A novel lipid-polymer system with unique properties has potential in drug delivery and biotechnology applications

TL;DR: DMPC-C8E5 mixtures retain interesting properties rendering them particularly advantageous in studies of membrane protein interactions and hold promise as vehicles for drug delivery.

Abstract: The utility of detergent micelle and bicelle systems has been demonstrated to be a valuable tool for the study of membrane protein interactions and in structural studies. Bicelles are distinguished from micelles in that they contain a lipid bilayer that mimics the plasma membrane of cells making it more native-like than its detergent micelle counter-part. Bicelles are typically comprised of a long-chain phospholipid such as 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) and either a short-chain phospholipid, typically 1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC), or a bile-salt derivative such as CHAPS or CHAPSO. In solution DMPC and DHPC bicelles assume a discoidal structure comprised of a heterogeneous arrangement where the short-chain lipids gather around the rim of the disk and the long-chain lipids form the flat, bilayer region of the bicelle. Aside from DHPC, CHAPS and CHAPSO few other detergents have reportedly been investigated for their ability to form bicelles with DMPC. In this study, the detergent, C8E5, was used to prepare mixtures with DMPC to determine if it adopts properties similar to DMPC-DHPC bicelles. Mixtures were evaluated using sedimentation equilibrium, 31P-phosphorus NMR, and light scattering and compared to DMPC-DHPC bicelles. Interestingly, mixtures of DMPC and C8E5 assumed a spherical-shaped micellar structure, not the predicted discoidal shape. DMPC-C8E5 mixtures retain interesting properties rendering them particularly advantageous in studies of membrane protein interactions and hold promise as vehicles for drug delivery.

Summary

1. Introduction

• It has been well-established that 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), a naturally derived phospholipid with a 14-carbon chain length and a choline head group, spontaneously assembles into discoidal lipid structures called bicelles when mixed with shortchain phospholipids such as 1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC), also naturally derived but containing a 6-carbon chain length with a choline head group 1.
• In practice certain lipid aggregates or mixed micellar systems can give rise to complexes with very high densities making density matching near impossible without using expensive reagents, i.e. D2O18, or the addition of large quantities of density modifiers, which can be susceptible to gradient formation under prolonged periods of centrifugation 16.
• In the DMPC-DHPC bicelle system, the authors found that DHPC imparts a substantial density to the entire bicellar aggregate necessitating the use of density modifiers.

2.1. Chemicals and reagents

• DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine) and NBD-DMPE (1,2-dimyristoyl- sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (ammonium salt)) were purchased from Avanti Polar Lipids (Alabaster, AL, USA).
• C8E5 (n-octylpentaoxyethylene) was purchased from Bachem (King of Prussia, PA, USA).
• 2. Preparation of sedimentation equilibrium samples 2.2.1 DMPC-DHPC bicelles Equilibrium absorbance measurements (464 nm) were collected from 10,000 rpm to 35,000 rpm stepping up in speed in 1,000 rpm increments.
• Meff was plotted against the percentage D2O concentration to evaluate the D2O required to density match the DMPC-C8E5 or produce Meff = 0.

2.3. Sedimentation equilibrium (AUC) density matching experiments

• The ratio of NBD-labeled lipid to unlabeled lipid was kept sufficiently low so that the label did not influence the physical properties of the aggregates (1:500 mole ratio of NBDDMPE:DMPC), but high enough that an appreciable absorbance signal could be acquired.
• The samples were briefly vortexed and spun in a microfuge.
• Samples were loaded into a 6-channel equilibrium charcoal-filled epon centerpiece with a pathlength of 1.2 cm. 120 μL of each sample and reference was loaded in the appropriate channels.

2.4 Samples for 31P-NMR analysis

• Samples with a 25% (w/w) total lipid composition were prepared for 31P-phosphorus NMR experiments.
• To two separate 1.5 mL Eppendorf tubes DMPC was added to a final concentration of 158 mM.
• The samples were vortexed to a milky, white suspension.
• This concentration was previously determined to saturate the lanthanide-phospholipid interaction making the DMPC and DHPC 31Pphosphorus signals distinguishable from one another in the NMR 24.
• The experiments were processed in TopSpin v. 1.3 (Bruker Corporation).

