drug delivery through the skin--molecular simulations of barrier lipids_huzil_2011

7/23/2019 Drug Delivery Through the Skin--Molecular Simulations of Barrier Lipids_Huzil_2011

1/14

Overview

Drug delivery through the skin:molecular simulations of barrier

lipids to design more effectivenoninvasive dermal andtransdermal delivery systems forsmall molecules, biologics, andcosmeticsJ. Torin Huzil1,2, Siv Sivaloganathan2, Mohammad Kohandel2

and Marianna Foldvari1

The delivery of drugs through the skin provides a convenient route of administra-tion that is often preferable to injection because it is noninvasive and can typicallybe self-administered. These two factors alone result in a significant reduction ofmedical complications and improvement in patient compliance. Unfortunately, asignificant obstacle to dermal and transdermal drug delivery alike is the resilientbarrier that the epidermal layers of the skin, primarily the stratum corneum,presents for the diffusion of exogenous chemical agents. Further advancementof transdermal drug delivery requires the development of novel delivery systemsthat are suitable for modern, macromolecular protein and nucleotide therapeuticagents. Significant effort has already been devoted to obtain a functional under-standing of the physical barrier properties imparted by the epidermis, specificallythe membrane structures of the stratum corneum. However, structural obser-vations of membrane systems are often hindered by low resolutions, making itdifficult to resolve the molecular mechanisms related to interactions between lipidsfound within the stratum corneum. Several models describing the molecular dif-fusion of drug molecules through the stratum corneum have now been postulated,where chemical permeation enhancers are thought to disrupt the underlying lipidstructure, resulting in enhanced permeability. Recent investigations using biphasicvesicles also suggested a possibility for novel mechanisms involving the formationof complex polymorphic lipid phases. In this review, we discuss the advantagesand limitations of permeation-enhancing strategies and how computational simu-lations, at the atomic scale, coupled with physical observations can provide insightinto the mechanisms of diffusion through the stratum corneum. 2011 John Wiley

& Sons, Inc.WIREs Nanomed Nanobiotechnol2011 3 449462 DOI: 10.1002/wnan.147

Correspondence to: [email protected] of Pharmacy, University of Waterloo, Waterloo, Ontario,Canada2Department of Applied Mathematics, University of Waterloo,Waterloo, Ontario, Canada

DOI: 10.1002/wnan.147

INTRODUCTION

The choice regarding a modality used for drugadministration is dependent on several factors,including the active substance in question, itspharmacokinetic profile, and the desired locationof action.1 Because they are noninvasive, theadministration of drugs by mouth, inhalation, or

Volume 3, September/October 20 11 2011 John Wiley & Son s, Inc. 449


2/14

Overview wires.wiley.com/nanomed

directly to the skin is often preferred to those requiringinjection into the circulatory system. However,while these delivery methods may be preferable,they do have several disadvantages over injections.For instance, oral administration requires efficientabsorption through the gastrointestinal (GI) tract

and a drug must therefore be resistant to the harshphysicochemical environment present in both thestomach and intestines. The oral delivery of drugs alsoexposes them to first-pass metabolism within the liver,often reducing their bioavailability significantly. Theinhalation of aerosol drugs eliminates both exposureto the GI track and first-pass metabolism; however,the difficulty to accurately meter dosage, coupledwith the requirement for complex delivery devices,generally limits its use to drugs targeting the lungsdirectly. Dermal delivery of drugs is complicatedbecause the skins natural role as a barrier to externalcontaminants often requires the use of some methodof physical or chemical disruption.

Many next-generation therapeutic agents,including recombinant proteins and other biologics,such as antibodies and nucleic acids are not amenableto oral, inhalational, or transdermal administrationbecause they are either susceptible to enzymatic degra-dation or cannot be absorbed through the epidermisinto the skin efficiently due to their molecular size.2,3

Many of these biological drugs can only be adminis-tered using injections, thereby avoiding the difficultiesassociated with oral, inhalational, and transdermaldelivery, while providing immediate bioavailability at

all locations in the body. However, because injectionsdamage the protective layers of the epidermis, thereis a strict requirement for asepsis in order to avoidthe introduction of pathogens, such as hepatitis, orhuman immunodeficiency virus (HIV). If not adminis-tered correctly, injections can also introduce contam-inants, such as insoluble drug or adjuvant particlesand potentially fatal air boluses into the circulatorysystem.

Numerous transdermal methodologies have nowbeen explored in an attempt to improve the nonin-vasive delivery of macromolecular drugs, including

proteins such as insulin,4

interferon- (IFN-),5

andDNA.6 Novel, transdermal approaches to the deliveryof these macromolecules include pulmonary, buccal,rectal, and vaginal administration methods. Intra-and transdermal, sustained-release delivery of macro-molecules would have many advantages comparedwith administration by injection and other short-acting dosage forms. However, the lack of efficient andsafe methods for compounds with molecular weightsgreater than 500 Da greatly limits the applicability ofmany of these, noninvasive, methods.

DERMAL AND TRANSDERMAL DRUG

DELIVERY

Delivery of drugs through the skin can be dividedinto two classes, which are associated with distinctpurposes. The first class involves delivery into the skin

itself for dermatological treatment, vaccination, orcosmetic applications, and is termed dermal delivery.Transdermal delivery, on the other hand, while alsousing the skin as the application site, introduces thedrug for transport into the circulatory system. Thedelivery of drugs through the skin provides a con-venient route of administration that bypasses the GItract, first-pass metabolism, and many of the com-plications associated with injectable drugs. However,because the skin is extremely effective at protecting thebody from external pathogens and toxins, both dermaland transdermal delivery systems must be designed tocircumvent its barrier properties.7 Despite the rapid

growth of new delivery technologies, the compli-cations associated with noninvasive introduction ofdrugs into the body are clearly evident by the fact thatas of 2008 only 20 transdermal drug formulationshad been approved by the Food and Drug Administra-tion (FDA).8 For dermal and transdermal delivery tobecome amenable for use with a wider range of next-generation therapeutic agents, a clear understandingof the mechanisms associated with the barrier proper-ties of the skin must first be developed and overcome.

SKIN STRUCTUREThe skin is composed of three primary layers: theepidermis, dermis, and subcutaneous tissue (Figure 1).The stratum corneum (SC) is the outmost layerof epidermis, providing the skin with its resilientabsorption barrier properties. The SC is composedof anywhere from 1060 layers of flattened, nonlivingcorneocytes which are almost entirely made up ofcross-linked keratin (7585% of the SC is keratin)surrounded by an intercellular matrix composedprimarily of long-chain ceramides, free fatty acids,triglycerides, cholesterol, cholesterol sulfate, and

sterol/wax esters.9,10 In the conceptualized bricksand mortar model of the SC, bricks correspond tohydrophilic corneocytes and the mortar is representedas intercellular spaces containing hydrophobiclipids.1114 It is generally thought that absorptionmay occur through (1) an intercellular route, typicalof lipophilic substances; (2) the appendages (hairfollicles and sweat ducts); or (3) an intracellularroute, more typical of hydrophilic substances. Recentstudies, describing the partitioning of lipophilicand hydrophilic molecules with corneocytes, showed

450 2 01 1 J oh n W il ey & S on s, I nc. V ol um e 3, S ep te mb er/O ct ob er 2 01 1


3/14

WIREs Nanomedicine and Nanobiotechnology Modeling drug delivery through the skin

Skin strucutre

Straum corneum Permeation pachwaysIntercellular

40 m

Intracellular

100 nmIntensity

corneo

cyte

40 nm

Viableepidermis

Dermis

Crystalline

0.1

Intensity(A.U.)

