amphiphile regulation of ion channel function by changes in the
TRANSCRIPT
Amphiphile regulation of ion channel functionby changes in the bilayer spring constantJens A. LundbĂŚka,b,1, Roger E. Koeppe IIc, and Olaf S. Andersena
aDepartment of Physiology and Biophysics, Weill Cornell Medical College, New York, NY 10065; bQuantum Protein Center, Department of Physics,Technical University of Denmark, Kongens Lyngby, DK-2800, Denmark; and cDepartment of Chemistry and Biochemistry, University of Arkansas,Fayetteville, AR 72701
Edited* by Christopher Miller, Brandeis University, Waltham, MA, and approved July 16, 2010 (received for review June 1, 2010)
Many drugs are amphiphiles that, in addition to binding to aparticular target protein, adsorb to cell membrane lipid bilayersand alter intrinsic bilayer physical properties (e.g., bilayer thick-ness, monolayer curvature, and elastic moduli). Such changescan modulate membrane protein function by altering the energeticcost (ÎGbilayer) of bilayer deformations associated with proteinconformational changes that involve the protein-bilayer interface.But amphiphiles have complex effects on the physical properties oflipid bilayers, meaning that the net change in ÎGbilayer cannot bepredicted from measurements of isolated changes in such proper-ties. Thus, the bilayer contribution to the promiscuous regulationof membrane proteins by drugs and other amphiphiles remainsunknown. To overcome this problem, we use gramicidin A (gA)channels as molecular force probes to measure the net effect ofamphiphiles, at concentrations often used in biological research,on the bilayer elastic response to a change in the hydrophobiclength of an embedded protein. The effects of structurally diverseamphiphiles can be described by changes in a phenomenologicalbilayer spring constant (HB) that summarizes the bilayer elasticproperties, as sensed by a bilayer-spanning protein. Amphiphile-induced changes in HB, measured using gA channels of a particularlength, quantitatively predict changes in lifetime for channels of adifferent lengthâas well as changes in the inactivation of voltage-dependent sodium channels in living cells. The use of gA channelsas molecular force probes provides a tool for quantitative, predic-tive studies of bilayer-mediated regulation of membrane proteinfunction by amphiphiles.
bilayer elasticity ⣠hydrophobic coupling ⣠hydrophobic matching
It long has been suspected that âmembrane-activeâ or âmem-brane-stabilizingâ drugs could regulate membrane protein func-
tion by partitioning into the host lipid bilayer and thereby alter itsphysical properties (1, 2). Indeed, numerous studies have shownthat amphiphiles, including many drugs, alter lipid bilayer physi-cal properties (for a recent review, see ref. 3). Moreover, mem-brane protein function involves conformational changes at theprotein-bilayer boundary (4), which due to hydrophobic couplingwill perturb the surrounding bilayer (Fig. 1A). The associatedbilayer deformation energy (ÎGbilayer) contributes to the free en-ergy of a protein conformational change (ÎGprot), meaning thatchanges in bilayer physical properties can alter protein functionby altering ÎGbilayer, cf. refs. 4â8. The promiscuous regulation ofmembrane proteins by amphiphiles therefore may be due toamphiphile-induced changes in ÎGbilayer. This mechanism wouldprovide a rationale for the observed correlations between theincreased hydrophobicity (or lipophilicity) of drugs and the like-lihood of adverse events (9) or attrition during drug development(10, 11).
Amphiphilic molecules alter many different bilayer properties[including intrinsic curvature (12), thickness (13), and elasticmoduli (14, 15)], some of which may have opposing effects onthe bilayer deformation energy (4). This complicates attemptsto predict even the sign of the changes in ÎGbilayer (12). The pro-blem can be overcome, however, by using gramicidin A (gA)
channels as probes to sense net changes in bilayer propertiesas experienced by a bilayer-spanning protein.
gA channels are dimers (D) formed by the transbilayer asso-ciation of monomeric subunits (M) from each bilayer leaflet(Fig. 1B), and channel gating is described by
2Mâk1
kâ1D;
where k1 and kâ1 are the association and dissociation rate con-stants for the monomer â dimer equilibrium. Channel formationin a bilayer with a hydrophobic thickness (d0) that exceeds thechannel hydrophobic length (l) involves a local bilayer deforma-tion (16â19) with an associated deformation energy. The bilayer,in response, exerts a disjoining force (Fdis) on the channel, themagnitude of which is determined by the bilayer elastic properties(Fig. 1B). Amphiphiles that decrease Fdis will increase k1, which isreported as an increase in channel appearance frequency (f ), anddecrease kâ1, which is reported as an increase in channel lifetime(Ď Âź 1âkâ1) (4, 12).
Results and DiscussionThe amphiphile Triton X-100 (TX100) increases Ď, meaning thatit decreases Fdis (12). To quantify the changes in the bilayer elasticproperties, we compare the changes in Ď for gA channels formedby subunits of different length and chirality; the 13-residue½des-Val1-Gly2ďż˝gAâ and the 15-residue ½Ala1ďż˝gA (20) [designatedgAâĂ°13Ă and AgA(15), respectively]. Fig. 2A shows current tracesthat illustrate the effects of 3 ÎźM TX100 on gAâĂ°13Ă and AgA(15) channels in diphytanoylphosphatidylcholine Ă°DPhPCĂân-decane bilayers. TX100 increases f and Ď for both channel types,with the larger effects on the shorter channels.
d0 l
BA
Fdis
Fdis
d0 l
Fig. 1. Hydrophobic coupling between a bilayer-embedded protein and itshost lipid bilayer. (A) A protein conformational change causes a local bilayerdeformation. (B) Formation of a gA channel involves local bilayer thinning.Modified from ref. 4.
Author contributions: J.A.L. and O.S.A. designed research; J.A.L. and O.S.A. performedresearch; R.E.K. contributed new reagents/analytic tools; J.A.L. and O.S.A. analyzed data;and J.A.L., R.E.K., and O.S.A. wrote the paper.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.1To whom correspondence should be addressed. E-mail: [email protected].
www.pnas.org/cgi/doi/10.1073/pnas.1007455107 PNAS ⣠August 31, 2010 ⣠vol. 107 ⣠no. 35 ⣠15427â15430
BIOPH
YSICSAND
COMPU
TATIONALBIOLO
GY
Fig. 2B shows the concentration-dependent effects of TX100and six other amphiphiles [capsaicin, capsazepine (12), daidzein,genistein, phloretin (21), and GsMTx4 (22)], on Ď for gA channelsformed by (left- or right-hand) subunits of 13 residues (Ď13) or 15residues (Ď15). At concentrations where these amphiphiles altermembrane protein function (summarized in the original articles),they all increase Ď for both channel typesâand all with the largereffects on the shorter channels; cf. refs. 4, 12, 21, 22. [At the con-centrations used in the gA channel experiments, the amphiphilesdo not alter membrane capacitance (12, 21, 23). Thus changes inbilayer thickness cannot account for the effects.]
The changes in channel lifetimes reflect changes in activationenergy for channel dissociation (ÎGâĄ):
ÎG⥠â ÎGâĄ
cntl Âź lnfĎâĎcntlg ¡ kBT; [1]
where kB and T are, respectively, Boltzmannâs constant and thetemperature in Kelvin, and âcntlâ denote results in the absence ofamphiphile. The larger changes in lifetime for the shorter chan-nels relative to the longer channels mean that the changes in ÎGâĄ
also are larger for the shorter channels than for the longer chan-nels (Fig. 3). Moreover, because the changes in Ď13 and Ď15 resultfrom the same underlying mechanism, the changes Ď (inÎGâĄ
13 andÎGâĄ
15) are likely to be correlated (5). Fig. 3A shows the changes inÎGâĄ
13 as a function of the changes in ÎGâĄ15 in experiments with
DPhPC as the bilayer-forming lipid. Despite the very differentstructures of the amphiphiles, the changes are described by ashared linear relation with a slope: ÎÎGâĄ
13âÎÎGâĄ
15 Ÿ 1.19�0.02 (�SE, r2 > 0.99). A similar relation is obtained using anumber of other amphiphiles with dioleoylphosphatidylcholine(DOPC) as the bilayer-forming lipid. Fig. 3B shows the effectsof TX100 and seven other amphiphiles [curcumin (24), docosa-
hexaenoic acid (25), 2,3-butanedione monoxime (26), reducedTriton X-100 (27); and the viral antifusion peptides Z-Gly-D-Phe, Z-Gly-Phe and Z-D-Phe-Phe-Gly (28)]. Again the resultscan be described by a shared linear relation: ÎÎGâĄ
13âÎÎGâĄ15 Âź
1.21ďż˝ 0.05 (ďż˝SE, r2 Âź 0.95). Indeed, the DPhPC and DOPC re-sults can be superimposed (Fig. 3C), with the overall fit:ÎÎGâĄ
13âÎÎGâĄ15 Âź 1.20ďż˝ 0.03 (ďż˝SE, r2 Âź 0.98). Fig. 3D shows
the distribution of the slopes (values of ÎÎGâĄ13âÎÎG
âĄ15) for
the different amphiphiles.The activation energy for gA channel dissociation is given by
ÎG⥠Ÿ ÎGâĄ
channel Ăž ÎGâĄ
bilayer; [2]
where ÎGâĄ
channel denotes an âintrinsicâ energy cost due to loss ofcontributions such as hydrogen bonds that stabilize the channeldimer, whereas ÎGâĄ
bilayer denotes the change in bilayer deforma-tion energy associated with separating the channel subunits by δ[a distance of âź0.16 nm (19, 29, 30)] to reach a transition statewhere the channel conductance is lost (Fig. 1B). A large body ofwork, summarized in ref. 4, has shown that the amphiphile-induced changes in Ď primarily reflect changes in ÎGâĄ
bilayer. Thisconclusion is reinforced by the results in Fig. 3 because the rela-tion between ÎGâĄ
13 and ÎGâĄ15 can be described by a shared func-
tion despite the different amphiphile structures, the differentlipid structure, and the opposite chiralities of the 13- and 15-residue channel-forming subunits. Specific amphiphile-channelinteractions thus do not contribute to the changes in Ď, whichmeans that the amphiphiles will have little effect on ÎGâĄ
channel.
Fig. 2. Effects of amphiphiles on gA channels in DPhPC/n-decane bilayers.(A) Current traces before and after addition of 3 ÎźM TX100 to both sidesof a bilayer doped with gAâĂ°13Ă and AgA(15). The red and blue lines denotethe current levels for gAâĂ°13Ă and AgA(15) channels. (B) Concentration-dependent effects of TX100, capsaicin, capsazepine, daidzein, genistein,phloretin, or GsMTx4 on lifetime of channels formed by monomeric subunitshaving 13 or 15 residues (except for TX100, based on results from refs. 12,21, and 22).
0
1
2
3
0 1 2 3
0 1 2 3 4
0 1 2 3 4
0
1
2
3
0
1
2
3
0 1 2 3
0
1
2
3
A
Ď/Ď
âG âG k T BâG
â Gk
TâG âG k T
Ď /ĎCâG
âGk
T
Ď/Ď
Ď /Ď
0 1 20
5
10
ââ ââ13 15/G G
D
Fig. 3. Effects of amphiphiles on lifetimes of channels formed by 13-residuesubunits (expressed as lnfĎâĎcntlg, left axis, or activation energy, ÎG⥠â ÎGâĄ
cntl ,right axis) vs. corresponding effects on channels formed by 15-residue sub-units (bottom or top axis). (A) Effects of TX100, capsaicin, capsazepine, daid-zein, genistein, phloretin, or GsMTx4 in DPhPCân-decane bilayers (except forTX100, results from refs. 12, 21 and 22). A subset of the results was publishedpreviously (5). (B) Effects of TX100, curcumin, docosahexaenoic acid (DHA),2,3-butanedione monoxime (BDM), reduced Triton X-100 (rTX100), Z-Gly-D-Phe (ZGdF), and Z-Gly-Phe (ZGF) (at pH 7 or pH 4) and Z-D-Phe-Phe-Gly(ZdFFG, at pH 7) in DOPCân-decane bilayers (except for TX100, results fromrefs. 24â28). (C) Superimposition of results obtained using DPhPC (red) orDOPC (black). (D) Distribution of ÎÎGâĄ
13âÎÎGâĄ15 for individual amphiphiles
using DPhPC (red) or DOPC (black). Mean � SEM (n ⼠3) or � range (n Ÿ 2).
15428 ⣠www.pnas.org/cgi/doi/10.1073/pnas.1007455107 LundbÌk et al.
The linear relation between ÎGâĄ13 and ÎGâĄ
15 therefore reflectsalso a linear relation between ÎGâĄ
13;bilayer and ÎGâĄ
15;bilayer.The shared linear relation between ÎGâĄ
13;bilayer and ÎGâĄ
15;bilayerindicates that the effects of amphiphiles can be described by asingle parameter that characterizes the changes in bilayer elasticproperties. Because the same linear relation applies to com-pounds that cause positive [TX100 (12)] and negative [capsaicin(12), curcumin (31), and docosahexaenoic acid (25)] changes inintrinsic curvature, the curvature-dependent contributions toÎGâĄ
bilayer are likely to be small. In this case, one can approximateÎGbilayer as
ÎGbilayer Âź HB ¡ Ă°l â d0Ă2; [3]
where the bilayer elastic properties are summarized by a singlephenomenological spring constant, HB (e.g., ref. 4). [Our conclu-sion, that the curvature contributions to ÎGbilayer are small, doesnot depend on the use of planar bilayers formed using decane.Using a fluorescence quench-based method (32) that employshydrocarbon-free large unilamellar vesicles to determine amphi-phile-induced changes in lipid bilayer properties, as sensed bygA channels, we find that TX100 and capsaicin have similareffectsâand that all compounds tested have similar effects whenassayed in the planar bilayer and the lipid vesicle system.]
The bilayer contribution to ÎG⥠then is given by
ÎGâĄ
bilayer Âź HB ¡ ½ðlĂž δ â d0Ă2 â Ă°l â d0Ă2�Ÿ HB ¡ δ ¡ ½2 ¡ Ă°l â d0Ă Ăž δ�; [4]
and, if the amphiphile effect on ÎGâĄ
bilayer can be described bychanges in HB:
ÎÎGâĄ
bilayer;13
ÎÎGâĄ
bilayer;15
Âź ÎHB ¡ δ ¡ ½2 ¡ Ă°l13 â d0Ă Ăž δ�ÎHB ¡ δ ¡ ½2 ¡ Ă°l15 â d0Ă Ăž δ� Âź
2 ¡ Ă°l13 â d0Ă Ăž δ
2 ¡ Ă°l15 â d0Ă Ăž δ;
[5]
which is independent ofÎHB. As a quantitative check of Eq. 5, wenote that the hydrophobic length of the AgA(15) channel, l15, isâź2.2 nm (16, 18, 33) and the channel length varies âź0.08 nmâresidue (e.g., ref. 4); therefore l13 âź 1.88 nm. Using these valuesand the estimates for ÎÎGâĄ
13âÎÎGâĄ15 from Fig. 3C, we find that
d0 âź 4.0 nm for bilayers formed using either DPhPC or DOPC.For comparison, the bilayer thickness at an applied potentialof 200 mV, as estimated from capacitance measurements (assum-ing a dielectric constant of 2), is âź4.2 nm for DPhPCân decane(21) and between 3.7â4.3 nm for DOPCân decane (24, 25). Theagreement between the measured bilayer thickness and the pre-diction based on Eq. 5 is striking and supports the use of Eqs. 1â4to describe the energetic coupling between gA channel functionand lipid bilayer properties.
It is thus possible to evaluate the changes in HB from thechanges in Ď, using the relation
ÎHB Âź kBT ¡lnfĎâĎcntlg
δ ¡ ½2 ¡ Ă°l â d0Ă Ăž δ� ; [6]
which is obtained from Eqs. 1 and 4. For l â d0 Âź 1.8 nm andδ Âź 0.16 nm, either 200 nM GsMTx4 or 3 ÎźM TX100 cause a4 kJâĂ°mole ¡ nm2Ă decrease in HB, which should be comparedto HB in unmodified phospholipid bilayers, âź56 kJâĂ°mole ¡nm2Ă (4). Even relatively modest changes in HB, thus, may causemeasurable changes in channel (membrane protein) function.
The amphiphile-induced changes in HB, as measured using gAchannels formed by 15-residue subunits, are transferable in thatthey provide for quantitative predictions of changes in ÎGâĄ
bilayerâ
and thus lifetimeâfor channels formed by 13-residue subunits ina bilayer of the same composition (and vice versa); cf. Fig. 3. Theyalso are scalable, in that they provide for quantitative descriptionsof the effects of amphiphiles on membrane protein function in-volving entirely different structural changes in plasma mem-branes. Fig. 4 shows the effects of five amphiphiles (TX100,reduced Triton X-100, β-octyl glucoside, Genapol X-100, and cap-saicin) on the membrane potential for 50% inactivation (V in) ofvoltage-dependent sodium channels in HEK293 cells, as functionof changes in HB measured using gA channels in DOPCân-decane bilayers (based on refs. 12 and 27). Despite the structuraldifferencesâand that capsaicin promotes opposite curvaturefrom the other moleculesâthe V in vs. HB relations vary littleamong the different amphiphiles.
Given the many different effects that amphiphilic compoundscan have on the physical properties of lipid bilayers, it is perhapssurprising that the effects of structurally different amphiphiles onthe bilayer elastic response can be characterized by changes in asingle parameter, a phenomenological bilayer spring constantHB. The finding represents an important conceptual simpli-fication of the bilayer-mediated effects of amphiphiles and, more-over, provides a quantitative measure (a number) that can beused in further mechanistic studies. That the amphiphiles de-crease HB (and the energetic cost of deforming the bilayer) mostlikely reflects that water-soluble amphiphiles reversibly adsorb tolipid bilayers and thereby decrease the bilayer elastic moduli (14).
Though long recognized, the regulation of membrane proteinfunction by membrane-active compounds has remained elusivebecause it has been difficult to obtain quantitative informationabout changes in the bilayer properties that are relevant for pro-tein function. Nevertheless, studies on amphiphile regulation ofmembrane proteins often are done at concentrations that alterbilayer physical properties (see ref. 4 for a list of examples).The issue becomes particularly important in drug discoveryand developmentâmost orally available drugs are amphiphiles(34), and increasing drug lipophilicity has become a major andrising cause of nonspecific drug actions and attrition (10, 11).The gA channel-based approach provides the advantage of adirect readout of amphiphile-induced changes in bilayer proper-ties, as experienced by an embedded protein. It provides a tool forquantitative and predictive explorations of the bilayer-mediated
-10 -8 -6 -4 -2 0
-10
-5
0
β
Vin-V
in,c
n tl
/ m
V
âH
Fig. 4. Amphiphile-induced changes in inactivation of voltage-dependentsodium channels in HEK293 cells as a function of changes in HB in DOPCân-decane bilayers. Shift in membrane potential for 50% inactivation(V in-V in;cntl) plotted vs. ÎHB (ÂźHB-HB;cntl) in DOPCân-decane bilayers. Amphi-philes: capsaicin, β-octyl puranoside (βOG), Genapol X-100 (GX100), reducedTriton X-100 (rTX100), and TX100. HEK293 cells were depolarized to Ăž20 mVfollowing 300-ms prepulses to potentials varying from â130 to Ăž50 mV.Based on results from refs. 12 and 27.