2.5 Samples for light scattering experiments

• Mixtures of DMPC-C8E5 were prepared according to the theoretical q - value used to characterize DMPC-DHPC bicelles.
• The samples were vortexed briefly until clear.
• The scattered light intensity was measured at a detector angle of 90 degrees for eight samples.

2.6 Curve fitting

• The shape of the lipid-detergent aggregates was deduced following the detailed method of Mazer and co-workers to characterize mixed micelle formation in bile salt-lecithin solutions.
• This method was used by Glover and co-workers to deduce the discoidal shape of DMPC-DHPC bicelles [3,23].
• The lipid-detergent solutions were treated as monodisperse and non-interacting where the mean scattering intensity, I, is given by the following equation: I = CMP where C is the concentration of the lipid-detergent (w/w), M is the molecular weight of the lipiddetergent aggregate, and P describes the scattering form factor.
• Using this approach, the product MP will have different values depending on the shape of Eq. 3 the lipid detergent aggregate.
• It follows that the quantity I/C, which is equivalent to MP, can be measured experimentally for DMPC-C8E5 aggregates and the shape deduced from a semi-log plot of the normalized I/C versus Rh.

3. Results and Discussion

• DMPC-C8E5 mixtures were prepared and characterized using analytical techniques each of which would allow us to determine if these mixtures were forming discoidal-type structures similar to the well-characterized DMPC-DHPC bicelles.
• Because membrane protein structure is highly influenced by the surrounding lipid environment, it is important to consider the choice of detergent or lipid system.
• Compared to DMPC-DHPC bicelles the solution viscosity of DMPC-C8E5 mixtures also increased substantially as the lipid concentration increased (Table 1).
• The product of the molecular weight and the form factor, MP, is equivalent to the experimentally determined scattering intensity, I, divided by the lipid-detergent concentration, C, for the DMPC-C8E5 structures: 𝐼 𝐶 = 𝑀𝑃 Eq. 4.
• In these two models the radius of the disk, r, and the length, L, of the rod-shaped aggregates were adjusted to the measured values of the radius for the DMPC-C8E5 aggregates.

4. Conclusions

• In this study DMPC-C8E5 lipid-detergent aggregates/mixtures were prepared at various molar ratios and total lipid compositions.
• These aggregates were evaluated using different bioanalytical techniques and the measured properties compared to the well-characterized DMPCDHPC bicellar system.
• Interestingly, these structures are predicted to form spherical aggregates rather than discoidalshaped aggregates like their DMPC-DHPC counterparts.
• This is the first reported instance where this type of mixture has been prepared and characterized.
• The authors can also conclude from these studies that there are unique conditions and properties that lipids and detergents molecules must have for their spontaneous assembly into bicelles and not all detergents are suited to adopt this geometric conformation.

Figures

Figure 7. Semi-log plot of normalized scattering intensity versus radius of hydration, Rh. The solid line represents the model of a spherical particle, the dashed line represents the model of a disk, the dotted line represents the model of a rod. The dotted-square line represents the DLS data. The DLS data fit the curve for the sphere indicating the shape of the DMPC / C8E5 aggregates are spherical.

Figure 6. Rh versus theoretical q. Increasing theoretical q corresponds to an increase in the hydrodynamic radius of the particle.

Figure 2. Meff versus D2O concentration. Approximately 18.4% D2O is required to density match DMPC-C8E5 lipid-detergent aggregates.

Figure 3. Density matching using three biocompatible density modifiers: A) D2O B) glycerol C) sucrose. Approximately 71.7% (v/v) D2O, 23.5 % (v/v) glycerol and 0.418 M sucrose is required to density match DMPC-DHPC bicelles.