0.2

q (-1)

q (-1)

biphasic vesicle-treated SC

liposome-treated SC

emulsion-treated SC

bare SC

0.3

Mixed polymorphic-to-cubic phase change

Pn3m D-surface

Lamellar-to-Lamellarphase change

Disordering, fluidizationor phase separation

Effect on SC lipidSAXS Pattern

Crystalline

Fluid

0.05 0.10

6.3 nm4.6 nm 3.3 nm 2.3 nm

0.15 0.20 0.25 0.30

bare SC

Proplylene glycol

EthanolOleic acidTranscutol

0.35

F I G U R E 1 | Human skin structure and effect of delivery systems and permeation enhancers on permeation of drug molecules. Three-dimensional

(3D) structure of human skin (left column): first-generation approach to transdermal delivery is limited primarily by the barrier posed by skinsoutermost layer called the stratum corneum, which is 1020 m thick. Underneath this layer is the viable epidermis, which measures 50100 m and

is avascular. Deeper still is the dermis, which is 12 mm thick, and contains a rich capillary bed for systemic drug absorption and nerve endings just

below the dermalepidermal junction. Chemical permeation enhancers and drug delivery systems interact with the intercellular lipids in the stratum

corneum. This interaction modifies the structural order of the lipids which are originally arranged in multiple bilayer stacks (left column) (electron

micrograph and bilayer model: Reprinted with permission from Ref 16. Copyright 2006 Elsevier). This modification, fluidization, disorder, or

rearrangement of the lipids can be monitored by small-angle X-ray scattering (SAXS)/wide-angle X-ray scattering (WAXS) (middle column).

Conventional permeation enhancers such as ethanol, oleic acid, propylene glycol and a marketed enhancer, Transcutol, cause disordering of the

lipids, but do not disrupt the bilayer configuration (right column). Among the lipid-based delivery systems, liposomes and a submicron emulsion also

have a disordering effect, whereas biphasic vesicles appear to cause rearrangement of the organization of the stratum corneum lipids into a Pn3m

cubic phase configuration (SAXS pattern: middle column, model: right column). This cubic phase could be an intercellular permeation nanopathway

that may explain the increased delivery of interferon-(IFN-) by biphasic vesicles.17

significant interactions with the cell surface.15

However, a precise description of the mechanismfor drug permeation requires the detailed analysisof molecular interactions in the skin.

Two prominent models have been developed toexplain the formation and molecular organization ofSC lipids. The Landmann model18 postulates thatlamellar granules, produced by granular cells in theviable epidermis, extrude into the extracellular spacesto form uninterrupted sheets in a bilayer configura-tion. This model was later confirmed when identicalpatterns of lamellar disks (stacks of flattened lipo-

somes), found inside granular cells, were observed inthe intercellular space.19 The bilayer structure of theseintercellular lipids is made up of two stacked lipidbilayers composed of different ceramide molecules(CER1CER8) in an interdigitated arrangement, hav-ing a periodicity of approximately 13 nm.20,21 Morerecent studies indicate that lipids extruded by granularcells during barrier lipid formation are also found inother polymorphic phases, including a cubic phase,which merges into the intercellular lamellar space.22

This is explained in the membrane folding model,

which implies that the flexibility of lipid arrangement,through estimation of energy requirements for inter-cellular lipid channel formation via cubiclamellarphase transition, is more favorable than via lamel-lar granule fusion.23 This model opened new areasof research into the mechanisms of drug permeationthrough the skin and how effectively lipids in the SCcan be manipulated to enhance drug delivery. As dis-cussed in the next section, most permeation enhancersdisturb, disorder, fluidize, integrate into, or extractlipids to various extents, which result in an increasein absorption of some drugs.

COMPOSITION OF STRATUM

CORNEUM LIPIDS

While its composition varies depending on age andlocation on the body, the SC consists primarily offree fatty acids (mainly C22 and C24, 1525%),long-chain ceramides (CER1CER8) (mostly C2426,3550%, and CER1 and CER4 with C3032),cholesterol (1525%), and cholesterol sulfate



4/14


(510%).10,24,25 Of the ceramide species, those with-hydroxy fatty acids, ester linked to linoleic acid, andamide linked to sphingosine [Cer(EOS)] predominate,and tend to be highly enriched in linoleic acid (C18:2),constituting a minimum of 2030% of the -esterifiedfatty acid.9,26 The predominant free fatty acids are

primarily saturated and range in chain lengths fromC14 to C36, with the longer chains present in higheramounts. It is thought that the long-chain lengths andcompact stacking of the ceramides and fatty acidsare the primary determinants of the extremely stablephysical properties of the SC lipid bilayers.

MOLECULAR ORGANIZATION

OF THE STRATUM CORNEUM LIPIDS

Understanding the effect that each of the ceramideand fatty acid classes play in the organization of the

SC is limited due to a lack of high-resolution electron

density maps below a resolution of approximately10 A.27 At this level of resolution, gross molecu-lar orientations and distinct phases of lipid packingare distinguishable; however, atom positions are notresolvable. Many other techniques, including infrared(IR) spectroscopy,28,29 atomic force, confocal, and

two-photon excitation fluorescence microscopy,30have been used to successfully study the physico-chemical structure of the lipid phases in the SC andinvestigate changes in their organization under differ-ing physical parameters. Arguably, the most advan-tageous technique for observing the phase behaviorof lipids has been X-ray scattering including small-(SAXS) and wide-angle (WAXS) methods (Figure 2).These techniques demonstrate that SC lipids are mostoften observed in a crystalline orthorhombic packinglattice,31 composed of two distinct lamellar phaseswith a repeat distance of approximately 6 and 13 nm

and a bilayer orientation that is parallel to the surface

CCDCamera

(a)

(b)

(c) (d)

Scattered beamIncident beam

Synchrotron ring

q=

d=

d=q

l

Intensity(

AU)

q()-10.05 0.10 0.15

3.3 nm2.3 nm

4.6 nm

6.6 nm

0.20 0.25 0.30

l

;

;

4p

sinq

2sin q

2p

Sample

F I G U R E 2 | Application of small-angle X-ray scattering (SAXS)wide-angle X-ray scattering (WAXS) for nanostructure analysis of stratum

corneum lipids. Schematic (a) and experimental setup (b) of a SAXS measurement (enlarged picture shows the multisample holder developed

in-house for the stratum corneum samples); typical scattering pattern (c); and scattering curve of human stratum corneum (d).



5/14


of the SC.16,21 There are two types of disorder thatcan be distinguished within the SC lipid regions: short-range disorder of the alkyl chains contained within asingle lipid bilayer and long-range disorder in thearrangement of lipids between multiple lamellae thatresults in changes to the space groups for lipid packing.

