated with fluid membranes. Biochimica Et Biophysica Acta-Biomembranes.

2014;1838:2341-9.

[5] Hughes AV, Holt SA, Daulton E, Soliakov A, Roser SJ, Charlton TR, et al. High Coverage

Fluid-Phase Floating Lipid Bilayers supported by Ω-Thiolipid SAMs. J Roy Soc Interface.

2014;11 20140447.

[6] Clifton LA, Holt SA, Hughes AV, Daulton EL, Arunmanee W, Heinrich F, et al. An

Accurate In Vitro Model of the E. coli Envelope. Angew Chem-Int Edit. 2015;54:11952-5.

[7] Lind TK, Wacklin H, Schiller J, Moulin M, Haertlein M, Pomorski TG, et al. Formation and

Characterization of Supported Lipid Bilayers Composed of Hydrogenated and Deuterated

Escherichia coli Lipids. Plos One. 2015;10.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

[8] *Paracini N, Clifton LA, Skoda MWA, Lakey JH. Liquid crystalline bacterial outer

membranes are critical for antibiotic susceptibility. Proceedings of the National Academy of

Sciences of the United States of America. 2018;115:E7587-E94.

An NR paper that clarifies a basic biological question and demonstrates that bacterial

outer membranes, a critical barrier to antiboitics, are not in the gel phase as

previously assumed.

[9] *Han M-L, Velkov T, Zhu Y, Roberts KD, Le Brun AP, Chow SH, et al. Polymyxin-Induced

Lipid A Deacylation in Pseudomonas aeruginosa Perturbs Polymyxin Penetration and

Confers High-Level Resistance. Acs Chemical Biology. 2018;13:121-30.

[10] *Han M-L, Shen H-H, Hansford KA, Schneider EK, Sivanesan S, Roberts KD, et al.

Investigating the Interaction of Octapeptin A3 with Model Bacterial Membranes. Acs

Infectious Diseases. 2017;3:606-19.

Two papers showing the application of NR data to the understanding of antibiotic

resistance and the development of new antibacterials.

[11] Nobre TM, Martynowycz MW, Andreev K, Kuzmenko I, Nikaido H, Gidalevitz D.

Modification of Salmonella Lipopolysaccharides Prevents the Outer Membrane Penetration

of Novobiocin. Biophysical Journal. 2015;109:2537-45.

[12] Clifton LA, Skoda MWA, Le Brun AP, Ciesielski F, Kuzmenko I, Holt SA, et al. Effect of

Divalent Cation Removal on the Structure of Gram-Negative Bacterial Outer Membrane

Models. Langmuir. 2015;31:404-12.

[13] Luchini A, Nzulumike ANO, Lind TK, Nylander T, Barker R, Arleth L, et al. Towards

biomimics of cell membranes: Structural effect of phosphatidylinositol triphosphate (PIP3) on

a lipid bilayer. Colloids and Surfaces B-Biointerfaces. 2019;173:202-9.

[14] Arunmanee W, Pathania M, Solovyova AS, Le Brun AP, Ridley H, Baslé A, et al. Gram-

negative trimeric porins have specific LPS binding sites that are essential for porin

biogenesis. Proceedings of the National Academy of Sciences. 2016;113:E5034-E43.

[15] Hossain KR, Holt SA, Le Brun AP, Al Khamici H, Valenzuela SM. X-ray and Neutron

Reflectivity Study Shows That CLIC1 Undergoes Cholesterol-Dependent Structural

Reorganization in Lipid Monolayers. Langmuir. 2017;33:12497-509.

[16] Wacklin HP, Bremec BB, Moulin M, Rojko N, Haertlein M, Forsyth T, et al. Neutron

reflection study of the interaction of the eukaryotic pore-forming actinoporin equinatoxin II

with lipid membranes reveals intermediate states in pore formation. Biochimica Et

Biophysica Acta-Biomembranes. 2016;1858:640-52.

[17] Clifton LA, Ciesielski F, Skoda MW, Paracini N, Holt SA, Lakey JH. The Effect of

Lipopolysaccharide Core Oligosaccharide Size on the Electrostatic Binding of Antimicrobial

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

Proteins to Models of the Gram Negative Bacterial Outer Membrane. Langmuir

2016;32:3485-94.

[18] *Jiang ZP, Heinrich F, McGlinchey RP, Gruschus JM, Lee JC. Segmental Deuteration of

alpha-Synuclein for Neutron Reflectometry on Tethered Bilayers. Journal of Physical

Chemistry Letters. 2017;8:29-34.