Figure 4. 31P-phosphorus NMR spectra of C8E5 (blue) and DMPC (red). The chemical shift of DHPC is 35.1 ppm and the chemical shift of DMPC appears at 42.0 ppm for DMPC in DMPC / DHPC bicelles and 42.5 ppm in DMPC / C8E5 bicelles. Panel A is the sample analyzed at 25 ⁰C and Panel B is the sample analyzed at 37 ⁰C. Note: Data was also collected at 31 ⁰C showing chemical shifts similar to 37 ⁰C. (spectrum not shown).

Table I. Measurements of viscosity and hydrodynamic radius in relation to increasing lipid concentration

Figure 5. Rh versus % total lipid (w/w). Particle size decreases as total lipid concentration increases.

Figure 1. A) Schematic diagram of a bicelle containing DMPC and DHPC. Samples were prepared replacing DHPC with C8E5 predicted to replace DHPC at the rim of bicelle. B) Dodecyl-β -D-maltoside (DDM) lipid with dense maltose sugar headgroup; glucose = 0.622 cm3 / g, 12-carbon chain = 0.990 cm3 / g. DDM has been used in mixed micelle systems to study the human adenosine A2A receptor expressed and isolated from S. cerevisiae.

Full text

1
Title: A novel lipid-polymer system with unique properties has potential in drug delivery
and biotechnology applications
Authors: Monica D. Rieth,
1
* and Kerney Jebrell Glover
2
Address:
1
Department of Chemistry, Southern Illinois University Edwardsville, 44 S. Circle
Dr. Edwardsville, IL 62026, U.S.A. Phone: 618-650-3561;
email: mrieth@siue.edu
2
Department of Chemistry, Lehigh University, 6 E. Packer Ave., Bethlehem, PA
18015, U.S.A. Phone: 610-758-5081; e-mail: kjg206@lehigh.edu
*Correspondence should be addressed to Monica D. Rieth, Ph.D., phone: 618-650-3561;
Fax: 1-618-650-3556; e-mail: mrieth@siue.edu
Abbreviations: DMPC, 1,2-dimyristoyl-sn-glycero-3-phosphocholine; DHPC, 1,2-dihexanoyl-sn-
glycero-3-phosphocholine; C
8
E
5
,
n-octylpentaoxyethylene; NBD-DMPE (1,2-dimyristoyl-sn-
glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (ammonium salt));
31
P-
NMR, phosphorus nuclear magnetic resonance spectroscopy; D
2
O, deuterium oxide; CHAPS, 3-
[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; CHAPSO, 3-([3-
Cholamidopropyl]dimethylammonio)-2-hydroxy-1-propanesulfonate; HEPES, 2-[4-(2-
hydroxyethyl)piperazin-1-yl]ethanesulfonic acid
.CC-BY-NC-ND 4.0 International licensea
certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under
The copyright holder for this preprint (which was notthis version posted October 7, 2018. ; https://doi.org/10.1101/437327doi: bioRxiv preprint

2
Abstract
The utility of detergent micelle and bicelle systems has been demonstrated to be a
valuable tool for the study of membrane protein interactions and in structural studies. Bicelles are
distinguished from micelles in that they contain a lipid bilayer that mimics the plasma membrane
of cells making it more native-like than its detergent micelle counter-part. Bicelles are typically
comprised of a long-chain phospholipid such as 1,2-dimyristoyl-sn-glycero-3-phosphocholine
(DMPC) and either a short-chain phospholipid, typically 1,2-dihexanoyl-sn-glycero-3-
phosphocholine (DHPC), or a bile-salt derivative such as CHAPS or CHAPSO. In solution DMPC
and DHPC bicelles assume a discoidal structure comprised of a heterogeneous arrangement
where the short-chain lipids gather around the rim of the disk and the long-chain lipids form the
flat, bilayer region of the bicelle. Aside from DHPC, CHAPS and CHAPSO few other detergents
have reportedly been investigated for their ability to form bicelles with DMPC. In this study, the
detergent, C
8
E
5
, was used to prepare mixtures with DMPC to determine if it adopts properties
similar to DMPC-DHPC bicelles. Mixtures were evaluated using sedimentation equilibrium,
31
P-
phosphorus NMR, and light scattering and compared to DMPC-DHPC bicelles. Interestingly,
mixtures of DMPC and C
8
E
5
assumed a spherical-shaped micellar structure, not the predicted
discoidal shape. DMPC-C
8
E
5
mixtures retain interesting properties rendering them particularly
advantageous in studies of membrane protein interactions and hold promise as vehicles for drug
delivery.
Keywords: bicelles; micelles; lipids; sedimentation equilibrium; density matching; light scattering
.CC-BY-NC-ND 4.0 International licensea
certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under
The copyright holder for this preprint (which was notthis version posted October 7, 2018. ; https://doi.org/10.1101/437327doi: bioRxiv preprint