The high concentration of long-chain ceramidesand fatty acids found in the SC suggests that thebulk of barrier lipids resides in the crystalline state,which is separated by a liquid crystalline or gelstate.32 It has been proposed that the distributionof crystalline and gel phases of the membrane lipidsallows diffusion of small molecules across the SC andinto the dermis.33 Membranes in the liquid-crystallinestate would allow water to pass through, whilebilayers in the gel state would provide an effectivebarrier to water penetration. Other models include theformation of polar pathways across the SC throughwhich hydrophilic molecules may freely diffuse.34

DIFFUSION ACROSS THE STRATUM

CORNEUM

It is generally accepted that molecular diffusionthrough the SC requires flow through the intercellular

lipid and constrained interlayer waters via some con-voluted path. This transport pathway is highly influ-enced by the structure and solubility of the moleculeas it passes through the heterogeneous lipid bilay-ers. Many different methods can be used to enhancethe permeation of drugs across the skin (Figure 3).

Physically bypassing the SC is the simplest technique,creating both macroscopic and microscopic pathwaysthrough which molecules can penetrate. Techniquessuch as electroporation35,36 and iontophoresis37 uti-lize voltage gradients to introduce a disruption ofthe SC. Pretreatment of the skin in this way isthought to enhance the passage of large, polarmolecules, such as peptides through the SC. Cur-rently, the most successful application of iontophore-sis is the intradermal administration of lidocaineas a local anesthetic prior to dermatological pro-cedures and administration of intravenous drugs.38

Unfortunately, the equipment required for these tech-niques are large, limiting their availability to clinicalapplications.

Alternatively, technologies that puncture orabrade the outer layers of the epidermis includemicroneedles (10100m in length), where the drugis coated on the microneedle surface to aid in rapidabsorption,39 dermal abrasion,40 and needle-free

Skin permeationenhancerstrategies

Delivery systemsOptimization of drug

and/or vehicle

Stratum corneum

manipulation

Stratum corneum

bypass

Removal by stripping,abrading or depilatories

Hydration

Solvent extraction oflipids

Prodrug approach

Supersaturation(concentration gradient)

Liposomes

Micro- andnanoemulsions

Nanoparticles

Polymeric systems

Targeting systems tohair follicles

Conjugation withhydrophobic moieties

Ion pairs

Eutectic mixturesChemical permeationenhancers

Ultrasound

Iontophoresis

MagnetophoresisThermoporation

Needle-free andballistic injections

Microneedles

Photomechanicalwave

Electroporation

Physical methods Complex formulations

Synergisticenhancers

Specializedpatches

Inkjet/microneedle

Physical methodand delivery

system

Pheroid particles

Biphasic vesicles

Micro- andnanoemulsions

Device andformulation

Combination

approaches

F I G U R E 3 | An overview of skin permeation enhancer strategies. A representative flowchart illustrating several skin permeation enhancer

strategies. Subcategories include stratum corneum (SC) bypass, SC manipulation, drug vehicle optimization, delivery systems, and combination

approaches. Methods in bold under SC manipulation are also considered physical methods.



6/14


high-pressure injection.41 Yet other technologies usemagnetophoresis, sonophoresis, and photomechanicalwaves42 to apply electromagnetic, ultrasonic, ormechanical energy to the skin; however, thesetechniques have met with limited success as methodsfor transdermal drug delivery43 and are often used

as complementary techniques to existing, passiveenhancement systems.

CHEMICAL PENETRATION

ENHANCERS

Chemical penetration enhancers (CPEs) are a groupof pharmacologically inactive compounds, whichreversibly alter the barrier properties of the skin.Current application of chemical enhancers for drugpermeation enhancement is based mostly on in vitroor in vivo screening studies that quantitatively mea-

sure the extent of enhancement. More than 300chemical enhancers can be classified into three essen-tial groups based on their mechanism of permeationenhancement.44,45 Group 1 enhancers extract skinlipids or damage the SC, thereby weakening the bar-rier. Examples include solvents (e.g., ethanol) andorganic acids (e.g., salicylic acid). Group 2 enhancersincrease drug solubility within the skin. Examplesinclude polyols (e.g., propylene glycol). Group 3enhancers disorder intercellular lipids. Examplesinclude terpenes, surfactants, fatty acids, fatty acidesters, Azone (1-dodecylazacycloheptan-2-one) andits derivatives, amides (e.g., dimethylformamide), andsulfoxides [e.g., dimethylsulfoxide (DMSO)]. TheAzone-like enhancer mechanism of action has beenmost extensively characterized, showing how the par-titioning of this molecule into the skin creates adisturbance of the lipid headgroups of skin ceramides,thereby enhancing permeation for many small drugmolecules.46 Surfactants also tend to penetrate intointercellular spaces, increasing lipid phase fluidity anddecreasing the resistance to permeation.4749 In gen-eral, absorption enhancement through lipid channelsdepends on the ability of the enhancer to integrate withthe existing lipids and create a perturbed microenvi-

ronment based on mechanisms, such as (1) alterationof lipid phase fluidity, (2) enhancement of solubil-ity characteristics of the skin for the drug to bedelivered, (3) creation of a disordering effect amongthe alkyl chains of skin lipids, and (4) localizedseparation of lipid domains to create hydrophilicpores.5052 One of the first-generation transdermaldelivery mechanisms used liposomes to encapsulatedrugs.53,54 Liposome-encapsulated drugs are able tomerge with skin lipids, facilitating the delivery of theirpayloads into the skin.14 Recently developed delivery

systems including deformable liposomes,55 synergisticcombinations of enhancers, nanoparticles,56 biphasicvesicles, and pheroids5658 (Figure 3). These systemsshow increasing potential for delivering a wider selec-tion of drugs including drugs with higher molecularweights.

As drugs are primarily transported through theintercellular lipid regions of the SC, the structuralmechanisms of lamellar disruption in the presenceof CPEs and delivery systems are now of greatinterest for the development of rational strategies toenhance drug penetration through the skin. However,as the overall barrier properties of the skin are notfundamentally changed during the use of CPEs, theyare commonly utilized to offer improvement to factorssuch as dose control, and not to widen the overallapplicability of transdermal drug delivery. While themolecular size of potentially deliverable drugs is likelyto increase with further research, larger compounds,such as those used in vaccines and gene therapy,continue to be problematic when used in combinationwith CPEs.

PHASE TRANSITIONS IN THE

STRATUM CORNEUM

Like the phospholipid bilayer surrounding a livingcell, the membranes found in the SC must pos-sess dynamic properties, otherwise they would beimpenetrable barriers across which not even watercould pass. While phase changes in membrane bilay-ers composed of phospholipids have been extensivelystudied, the phase behaviors of lipids found in theSC have only been studied since the late 1970s.59 Agreat deal of what we now know about the molec-ular organization and structure of lipid bilayers andhow they respond to the introduction of exogenouschemicals comes from the development of syntheticsystems such as liposomal bilayers.13 The most com-mon phases found in hydrated phospholipid andceramide artificial membrane systems include fourliquid-crystalline and crystalline lamellar phases andseveral nonlamellar phases. The nonlamellar phases

include both hexagonal and cubic phases, wherehexagonal phases are formed by tubular aggregatesand are either normal or reverse hexagonal liquid-crystalline phases and cubic phases, including cubic,rhombic, and tetragonal, that are formed by the inter-action of curved bilayers. Recent studies have shownthat lipids extruded by granular cells during barrierlipid formation in the epidermis may also exist in otherpolymorphic states such as a cubic phase, which thenmerge into the lamellar intercellular space.22,23 Theseobservations support the idea that a transition from



7/14


lamellar to these nonlamellar phases could also occurwithin the SC lipids themselves, producing aqueouschannels through which drug molecules could freelydiffuse.