LundbÌk et al. PNAS ⣠August 31, 2010 ⣠vol. 107 ⣠no. 35 ⣠15429
BIOPH
YSICSAND
COMPU
TATIONALBIOLO
GY
regulation of membrane protein function by drug candidates andother amphiphiles.
Materials and MethodsGramicidin analogues ½des-Val1-Gly2ďż˝gAâ and ½Ala1ďż˝gA were synthesized asdescribed by ref. 20. Single-channel measurements using Triton X-100 weredone at 25 °C in DPhPC or DOPC Ă°Avanti Polar LipidsĂân-decane bilayers,separating 1.0 M NaCl, 10 mM Hepes, pH 7 solutions using the bilayer punchmethod, as described previously (4). The applied potential wasďż˝200 mV. The
gA analogues and amphiphile were added to both sides of the bilayer.Survivor plots of channel lifetimes were fitted by a single exponential distri-bution, NĂ°tĂâNĂ°0Ă Âź expfâtâĎg, where NĂ°tĂ is the number of channels withduration longer than time t, and Ď is the average lifetime (12, 22, 24â28).
ACKNOWLEDGMENTS. We thank Denise V. Greathouse for the peptides andKevin Lum for assistance with the single-channel measurements. This workwas supported by National Institutes of Health Grants GM021342 andRR015569.
1. Seeman P (1972) The membrane actions of anesthetics and tranquilizers. PharmacolRev 24:583â655.
2. Sackmann E (1984) Biological Membranes, ed D Chapman (Academic, London),pp 105â143.
3. Seddon AM, et al. (2009) Drug interactions with lipid membranes. Chem Soc Rev38:2509â2519.
4. LundbĂŚk JA, Collingwood SA, IngĂłlfsson HI, Kapoor R, Andersen OS (2010) Lipidbilayer regulation of membrane protein function: Gramicidin channels as molecularforce probes. J R Soc Interface 7:373â395.
5. LundbĂŚk JA (2006) Regulation of membrane protein function by lipid bilayer elasti-city: A single molecule technology to measure the bilayer properties experienced byan embedded protein. J Phys Condens Matt 18:S1305âS1344.
6. Andersen OS, Koeppe RE, II (2007) Bilayer thickness and membrane protein function:An energetic perspective. Annu Rev Biophys Biomol Struct 36:107â130.
7. Marsh D (2008) Protein modulation of lipids, and vice-versa, in membranes. BiochimBiophys Acta 1778:1545â1575.
8. Phillips R, Ursell T, Wiggins P, Sens P (2009) Emerging roles for lipids in shapingmembrane-protein function. Nature 459:379â385.
9. Hughes JD, et al. (2008) Physiochemical drug properties associated with in vivotoxicological outcomes. Bioorg Med Chem Lett 18:4872â4875.
10. Leeson PD, Springthorpe B (2007) The influence of drug-like concepts on decision-making in medicinal chemistry. Nat Rev Drug Discov 6:881â890.
11. KeserĂź GM, Makara GM (2009) The influence of lead discovery strategies on theproperties of drug candidates. Nat Rev Drug Discov 8:203â212.
12. LundbĂŚk JA, et al. (2005) Capsaicin regulates voltage-dependent sodium channels byaltering lipid bilayer elasticity. Mol Pharmacol 68:680â689.
13. Ebihara L, Hall JE, MacDonald RC, McIntosh TJ, Simon SA (1979) Effect of benzylalcohol on lipid bilayers. A comparisons of bilayer systems. Biophys J 28:185â196.
14. Evans E, Rawicz W, Hofmann AF (1995) Bile Acids in Gastroenterology: Basic andClinical Advances, eds AF Hofmann, G Paumgartner, and A Stiehl (Kluwer AcademicPublishers, Dordrecht), pp 59â68.
15. Ly HV, Longo ML (2004) The influence of short-chain alcohols on interfacial tension,mechanical properties, area/molecule, and permeability of fluid lipid bilayers. BiophysJ 87:1013â1033.
16. Huang HW (1986) Deformation free energy of bilayer membrane and its effect ongramicidin channel lifetime. Biophys J 50:1061â1070.
17. Nielsen C, Goulian M, Andersen OS (1998) Energetics of inclusion-induced bilayerdeformations. Biophys J 74:1966â1983.
18. Harroun TA, Heller WT, Weiss TM, Yang L, Huang HW (1999) Experimental evidencefor hydrophobic matching and membrane-mediated interactions in lipid bilayerscontaining gramicidin. Biophys J 76:937â945.
19. LundbĂŚk JA, Andersen OS (1999) Spring constants for channel-induced lipid bilayerdeformationsâestimates using gramicidin channels. Biophys J 76:889â895.
20. Greathouse DV, Koeppe RE, II, Providence LL, Shobana S, Andersen OS (1999) Designand characterization of gramicidin channels. Method Enzymol 294:525â550.
21. Hwang TC, Koeppe RE, II, Andersen OS (2003) Genistein can modulate channelfunction by a phosphorylation-independent mechanism: importance of hydrophobicmismatch and bilayer mechanics. Biochemistry 42:13646â13658.
22. Suchyna TM, et al. (2004) Bilayer-dependent inhibition of mechanosensitive channelsby neuroactive peptide enantiomers. Nature 430:235â240.
23. LundbĂŚk JA, Birn P, Girshman J, Hansen AJ, Andersen OS (1996) Membrane stiffnessand channel function. Biochemistry 35:3825â3830.
24. IngĂłlfsson HI, Koeppe RE, II, Andersen OS (2007) Curcumin is a modulator of bilayermaterial properties. Biochemistry 46:10384â10391.
25. Bruno MJ, Koeppe RE, II, Andersen OS (2007) Docosahexaenoic acid alters bilayerelastic properties. Proc Natl Acad Sci USA 104:9638â9643.
26. Artigas P, et al. (2006) 2,3-butanedione monoxime affects cystic fibrosis transmem-brane conductance regulator channel function through phosphorylation-dependentand phosphorylation-independent mechanisms: The role of bilayer material proper-ties. Mol Pharmacol 70:2015â2026.
27. LundbĂŚk JA, et al. (2004) Regulation of sodium channel function by bilayer elasticity:The importance of hydrophobic coupling: Effects of micelle-forming amphiphiles andcholesterol. J Gen Physiol 123:599â621.
28. Ashrafuzzaman Md, Lampson MA, Greathouse DV, Koeppe RE, II, Andersen OS (2006)Manipulating lipid bilayer material properties using biologically active amphipathicmolecules. J Phys Condens Matt 18:S1235âS1255.
29. Durkin JT, Koeppe RE, II, Andersen OS (1990) Energetics of gramicidin hybrid channelformation as a test for structural equivalence. Side-chain substitutions in the nativesequence. J Mol Biol 211:221â234.
30. Miloshevsky GV, Jordan PC (2004) Gating gramicidin channels in lipid bilayers:Reaction coordinates and the mechanism of dissociation. Biophys J 86:92â104.
31. Barry J, et al. (2009) Determining the effects of lipophilic drugs onmembrane structureby solid-state NMR spectroscopy: The case of the antioxidant curcumin. J Am Chem Soc131:4490â4498.
32. Ingolfsson HI, Andersen OS (2010) Screening for small moleculeâs bilayer-modifyingpotential using a gramicidin-based fluorescence assay. Assay Drug Dev Techndoi:10.1089/adt.2009.0250.
33. Elliott JR, Needham D, Dilger JP, Haydon DA (1983) The effects of bilayer thickness andtension on gramicidin single-channel lifetime. Biochim Biophys Acta 735:95â103.
34. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (1997) Experimental and com-putational approaches to estimate solubility and permeability in drug discovery anddevelopment settings. Adv Drug Deliver Rev 23:3â26.
15430 ⣠www.pnas.org/cgi/doi/10.1073/pnas.1007455107 LundbÌk et al.