3
1. Introduction
It has been well-established that 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), a
naturally derived phospholipid with a 14-carbon chain length and a choline head group,
spontaneously assembles into discoidal lipid structures called bicelles when mixed with short-
chain phospholipids such as 1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC), also naturally
derived but containing a 6-carbon chain length with a choline head group
1
. These bilayered
structures can also be prepared using DMPC and bile salt detergents such as CHAPS or
CHAPSO in lieu of a short-chain phospholipid [1-3]. DMPC-DHPC bicelles are among the most
extensively characterized bicellar structures, and they have been used successfully as membrane
mimics in solution NMR studies of membrane protein structure [3,6]. These structures can be
tailored to different applications by adjusting the type of lipid used. For example, the lipid tail
length and the ratio of DMPC to DHPC, called the q  value, can affect the resulting physical
properties of bicelles. Anionic lipids can also be included to introduce a negative charge at the
bicellar surface
7
. Collectively, these properties affect the thickness and the planar length of the
bicellar structure
8
. Bicelles are especially interesting for membrane protein studies because,
unlike their detergent counterpart, they contain a bilayered planar region, which captures a true
membrane-like environment [3,4]. Membrane protein structure and function is dependent on the
surrounding detergent or lipid environment, therefore, choosing the optimal detergent or lipid
system is critical and usually necessitates a screening of potential candidates due to the fact that
not all systems are universally adaptable to all membrane proteins [9,12]. In this study, we
present a novel detergent-lipid aggregate system prepared with DMPC, a long-chain
phospholipid, and a polymer-based detergent system, C
8
E
5
. We compared and contrasted this
system with the DMPC-DHPC bicellar system to determine if some of the important properties
have been retained in this new system. We highlight several advantages of this new
.CC-BY-NC-ND 4.0 International licensea
certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under
The copyright holder for this preprint (which was notthis version posted October 7, 2018. ; https://doi.org/10.1101/437327doi: bioRxiv preprint

4
detergent/lipid system, particularly in the study of membrane protein interactions and vehicles for
drug delivery.
Sedimentation equilibrium (analytical ultracentrifugation) is a method that can be used to
characterize the oligomeric state of membrane proteins based on a technique called density
matching. In this technique the density of a buffered solution is matched to a lipid aggregate or
detergent micellar solution using one of three density modifiers: D
2
O, glycerol, or sucrose. The
amount of the density modifier required is dependent on the choice of detergent or lipids used as
well as the composition for mixed micellar systems. This approach is especially useful in studies
of membrane protein oligomerization where direct knowledge of detergent binding is lacking
[13,14]. Detergent micelles have been successfully density-matched in sedimentation equilibrium
experiments using the density modifier, deuterium oxide (D
2
O) and to our knowledge this has not
been done comprehensively for DMPC-DHPC bicelles until now using three biocompatible density
modifiers; D
2
O, glycerol and sucrose
15
. In practice certain lipid aggregates or mixed micellar
systems can give rise to complexes with very high densities making density matching near
impossible without using expensive reagents, i.e. D
2
O
18
,
or the addition of large quantities of
density modifiers, which can be susceptible to gradient formation under prolonged periods of
centrifugation
16
. We sought to find an alternative to DMPC-DHPC bicelles and other mixed
micellar systems, for example, the DDM/ CHAPS/ CHS (n-dodecyl--D-maltopyranoside/ 3-[(3-
cholamidopropyl)dimethylammonio]-1-propanesulfonate/ cholesteryl hemisuccinate) mixed
micelle system, which has reportedly been used for the preparation and purification of the human
adenosine A
2A
    Saccharomyces cerevisiae [17,19]. Due to the
presence of the maltose headgroup on DDM in this reported mixed micelle system, we predict a
lipid aggregate comprised entirely of DDM to have a specific volume of approximately 0.806 cm
3
/ g arising from the constituent molecular component densities of 0.622 cm
3
/ g from glucose
(maltose is a disaccharide of two glucose units) and 0.990 cm
3
/ g arising from the reported
.CC-BY-NC-ND 4.0 International licensea
certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under
The copyright holder for this preprint (which was notthis version posted October 7, 2018. ; https://doi.org/10.1101/437327doi: bioRxiv preprint