We have recently described a novel formula-tion consisting of biphasic vesicles designed for the

delivery of IFN-, a 19-kDa protein for the topicaltreatment of human papillomavirus infections.60 Aspart of the mechanistic studies related to these sys-tems, we have conducted SAXS and WAXS analysisof the SC lipids structural arrangement after treat-ment with biphasic vesicles.17 These studies showeda previously unobserved effect on the structural orderof the SC lipids, namely the induction of a poly-morphic rearrangement of the lipids as a result ofbiphasic vesicle interaction (Figure 1). SAXS andWAXS studies showed that biphasic vesicles inducedthe formation of a three-dimensional bicontinuousPn3m cubic phase in the SC lipids. Formation ofthis cubic phase seems to be unique to biphasicvesicles, as their subcomponents, liposomes and sub-micron emulsion, or commonly used CPEs do notinduce such a polymorphic phase change (Figure 1).The cubic phase is a complex three-dimensional net-work of independent aqueous channels within a lipidmatrix on an infinite periodic minimal surface. Theinteraction of biphasic vesicles also induces the con-version of the more commonly observed orthorhombiclipid-packing configuration (which provides a strongbarrier to permeation of substances) to a more per-meable hexagonal configuration or liquid state. The

combination of these two effects may be responsi-ble for increasing the permeability of IFN- acrossthe SC.

Ultimately, the presence of a particular spacegroup in the supramolecular organization of the SCdepends on several factors: (1) the chemical structureand amount of lipid present, (2) the water contentof the system, (3) the presence of other solutes and(4) temperature. Because of the relationship betweenlipid structure and aggregate form, it is not surpris-ing that complex lipid structures behave in unusualways. The tendency to produce these larger transitions

also depends on the local curvature of the mem-brane, which ultimately depends on the local lipidcomposition and fluidity of the membrane. Alteringindividual physical parameters such as temperature,ceramide composition, and cholesterol or fatty acidconcentrations results in the generation of many ofthese complex phases in SC lipid lamellae.17,23,61

For instance, the addition of cholesterol to a fluid-phase bilayer results in its intercalation betweenlipid molecules, filling in free space and decreas-ing the flexibility of surrounding lipid chains.62

Unfortunately, owing to its relatively disorderednature, a lack of available molecular detail relatedto the physical properties of the SC makes develop-ing models and understanding the diffusion processdifficult.

MOLECULAR MODELING

To advance our understanding of the physicochem-ical properties related to drug diffusion, includingthe effect of both drug and vehicle on permeationacross the SC, it is essential to first determine the pre-dominant mechanisms for drug penetration throughthe SC. How this knowledge is obtained depends onseveral factors. Technological advancements in molec-ular imaging will undoubtedly increase the overallresolution at which researchers can observe lipids inthese systems. However, obtaining data related to the

atomic structures of these systems requires technologynot yet available. Because molecules are dynamic, itis often difficult to extrapolate atomic motions fromexperimentally derived structures. This applies to themovement of membrane leaflets, including motionsbetween leaflets, the bulk motion of lipid molecules,and motion of atoms within the lipid itself.63 Onepossible way of observing these interactions is thedevelopment of computational models where physi-cal parameters available in the literature can provideinsight into the intrinsic molecular behavior of thesecomplex lipid systems.64 Motions of each atom withinthe system affect the energy of the molecules, whichcan be calculated at a given time, given the relativeatom positions using classical simulation techniquessuch as molecular mechanics (MM). Molecular mod-eling allows for the observation of molecular con-formations at timescales that are difficult to obtainthrough experimental analysis alone. The principalsimulation technique used to examine realistic molec-ular systems is known as molecular dynamics (MD).MD has its underpinnings in statistical mechanics,where it is assumed that statistical ensemble aver-ages are identical to time averages of the system. Byutilizing several, well-established, approximations, a

real-time depiction of atomic motion in these sys-tems can be collected. Once a statistically adequatesample has been obtained, detailed atomic interac-tions within the system can be evaluated and energiescalculated.

Modeling lipid interactions within SC mem-branes is the first step toward obtaining a clearerunderstanding of the free-energy landscapes con-tributing to both the stabilization and destabilizationof these macromolecular complexes. The acquisitionof simulation data relating to phase transitions



8/14


associated with various lipids and penetrationenhancers will provide data required for the develop-ment of structure activity relationships (SARs) relatingthe structure of CPEs to their role in the formation ofcomplex phase transitions.

MODELING MEMBRANE SYSTEMS

To date, the majority of simulations examininglipids are used to study the structure, dynamics, andinteractions of membrane proteins and peptides.65,66

A significant number of models have been proposedthat attempt to describe the behavior of hydropho-bic drug molecules that enter the cell through passivediffusion across the phospholipid membrane.6769 So,what can researchers expect to obtain by perform-ing MD simulations targeted toward these types ofmembranes? In principle, through the use of sta-

tistical mechanics, such simulations can provide acomplete picture describing the motion of lipids andother molecules in the system, as well as access tothermodynamic properties.

It is not obvious how one relates bulk ther-modynamic partitioning experiments to partitioninginto bilayers, but detailed computer simulations canbe used to interpret these experiments, elucidate therelative importance of entropic and enthalpy con-tributions, and relate these to properties such aschain ordering, free volume distribution, polarity, andshape of the small molecules being investigated.70 Itis possible to obtain sufficiently accurate experimen-tal information to validate simulations by comparingaverage properties to experimentally measured val-ues. Simulating the aggregation of randomly dissolvedphospholipids into bilayers takes on the order of tensof nanoseconds, whereas the formation of small phos-pholipid vesicles has been observed in near-atomicdetail using simulations only 100 ns in length.71 Cubiclipid phases have been studied by MD, and a phasetransition between a cubic and an inverted hexagonalphase have been observed.72 Similar methods havealso been applied to understanding the detailed mech-anisms of ion permeation through ion channels,7375

proton exclusion in aquaporin,76,77 and the mecha-nisms of bilayer fusion.78

There are technical limitations related to the sizeand accuracy of the systems that can be simulated, anddifficulties with accurately incorporating importantvariables such as pH, transmembrane potential dif-ferences, and low concentrations of ions. In addition,because there is limited data related to the startingconfigurations of many of these lipid simulations, thechoice of initial conditions may also bias results inundesirable ways.

UMBRELLA AND REPLICA EXCHANGE

SAMPLING OF MEMBRANE SYSTEMS

A significant challenge when designing simulationsis the observation of conformational changes thatare too slow to detect on a timescale of only tensto hundreds of nanoseconds. Fortunately, extensionsto the basic simulation methods make it possibleto calculate free-energy differences from these typesof simulations.79,80 Techniques including umbrellasampling and replica exchange molecular dynamics(REMD)81 are ideally suited for the study ofmembrane systems and allow for the simulationof molecular processes that normally occur onlong timescales. However, these techniques requireprevious knowledge of a specific reaction coordinateand the convergence of all motions that are not partof the reaction coordinate.