A good demonstration of how selective protein deuteration methods can be used to

increase the resolution of NR experiments

[19] O'Neil L, Andenoro K, Pagano I, Carroll L, Langer L, Dell Z, et al. HIV-1 matrix-31

membrane binding peptide interacts differently with membranes containing PS vs. PI(4,5)P-

2. Biochimica Et Biophysica Acta-Biomembranes. 2016;1858:3071-81.

[20] Barros M, Heinrich F, Datta SAK, Rein A, Karageorgos I, Nanda H, et al. Membrane

Binding of HIV-1 Matrix Protein: Dependence on Bilayer Composition and Protein Lipidation.

Journal of Virology. 2016;90:4544-55.

[21] Ding Y, Chow SH, Liu GS, Wang B, Lin TW, Hsu HY, et al. Annexin V-containing

cubosomes for targeted early detection of apoptosis in degenerative retinal tissue. J Mat

Chem B. 2018;6:7652-61.

[22] Zimmermann K, Eells R, Heinrich F, Rintoul S, Josey B, Shekhar P, et al. The cytosolic

domain of T-cell receptor zeta associates with membranes in a dynamic equilibrium and

deeply penetrates the bilayer. Journal of Biological Chemistry. 2017;292:17746-59.

[23] Rondelli V, Del Favero E, Brocca P, Fragneto G, Trapp M, Mauri L, et al. Directional K+

channel insertion in a single phospholipid bilayer: Neutron reflectometry and

electrophysiology in the joint exploration of a model membrane functional platform.

Biochimica Et Biophysica Acta-General Subjects. 2018;1862:1742-50.

[24] *Soranzo T, Martin DK, Lenormand J-L, Watkins EB. Coupling neutron reflectivity with

cell-free protein synthesis to probe membrane protein structure in supported bilayers.

Scientific Reports. 2017;7.

A clear demonstration of a different approach to inserting membrane proteins into

model bilayers for biophysical study. (see also 29).

[25] Heinrich F. Deuteration in Biological Neutron Reflectometry. In: Kelman Z, editor.

Isotope Labeling of Biomolecules - Applications2016. p. 211-30.

[26] Le Brun AP, Clifton LA, Holt SA, Holden PJ, Lakey JH. Deuterium Labeling Strategies

for Creating Contrast in Structure-Function Studies of Model Bacterial Outer Membranes

Using Neutron Reflectometry. In: Kelman Z, editor. Isotope Labeling of Biomolecules -

Applications2016. p. 231-52.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

[27] Haertlein M, Moulin M, Devos JM, Laux V, Dunne O, Forsyth VT. Biomolecular

Deuteration for Neutron Structural Biology and Dynamics. In: Kelman Z, editor. Isotope

Labeling of Biomolecules - Applications. San Diego: Elsevier Academic Press Inc; 2016. p.

113-57.

[28] Waldie S, Lind TK, Browning K, Moulin M, Haertlein M, Forsyth VT, et al. Localization of

Cholesterol within Supported Lipid Bilayers Made of a Natural Extract of Tailor-Deuterated

Phosphatidylcholine. Langmuir. 2018;34:472-9.

[29] *Moulin M, Strohmeier GA, Hirz M, Thompson KC, Rennie AR, Campbell RA, et al.

Perdeuteration of cholesterol for neutron scattering applications using recombinant Pichia

pastoris. Chemistry and Physics of Lipids. 2018;212:80-7.

A clear demonstration of a different approach to inserting membrane proteins into

model bilayers for biophysical study. (see also 24)

[30] Luchini A, Delhom R, Deme B, Laux V, Moulin M, Haertlein M, et al. The impact of

deuteration on natural and synthetic lipids: A neutron diffraction study. Colloids and Surfaces

B-Biointerfaces. 2018;168:126-33.

[31] *Maccarini M, Gayet L, Alcaraz J-P, Liguori L, Stidder B, Watkins EB, et al. Functional

Characterization of Cell-Free Expressed OprF Porin from Pseudomonas aeruginosa Stably

Incorporated in Tethered Lipid Bilayers. Langmuir. 2017;33:9988-96.

[32] Dupuy FG, Pagano I, Andenoro K, Peralta MF, Elhady Y, Heinrich F, et al. Selective

Interaction of Colistin with Lipid Model Membranes. Biophysical Journal. 2018;114:919-28.

[33] Andersson J, Fuller MA, Wood K, Holt SA, Koper I. A tethered bilayer lipid membrane

that mimics microbial membranes. Physical Chemistry Chemical Physics. 2018;20:12958-

69.

