Temperature-Dependent Partitioning of C152 in Binary

Jul 20, 2017 - Time-resolved fluorescence indicated that C152 partitioning into ..... The enthalpy of transition for DMPE is 24.3 kJ/mol, and Tm = 49...
0 downloads 0 Views 733KB Size
Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Article

Temperature Dependent Partitioning of C152 in Binary Phosphatidylcholine Membranes and Mixed Phosphatidylcholine/Phosphatidylethanolamine Membranes Christine A. Gobrogge, Victoria A. Kong, and Robert Allan Walker J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b04831 • Publication Date (Web): 20 Jul 2017 Downloaded from http://pubs.acs.org on July 31, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Temperature Dependent Partitioning of C152 in Binary Phosphatidylcholine membranes and Mixed Phosphatidylcholine/Phosphatidylethanolamine Membranes Christine A. Gobrogge, Victoria A. Kong, Robert A. Walker* Department of Chemistry and Biochemistry, Montana State University, Bozeman, Montana 59717, United States

ABSTRACT

Time-resolved fluorescence and differential scanning calorimetry were used to determine the partitioning of coumarin 152 (C152) into large unilamellar vesicles composed of binary mixtures of two phosphatidylcholines (12:0/12:0 DLPC and 14:0/14:0 DMPC) and vesicles composed of binary mixtures of a phosphatidylcholine and a phosphatidylethanolamine (14:0/14:0 DMPC and 14:0/14:0 DMPE).

Differential scanning calorimetry showed that both DLPC/DMPC and

DMPC/DMPE are miscible in lipid vesicles. Time-resolved fluorescence indicated that C152 partitioning into DLPC/DMPC mixtures showed nearly ideal behavior that was described with weighted contributions from C152 partitioning into pure DLPC and pure DMPC vesicles. contrast, C152 partitioning into DMPC/DMPE mixtures was distinctly non-ideal.

In For

DMPC/DMPE lipid vesicles having DMPC mole fractions between 10–80%, C152 partitioning into the bilayer was measurably enhanced near the melting temperature, relative to expectations based simply on weighted contributions from C152 partitioning into vesicles comprised of pure

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

lipids. The origin of this behavior remains uncertain. For vesicles comprised of pure DMPE< C152 shows almost no partitioning into the membrane with ≥80% of the solute remaining in the buffer solution at temperatures between 10–50 °C.

Introduction Since the “fluid-mosaic model” describing cell membrane structure was first proposed in 1972,1 scientists have been investigating the influence of lipid composition on cellular membrane structure and function.2 Of particular interest is how lipids are distributed within and between the outer and inner leaflets of the membrane bilayer.3,4

Entropic considerations predict a

homogeneous distribution of lipid mixtures with each leaflet having equivalent compositions. However in mammalian cells, distinct heterogeneity is observed between inner and outer leaflets. For example, the inner leaflet of the cell membrane generally is enriched in phosphatidylserine and phosphatidylethanolamine, and the outer leaflet is enriched with phosphatidylcholine. Deviation in standard physiological lipid distributions is often an indicator for other cellular processes.

For example, an enrichment of phosphatidylserine in the outer leaflet is a

susceptibility signal for phagocytosis and a propagation signal for blood coagulation.5-7 Transmembrane lipid arrangement is yet another example of local lipid ordering in lipid bilayers. Fluid phase binary mixtures of lipids differing by only 2 methylenes per tail have been shown to anti-correlate in positions across a bilayer in a phenomenon termed “transbilayer complementarity.”8 Lipids are also heterogeneously mixed within individual membrane leaflets.

Lipid

domains (“rafts”) with micrometer to nanometer diameters2,9,10 are formed in fluid membranes creating lateral heterogeneity.11 Directly related to lipid domain size and composition is their cellular function, as lipid domains have been found to be instrumental in gene expression,

ACS Paragon Plus Environment

2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

enzyme binding, enzyme activation, and cell division.4,6,11-13 Heterogeneity is even observed in lipid membranes composed of one type of lipid.

Near the gel-liquid crystalline transition

temperature, clusters of 2˚ phases form in the bulk equilibrium phase. In other words, below the transition temperature, liquid-crystalline phase domains form in the bulk gel phase, and above the phase transition temperature, gel phases form in the bulk liquid-crystalline phase.14 Near the membrane’s transition temperature, domains vary significantly in size and can be as small as 10 Å and as large as 1000 Å depending on the precise temperature of the system as well as the characteristics of the lipid (or lipids) comprising the domains.15,16 Phosphatidylcholines (PCs) and phosphatidylethanolamines (PEs) differ only by their headgroup structures. However, significant structural differences are observed in membranes composed of either type of lipid.

Elaborate networks of interconnected lipids and water

molecules are formed in PC bilayers, while PE membranes include far fewer interactions with interfacial water.17-20 Computational simulations of interfacial water at pure liquid-crystalline phase DMPC membranes show that water bridges link 76% of DMPC molecules, while Coulomb associations are observed between 93% of the DMPC lipids.19,20 (We were unable to find comparable data for DMPE lipid membranes.) PE membrane stability is attributed to the strong interactions between neighboring PE lipids, including hydrogen bond donating and accepting interactions as well as strong Coulomb associations.17,18,21-25 These strong Coulomb interactions are largely responsible for transition temperatures that are, on average, 20–30 ˚C higher in PE bilayers composed of lipids having equivalent chain lengths. Lipid domain formation and lipid mixing phenomena are of great interest to the pharmaceutical industry, as membrane lipids often control the ADMET (absorption, distribution, metabolism, excretion, and toxicology) of drugs.26 Overwhelming evidence has shown that the

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

permeation of certain solutes is enhanced near the phase transition temperature of PC membranes.27-31

Partitioning enhancement near the melting temperature is, however, not

observed in vesicles composed of PEs.32 In PC membranes, enhanced permeation is thought to be a consequence of unique solvation environments resulting from interfacial boundaries of lipid domains that are present only near the melting temperature.32-34 The physical structures of lipid domains and their interfaces are dependent on many factors including lipid headgroup, acyl chain length, and the precise temperature of the system.15 Although enhancement of solute permeation near the phase transition temperature is well documented and generally accepted to be a consequence of lipid domain formation, reports in the literature generally lack molecularly specific descriptions of how solutes partition into single component and mixed lipid membranes.32,33 A variety of techniques have been employed to study lipid miscibility including computational strategies,35-37 DSC,13,35,38-42

fluorescence,43-45 electron microscopy,46-49 and

NMR.50,51 As a general trend, binary mixtures of saturated phosphatidylcholines appear miscible in all phases if the difference in acyl chain length is no more than two carbons. When the acyl chains of two PCs differ by four or more carbons, non-ideal mixing (lateral phase segregation) is observed.52-54 DSC experiments of DMPC/DMPE lipid mixtures performed by Silvius et al. have indicated that the two components are completely miscible in the liquid crystalline phase, while limited miscibility was observed in the gel phase.38 Arnold et al. studied the same system using 31P-NMR and found the lipids are immiscible in all phases.50 Studies described in this work examine partitioning of an organic solute, coumarin 152 (C152), into vesicle bilayers having mixed lipid compositions. Binary lipid mixtures examined include membranes composed of two phosphatidylcholines (12:0/12:0 DLPC and 14:0/14:0

ACS Paragon Plus Environment

4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

DMPC) and membranes composed of a phosphatidylcholine and a phosphatidylethanolamine (14:0/14:0 DMPC and 14:0/14:0 DMPE). DSC experiments measured how bilayer transition temperatures changed as a function of composition, and time-correlated single photon counting was used quantify and characterize C152 partitioning into model membranes as a function of composition and temperature.

Previous work characterizing C152 partitioning into vesicle

bilayers composed of single PCs quantified the dramatic enhancement in partitioning at the gelliquid crystalline transition temperature with nearly 80% of all available solutes localizing in the polar headgroup region. Temperature dependent data, however, showed that solute partitioning into the membrane near the transition temperature is exothermic, so raising the temperature even further leads to C152 exsolvation back into bulk solution.27 Findings described in detail below show that DLPC and DMPC exhibit near ideal miscibility in all phases, while DMPC/DMPE mixtures show slightly less ideal mixing behavior. Furthermore, C152 partitioning into vesicles composed of DLPC/DMPC mixtures is ideal, meaning that the behavior can be described simply by considering weighted contributions from C152 partitioning into pure DLPC and DMPC vesicles. In contrast, C152 partitioning into DMPC/DMPE mixtures is distinctly non-ideal. A larger amount of C152 partitions into the polar headgroup region of DMPC/DMPE lipid membranes than is predicted assuming ideal mixing behavior, supporting a model describing C152 partitioning as being dependent on domain formation. MATERIALS The phospholipids 1,2-dilauroyl-sn-glycero-3-phosphocholine (12:0 DLPC), 1,2dimyristoyl-sn-glycero-3-phosphocholine (14:0 DMPC), and 1,2-dimyristoyl-sn-glycero-3phosphoethanolamine (14:0 DMPE) were purchased as powder from Avanti Polar Lipids (Alabaster, AL). Laser grade coumarin 152 (C152) was purchased from Exciton (Dayton, OH).

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

All materials were used as received without further purification. Solvents were HPLC grade and received from Sigma (Milwaukee, WI). Phosphate buffered saline solutions for vesicles were prepared using Millipore water (18.2 MΩ).

LUVS Large unilamellar vesicles (LUVs) were formed from solid lipids purchased from Avanti Polar Lipids (Alabaster, AL). Following standard protocols,55-57 ~1 mg/mL lipid/chloroform was mixed thoroughly and the chloroform was removed via rotary evaporation (>60 °C) leaving behind a lipid film on the round-bottom flask. The sheets were rehydrated using phosphate buffered saline (PBS, 10 mM, pH = 7). Vesicles prepared for fluorescence experiments were prepared with a 1.5 mM concentration, while vesicles prepared for DSC experiments were prepared at 20 mM. Upon rehydration, solutions were sonicated for 30 min to produce LUVs. Solutions prepared for DSC measurements were used as produced from this procedure (DSC experimental details shown in the Supporting Information).

LUVs used in fluorescence

experiments were filtered through a 450 nm pore PTFE syringe filter. Vesicle solutions were spiked with PBS solutions containing coumarins. This final solution was filtered 12 times with a miniextruder (Avanti Polar Lipids) through a 200 nm pore membrane.

Extrusions were

performed above the melting temperature of the lipid bilayer. Dynamic light scattering was used to determine average vesicle diameter.

Results from these experiments showed a bimodal

distribution of 60 ± 10 nm vesicles (~20%) and 200 ± 25 nm vesicles (~80%).27 Partitioning experiments using 50 nm vesicles (prepared in the same manner described here, replacing filtration through 50 nm pore membrane) were also performed, and results from fluorescence experiments using these smaller vesicles yielded results that were identical to within

ACS Paragon Plus Environment

6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

experimental uncertainty to reported findings from experiments using 200 nm vesicles. (Data not shown.) Time-Resolved Fluorescence Time-resolved fluorescence emission data were collected using Picoquant PicoHarp 300 and FluoTime 200 software. An APE Autotracker containing a SHG crystal situated after a Chameleon laser (Coherent, 80 MHz, 680−1080 nm) produced excitation pulses (85 fs pulsewidth) at a wavelength of 400 nm. A Conoptics Model 350−105 modulator reduced the repetition rate to 4 MHz. Samples were equilibrated at reported temperatures for approximately 5 min. Extending the equilibration time led to no observable changes in the measured lifetimes or relative amplitudes. The sample temperature was monitored by a Quantum Northwest TC152 control. A long-pass filter (80% transmission >455 nm) was placed after the sample to reduce scattering from the vesicles. Photon emission was collected at 500 nm. Additional details about this assembly can be found in previous reports.27,58,59 In a given time-correlated single photon counting (TCSPC) experiment, signal was collected until the maximum intensity at early times reached a threshold of 8000 counts. Depending on the local solvation environment of C152, the time required for each experiment varied with temperature. The instrument response function (FWHM 50% DMPC.

In pure DMPC vesicles, the peak at 14 °C is assigned to the

pretransition, where vesicles change from their rigid gel structure to a rippled gel phase. In mixtures composed of 80% and 90% DMPC, the secondary peak occurs at lower temperatures. We can not definitively assign this feature, however, the absence of a similar peak in the 95% DMPC/ 5% DLPC mixture leads us to believe this peak is due to the presence of liquidcrystalline DLPC domains forming in the (primarily) gel DMPC bilayer.

Notably, the

95% DMPC/5% DLPC vesicles also show no evidence of the gel-to-rippled transition. Miscibility in vesicles composed of binary PC mixtures was further assessed by comparing phase diagrams derived from the experimental data with expectations based on ideal mixing behavior. In binary mixtures of lipids, the presence of one main calorimetric peak indicates lipid miscibility, while broad phase transitions indicate some degree of non-ideal mixing behavior and lipid domain formation. Theoretical mixing curves

(Figure 2) were

calculated assuming ideal behavior according to Sturtevant et al.39 The dashed lines in Figure 2 are the ideal solidus (s) and liquidus curves (l), calculated using the following expressions: 

XDMPC =  (l)

(2)

XDMPC =  XDMPC (s)

(l)

(3)

ACS Paragon Plus Environment

10

Page 11 of 31

where XDMPC(l) and X DMPC(s) are the mole fractions of DMPC in the liquidus and solidus phases, and α and β are described by:  = exp

∆HDLPC

 = exp

R

∆HDMPC R

- 1

1

T

TDLPC



(4)

- 1

1

T

TDMPC



(5)

where ∆HDLPC is the enthalpy of the gel to liquid-crystalline transition of pure DLPC (+7.1 kJ/mol), ∆HDMPC is the enthalpy of the gel to liquid-crystalline transition of pure DMPC (+22.8 kJ/mol), TDLPC is the melting temperature of pure DLPC (-3.3 °C), and TDMPC is the melting temperature of pure DMPC (22.8 °C).39 25 Temperature (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

20 15 10 5 0 -5 0.0

0.2

0.4 0.6 XDMPC

0.8

1.0

Figure 2. Calculated transition curves (---) using Eqns. 2-5. Experimental data are shown by markers: peak melting temperature (black triangles), transition start (red circles), and transition end (blue circles). Despite clear departure from the predicted melting temperatures for ideally mixed binary lipid bilayers, data in Figure 2 show that DLPC and DMPC are miscible in all proportions as expected. These results are consistent with those reported by Sturtevant and others. 39-41

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry

Calorimetric traces of vesicles composed of DMPC/DMPE are shown in Figure 3 and summarized in Table 2. (Additional data are shown in the Supporting Information).

DMPC -0.15

Heat Flow (W/g)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

-0.10

-0.05

DMPE 0.00

25

30

35

40

45

50

55

Temperature (ºC)

Figure 3. Calorimetric traces of DMPC vesicles (red), DMPE vesicles (black), and binary mixtures of DLPC and DMPC: 0.99 DMPC (orange), 0.95 DMPC (yellow), 0.90 DMPC (light green), 0.80 DMPC (dark green), 0.50 DMPC (light blue), 0.10 DMPC (dark blue), 0.05 DMPC (light purple), 0.01 DMPC (dark purple). Table 2. Melting temperatures of large unilamellar vesicles composed of binary mixtures of DMPC and DMPE. Uncertainties in measured temperatures are ± 0.5˚C. Mole Ratio DMPE: DMPC

Tm (°C)

Tm, ideal (°C)

0:100

22.8

22.8

10:90

23.2

26.7

20:80

25.5-30.8

29.2

50:50

33.0-40.6

36.7

ACS Paragon Plus Environment

12

Page 13 of 31

80:20

44.8-47.9

43.9

90:10

49.5

45.9

100:0

56.3*

49.5

Complicating the analysis of DMPC/DMPE mixtures are the two crystal phases of DMPE. Largely dependent on sample history and preparation technique, the highly ordered crystalline phase melts directly into the liquid crystalline phase at ~56 °C, while the “less stable” crystalline phase (gel phase) melts into the liquid-crystalline phase at ~50 °C.17,38 The stable, highly ordered crystal phase has been reported previously, however, formation of the Lc phase generally occurs only after a long (3–4 week) incubation periods at –10 °C.17,60 Upon addition of C152 to pure DMPE membranes, the stable crystalline phase transition is less pronounced and the gel phase transition is observed as the main calorimetric event (data shown in Supporting Information). These observations raise questions about whether or not solutes can associate with membranes in discrete, well-defined ways to change lipid-lipid interactions within the membrane itself.

55

Temperature (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

50 45 40 35 30 25 0.0

0.2

0.4

0.6

0.8

1.0

XDMPE

Figure 4. Calculated transition curves (---) of DMPC/DMPE mixtures using Eqns. 2-5. The enthalpy of transition for DMPE is 24.3 kJ/mol and the Tm= 49.5 °C.61 Experimental data are shown by markers: peak melting temperature (black triangles), transition start (red circles), and

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

transition end (blue circles). At XDMPE =1, the peak melting temperature of the stable gel phase (unobserved in systems with C152 present) is denoted by an asterisk. The calorimetric data for DMPC/DMPE mixtures slightly deviate from the calculated ideal mixing transition curves in Figure 4. Interestingly, at low mole ratios of DMPE, the transition temperature is lower than expected if ideal mixing were occurring, however, beyond 80 mol percent DMPE, the melting temperatures are higher than predicted were ideal mixing occurring. A nonlinear trend in melting temperature of DMPC/DMPE lipid mixtures has also been observed by Melchior et al., who attributed the asymmetry and broad calorimetric traces to fractional crystallization.62 PARTITIONING STUDIES A. Partitioning in DLPC/DMPC vesicles Fluorescence absorbance and emission of C152 in bulk solvents have been previously reported27 and are summarized in Table 3. Four solvents were chosen to represent the different fluorescence environments may encounter in a lipid vesicle solution: cyclohexane (hydrophobic membrane interior), acetonitrile and methanol (polar regions near the lipid headgroups), and buffer (unassociated with the vesicle or fully solvated within the water pool inside the vesicle). Fluorescence lifetimes of C152 vary systematically with solvent polarity. C152 in buffer is fit to two exponential decays, with a large contribution due to a fast (~0.5 ns) decay. In polar solvents such as methanol and acetonitrile, the fluorescence lifetime is longer (~1-2 ns), while in nonpolar solvents, fluorescence decay is significantly longer (~4 ns). When C152 is added to vesicle containing buffer solutions, C152’s fluorescence lifetime behavior changes dramatically, indicating that vesicles are providing C152 with heterogeneous solvation environments. At temperatures from 10 – 50 °C, fluorescence decays of C152 in vesicle solutions are fit to a short

ACS Paragon Plus Environment

14

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

lifetime (τ1= ~0.5 ns), an intermediate lifetime (τ2= ~1.5 ns), and a long lifetime (τ3= 4 ns). Due to their similarities with lifetimes measured in bulk solvents over the range of temperatures sampled, τ1 and τ3 are assigned to C152 solvated in bulk aqueous solution and in the nonpolar, hydrophobic bilayer interior, respectively. The intermediate lifetime has been assigned to solutes sampling the polar lipid interior in the lipid headgroup region. Using the radiative rate corrected, normalized fluorescence amplitudes of C152 in vesicle solutions, the relative percent of C152 in each environment in lipid vesicles can be estimated.

Table 3. Photophysical Properties of C152 in Selected Solvents at 10 ºC.

Solvent

λem λexc a (nm) (nm)a

kf (107 s-1)

φf

τf (ns)d

cyclohexane

372

426

25.1

0.97b

3.86

acetonitrile

396

502

9.8

0.23b

2.34

methanol

397

515

9.0

0.09c

1.09

0.01 M PBS buffer

404

527

8.1

0.05c

0.62 (0.83), 3.75 (0.17)

a Measured at room temperature. b Quantum yields as reported by Pal et al 63. c Quantum yields measured in this work. Experiments were conducted at room temperature, and remain virtually unchanged from 10 – 70 °C. d Uncertainties in reported lifetimes are ± 0.10 ns. Uncertainties in reported amplitudes (for the PBS buffer solution) are ±0.05.

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry

The partitioning of C152 in pure DLPC and pure DMPC lipid vesicles have been described in detail previously.27 Time resolved fluorescence emission traces of C152 in pure DLPC vesicles, pure DMPC vesicles, and vesicles composed of 50% DLPC / 50% DMPC at 10 °C is shown in Figure 5.

Similar traces were recorded for all lipid mixtures over a

temperature range of 10–30 °C.

10

Counts

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

10

10

10

4

3

DLPC DMPC 50% DLPC / 50% DMPC

2

1

0

2

4

6 8 Time (ns)

10

12

Figure 5. Time resolved fluorescence emission from C152 in DLPC vesicles (red), DMPC vesicles (green), and vesicles composed of 50% DLPC/ 50% DMPC (blue). Fluorescence was recorded at 10 °C. Solid black lines are fits from analyzing the traces according to Eq. 1. Also included is a representative instrument response function (IRF) from a scattering solution. Typical IRF traces measured ~100 ps. Once fit to Eq. 1, fluorescence contributions of C152 in buffer, the polar headgroup region, or embedded in the membrane interior were calculated (Supporting Information). Fluorescence contributions of C152 in select mixtures are shown in Figure 6, and mixtures of additional mole ratios are shown in the Supporting Information. Dashed lines in Figure 6 represent ideal C152 partitioning assuming that partitioning into the mixed lipid bilayer can be represented simply as a weighted average with contributions from C152 partitioning into pure DLPC and DMPC bilayers:

ACS Paragon Plus Environment

16

Page 17 of 31

AiXDMPC (ideal) = XDMPC ·AiDMPC XDLPC ·AiDLPC

(6)

where XDMPC is the mole fraction of DMPC and XDMPC + XDLPC = 1 and Ai is the lifetime contribution where i is the buffer, polar, or nonpolar regions. (Similar plots for all DLPC/DMPC ratios studied are shown in Supporting Information.)

100% DLPC

0.8 0.6 0.4 0.2 0.0 10

15

25 30 25 20 Temperature (ºC)

15

0.8 0.6 0.4 0.2 0.0

10

10

50% DLPC / 50% DMPC

15

0.8 0.6 0.4 0.2 0.0

20

25 30 25 20 Temperature (ºC)

15

10

20% DLPC / 80% DMPC

1.0 Lifetime Contribution

1.0 Lifetime Contribution

20

80% DLPC / 20% DMPC

1.0 Lifetime Contribution

Lifetime Contribution

1.0

0.8 0.6 0.4 0.2 0.0

10

15

20

1.0 Lifetime Contribution

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

25 30 25 20 Temperature (ºC)

15

10

10

15

20

25 30 25 20 Temperature (ºC)

15

10

100% DMPC

0.8 0.6 0.4 0.2 0.0 10

15

20

25 30 25 20 Temperature (ºC)

15

10

Figure 6. Lifetime contribution of C152 in buffer (red circles), polar (black squares), and nonpolar (blue triangles) environments in vesicles composed of pure DLPC (top left), 20% DMPC/ 80% DLPC (top right), 50% DMPC/ 50% DLPC (middle left), 80% DMPC/ 20% DLPC

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

(middle right), and pure DMPC (bottom left). Note that these data show a cycle with temperature increasing and subsequently decreasing on the x-axis.

Table 4. Percent of C152 in the buffer, polar region, and nonpolar region of vesicles at selected temperatures of 10, 20, and 30 °C. Reported uncertainties reflect one standard deviations and are calculated from the number of trials displayed next to each vesicle solution. Temp.

%C152:

% C152:

%C152:

(°C)

Buffer

Polar

Nonpolar

10

7±2

78 ± 5

15 ± 5

20

11 ± 1

85 ± 2

4±1

30

14 ± 1

82 ± 2

3±1

80 DLPC: 20 DMPC

10

7±2

87 ± 4

6±2

20

13 ± 5

85 ± 5

2±1

(n=3)

30

17 ± 6

81 ± 6

2±1

50 DLPC: 50 DMPC

10

22 ± 6

61 ± 11

16 ± 5

20

18 ± 6

76 ± 8

6±3

(n=4)

30

21 ± 7

75 ± 8

4±3

20 DLPC: 80 DMPC

10

27 ± 6

64 ± 1

8±4

20

20 ± 5

76 ± 3

4±2

(n=3)

30

16 ± 5

82 ± 5

2±1

Pure DMPC

10

31 ± 4

54 ± 4

13 ± 2

20

23 ± 3

69 ± 5

5±2

(n=7)

30

9±2

87 ± 3

4±2

Mole Ratios

Pure DLPC (n=4)

Partitioning of C152 in mixed DLPC/DMPC vesicle shows only slight deviations from ideal behavior (as determined by Eqn. 6). DLPC exists in its liquid crystalline state over all the

ACS Paragon Plus Environment

18

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

entire sampled temperature range. In pure DLPC vesicles at 10 °C, ~10-15% of C152 partitions into the nonpolar region of the lipid membrane, ~5-10% of C152 is solvated in the buffer, and the majority of C152 (~80%) partitions into the polar lipid headgroup region (~80%). At 20 °C, C152 is exsolvated from the membrane interior, and the population of C152 associated with the lipid headgroups and C152 in bulk aqueous solution increases accordingly. Nearly ideal C152 partitioning behavior is observed in all vesicles composed of binary mixtures of DLPC and DMPC (data shown in Supporting Information). Ideal C152 partitioning behavior manifests differently in vesicles composed of equimolar DLPC and DMPC. As determined by DSC, the phase transition occurs over a broad temperature range (7.3–14.9 °C). At the lowest temperature sampled in these partitioning studies (10 °C), the largest population of C152 is in the polar region of the lipid headgroups (~60%), while 20% of C152 is in the membrane interior and the remaining 20% of C152 is not associated with the membrane. At 20 °C, the population of C152 solvated in the membrane interior decreases to 80% DMPC, C152 partitioning is nearly ideal and follows previously reported trends of C152 partitioning into DMPC and DPPC vesicles.27 Most evident in these mixtures is the increase in C152 association with the polar lipid headgroups near the main phase transition temperature.

ACS Paragon Plus Environment

19

The Journal of Physical Chemistry

Of interest is the relative amount of C152 associated with the membrane (Kpart) as a function of the reduced temperature (T/Tm) of the vesicle system. Although membranes with lower compositions 1.00 is observed for pure DMPC, 95% DMPC, and 90%

ACS Paragon Plus Environment

20

Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

DMPC vesicles. The change in partition coefficient is more gradual and smaller in magnitude for vesicles composed of both 95% and 80% DMPC. The origin of this non-monotonic behavior is not clear and no ready explanation emerges from DSC measurements in Figure 1. Vesicles comprised of equimolar DLPC/DMPC mixtures showed a relatively small change in partitioning as a function temperature, consistent with membranes with this composition not having a welldefined melting transition temperature. One unexpected but reproducible observation is that partition coefficients of C152 in pure DMPC vesicles and 90% DMPC vesicles just above their respective melting temperatures are almost double that of vesicles with DMPC compositions of 95%, 80%, and 50%. The reason for this dramatic increase in partition coefficient at 90% DMPC and 100% DMPC (but not 95% DMPC) is unknown.

In addition, results from

simulations performed with mixed lipid systems could prove very enlightening. B. Partitioning in DMPC/DMPE vesicles C152 partitioning in vesicles composed of mixtures of DMPC and DMPE was analyzed in a similar manner to partitioning analysis of C152 in DLPC/DMPC mixtures.

The lifetime

contribution of C152 in buffer, polar, and nonpolar regions over a temperature range of 10–50 °C is shown in Figure 8 and summarized in Table 5.

ACS Paragon Plus Environment

21

The Journal of Physical Chemistry

1.0

100% DMPC Lifetime Contribution

Lifetime Contribution

1.0 0.8 0.6 0.4 0.2 0.0 20

1.0

30

40 50 40 30 Temperature (ºC)

20

0.8 0.6 0.4 0.2

10

10

50% DMPC / 50% DMPE

20

0.8 0.6 0.4 0.2 0.0

30

40 50 40 30 Temperature (ºC)

20

10

20% DMPC / 80% DMPE

1.0 Lifetime Contribution

Lifetime Contribution

80% DMPC / 20% DMPE

0.0 10

0.8 0.6 0.4 0.2 0.0

10

20

30

40 50 40 30 Temperature (ºC)

20

10

10

20

30

40 50 40 30 Temperature (ºC)

20

10

100% DMPE

1.0 Lifetime Contribution

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

0.8

Buffer ( τ1)

0.6

Polar (τ2)

0.4

Nonpolar ( τ3)

0.2 0.0 10

20

30

40 50 40 30 Temperature (ºC)

20

10

Figure 8. Lifetime contribution of C152 in buffer (red circles), polar (black squares), and nonpolar (blue triangles) environments in vesicles composed of pure DMPC (top left), 80% DMPC/ 20% DMPE (top right), 50% DMPC/ 50% DMPE (middle left), 20% DMPC/ 80% DMPE (middle right), and pure DMPE (bottom left). Note that these data show a cycle with temperature increasing and subsequently decreasing on the x-axis.

ACS Paragon Plus Environment

22

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Table 5. Percent of C152 in the buffer, polar region, and nonpolar region of vesicles at 10, 20, and 30 °C* DMPC: DMPE Pure DMPC (n=7) 80 DMPC: 20 DMPE (n=4) 50 DMPC: 50 DMPE (n=3) 20 DMPC: 80 DMPE (n=3) Pure DMPE (n=7)

Temp. (°C) 10 20 30 50 10 20 30 50 10 20 30 50 10 20 30 50 10 20 30 50

%C152 Buffer 31 ± 4 23 ± 3 9±2 25 ± 7 53 ± 1 50 ± 2 20 ± 2 21 ± 1 77 ± 2 71 ± 1 60 ± 12 32 ± 8 77 ± 5 77 ± 1 76 ± 3 38 ± 12 82 ± 2 82 ± 1 86 ± 5 78 ± 4

%C152 Polar 54 ± 4 69 ± 5 87 ± 3 71 ± 7 29 ± 1 41 ± 4 76 ± 4 76 ± 3 12 ± 1 23 ± 1 35 ± 13 66 ± 6 14 ± 6 15 ± 1 19 ± 2 58 ± 10 14 ± 5 13 ± 1 10 ± 5 16 ± 6

%C152 Nonpolar 13 ± 2 5±2 4±2 4±1 17 ± 1 10 ± 1 5±2 3±2 11 ± 1 5±1 5±2 6±2 8±1 8±1 4±2 5±2 5±3 5±1 4±2 6±2

*Standard deviations are calculated from the number of trials displayed next to each vesicle solution.

Similarities in DSC traces (Figure 3, Table 2) of DMPC rich vesicles indicate that relatively small amounts (