In umbrella sampling, a biasing potential is used

to restrict a system to sample phase space withina specified region. By placing windows along thereaction coordinate, one can generate a free-energyprofile (also called potential of mean force), whichquantitatively describes why one region of space ismore favorable than another (Figure 4). By restraininga molecule at different depths within the bilayer(effectively forcing the drug molecule to spend timein unfavorable regions of the system), an energydistribution can be calculated that depends both onthe restraining potential and the actual distributionwithout the restraining potential. By utilizing a set

of such simulations to reconstruct the distributionthroughout the entire bilayer, it is possible to correctfor the addition of restraining potentials. The result isan accurate sampling of the available conformationalspace, even for cases where there is a very largedifference in free energy between the least and themost favorable positions.

REMD, also referred to as parallel tempering,is a simulation method aimed at improving thedynamic properties of Monte Carlo (MC) methodsimulations of physical systems. Typically, an MCsimulation consists of a stochastic step that evaluatesthe energy of the system and accepts/rejects updates

based on the temperature (T).82,83 If two identicalsimulations are performed at temperatures separatedby T, one would observe that if T is smallenough, the energy histograms obtained by collectingvalues of energies over a set of n MC steps willcreate two distributions that overlap. We can interpretthis overlap by assuming that system configurationssampled at temperatureT1are likely to appear duringa simulation at T2. Because the Markov chain has nomemory of its past, we can create a new update for thesystem composed of the two systems atT1and T2. At



9/14


E

yposition

1

2

3

4

5

5

4

3

2

1

F I G U R E 4 | Schematic representation of the umbrella sampling technique. To perform umbrella sampling, one must first generate a series of

configurations along a predetermined reaction coordinate. In this example, the reaction coordinate is defined by applying a constant force (y) to a

permeation enhancer and pulling it through the (x,z) plane of a membrane composed of ceramides, cholesterol and free fatty acids (dashed arrow).

Configurations generated in this way serve as the starting points for the umbrella sampling windows, which are run in independent simulations.Configurations of the system which are generated during the pulling step are extracted after the initial simulation is complete. The middle image

corresponds to the independent simulations conducted within each sampling window, with the center of mass of the permeation enhancer restrained

in that particular window by an umbrella biasing potential. The right panel illustrates an ideal histogram of configurations, with neighboring windows

overlapping such that a continuous energy function with respect to the passage of the permeation enhancer through the bilayer can later be derived

from these simulations.

any given MC step, we can update the global systemby swapping the configuration of the two systems oralternatively trading the two temperatures. Throughthe careful choice of temperatures and number ofreplicas, we can achieve an improvement in the

sampling properties of an MC simulation that greatlyexceeds the additional computational cost of runningparallel simulations.

MULTISCALE AND CONTINUUM

SIMULATIONS

Through the use of a hierarchical approach, based onparameters obtained from both experimental and MDsimulation, it is also possible to create multiscale sim-ulations that can represent extremely large membrane

systems at timescales not amenable to the previouslymentioned atomistic simulations.84,85 Multiscale sim-ulations are generally based on a three-tiered modelingapproach consisting of the generation of (1) atomisticMD simulations, (2) coarse-grained mathematical

models, and (3) validation through correlation withexperimental data. The continuum approach rep-resents biochemical systems as sets of differen-tial equations where the concentrations of chemi-cal species are assumed to vary continuously andsmoothly across the reaction space. This approachdraws on the rather extensive foundation of mathe-matical and engineering theory that exists to solvesuch systems, but does not deal with molecularspecies explicitly. For instance, the utilization ofMD simulations to obtain the microscopic diffusion



10/14


coefficient within SC lipid bilayers has been coupledwith a course-grained approach to obtain effectivemacroscopic diffusion coefficients that are in goodagreement with experimental data.86,87 Molecularmodeling procedures such as those described herewill allow (1) the elucidation of mechanisms of inter-

action between carrier molecules and drugs and(2) extraction of molecular energetic and structuralparameters to be employed in the formulation ofmathematical models of diffusion through the SC.

MODELING THE STRATUM

CORNEUM

The systematic simulation of lipids observed in the SCis limited, in part, because of the limited structural datadescribing the primary lipid composition and theirrespective role in the barrier properties of the skin. Pre-

liminary MD simulations have been targeted at study-ing components of the SC, and generally consider freefatty acids and cholesterol, but contained no ceramidemolecules due to a lack of adequate parameters.88

Because there is no universal atomic force field con-taining parameters for simulating all proteins, nucleicacids, lipids, carbohydrates, and arbitrary smallmolecules, selection of correct atomic parameters isessential. The most widely used parameters for thelipid component of typical membrane protein simula-tions in simulation packages such as GROMACS areknown as the Berger lipids.89 Because ceramides lackthe phosphate headgroup found in standard phospho-lipid parameters, new parameters need to be tested fortheir ability to replicate experiment results adequately.Initial parameters for ceramides were first introducedin simulations of single homogeneous ceramide bilay-ers composed of C16:0CER[NS].90 Here, the partialcharges for atoms within the ceramide headgroupwere taken from the side chain atoms of serine.91

Atomistic simulations of SC bilayers are now pro-gressing quickly, the first simulations involving CPEsexamined DMSO on a hydrated gel-phase bilayercomposed of C24:0CER[NS].92 In these simulations,the authors showed that DMSO accumulated in the

ceramide headgroup region, weakening lateral forcesbetween them. At high concentrations of DMSO,ceramide bilayers were observed going through aphase transition from the gel phase to the liquid-crystalline phase. Because the liquid-crystalline phaseis thought to be markedly more permeable to solutesthan the gel phase, these simulations correlated wellwith experimental results showing that high concen-trations of DMSO fluidize the SC lipids, therebyenhancing permeability. Simulations involving fullyhydrated bilayers composed of ceramide 24:0[NS],

Upper hydrationshell

Upper bilayer

composed ofceramide andcholesterol

Lower bilayercomposed ofceramide andcholesterol

Intermembranewaters

Lower hydrationshell

F I G U R E 5 | Construction of two ceramide bilayers separated by a

thin layer of water. To observe the ability of a permeation enhancer to

cross the stratum corneum and interactions at the intermembrane

interface, a bilayer assembly was created using the CHARMM c33b1

package. To aid computational efficiency, a segment having a small

cross-sectional area of 5050 A was chosen. For the initial selection

of the parameters required for ceramide equilibration, a ratio of 2:1

ceramide C15:0CER[NS] to cholesterol was used. These initial

parameters produced an upper and lower bilayer containing 72

ceramide molecules (green sticks) and 36 cholesterol molecules (blue

sticks) each. The two bilayers are separated by a 5 A layer of water

and are surrounded by a 12 A layer of water on either side of the

leaflet.

lignoceric acid C24:0, and cholesterol (Figure 5) havebeen performed to address the effect of these differ-ent SC components on its structural properties.93,94

Information collected from these simulations showedthat at physiological temperatures, the lipids were inthe gel phase with ordered lipid tails. Interestingly,the large asymmetry in the tail lengths of ceramidemolecules resulted in a fluid-like environment at thebilayer midplane, a detail proposed in earlier models

of SC permeability.95 These results have now laid thegroundwork for the simulation of multicomponentSC bilayers in the presence of libraries of CPEs.Techniques such as umbrella sampling and REMDwill undoubtedly provide an image of the structure andfunctional relationships between all the componentsof the SC lipid bilayers and their interplay duringphase changes, facilitating the rational design of new,passive transdermal delivery modalities for a widerrange of macromolecules than is currently availabletoday (Figure 6).



11/14


Experimental Computational

Obtain samples of stratumcorneum from skin orcreate synthetic systems

Assay samples and applypenetration enhanceror delivery system

Slowprocess

Using data collected fromexperiment, constructmodel stratum corneumbilayers at varying ratios

Equilibrate model systemsand introduce penetrationenhancers to the models

Fast reversibleprocess

Obtain physical parametersfor lipid phases (cubic,lamellar and hexagonal)

Observe correlationsbetween empirial andcomputational paramenters

Develop an atomic scalemodel of lipid/ enhancerinteractions

Develop a penetrationenhancement structureactivity relationship library

Saxs Waxs Molecular dynamics

Feedbackin

toexp

erim

ent

F I G U R E 6 | Synergistic feedback of computational simulations into

predictive enhancement of the experimental observation of permeation

enhancer effects. Using computational techniques it is possible to

simulate the effects that drugs and penetration enhancers may have on

components of the stratum corneum. Several parameters can be

introduced and a predictive library can be constructed. This library can

then be used to guide subsequent experimental design in an effort to

eliminate unproductive results and speed combinatorial screening of

large chemical libraries.

CONCLUSIONS

Research into the life sciences cannot achieve itsfull social, economic, and scientific potential until it

is able to explore the underlying structural designprinciples of life and how they impact an entireorganism. Predicting cellular behavior by compu-tational means must become an overarching andunifying concept that integrates the recent advancesin technology that include proteomics, genomics,

bioinformatics, and nanotechnology as necessarycomponents. Today, researchers have the advantagethat the chemical/physical properties of biomoleculescan be rigorously described, quantified, and simu-lated computationally. This advantage also extendsto the simulation of structures such as membranesand lipid mixtures, for which atomic resolution dataare not yet available. Simulations can bridge the gapthat exists between available structures and atomicdetails relating to functional relationships in lipidsystems.

An understanding of the mechanisms associ-

ated with phase transitions in membranes of theSC will undoubtedly aid in the design of new deliv-ery systems. The creation of large databases of lipidinteractions with penetration enhancers and their n-ary mixtures will support both computational andexperimental model development by limiting theoverall search space required. There is a uniqueopportunity to work with existing developmentalstage drugs that had previously been shelved dueto excessive difficulties when administered orally.We believe that new noninvasive transdermal tech-nologies suitable for difficult-to-deliver moleculescan be designed and that we are approaching a

new frontier, where the next generation of deliv-ery systems will provide enhancement to a broadrange of compounds through a structured reorderingof the SC.

ACKNOWLEDGMENTS

This article was supported by grants from the Canadian Institutes of Health Research and the Natural Sciencesand Engineering Research Council of Canada (NSERC). Research involving SAXS and WAXS was conductedat the National Synchrotron Light Source, Brookhaven National Laboratory, Upton, NY, which is supportedby the U.S. Department of Energy, Division of Materials Sciences and Division of Chemical Sciences, underContract No. DE-AC02-98CH10886. The generous support of the Canada Research Chairs Program, theCanada Foundation for Innovation, and the Ontario Research Fund for M. Foldvari is gratefully acknowledged.

REFERENCES

1. Brocks DR, Mehvar R. Rate and extent of drug accu-

mulation after multiple dosing revisited. Clin Pharma-

cokinet2010, 49:421438.

2. Xu L, Anchordoquy T. Drug delivery trends in clin-

ical trials and translational medicine: challenges and

opportunities in the delivery of nucleic acid-based ther-

apeutics.J Pharm Sci2011, 100:3852.

3. Hamman JH, Enslin GM, Kotze AF. Oral delivery of

peptide drugs: barriers and developments. BioDrugs

2005, 19:165 177.



12/14


4. Strack T. The pharmacokinetics of alternative insulindelivery systems. Curr Opin Investig Drugs 2010,11:394401.

5. Foldvari M, Baca-Estrada ME, He Z, Hu J, Attah-Poku S, King M. Dermal and transdermal deliveryof protein pharmaceuticals: lipid-based delivery sys-tems for interferon .Biotechnol Appl Biochem 1999,30:129137.

6. Yang P, Singh J, Wettig S, Foldvari M, Verrall RE,Badea I. Enhanced gene expression in epithelialcells transfected with amino acid-substituted gem-ini nanoparticles. Eur J Pharm Biopharm 2010, 75:311320.

7. Barry BW. Breaching the skins barrier to drugs. NatBiotechnol2004, 22:165167.

8. Prausnitz MR, Langer R. Transdermal drug delivery.Nat Biotechnol2008, 26:12611268.

9. Gray GM, White RJ, Williams RH, Yardley HJ. Lipidcomposition of the superficial stratum corneum cells of

pig epidermis.Br J Dermatol1982, 106:5963.

10. Lampe MA, Williams ML, Elias PM. Human epider-mal lipids: characterization and modulations duringdifferentiation.J Lipid Res1983, 24:131140.

11. Elias PM. Epidermal lipids, barrier function, anddesquamation.J Invest Dermatol1983, 80:44s49s.

12. Tojo K. Random brick model for drug transport acrossstratum corneum.J Pharm Sci1987, 76:889891.

13. Abraham W, Downing DT. Lamellar structures formedby stratum corneum lipids in vitro: a deuterium nuclearmagnetic resonance (NMR) study. Pharm Res 1992,9:14151421.

14. El Maghraby GM, Barry BW, Williams AC. Liposomesand skin: from drug delivery to model membranes. Eur

J Pharm Sci2008, 34:203222.

15. Hansen S, Naegel A, Heisig M, Wittum G, Neumann D,Kostka KH, Meiers P, Lehr CM, Schaefer UF. The roleof corneocytes in skin transport revised-a combinedcomputational and experimental approach.Pharm Res2009, 26:13791397.

16. Bouwstra JA, Ponec M. The skin barrier in healthyand diseased state. Biochim Biophys Acta 2006,1758:20802095.

17. Foldvari M, Badea I, Wettig S, Baboolal D, Kumar P,Creagh AL, Haynes CA. Topical delivery of interferon by biphasic vesicles: evidence for a novel nanopath-way across the stratum corneum. Mol Pharm 2010,7:751762.

18. Landmann L. Epidermal permeability barrier: transfor-mation of lamellar granule-disks into intercellular sheetsby a membrane-fusion process, a freeze-fracture study.

J Invest Dermatol1986, 87:202209.

19. Swartzendruber DC, Wertz PW, Kitko DJ, MadisonKC, Downing DT. Molecular models of the intercel-lular lipid lamellae in mammalian stratum corneum.

J Invest Dermatol1989, 92:251257.

20. Robson KJ, Stewart ME, Michelsen S, Lazo ND, Down-ing DT. 6-Hydroxy-4-sphingenine in human epidermalceramides.J Lipid Res1994, 35:20602068.

21. Bouwstra JA, Honeywell-Nguyen PL, Gooris GS,Ponec M. Structure of the skin barrier and its mod-ulation by vesicular formulations. Prog Lipid Res2003,42:136.

22. Norlen L. Skin barrier structure and function: the sin-gle gel phase model. J Invest Dermatol 2001, 117:830836.

23. Norlen L. Skin barrier formation: the membrane foldingmodel.J Invest Dermatol2001, 117:823829.

24. Wertz PW. Lipids and barrier function of the skin.ActaDerm Venereol2000, 208:711.

25. Bouwstra JA, Gooris GS, Weerheim A, Kempenaar J,Ponec M. Characterization of stratum corneum struc-ture in reconstructed epidermis by X-ray diffraction.

J Lipid Res 1995, 36:496504.

26. Jungersted JM, Hgh JK, Hellgren LI, Jemec GBE,Agner T. Ethnicity and stratum corneum ceramides. Br

J Dermatol2010, 163:11691173.

27. Groen D, Gooris GS, Bouwstra JA. New insights intothe stratum corneum lipid organization by X-raydiffraction analysis.Biophys J2009, 97:22422249.

28. Merle C, Baillet-Guffroy A. Physical and chemical per-turbations of the supramolecular organization of thestratum corneum lipids: in vitro to ex vivo study.Biochim Biophys Acta2009, 1788:10921098.

29. Mendelsohn R, Flach CR, Moore DJ. Determination ofmolecular conformation and permeation in skin viaIR spectroscopy, microscopy, and imaging. Biochim

Biophys Acta2006, 1758:923933.30. Norlen L, Plasencia I, Bagatolli L. Stratum corneum

lipid organization as observed by atomic force, confocaland two-photon excitation fluorescence microscopy. Int

J Cosmet Sci 2008, 30:411.

31. Bouwstra JA, Gooris GS, van der Spek JA, Bras W.Structural investigations of human stratum corneum bysmall-angle X-ray scattering. J Invest Dermatol1991,97:10051012.

32. Bouwstra JA, Dubbelaar FE, Gooris GS, Weerheim AM,Ponec M. The role of ceramide composition in the lipidorganisation of the skin barrier. Biochim Biophys Acta1999, 1419:127 136.

33. Forslind B, Engstrom S, Engblom J, Norlen L. A novelapproach to the understanding of human skin barrierfunction.J Dermatol Sci1997, 14:115125.

34. Sznitowska M, Janicki S, Williams AC. Intracellular orintercellular localization of the polar pathway of pen-etration across stratum corneum. J Pharm Sci 1998,87:11091114.

35. Weaver JC, Vaughan TE, Chizmadzhev Y. Theory ofelectrical creation of aqueous pathways across skintransport barriers. Adv Drug Deliv Rev 1999, 35:2139.



13/14


36. Sugar IP, Neumann E. Stochastic model for electricfield-induced membrane pores. Electroporation. Bio-

phys Chem 1984, 19:211225.

37. Turner NG, Kalia YN, Guy RH. The effect of currenton skin barrier function in vivo: recovery kinetics post-iontophoresis.Pharm Res1997, 14:12521257.

38. Greenbaum SS, Bernstein EF. Comparison of ion-tophoresis of lidocaine with a eutectic mixture oflidocaine and prilocaine (EMLA) for topically admin-istered local anesthesia. J Dermatol Surg Oncol1994,20:579583.

39. Henry S, McAllister DV, Allen MG, PrausnitzMR. Microfabricated microneedles: a novel approachto transdermal drug delivery. J Pharm Sci 1998,87:922925.

40. Fang JY, Lee WR, Shen SC,Fang YP, Hu CH. Enhance-ment of topical 5-aminolaevulinic acid delivery byerbium:YAG laser and microdermabrasion: a com-parison with iontophoresis and electroporation. Br

J Dermatol2004, 151:132140.41. Burkoth TL, Bellhouse BJ, Hewson G, Longridge DJ,

Muddle AG, Sarphie DF. Transdermal and transmu-cosal powdered drug delivery. Crit Rev Ther DrugCarrier Syst1999, 16:331384.

42. Lee S, Kollias N, McAuliffe DJ, Flotte T, Doukas A.Topical delivery in humans with a single photome-chanical wave.Pharm Res1999, 16:1717 1721.

43. Barry BW. Novel mechanisms and devices to enablesuccessful transdermal drug delivery. Eur J Pharm Sci2001, 14:101 114.

44. Williams AC, Barry BW. Skin absorption enhancers.

Crit Rev Ther Drug Carrier Syst1992, 9:305353.45. Chattaraj S, Walker R. Penetration enhancer classifica-

tion. In: Smith EW, Maibach HI, eds. PercutaneousPenetration Enhancers. Boca Raton, FL: CRC Press;1995, 520.

46. Hadgraft J, Pugh WJ. The selection and design of topi-cal and transdermal agents: a review.J Investig Derma-tol Symp Proc1998, 3:131135.

47. Cocera M, Lopez O, Coderch L, Parra JL, de laMaza A. Influence of the level of ceramides on the per-meability of stratum corneum lipid liposomes causedby a C12-betaine/sodium dodecyl sulfate mixture. Int

J Pharm 1999, 183:165173.

48. Shokri J, Nokhodchi A, Dashbolaghi A, Hassan-ZadehD, Ghafourian T, Barzegar Jalali M. The effect ofsurfactants on the skin penetration of diazepam Int

J Pharm 2001, 228:99107.

49. Honeywell-Nguyen PL, Bouwstra JA. The in vitro trans-port of pergolide from surfactant-based elastic vesiclesthrough human skin: a suggested mechanism of action.

J Control Release2003, 86:145156.

50. Hadgraft J, Walters KA, Guy RH. Epidermal lipidsand topical drug delivery. Semin Dermatol1992, 11:139144.

51. Hadgraft J. Recent developments in topical and trans-dermal delivery. Eur J Drug Metab Pharmacokinet1996, 21:165 173.

52. Marjukka Suhonen T, Bouwstra JA, Urtti A. Chemicalenhancement of percutaneous absorption in relationto stratum corneum structural alterations. J ControlRelease1999, 59:149161.

53. Mezei M, Gulasekharam V. Liposomesa selectivedrug delivery system for the topical route of admin-istration. I. Lotion dosage form. Life Sci 1980, 26:14731477.

54. Mezei M, Gulasekharam V. Liposomes. A selectivedrug delivery system for the topical route of admin-istration: gel dosage form. J Pharm Pharmacol 1982,34:473474.

55. Elsayed MM, Abdallah OY, Naggar VF, KhalafallahNM. Deformable liposomes and ethosomes as car-riers for skin delivery of ketotifen. Pharmazie 2007,62:133137.

56. Kuchler S, Radowski MR, Blaschke T, Dathe M, PlendlJ, Haag R, Schafer-Korting M, Kramer KD. Nanopar-ticles for skin penetration enhancementa compari-son of a dendritic core-multishell-nanotransporter andsolid lipid nanoparticles.Eur J Pharm Biopharm2009,71:243250.

57. Karande P, Jain A, Ergun K, Kispersky V, Mitragotri S.Design principles of chemical penetration enhancers fortransdermal drug delivery. Proc Natl Acad Sci U S A2005, 102:46884693.

58. Gerber M, Breytenbach JC, du Plessis J. Transdermalpenetration of zalcitabine, lamivudine and synthe-sised N-acyl lamivudine esters. Int J Pharm 2008,

351:186193.

59. Jain MK, Wu NY, Wray LV. Drug-induced phasechange in bilayer as possible mode of action of mem-brane expanding drugs.Nature1975, 255:494496.

60. Kumar P, Batta R, LaBine G, Shen J, Gaspar K,Docherty J, Foldvari M. Stabilization of interferon-2bin a topical cream.Pharm Technol2009, 33:8086.

61. Bouwstra JA, Cheng K, Gooris GS, Weerheim A,Ponec M. The role of ceramides 1 and 2 in the stra-tum corneum lipid organization.Biochim Biophys Acta1996, 1300:177 186.

62. Needham D, Nunn RS. Elastic deformation and fail-

ure of lipid bilayer membranes containing cholesterol.Biophys J1990, 58:9971009.

63. Hamilton JA, Cordes EH. Molecular dynamics of lipidsin human plasma high density lipoproteins. A high field13C NMR study.J Biol Chem 1978, 253:51935198.

64. Hadgraft J. Skin deep. Eur J Pharm Biopharm 2004,58:291299.

65. Tieleman DP, Marrink SJ, Berendsen HJ. A computerperspective of membranes: molecular dynamics studiesof lipid bilayer systems. Biochim Biophys Acta 1997,1331:235270.



14/14


66. Ash WL, Stockner T, MacCallum JL, Tieleman DP.Computer modeling of polyleucine-based coiled coildimers in a realistic membrane environment: insightinto helix-helix interactions in membrane proteins.Bio-chemistry2004, 43:9050 9060.

67. Egberts E, Marrink SJ, Berendsen HJ. Moleculardynamics simulation of a phospholipid membrane. EurBiophys J1994, 22:423436.

68. Smondyrev AM, Berkowitz ML. Molecular dynamicssimulation of DPPC bilayer in DMSO.Biophys J1999,76:24722478.

69. Tieleman DP, Leontiadou H, Mark AE, Marrink SJ.Simulation of pore formation in lipid bilayers bymechanical stress and electric fields. J Am Chem Soc2003, 125:63826383.

70. Kim EB, Lockwood N, Chopra M, Guzman O, AbbottNL, de Pablo JJ. Interactions of liquid crystal-formingmolecules with phospholipid bilayers studied bymolecular dynamics simulations. Biophys J 2005,89:31413158.

71. de Vries AH, Mark AE, Marrink SJ. Molecular dynam-ics simulation of the spontaneous formation of a smallDPPC vesicle in water in atomistic detail. J Am ChemSoc2004, 126:44884489.

72. Marrink SJ, Tieleman DP. Molecular dynamics sim-ulation of spontaneous membrane fusion during acubic-hexagonal phase transition. Biophys J 2002,83:23862392.

73. Tieleman DP, Biggin PC, Smith GR, Sansom MS. Sim-ulation approaches to ion channel structure-functionrelationships.Q Rev Biophys2001, 34:473561.

74. Beckstein O, Biggin PC, Bond P, Bright JN, Domene C,

Grottesi A, Holyoake J, Sansom MSP. Ion channel gat-ing: insights via molecular simulations. FEBS Lett2003,555:8590.

75. Aqvist J, Luzhkov V. Ion permeation mechanism of thepotassium channel.Nature2000, 404:881 884.

76. de Groot BL, Grubmuller H. Water permeation acrossbiological membranes: mechanism and dynamics ofaquaporin-1 and GlpF.Science2001, 294:2353 2357.

77. Chakrabarti N, Tajkhorshid E, Roux B, Pom`es R.Molecular basis of proton blockage in aquaporins.Structure2004, 12:6574.

78. Marrink SJ, Mark AE. The mechanism of vesicle fusionas revealed by molecular dynamics simulations. J Am

Chem Soc2003, 125:11144 11145.

79. Kollman P. Free-energy calculations - applications tochemical and biochemical phenomena. Chem Rev 1993,93:23952417.

80. Northrup SH, Pear MR, Lee CY, McCammon JA,Karplus M. Dynamical theory of activated processes

in globular proteins.Proc Natl Acad Sci U S A 1982,79:40354039.

81. Mitsutake A, Sugita Y, Okamoto Y. Generalized-ensemble algorithms for molecular simulations ofbiopolymers.Biopolymers2001, 60:96123.

82. Metropolis N, Ulam S. The Monte Carlo method.J Am

Stat Assoc1949, 44:335341.83. Hastings WK. Monte Carlo sampling methods using

Markov chains and their applications. Biometrika 1970,57:97109.

84. Hills RD, Lu L, Voth GA. Multiscale coarse-graining ofthe protein energy landscape.PLoS Comput Biol2010,6:e1000827.

85. Chao SD, Kress JD, Redondo A. Coarse-grained rigidblob model for soft matter simulations. J Chem Phys2005, 122:234912.

86. Rim JE, Pinsky PM, van Osdol WW. Multiscale mod-eling framework of transdermal drug delivery. Anals

Biomed Eng2009, 37:1217 1229.87. Schumm P, Scoglio CM, van der Merwe D. A network

model of successive partitioning-limited solute diffu-sion through the stratum corneum. J Theor Biol2010,262:471477.

88. Holtje M, Forster T, Brandt B, Engels T, von RybinskiW, Holtje HD. Molecular dynamics simulations of stra-tum corneum lipid models: fatty acids and cholesterol.Biochim Biophys Acta2001, 1511:156 167.

89. Van Der Spoel D, Lindahl E, Hess B, Groenhof G,Mark AE, Berendsen HJC. GROMACS: fast, flexible,and free.J Comput Chem2005, 26:1701 1718.

90. Pandit SA, Scott HL. Molecular-dynamics simulation ofa ceramide bilayer.J Chem Phys2006, 124:14708.

91. Mombelli E, Morris R, Taylor W, Fraternali F.Hydrogen-bonding propensities of sphingomyelin insolution and in a bilayer assembly: a molecular dynam-ics study.Biophys J2003, 84:15071517.

92. Notman R, den Otter WK, Noro MG, Briels WJ,Anwar J. The permeability enhancing mechanism ofDMSO in ceramide bilayers simulated by moleculardynamics.Biophys J2007, 93:20562068.

93. Das C, Noro MG, Olmsted PD. Simulation studies ofstratum corneum lipid mixtures. Biophys J 2009,97:19411951.

94. Das C, Olmsted PD, Noro MG. Water permeationthrough stratum corneum lipid bilayers from atomisticsimulations.Soft Matter2009, 5:4549 4555.

95. Bouwstra JA, Dubbelaar FE, Gooris GS, Ponec M. Thelipid organisation in the skin barrier. Acta DermVenereol Suppl2000, 208:2330.


drug delivery through the skin--molecular simulations of barrier lipids_huzil_2011

Documents