[34] *Rodriguez-Loureiro I, Latza VM, Fragneto G, Schneck E. Conformation of Single and

Interacting Lipopolysaccharide Surfaces Bearing O-Side Chains. Biophysical Journal.

2018;114:1624-35.

Demonstrates the ability to observe extended, hydrated polysacharide structures

which occur on the surfaces of cells. Such molecules are unlikley to be accurately

observed by other methods and are critical for many intercellular interactions in both

eukaryotes and prokaryotes.

[35] Herrmann M, Schneck E, Gutsmann T, Brandenburg K, Tanaka M. Bacterial

lipopolysaccharides form physically cross-linked, two-dimensional gels in the presence of

divalent cations. Soft Matter. 2015;11:6037-44.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

[36] Schneck E, Schubert T, Konovalov OV, Quinn BE, Gutsmann T, Brandenburg K, et al.

Quantitative determination of ion distributions in bacterial lipopolysaccharide membranes by

grazing-incidence X-ray fluorescence. Proceedings of the National Academy of Sciences of

the United States of America. 2010;107:9147-51.

[37] Skoda MWA, Thomas B, Hagreen M, Sebastiani F, Pfrang C. Simultaneous neutron

reflectometry and infrared reflection absorption spectroscopy (IRRAS) study of mixed

monolayer reactions at the air-water interface. Rsc Advances. 2017;7:34208-14.

[38] Michel JP, Wang YX, Kiesel I, Gerelli Y, Rosilio V. Disruption of Asymmetric Lipid

Bilayer Models Mimicking the Outer Membrane of Gram-Negative Bacteria by an Active

Plasticin. Langmuir. 2017;33:11028-39.

[39] Hoogerheide DP, Noskov SY, Kuszak AJ, Buchanan SK, Rostovtseva TK, Nanda H.

Structure of voltage-dependent anion channel-tethered bilayer lipid membranes determined

using neutron reflectivity. Acta Crystallographica Section D-Structural Biology.

2018;74:1219-32.

[40] Holt SA, Le Brun AP, Majkrzak CF, McGillivray DJ, Heinrich F, Lösche M, et al. An ion

channel containing model membrane: structural determination by magnetic contrast neutron

reflectometry Soft Matter. 2009;5 2576 - 86.

[41] Hsia CY, Chen LX, Singh RR, DeLisa MP, Daniel S. A Molecularly Complete Planar

Bacterial Outer Membrane Platform. Scientific Reports. 2016;6.

[42] **Pace HP, Hannestad JK, Armonious A, Adamo M, Agnarsson B, Gunnarsson A, et al.

Structure and Composition of Native Membrane Derived Polymer-Supported Lipid Bilayers.

Analytical Chemistry. 2018;90:13065-72.

An excellent demonstration of the possibilities to move closer to the native state in

membrane models and be able to analyse them with a range of biophysical methods

[43] **Koutsioubas A, Appavou M-S, Lairez D. Time-Resolved Neutron Reflectivity during

Supported Membrane Formation by Vesicle Fusion. Langmuir. 2017;33:10598-605.

Uses both specular and off specular methods to follow kinetics of vesicle fusion at a

solid surface

[44] Adlmann FA, Gutfreund P, Ankner JF, Browning JF, Parizzi A, Vacaliuc B, et al.

Towards neutron scattering experiments with sub-millisecond time resolution. Journal of

Applied Crystallography. 2015;48:220-6.

[45] Le Brun AP, Clifton LA, Halbert CE, Lin B, Meron M, Holden PJ, et al. Structural

Characterization of a Model Gram-Negative Bacterial Surface Using Lipopolysaccharides

from Rough Strains of Escherichia coli. Biomacromolecules. 2013;14:2014-22.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

[46] Schneck E, Oliveira RG, Rehfeldt F, Deme B, Brandenburg K, Seydel U, et al.

Mechanical properties of interacting lipopolysaccharide membranes from bacteria mutants

studied by specular and off-specular neutron scattering. Phys Rev E. 2009;80:9.

[47] Nouhi S, Hellsing MS, Kapaklis V, Rennie AR. Grazing-incidence small-angle neutron

scattering from structures below an interface. Journal of Applied Crystallography.

2017;50:1066-74.

[48] Muller-Buschbaum P. GISAXS and GISANS as metrology technique for understanding

the 3D morphology of block copolymer thin films. European Polymer Journal. 2016;81:470-

93.

[49] Wolff M, Herbel J, Adlmann F, Dennison AJC, Liesche G, Gutfreund P, et al. Depth-

resolved grazing-incidence time-of-flight neutron scattering from a solid-liquid interface.

Journal of Applied Crystallography. 2014;47:130-5.