5
specific volume of the 12-carbon lipid tail (Figure 1B). These values can be obtained from the
method described by Durchschlag et al. and in actuality these values vary depending on the
organization and assembly of the lipid aggregate [20, 21]. Although these values are only
estimates, they are sufficiently valid for our study. In the DMPC-DHPC bicelle system, we found
that DHPC imparts a substantial density to the entire bicellar aggregate necessitating the use of
density modifiers. Using the density matching approach in studies of membrane protein
interactions, the contribution of the lipid aggregate to the measured buoyant molecular weight of
an incorporated membrane protein is negated and the oligomeric state of the target membrane
protein can be determined. We sought to investigate the physical properties of a novel lipid
aggregate system with a lower density, yet preserving the bilayer properties of the bicellar
mixtures using DMPC and the detergent, n-octylpentaoxyethylene (C
8
E
5
). C
8
E
5
has a reported
specific volume of 0.993 cm
3
/ g
22
. C
8
E
5
has been used to study membrane protein interactions
in the absence of DMPC in sedimentation equilibrium experiments. With a partial specific volume
(inverse density) substantially similar to water, C
8
E
5
requires little need for density matching
13
.
In the following study, DMPC-C
8
E
5
mixtures were prepared and investigated and their physical
properties compared to DMPC-DHPC bicelles. The size and shape of these lipid-detergent
structures was also investigated using the method of Mazer and co-workers 1980, and later used
by Glover and co-workers to evaluate the structure of DMPC-DHPC bicelles [3,23]. This approach
along with analysis by light scattering was used to predict the shape of DMPC-C
8
E
5
lipid-detergent
aggregates to determine if they assume a discoidal arrangement of lipids and detergent molecules
similar to DMPC-DHPC bicelles. We found that interestingly, these lipid aggregates assume a
shape that is closer to spherical than to discoidal and the properties we observe for DMPC-DHPC
bicelles are not preserved in DMPC-C
8
E
5
aggregates. In and of themselves, DMPC-C
8
E
5
aggregates have very interesting properties and warrant further study for their potential as a
platform in membrane protein analysis and as a vehicle for drug delivery.
.CC-BY-NC-ND 4.0 International licensea
certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under
The copyright holder for this preprint (which was notthis version posted October 7, 2018. ; https://doi.org/10.1101/437327doi: bioRxiv preprint

Frequently asked questions

Q1. What contributions have the authors mentioned in the paper "A novel lipid-polymer system with unique properties has potential in drug delivery and biotechnology applications" ?

In this paper, a detergent-lipid aggregate system was proposed for drug delivery using a mixture of detergent and long-chain phospholipids. 

Q2. What are the future works mentioned in the paper "A novel lipid-polymer system with unique properties has potential in drug delivery and biotechnology applications" ?

This lipid-detergent system can be especially advantageous when evaluating small membrane proteins, which can be more challenging to study given their low density relative to water and requiring greater centrifugal speeds in sedimentation equilibrium analyses. These mixtures are unique to themselves and warrant further study to better understand their behavior at different temperatures and under different physical conditions. The authors can also conclude from these studies that there are unique conditions and properties that lipids and detergents molecules must have for their spontaneous assembly into bicelles and not all detergents are suited to adopt this geometric conformation.