[50] Wellert S, Hubner J, Boyaciyan D, Ivanova O, von Klitzing R, Soltwedel O, et al. A

grazing incidence neutron spin echo study of near surface dynamics in p(MEO(2)MA-co-

OEGMA) copolymer brushes. Colloid and Polymer Science. 2018;296:2005-14.

[51] Baboolal TG, Conroy MJ, Gill K, Ridley H, Visudtiphole V, Bullough PA, et al. Colicin N

binds to the periphery of its receptor and translocator, outer membrane protein F. Structure.

2008;16:371-9.

Figure legends

Figure 1. Biological membrane features addressable by neutron reflection methods

Studies of simple lipid bilayers can reveal features such as the hydration of the interior which

can be caused by incomplete assembly of the bilayer or physiologically important effects of

certain lipid mixtures. Neutron reflection is uniquely sensitive to replacement of H2O by D2O

and this can probe hydration of the interior and surface headgroups. As shown, this lipid

packing can be affected by the insertion of amphiphilic molecules such as detergents or

lipopeptide antibiotics. The physical state of the lipids, shown on the right as a phase

change, can also be observed by changes in thickness or hydration. B) More complex lipid

mixtures can be studied using differences in headgroups size to follow the components but

clearly if certain lipids can be deuterated (blue) then variations in symmetry or other

distributions are easier to follow. Protein-lipid interactions (green) are also an area yet to be

fully addressed by NR. C) Proteins can interact either with the surface of the model

membrane or with integral membrane proteins acting as receptors. Ion channel proteins

(centre) are a particular challenge since their water filled centres can look like gaps in the

membrane seen in (A). Finally protein-protein structure in the plane of the membrane will

require data from GISANS and related methods (see conclusions).

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

Figure 2. Gram-negative outer membrane and models. (A) Schematic representation of

the Gram-negative cell envelope. The cytoplasm is enclosed by a phospholipid inner

membrane (proteins not shown) surrounded by a thin peptidoglycan cell wall. The outer layer

is the asymmetric outer membrane composed of phospholipids in the inner leaflet and

lipopolysaccharides in the outer leaflet. It contains water-filled channel proteins (porins)

through which most antibiotics enter the cell. The LPS is shown as two forms, a rough LPS

(red) containing the lipid A and core oligosaccharide, and variable lengths of O-antigen

present in smooth strains of Gram-negative bacteria in orange. Stabilizing divalent cations

are shown in green. (B) Simple OM model formed by a hybrid bilayer membrane containing

a monolayer of LPS self-assembled on a hydrophobic monolayer. (C) Asymmetric OM model

assembled in vitro on a silicon substrate by sequential deposition of synthetic phospholipids

and purified LPS using a Langmuir trough. (D) Chemical structure of polymyxin B.

Reproduced from [8].

Figure 3. Study of binding of the antibacterial protein colicin N to a model bacterial

outer membrane. The model was composed of short chain Rd-LPS and deuterated

dipalmitoyl-phosphatidylcholine (d-DPPC) on a silicon substrate. Neutron reflectivity profiles

and model data fits (A−C) and the scattering length density profiles these fits describe (D) for

equilibrium Colicin N adsorbed asymmetric DPPC (inner leaflet)/Rd LPS (outer leaflet)

bilayers in the presence of 20 mM HEPES pH 7.2 buffer with 20 µM CaCl2. The three

simultaneously fitted isotopic contrasts shown are (A) d-DPPC/Rd LPS in D2O (red line), (B)

d-DPPC/Rd LPS in a mixture of H2O/D2O that has an SLD equal to that of the silicon

substrate, effectively removing this interface from the data (black line), and (C) d-DPPC/Rd

LPS in H2O (green line). A representation of the interfacial structure determined from these

fits is shown (E) with the silicon substrate to the left (Black) covered with the SiO2 oxide layer

(grey) , lipid head group (green), deuterated tails (red), hydrogenous LPS tails (brown), LPS

core head group (green) and colicin N protein (pink). Reproduced from [17]

Figure 4. In plane data from neutron reflection experiments on model membranes.

Most current NR experiments collect data in the specular range (solid arrows) which

provides profiles of material distribution across the z axis of the membrane (see fig 2).

Methods which acquire data from the y-axis can provide information on the arrangement of

molecules, such as the outer membrane protein F shown here which forms lattice structures

observable by electron microscopy [51], scale bar 10 nm. Such structures are often in

dynamic equilibrium between different forms and regulate many processes in the cell.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

Neutron reflection could provide a unique view of how the individual molecular components

rearrange and contribute to these equilibria.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT