Effects of Polyhexamethylene Biguanide and Polyquaternium-1 on

Article pubs.acs.org/JPCB

Effects of Polyhexamethylene Biguanide and Polyquaternium‑1 on Phospholipid Bilayer Structure and Dynamics Ian J. Horner,† Nadine D. Kraut,‡ Jerod J. Hurst,† Alyssa M. Rook,† Crystal M. Collado,† G. Ekin Atilla-Gokcumen,† E. Peter Maziarz,§ X. Michael Liu,§ Mohinder M. Merchea,∥ and Frank V. Bright*,† †

Department of Chemistry, Natural Sciences Complex, SUNY-Buffalo, Buffalo, New York 14260-3000, United States PPG Industries Inc., 440 College Park Dr., Monroeville, Pennsylvania 15146, United States § Pfizer Consumer Healthcare, 1211 Sherwood Ave., Richmond, Virginia 23220, United States ∥ LENSAR, Inc., 2800 Discovery Dr., Orlando, Florida 32826, United States Downloaded by UNIV OF CAMBRIDGE on September 1, 2015 | http://pubs.acs.org Publication Date (Web): August 7, 2015 | doi: 10.1021/acs.jpcb.5b07162

‡

ABSTRACT: Multipurpose solutions (MPS) are a single solution that functions to simultaneously rinse, disinfect, clean, and store soft contact lenses. Several commercial MPS products contain polyhexamethylene biguanide (PHMB) and/or polyquaternium-1 (PQ-1) as antimicrobial agents. In this paper we have created an in vitro small unilamellar vesicle (SUV) model of the corneal epithelial surface, and we have assessed the interactions of PHMB and PQ-1 with several model biomembranes by using fluorescence spectroscopy, dynamic light scattering (DLS), and liquid chromatography− mass spectrometry (LC-MS). Steady-state and time-resolved fluorescence were used to assess the membrane acyl chain and polar headgroup region local microenvironment as a function of added PHMB or PQ-1. DLS was used to detect and quantify SUV aggregation induced by PHMB and PQ-1. LC-MS was used to determine the liposomal composition from any precipitated materials in comparison to the as-prepared SUVs. The results are consistent with PHMB adsorbing onto and PQ-1 intercalating into the biomembrane structure. The differences between the two interaction mechanisms have substantial impacts on the biomembrane dynamics and stability.

■

INTRODUCTION It is well-known that cell membrane structure and dynamics are crucial to their fundamental physiological functions.1 As such, knowledge of phospholipid membranes is of tremendous importance in myriad areas.2−7 Over the years, researchers have explored many aspects of phospholipid membrane behavior including quantifying the effects of, for example, antimicrobial peptides,8 cholesterol,9 CO2,10 drugs,11 and spermicides12 on the phospholipid bilayer dynamics, stability, and structure. In the eye care industry, multipurpose solutions (MPS) are used extensively for soft contact lens disinfection, cleaning, and storage.13−15 Commercial MPS are composed of a buffer system, at least one antimicrobial agent, and other additives such as surfactants or humectants used to provide lens comfort and improve performance. Polyhexamethylene biguanide (PHMB) and polyquaternium-1 (PQ-1) are among the most widely used MPS antimicrobial agents.13−15 After a contact lens is soaked and then removed from a MPS and applied to the ocular surface, antimicrobial agents begin to desorb from the contact lens at a rate that depends on the agent and the contact lens material.16−18 How these antimicrobial agents might impact the corneal surface represents a complex © 2015 American Chemical Society

process involving, at the minimum, corneal epithelial surface (bio)chemistry, the instantaneous local agent concentration at the cornea surface, flow rates between the contact lens and corneal surface, and any association between the agent and the corneal surface constituents. Within the human precorneal tear film, PHMB and PQ-1 exist as polycations.19−21 As such, it is not surprising to learn that PHMB interacts with anionic functional groups within peptides (e.g., carboxylates)22 and nucleic acids (e.g., phosphates),23 with cellulose (e.g., carboxylates and sulfonates),19,24 and with highly glycosylated mucins,25 some phospholipids,26 and albumin.27 Similarly, PQ1 is known to interact with anionic dyes,28 polyelectrolytes,29 and cell surface residues.21 Thus, strong evidence for ion−ion interactions between PHMB and PQ-1 and oppositely charged functionalities present at cell surfaces is well established. Further, we recently showed that PHMB and PQ-1 can associate with phospholipid vesicles (i.e., bilayers) that were designed to mimic key aspects of the average human corneal epithelial cell membrane.30 There was no detectable effect on the mimic phospholipid vesicle gel-to-liquid phase transition Received: July 23, 2015 Published: August 3, 2015 10531

DOI: 10.1021/acs.jpcb.5b07162 J. Phys. Chem. B 2015, 119, 10531−10542

Article

Downloaded by UNIV OF CAMBRIDGE on September 1, 2015 | http://pubs.acs.org Publication Date (Web): August 7, 2015 | doi: 10.1021/acs.jpcb.5b07162

The Journal of Physical Chemistry B

elucidate the effects of PHMB, PQ-1, and SUV composition on the biomembrane: (i) gel-to-liquid phase transition temperature, (ii) van’t Hoff enthalpy (phospholipid cooperativity), (iii) acyl chain region dielectric properties, (iv) acyl chain region order parameter, wobbling-in-cone angle and wobbling diffusion coefficient within the acyl chain region, (v) acyl chain region local microviscosity, and (vi) local microenvironment at the polar headgroup region. We have also carried out dynamic light scattering (DLS) and liquid chromatography−mass spectrometry (LC-MS) experiments on fluorophore-free SUV systems to investigate the effects of PHMB and PQ-1 on biomembrane−antimicrobial interactions and aggregation and the SUV composition after antimicrobial interaction, respectively.

temperature in the presence of up to 1000 ppm PHMB. In the case of PQ-1 interacting with mimic phospholipid vesicles, we saw similar behavior up to 7 ppm PQ-1. In this paper we have created several types of phospholipidbased small unilamellar vesicles (SUVs) that are relevant to the average human corneal epithelial cell membrane composition,31 and we focus with more detail exclusively on the PHMB and PQ-1 concentration range between 0 and 10 ppm. This specific concentration range was selected to avoid instability regions seen in our earlier research,30 and it allows us to focus on and identify the root interactions between PHMB and PQ-1 with the phospholipid bilayer structures over more clinically relevant concentration ranges. To facilitate our studies, we have incorporated two fluorescent probe molecules, 1,6-diphenyl1,3,5-hexatriene (DPH) and 6-dodecanoyl-2-dimethylaminonapthalene (Laurdan), within the bilayer structure. DPH and Laurdan are small, fluorescent probe molecules that selectively distribute within and report from the vesicle bilayer acyl chain region (DPH) or the vesicle headgroup region (Laurdan) (Figure 1). DPH and Laurdan have been used extensively in the past to assess biomembrane dynamics and structure.32−49 Here we use steady-state and time-resolved fluorescence spectroscopy to

■

EXPERIMENTAL SECTION Reagents. The following reagents were used: ethanol (Pharmco); ammonium formate (Fluka Analytical); chloroform (Honeywell); KCl, TRIZMA hydrochloride, TRIZMA base, formic acid, ammonium hydroxide, and dimethyl-1,4-bis(5phenyl-2-oxazolyl)benzene (Me2POPOP) (Sigma-Aldrich); PHMB (Mw = 2500 Da, Mn = 1300 Da) and PQ-1 (Mw = 8700 Da, Mn = 4850 Da) (Bausch + Lomb); isopropanol and methanol (MS grade, EMD Millipore); DPH and Laurdan (Molecular Probes/Invitrogen); 1,2-dipalmitoyl-sn-glycero-3phosphocholine (DPPC), 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine (DPPE), 1,2-dipalmitoyl-sn-glycero-3-phosp h o - L -se ri n e ( DP PS) a nd N -p almi toyl- D -e ry t hr osphingosylphosphorylcholine (SM) (Avanti Polar Lipids); and 1,2-dipalmitoylphosphatidylinositol (DPPI) (Cayman Chemicals). LC-MS columns were purchased from Phenomenex. Water was purified using a Milli-Q water purification system (EMD Millipore). Lipid Vesicle Preparation. SUVs were prepared by using previously published protocols.26,30,32−39,50 Four SUV types were prepared: (i) human corneal epithelial mimic (mimic), (ii) control, (iii) control + DPPS (control w/PS), and (iv) control + DPPI (control w/PI). The mimic SUVs were composed of the major phospholipids found within the average human corneal epithelium:31 DPPC (45%), DPPE (13%), DPPS (8%), DPPI (8%), and SM (26%). The mimic SUV phospholipid composition produces a biomembrane surface that is composed of anionic and zwitterionic headgroups. The control SUVs contained DPPC (50%), DPPE (18%), and SM (32%); the phospholipids with anionic headgroups are omitted, resulting in only zwitterionic headgroups at the control bilayer surface. The control w/PI SUVs contained DPPC (48%), DPPE (16%), SM (28%), and DPPI (8%). The control w/PI SUV phospholipid composition produces a bilayer surface with zwitterionic headgroups with DPPI being the only anionic headgroup. The control w/PS SUVs contained DPPC (48%), DPPE (16%), SM (28%), and DPPS (8%). The control w/PS SUV phospholipid composition produces a bilayer surface with zwitterionic headgroups with DPPS being the only anionic headgroup. To create SUVs, the appropriate amounts of each phospholipid were individually dissolved in CHCl3, mixed together within a clean 1 dram glass vial, and the organic solvent was removed with a gentle N2 gas stream. The phospholipids were then resuspended in 2 mL of warm 20 mM pH 7.4 TRIS buffer containing 100 mM KCl. Vesicles were created by sonicating the phospholipid mixtures with a model 4710 ultrasonic homogenizer (Cole-Parmer Instrument) while

Figure 1. Geometry and structure of small unilamellar vesicles (SUVs) containing the fluorescent reporter molecules DPH and Laurdan. z and z′ represent the coordinates describing the vesicle surface normal and the DPH major axis, respectively. The chemical structures of 1,2dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-dipalmitoyl-snglycero-3-phosphoethanolamine (DPPE), 1,2-dipalmitoyl-sn-glycero-3phospho-L-serine (DPPS), N-palmitoyl-D-erythro-sphingosylphosphorylcholine (SM), and 1,2-dipalmitoylphosphatidylinositol (DPPI); DPH and Laurdan are shown for comparison along with the DPH absorbance and emission transition moments (λ ex and λ em, respectively). 10532


Article


The Journal of Physical Chemistry B the aqueous sample was maintained at 60−80 °C. The sonication process (1 min on, 3 min rest, 4 cycles) was performed with the controller set to level “2”. The vesiclecontaining samples were then allowed to cool naturally to room temperature after the final cycle and then centrifuged for 1 h at 13 000 rpm (14 000g) (Galaxy 14D, VWR) to remove large/ multilamellar vesicles. The resulting supernatant containing the desired SUVs was used for experimentation. DPH-doped SUVs were created by adding the appropriate amounts of DPH in EtOH (lipid to DPH mol:mol ratio of 150−200:1) to the initial phospholipid mixtures prior to sonication and SUV formation. (The recovered results from 500:1 preparations were statistically equivalent to those recovered at the higher DPH loadings (n = 65, p > 0.45).) Laurdan-doped SUVs were prepared by creating fluorophorefree SUVs (vide supra) and then allowing Laurdan to spontaneously intercalate into the lipid bilayer. Here, appropriate amounts of Laurdan dissolved in EtOH (lipid to Laurdan mol:mol ratio of 950−1000:1) was added to the SUV solution, and the solution was heated above the SUVs phase transition temperature (60 °C). To ensure Laurdan incorporation into the biomembrane architecture, the fluorescence was monitored over time until the Laurdan emission intensity no longer increased (typically less than 1 h). Antimicrobial agents were added to the so-formed SUVs at physiological temperature (37 °C). The agent concentrations listed here refers to the total concentration in the sample. SUVcontaining samples and blanks (no SUV; SUV, no DPH or Laurdan) were allowed to equilibrate for 1 h after antimicrobial agent addition before beginning experiments. Individual SUV/ agent samples were prepared and studied to avoid any possible memory effects. Fluorescence Measurements. Standard 1 cm2 quartz fluorescence cuvettes (Starna) were used. Sample temperature was controlled by using a refrigerated bath circulator (RTE-111, Neslab). Samples were allowed to equilibrate for at least 15 min between temperature adjustments. Steady-state measurements were performed by using a SLMAMINCO model 8100 spectrofluorometer with a 450 W Xe arc lamp and Glan-Thompson calcite polarizers and photomultiplier tube (PMT) detectors. Double- and single-grating monochromators served as the excitation and emission wavelength selection devices, respectively. The excitation/ emission wavelengths were set to 358 nm/428 and 350 nm/ 450 nm for DPH and Laurdan, respectively, and the spectral bandpasses were adjusted to 4 nm. Fluorescence anisotropy measurements were performed in the “L” format. Frequency-domain measurements were carried out on the DPH-loaded SUVs by using an SLM model 48000 MHF spectrofluorometer with Glan-Thompson calcite polarizers and PMT detectors. A CW argon ion laser (Innova 90-6, Coherent) operating at 351.1 nm was used as the excitation source. Magic angle polarization51 was used for all phase-modulation measurements to avoid artifacts associated with rotational reorientation. Emission was detected through a 420 nm long pass filter (Oriel). Me2POPOP dissolved in ethanol served as the reference fluorescence lifetime standard; its lifetime is 1.45 ns.52 For all experiments, the Pockels cell was operated at a repetition rate of 5 MHz, and data were recorded at 5 MHz intervals between 5 and 250 MHz. Fluorescence experiments were performed on at least three occasions using separate samples, and each datum is the result

of at least 15 individual measurements. Average results are reported along with the corresponding measurement standard deviation. There was no detected evidence of DPH or Laurdan photodecomposition or hysteresis under our experimental conditions. Dynamic Light Scattering (DLS) Measurements. DLS measurements were carried out at 37 °C in 1 cm2 quartz fluorescence cuvettes by using a 90Plus particle size analyzer (Brookhaven Instruments Corp.). DLS experiments were performed on at least three occasions using separate samples. Liquid Chromatography−Mass Spectrometry (LC-MS) Measurements. LC-MS measurements were performed by using an Agilent 6530 QTOF-Accurate Mass system. The positive ionization mode was used to detect DPPC, DPPE, and SM phospholipids; negative ionization mode was used to detect DPPS and DPPI phospholipids. Mobile phase A was composed of 95:5 (water:methanol). Mobile phase B was composed of 60:35:5 (isopropanol:methanol:water). 0.1% (v/v) of formic acid and 0.1% (v/v) of ammonium formate were added to the mobile phases in the positive ionization mode. 0.1% (mol/v) of ammonium hydroxide was added to the mobile phases in the negative ionization mode. For negative ionization mode experiments a Gemini C18 reversed phase column (5 μm, 4.6 mm × 50 mm) with a Gemini C18 reversed phase guard cartridge was used. For positive ionization mode experiments a Luna C5 reversed phase column (5 μm, 4.6 mm × 50 mm) with a Luna C5 reversed phase guard cartridge was used. Pure solutions of individual phospholipids were used for LC calibration purposes. The mobile phase flow rate began at 0.1 mL/min for the first 5 min, and it was then increased to 0.5 mL/min for the duration of the run. The gradient started at 5 min with 100% mobile phase A while increasing linearly to 100% mobile phase B over 20 min. Each LC-MS run was repeated twice. Data Analysis and Statistics. All frequency-domain data were analyzed by using Globals WE (Globals Unlimited) with the actual uncertainty in each datum serving as the associated measurement imprecision. For all data, statistical significance was assessed by ANOVA at the 95% confidence level with pairwise comparison (Holm-Sidak test) (p ≤ 0.05 being significant). In all cases the power of performance test exceeded 0.96.

■

THEORY SECTION

The DPH and Laurdan experiments report on the SUV bilayer acyl chain and headgroup regions, respectively.32−49 The steady-state fluorescence anisotropy, rss, is given by52 rss = (IVV − IVH)/(IVV + GIVH)

(1)

where I represents the fluorescence intensity, H and V denote the horizontal and vertical polarizer orientations (e.g., IVH represents the intensity when the excitation and emission polarizers are in the V and H orientation, respectively), and the G term corrects for depolarization caused by the gratings or any other optical system components. It is well-known that phospholipid bilayers undergo reversible temperature-dependent gel-to-liquid phase transitions.32−39 These transitions are described by a characteristic “melting” temperature, Tm (defined by the midpoint in the rss vs T profile32−39), and by an associated van’t Hoff enthalpy 10533


Article

The Journal of Physical Chemistry B R = 1/6θ

(ΔHVH) that reflects subunit (phospholipid) cooperativity during the gel-to-liquid phase transition process:53,54 ΔHVH = 4RTm 2(δα(T )/δT )Tm

Finally, one can also assess the local microenvironment surrounding the DPH reporter molecule within the bilayer acyl chain region in terms of a local microviscosity, ηlocal:32−39

(2)

where α(T), the fraction of SUV in the liquid phase, is given by


α(T ) = [rss,max − rss(T )]/Δrss,total

ηlocal = kBT /6D W Veff

(3)

GP = (I440 − I490)/(I440 + I490)

(4)

∫0

∞

f (τ )τ −1e−t / τ dτ

■

(5)

RESULTS AND DISCUSSION Common MPS contain ≤1 ppm PHMB and/or ≤10 ppm PQ1.60,61 In previous research using DPH within the bilayer acyl chain region30 we showed that (i) ≤1000 ppm PHMB had no detectable effect on control or mimic liposome Tm values, (ii) ≤1000 ppm PQ-1 had no effect on control liposome Tm values, and (iii) ≥7 ppm PQ-1 began to alter the mimic liposome Tm. Given these previous results, the current experiments focus on the antimicrobial agent concentration range between 0 and 10 ppm PHMB and PQ-1, allowing us to focus on the root interactions between PHMB and PQ-1 with the phospholipid bilayer acyl chain (DPH) and headgroup (Laurdan) regions and on the issue of SUV aggregation (DLS). DPH (Acyl Chain Region). Figure 2 presents typical temperature-dependent steady-state fluorescence anisotropy profiles for DPH-loaded control (A) and mimic (B) SUVs at several PHMB and PQ-1 concentrations. The current results are in full agreement with our earlier results that covered the 0 to 1000 ppm PHMB and PQ-1 range.30 The results in Figure 2 show a single well-defined gel-toliquid phase transition temperature in each case. The presence of 6 ppm PHMB or PQ-1 had no detectable effect on the control SUV Tm (p = 0.56). In the mimic SUVs, 6 ppm PHMB did not affect Tm (p = 0.41). In contrast, when the mimic SUVs are challenged with 4 ppm PQ-1, Tm decreased (p = 0.03). Figure 3 summarizes the effects of PHMB, PQ-1, and SUV composition on the recovered van’t Hoff enthalpies for DPHloaded SUVs. In the absence of PHMB and PQ-1, the ΔHVH values for the control and mimic SUVs were equivalent to each other (p = 0.92, 500 ± 30 kJ mol−1) and similar to values reported by other authors for pure DPPC vesicles (320−490 kJ mol−1).62 The addition of up to 6 ppm PQ-1 to the control SUVs had no effect on ΔHVH (p > 0.71), suggesting that the extent of phospholipid cooperativity was unaffected in the control SUV

The DPH time-resolved fluorescence anisotropy decay, r(t), in vesicle systems is often described by using a hindered rotor model:32−39 r(t ) = (r0 − r∞) exp( −t /θ ) + r∞

(6)

where r0, r∞, and θ represent the limiting fluorescence anisotropy in the absence of rotational diffusion, the hindered fluorescence anisotropy at long times following excitation, and the rotational reorientation time, respectively. Under our excitation conditions r0 was 0.390 ± 0.005, demonstrating that the DPH absorbance and emission transition moments (μex, μem) are nearly colinear, lying along the DPH major axis (Figure 1). This result agrees with previous literature reports on DPH in phospholipid vesicles.32−39 In previous DPH/phospholipid vesicle studies, researchers have shown that r∞ is related to the average DPH order parameter, S, within the vesicle bilayer:32−39 S = (r0/r∞)1/2

(7)

One can also assess the DPH reporter range of motion within the context of a wobbling-in-cone model32−39 where the DPH motion is envisioned to undergo free rotation within a conically shaped domain within the bilayer acyl chain region that is described by an angle, Γcone, symmetrically distributed about an axis normal to the bilayer:32−39 Γcone = cos−1((2S2 + 1)/3)1/2

(8)

Further, one can assign a wobbling diffusion coefficient, DW, to describe the DPH dynamics within the conical domain sensed by the DPH reporter molecule:32−39 D W = R(r0 − r∞)/r0

(12)

where I440 is the fluorescence intensity at λem = 440 nm and I490 is the fluorescence intensity at λem = 490 nm. GP measurements were used in the current research to assess the bilayer headgroup microenvironment43,46 as a function of temperature and antimicrobial agent chemistry and concentration.

In this expression, τ represents the fluorescence lifetime, τM denotes the mean fluorescence lifetime, W is the distribution width at half-height and A is obtained from the normalization condition: ∫ ∞ 0 f(τ) dτ = 1. I(τ) is then given by I (t ) =

(11)

In this expression kB, T (in K), and Veff are the Boltzmann constant, Kelvin temperature, and effective DPH molar volume (360 Å3),32−39 respectively. The fluorescent reporter, Laurdan, selectively locates to and reports from the liposome bilayer outer surface headgroup region, and the Laurdan emission spectrum is highly sensitive to its local microenvironment polarity (Figure 1).40−43,46−49 Specifically, Laurdan exhibits a large red-shift in its emission spectrum in polar media in comparison to the spectrum observed in nonpolar solvents. This red-shift arises from dipolar solvent relaxation.40,44,45 To quantify the Laurdan spectral shift, Gratton and co-workers44 established the generalized polarization (GP) concept:

In eqs 2 and 3, R is the ideal gas constant (8.314 J/(mol K)), rss,max is the maximum steady-state fluorescence anisotropy, rss(T) is the steady-state fluorescence anisotropy at any T (in K), and Δrss,total is the difference between the maximum and minimum steady-state fluorescence anisotropy. The DPH time-resolved fluorescence intensity decay, I(t), in phospholipid vesicle systems has been described by multiexponential and/or continuous lifetime distribution models.55−58 In this research, our multifrequency phase-modulation data were fit well (based on χ2 ( 0.56). The increase in ΔHVH shows that the extent of phospholipid cooperativity (i.e., the melting unit size) progressively increased when PHMB was added to either SUV type, but this process did not affect the overall phospholipid bilayer Tm. This suggests that PHMB association with the SUV bilayer surface30 caused additional phospholipids to act in concert during the gel-to-liquid phase transition in comparison to the SUVs in the absence of antimicrobial agent. This result is consistent with PHMB molecules binding to the SUV bilayer surface in such a way that individual PHMB molecules interact with multiple phospholipids simultaneously causing further correlated interaction among the phospholipids within the bilayer. This is not 10535


Article



the magnitude of this decrease was lower in comparison to the control SUV bilayer. Thus, the ability of PHMB to exclude water from the bilayer acyl chain region was actually greater in the control SUV bilayer in comparison to the mimic SUV bilayer. Together, these results suggest that PHMB may preferentially interact with the zwitterionic phospholipids over the anionic phospholipids. This behavior could result from the exact spatial distribution of these species at the bilayer interface, the interphospholipid headgroup distances in comparison to the PHMB repeat unit distance, and/or some other type of steric effect. Finally, when PQ-1 was added to the mimic SUVs, we saw the DPH τM and W decrease dramatically. The relative standard deviation in the DPH lifetime distribution is essentially unchanged with added PQ-1, suggesting that intrinsic extent of heterogeneity is unchanged in the mimic SUV bilayer upon adding up to 6 ppm PQ-1. The decrease in DPH τM suggests that PQ-1 association to mimic bilayer leads to an increase in the local ε surrounding the DPH molecule. This behavior could derive from PQ-1 inducing additional water to penetrate the mimic SUV bilayer acyl chain region and is consistent with PQ-1 intercalating within the mimic SUV bilayer or otherwise opening the biomembrane and facilitating water penetration into the bilayer. Together, these results suggest that PQ-1 preferentially interacts with the anionic phospholipids over the zwitterionic phospholipids. This could result from the exact spatial distribution of these species at the bilayer interface, the interphospholipid headgroup distance in comparison to the PQ-1 repeat unit distance, and/or some other type of steric effect. Figure 6 summarizes the effects of PHMB and PQ-1 concentration and SUV composition at 37 °C on the recovered DPH rotational reorientation time (A) and hindered anisotropy (B) within the acyl chain region.

Figure 5. Antimicrobial agent-dependent mean excited-state DPH fluorescence lifetimes (A) and distribution width at half maxima (B) for control and mimic SUV bilayers at 37 °C. A typical error bar is also shown.

(p = 0.03), and W was more narrow for the control SUV in comparison to the mimic SUV (p = 0.03). This suggests that the local ε surrounding the DPH reporter molecules in the control SUV bilayer acyl chain region was lower in comparison to ε in the mimic SUV bilayer. This result is consistent with the larger fraction of anionic phospholipids present in the mimic SUVs in comparison to the control SUVs and/or an increase in the water accessing the mimic SUV bilayer acyl chain region. In addition, the local microheterogeneity surrounding the DPH reporter molecule in the control SUV bilayer was less in comparison to the mimic SUV bilayer. When PQ-1 was added to the control SUVs, there was no detectable change in the DPH τM or W (p > 0.62). This result is consistent with the results in Figures 2 and 3, and it argues that PQ-1 does not affect the control SUV bilayer acyl chain region. When PHMB was added to the control SUVs, we observed a dramatic increase in the DPH τM and concomitant decrease in W. The increase in DPH τM is consistent with PHMB association to the control SUV bilayer, causing a decrease in the local ε surrounding the DPH molecule. The decrease in W is consistent with the local microenvironment surrounding the DPH reporter molecule becoming more homogeneous. Together these results are consistent with PHMB-to-control bilayer binding leading to significant exclusion of water from the SUV bilayer acyl chain region. When PHMB was added to the mimic SUVs, we saw that τM and W both increased. The relative standard deviation of the distribution was essentially PHMB independent, suggesting that the intrinsic extent of heterogeneity is unchanged in the mimic SUV bilayer. The increase in DPH τM was less in comparison to the behavior seen with the mimic SUVs. This result argues that PHMB association to mimic SUV bilayer surface lowers the local ε surrounding the DPH molecule within the bilayer structure, but

Figure 6. Antimicrobial agent-dependent DPH rotational reorientation times (A) and hindered fluorescence anisotropies (B) for control and mimic SUV bilayers at 37 °C. A typical error bar is also shown. 10536


Article


The Journal of Physical Chemistry B Several aspects of these data merit additional discussion. In the absence of antimicrobial agent, the DPH θ in the control SUV was slightly greater in comparison to the mimic SUV (p = 0.04) and r∞ was equivalent for the two SUV types (p = 0.68). This suggests that the dynamics surrounding DPH in the control SUV bilayer were slightly slower in comparison to the mimic SUV bilayer. When PQ-1 was added to the control SUVs, there was no detectable change in the DPH θ or r∞ (p > 0.7). This result is consistent with the results in Figures 2, 3, and 5, and it again argues that PQ-1 does not affect the control SUV bilayer acyl chain region. When PHMB was added to the control SUVs, we see dramatic increases in the DPH θ and r∞ values. The increase in DPH θ is consistent with PHMB association to the control SUV bilayer, causing a decrease in the average DPH molecule dynamics within the acyl chain region. The increase in r∞ is consistent with the local microenvironment surrounding the DPH reporter molecule becoming more restrictive to DPH motion. When PHMB was added to the mimic SUVs, we saw that θ and r∞ both increased. The increase in DPH θ was less in comparison to the behavior seen with the control SUVs. This result argues that PHMB association to mimic SUV bilayer surface decreased the DPH local dynamics within the acyl chain region, but the magnitude of this decrease was less in comparison to the control SUV bilayer. Thus, PHMB impacts the control SUV bilayer acyl chain region more so in comparison to the mimic SUV bilayer acyl chain region. Unlike the behavior in any other system studied, when PQ-1 was added to the mimic SUVs, we saw that the DPH θ and r∞ both decrease dramatically. The decrease in DPH θ suggests that PQ-1 association to mimic SUV bilayer leads to an increase in the DPH local dynamics within the acyl chain region. This result is consistent with PQ-1-mimic SUV bilayer binding allowing additional water to penetrate the mimic SUV bilayer and the aforementioned PQ-1 intercalation concept (vide supra). Figures 7, 8, and 9 summarize the effects of PHMB and PQ-1 concentration and SUV composition at 37 °C on the DPH

Figure 8. Antimicrobial agent-dependent DPH wobble-in-cone angle from within control and mimic SUV bilayers at 37 °C. A typical error bar is also shown.

Figure 9. Antimicrobial agent-dependent DPH wobbling diffusion coefficient from within control and mimic SUV bilayers at 37 °C. A typical error bar is also shown.

phospholipid acyl chains to restrict the DPH reporter range of motion (S, Γcone) and decrease the DPH rotational dynamics (DW). This behavior is consistent with PHMB molecules adsorbing to the bilayer surface (vide inf ra). When PQ-1 was added to the mimic SUV, S decreased, Γcone increased, and DW increased. This behavior is consistent with PQ-1 binding to the mimic SUV bilayer inducing the acyl chains to “open” around the DPH reporter molecules, increasing the DPH reporter molecule range of motion (S, Γcone) and simultaneously increasing the rotational dynamics (DW) within the acyl chain region. This behavior is consistent with PQ-1 molecules intercalating into the bilayer (vide inf ra). In Figure 10, the DPH dynamics within the acyl chain region are assessed in terms of an average local microviscosity surrounding the DPH reporter molecules within the SUV bilayers. In the absence of PHMB and PQ-1, ηlocal was independent (p > 0.52) of SUV composition (4.1 ± 0.2 cP). When PQ-1 was added to the control SUV, there was no detectable change in ηlocal. This result is consistent with a lack of interaction between the control SUV bilayer and PQ-1 as discussed previously. When PHMB was added to the control or mimic SUV, we saw ηlocal increase. The increase in ηlocal is larger (p < 0.03) for the control SUVs in comparison to the mimic SUVs. This result is consistent with PHMB adsorbing onto the bilayer surface (vide inf ra). When PQ-1 was added to the mimic SUV, we saw ηlocal decrease. This is consistent with PQ-1 association with the mimic SUV leading to PQ-1 intercalation into the bilayer,

Figure 7. Antimicrobial agent-dependent DPH order parameter within control and mimic SUV bilayers at 37 °C. A typical error bar is also shown.

order parameter, wobble-in-cone angle, and wobbling diffusion coefficient, respectively, within the acyl chain region. In the absence of PHMB and PQ-1, S, Γcone, and DW are independent of SUV composition (p > 0.74). When PHMB was added to the control or mimic SUVs, S increased, Γcone decreased, and DW decreased. This behavior was found to be independent of SUV composition (p > 0.33). This behavior is consistent with PHMB binding to the SUV bilayer inducing the 10537


Article


Making the assumption that these temperature-dependent GP profiles can be interpreted using eqs 2 and 3, we can recover ΔHVH values that can be associated with the SUV polar headgroup region (Figure 11B). Inspection of these results shows that the mimic ΔHVH values in the absence of antimicrobial agents using DPH and Laurdan probes are statistically equivalent (p = 0.92) to each other (vide supra). Addition of PHMB to the mimic SUVs caused ΔHVH for the polar headgroup region to systematically increase. The increase in ΔHVH shows that the extent of phospholipid cooperativity (i.e., the melting unit size) progressively increased when PHMB was added to the mimic SUVs, and this result parallels our earlier DPH findings from the acyl chain regions. This supports that PHMB association with the SUV bilayer surface30 causes additional phospholipids to act in concert during the gel-toliquid phase transition in comparison to the SUVs in the absence of antimicrobial agent. This result is consistent with PHMB molecules adsorbing/binding to the SUV bilayer surface in such a way that individual PHMB molecules interact with multiple phospholipids simultaneously causing further correlated interaction among the phospholipids within the bilayer. The addition of up to 6 ppm PQ-1 to the mimic SUVs had no detectable effect on ΔHVH for the polar headgroup region (p > 0.59), suggesting that the extent of phospholipid cooperativity from the perspective of the lipid exterior headgroup region was unaffected in the mimic SUV bilayer at up to 6 ppm PQ-1. This result is to be contrasted with the behavior seen with DPH and the acyl chain region (Figure 3) where cooperativity decreased as PQ-1 was added. Figure 12 summarizes the effects of antimicrobial agent concentration on the Laurdan-loaded mimic SUV Tm (A) and the GP at 37 °C (B). As PHMB is added, there is a steady increase in the mimic SUV Tm (p = 0.047). These results are consistent with PHMB adsorbing/binding onto the bilayer surface, increasing the


Figure 10. Antimicrobial agent-dependent local microviscosity surrounding DPH within control and mimic SUV bilayers at 37 °C. A typical error bar is also shown.

which in turn opens up the site surrounding the DPH reporter molecules within the acyl chain region (vide inf ra). Laurdan (Exterior Lipid Headgroup Region). Figure 11A presents typical temperature-dependent GP profiles for Laurdan-loaded mimic SUVs without and with PHMB (10 ppm) and PQ-1 (7 ppm).

Figure 11. Typical Laurdan fluorescence results from the exterior polar headgroup region of mimic SUVs. (A) Temperature-dependent steady-state fluorescence Laurdan GP profiles. (B) Antimicrobial agent-dependent van’t Hoff enthalpies for mimic SUV bilayers. The Tm is denoted by the midpoint of each temperature-dependent GP profile. A typical error bar is also shown.

The temperature-dependent GP results show a single welldefined gel-to-liquid phase transition temperature in each case similar to the temperature-dependent rss results found in Figure 2 for DPH reporting from the acyl chain region. In the Laurdan-loaded mimic SUVs, 10 ppm PHMB lead to an increase in Tm (p = 0.047) in comparison to the system without PHMB. In contrast, up to 7 ppm PQ-1 had no detectable effect on the Laurdan-loaded mimic SUV Tm (p = 0.95).

Figure 12. Antimicrobial agent-dependent Tm (A) and the antimicrobial agent-dependent steady-state fluorescence Laurdan GP at 37 °C (B) for mimic SUV bilayers. A typical error bar is also shown. 10538


Article


The Journal of Physical Chemistry B bilayer stability as described previously (vide supra). In contrast, addition of PQ-1 had no detectable effect on the mimic SUV Tm (p = 0.95). Thus, the exterior headgroup and acyl chain region behaviors are different in the presence of PQ-1. The GP value reported by Laurdan in our mimic SUV bilayer in the absence of preservative agent is ∼0.56 (Figure 12B). This value is consistent with values reported by Gratton for other DPPC-based SUV systems.43 At 37 °C, when PHMB was added to the mimic SUV, we saw a statistically relevant increase in GP (polarity surround the Laurdan decreases) above 5 ppm (p < 0.001). This result is consistent with PHMB-to-mimic bilayer binding leading to significant exclusion of water from the SUV headgroup and a concomitant decrease of water from within the acyl chain regions (cf. Figure 5). When PQ-1 was added to the mimic SUV, we observed no statistical change in the GP between 0 and 7 ppm PQ-1 (p > 0.065). This result gave us no additional insights into how PQ-1 interacts with the mimic bilayer external headgroup region and is most readily explained by the Laurdan reporter molecules, which are dilute in comparison to lipids, not being physically located at sites where PQ-1 intercalates into/associated with the biomembrane surface. Dynamic Light Scattering and LC-MS. In our original offering on antimicrobial agent interactions with mimic and control SUVs,30 we saw evidence for SUV precipitation under certain conditions. Figure 13 summarizes the effects of SUV composition, antimicrobial agent, and agent concentration on the average SUV hydrodynamic diameter.

Figure 13A summarizes the effects of PHMB and PQ-1 on the hydrodynamic diameter of mimic SUVs. In the absence of antimicrobial agent, the mimic SUV diameter is 170 ± 1 nm. As PHMB is added, the hydrodynamic diameter systematically increased by 15 ± 2 nm/ppm PHMB (R2 = 0.95). This result is consistent with PHMB adsorbing and layering onto the SUV bilayer surface. In comparison, when PQ-1 was added to the mimic SUVs, the hydrodynamic diameter increased dramatically by 1.4 ± 0.2 μm/ppm PQ-1 (R2 = 0.82). This result is consistent with PQ-1 causing SUVs to aggregate with each other and eventually precipitate from solution.30 These DLS studies show clearly that PQ-1 induces the mimic SUVs to aggregate, and it is consistent with PQ-1 intercalation into/ interaction with the mimic SUV bilayer and between bilayers of different mimic SUVs. We hypothesize that the terminal ends of the PQ-1 molecule intercalate into or bind to anionic functionalities on the biomembrane surface on separate mimic SUVs which results in significant SUV aggregation leading to a critical point when mimic SUV aggregates precipitate from solution. This scenario is consistent with polyelectrolyteinduced liposome aggregation seen by other researchers.64−75 In Figure 13B, we explore the effects of SUV composition and antimicrobial agent concentration on the SUV hydrodynamic diameter. For ease of comparison, we repeat the results from Figure 13A (dashed traces) and then show results for three additional SUVs as a function of added PQ-1: (i) control SUV where all the anionic phospholipids have been removed, (ii) control w/PI SUV where DPPI is the only anionic phospholipid in the SUV, and (iii) control w/PS SUV where DPPS is the only anionic phospholipid in the SUV. In the absence of DPPS and/or DPPI the average SUV hydrodynamic diameter is 80 ± 1 nm, substantially smaller in comparison to the mimic SUV with DPPI and DPPS (p < 0.001). When PQ-1 was added to the control SUVs, the hydrodynamic diameter increased slightly by 1.2 ± 0.3 nm/ ppm PQ-1 (R2 = 0.92). When PQ-1 was added to control w/PI SUVs, we saw no significant increase in hydrodynamic diameter in comparison to the control SUVs (p = 0.90). This result shows that the PQ-1-to-mimic-bilayer intercalation and aggregation does not depend on the anionic DPPI headgroup alone. When PQ-1 was added to the control w/PS SUVs, we saw the hydrodynamic diameter increased by 25 ± 1 nm/ppm PQ-1 (R2 = 0.91). This size increase was significantly greater in comparison to the other two control SUVs (p < 0.001), but it was substantially less in comparison to the mimic SUVs that had both DPPI and DPPS (p < 0.001). This result shows that PQ-1-to-mimic-bilayer intercalation and aggregation depend somewhat on the anionic DPPS headgroup. Taken together these results show dramatic synergy between the DPPS and DPPI phospholipids within the mimic SUV bilayer that results in PQ-1 intercalation/interaction and the substantial SUV aggregation. Given the results from the previous section, we questioned if PQ-1 was selectively precipitating DPPI and/or DPPS from the mimic SUVs or was the precipitate compositionally equivalent to the mimic SUVs, indicating that intact SUVs are being directly precipitated from solution. Figure 14 presents LC-MS results for as-prepared mimic SUVs and SUVs that were precipitated by PQ-1. Figure 14A presents the extracted ion chromatograms showing the presence of each phospholipid. Figure 14B presents a typical mass spectrum for DPPC. Figure 14C presents fractional contribution based on the abundance for

Figure 13. Dynamic light scattering results for SUVs in the presence of PHMB and PQ-1. (A) Antimicrobial agent-dependent hydrodynamic diameters for mimic SUVs in the presence of PHMB or PQ-1 at 37 °C. (B) Antimicrobial agent-dependent hydrodynamic diameters for mimic SUVs and control SUVs with DPPI or DPPS anionic phospholipids only in the presence of PHMB or PQ-1 at 37 °C. The uncertainty in the hydrodynamic diameter is substantially less in comparison to the datum point dimensions. 10539


Article



Figure 15. Model describing the effects of PHMB and PQ-1 on mimic SUV bilayers at 37 °C.

bilayers of different mimic SUVs. Overall, PHMB and PQ-1 interact with the mimic SUV biomembrane, but they do so by two very unique mechanisms.

Figure 14. Typical LC-MS results for SUVs. (A) Extracted ion chromatograms for each phospholipid. (B) Mass spectrum for DPPC. (C) Fractional contribution for each phospholipid in as-prepared mimic SUVs and mimic SUVs that have been precipitated by PQ-1.

■

AUTHOR INFORMATION

Corresponding Author

*Tel 716-645-4180, Fax 716-645-6963, e-mail chefvb@buffalo. edu (F.V.B.).

each phospholipid from the as-prepared mimic SUVs (black) and SUVs that were precipitated by 10 ppm PQ-1 (red). Figure 14C shows that all phospholipid fractions present in the asprepared mimic SUVs are equivalent to the same phospholipid fractions after PQ-1-induced precipitation (p > 0.07). This result demonstrates that the precipitate formed as a result of PQ-1 addition is compositionally equal to the as-prepared mimic SUVs and reaffirms that PQ-1 intercalates into the mimic SUV bilayer and induces intact mimic SUVs to aggregate and eventually precipitate at higher PQ-1 concentrations; there is not any selective DPPI or DPPS extraction by PQ-1 from the mimic SUVs.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the Prasad research group at the University at Buffalo for the use of their 90Plus particle size analyzer for all DLS experiments. This research is based in part upon work supported by the National Science Foundation under Grants CHE-0848171, CHE-1126301, and CHE-1411435, the Department of Energy under Grant DEFG0290ER14143, and an unrestricted gift from Bausch + Lomb. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding sources.

■

CONCLUSIONS We report the effects of the two most common MPS preservative agents (PHMB and PQ-1) on SUV bilayers that mimic important features of the human corneal epithelial cell membrane chemistry by using fluorescent probes that report from the biomembrane acyl chain and exterior polar headgroup regions, dynamic light scattering, and LC-MS. The results are interpreted using the model shown in Figure 15. Briefly, when PHMB interacts with the mimic SUV bilayers, the phospholipid−phospholipid cooperativity increases, water is excluded from the bilayer acyl chain and polar headgroup regions, the acyl chain region becomes more restrictive/less dynamic, and the PHMB adsorbs by layering onto the bilayer surface. When PQ-1 interacts with the mimic SUV bilayers, the phospholipid−phospholipid cooperativity decreases, the water content within the acyl chain region increases, the acyl chain region becomes less restrictive/more dynamic, and the PQ-1 intercalates within/interacts with the bilayer and between

■

REFERENCES

(1) Sperelakis, N. Cell Physiology Source Book; Academic: San Deigo, 1995. (2) Gregoriadis, G. Liposome Preparation and Related Techniques, 3rd ed.; Informa Healthcare USA, Inc.: New York, 2007; Vol. 1. (3) Lasie, D. D. Liposomes: From Physics to Applications; Elsevier: Amsterdam, Netherlands, 1993. (4) Mouritsen, O. G. Theoretical Models of Phospholipid Phase Transitions. Chem. Phys. Lipids 1991, 57, 179−194. (5) Papo, N.; Shai, Y. Can We Predict Biological Activity of Antimicrobial Peptides From Their Interactions With Model Phospholipid Membranes? Peptides 2003, 24, 1693−1703. (6) Rand, R. P.; Parsegian, V. A. Hydration Forces Between Phospholipid Bilayers. Biochim. Biophys. Acta, Rev. Biomembr. 1989, 988, 351−376. 10540


Article


The Journal of Physical Chemistry B (7) Seelig, J.; Seelig, A. Lipid Conformation in Model Membranes and Biological Membranes. Q. Rev. Biophys. 1980, 13, 19−61. (8) Thennarasu, S.; Huang, R.; Lee, D.-K.; Yang, P.; Maloy, L.; Chen, Z.; Ramamoorthy, A. Limiting an Antimicrobial Peptide to the LipidWater Interface Enhances Its Bacterial Membrane Selectivity: A Case Study of MSI-367. Biochemistry 2010, 49, 10595−10605. (9) Stott, B. M.; Vu, M. P.; McLemore, C. O.; Lund, M. S.; Gibbons, E.; Brueseke, T. J.; Wilson-Ashworth, H. A.; Bell, J. D. Use of Fluorescence to Determine the Effects of Cholesterol on Lipid Behavior in Sphingomyelin Liposomes and Erythrocyte Membranes. J. Lipid Res. 2008, 49, 1202−1215. (10) Bothun, G. D.; Knutson, B. L.; Strobel, H. J.; Nokes, S. E. Liposome Fluidization and Melting Point Depression by Pressurized CO2 Determined by Fluorescence Anisotropy. Langmuir 2005, 21, 530−536. (11) Bhandary, S.; Basu, R.; Das, S.; Nandy, P. Effect of Some Statin Group of Drugs on the Phase Profile of Liposomal Membrane - A Fluorescence Anisotropy Study. Phase Transitions 2009, 82, 821−830. (12) Troup, G. M.; Wrenn, S. P.; Apel-Paz, M.; Doncel, G. F.; Vanderlick, T. K. A Time-Resolved Fluorescence Diphenylhexatriene (DPH) Anisotropy Characterization of a Series of Model Lipid Constructs for the Sperm Plasma Membrane. Ind. Eng. Chem. Res. 2006, 45, 6939−6945. (13) Missel, P. J.; Lang, J. C.; Rodeheaver, D. P.; Jani, R.; Chowhan, M. A.; Chastain, J.; Dagnon, T. Design and Evaluation of Ophthalmic Pharmaceutical Products. Drugs Pharm. Sci. 2009, 189, 101−189. (14) Rakow, P. L. Current Contact Lens Care Systems. Ophthalmol. Clin. North Am. 2003, 16, 415−432. (15) Robertson, D. M.; Petroll, W. M.; Jester, J. V.; Cavanagh, H. D. The Role of Contact Lens Type, Oxygen Transmission, and CareRelated Solutions in Mediating Epithelial Homeostasis and Pseudomonas Binding to Corneal Cells: An Overview. Eye Contact Lens 2007, 33, 394−398. (16) Willcox, M. D. P.; Phillips, B.; Ozkan, J.; Jalbert, I.; Meagher, L.; Gengenbach, T.; Holden, B.; Papas, E. Interactions of Lens Care With Silicone Hydrogel Lenses and Effect on Comfort. Optom. Vis. Sci. 2010, 87, 839−846. (17) Powell, C. H.; Lally, J. M.; Hoong, L. D.; Huth, S. W. Lipophilic Versus Hydrodynamic Modes of Uptake and Release by Contact Lenses of Active Entities Used in Multipurpose Solutions. Cont. Lens Anterior Eye 2010, 33, 9−18. (18) Dassanayake, N. L.; Garofalo, R. J.; Carey, R.; David, R.; Meadows, D.; Stone, R. Correlating Biocide Uptake and Release Profiles with Corneal Staining and Subjective Symptoms. Invest. Ophthalmol. Vis. Sci. 2005, 46, E-Abstract 915. (19) Blackburn, R. S.; Harvey, A.; Kettle, L. L.; Payne, J. D.; Russell, S. J. Sorption of Poly(hexamethylenebiguanide) on Cellulose: Mechanism of Binding and Molecular Recognition. Langmuir 2006, 22, 5636−5644. (20) Gilbert, P.; Moore, L. E. Cationic Antiseptics: Diversity of Action Under a Common Epithet. J. Appl. Microbiol. 2005, 99, 703− 715. (21) Bruinsma, G. M.; Rustema-Abbing, M.; van der Mei, H. C.; Lakkis, C.; Busscher, H. J. Resistance to a Polyquaternium-1 Lens Care Solution and Isoelectric Points of Pseudomonas Aeruginosa Strains. J. Antimicrob. Chemother. 2006, 57, 764−766. (22) Bromberg, L.; Chang, E. P.; Hatton, T. A.; Concheiro, A.; Magarinos, B.; Alvarez-Lorenzo, C. Bactericidal Core-Shell Paramagnetic Nanoparticles Functionalized with Poly(hexamethylene biguanide). Langmuir 2011, 27, 420−429. (23) Allen, M. J.; Morby, A. P.; White, G. F. Cooperativity in the Binding of the Cationic Biocide Polyhexamethylene Biguanide to Nucleic Acids. Biochem. Biophys. Res. Commun. 2004, 318, 397−404. (24) Kawabata, A.; Taylor, J. A. The Effect of Reactive Dyes Upon the Uptake and Antibacterial Action of Poly(hexamethylene biguanide) on Cotton. Part 2: Uptake of Poly(hexamethylene biguanide) on Cotton Dyed with β-sulphatoethylsulphonyl Reactive Dyes. Dyes Pigm. 2006, 68, 197−204.

(25) Muya, L.; Mowrey-McKee, M. F.; Minno, G.; Meadows, F. Determination of Binding of PHMB to Mucin. Invest. Ophthalmol. Vis. Sci. 2008, 49, E-Abstract 4869. (26) Ikeda, T.; Tazuke, S.; Watanabe, M. Interaction of Biologically Active Molecules with Phospholipid Membranes. I. Fluorescence Depolarization Studies on the Effect of Polymeric Biocide Bearing Biguanide Groups in the Main Chain. Biochim. Biophys. Acta, Biomembr. 1983, 735, 380−386. (27) Riordan, W. T.; Heilmann, S. M.; Brorson, K.; Seshadri, K.; Etzel, M. R. Salt Tolerant Membrane Adsorbers for Robust Impurity Clearance. Biotechnol. Prog. 2009, 25, 1695−1702. (28) Stevens, L. E.; Eckardt, J. I. Spectrophotometric Determination of Polyquaternium-1 with Trypan Blue by a Difference Procedure. Analyst 1987, 112, 1619−1621. (29) Ketelson, H. A.; Desousa, R.; Dassanayake, N. L.; Luck, M. E. Interaction Control of Cationic Biocides Using Labile Anionic Polyelectrolytes. U.S. Patent WO2009123890A2, August 10, 2009. (30) Bright, F. V.; Merchea, M. M.; Kraut, N. D.; Maziarz, E. P.; Liu, X. M.; Awasthi, A. K. A Preservative-and-Fluorescein Interaction Model for Benign Multipurpose Solution-Associated Transient Corneal Hyperfluorescence. Cornea 2012, 31, 1480−1488. (31) Merchant, T. E.; Lass, J. H.; Roat, M. I.; Skelnik, D. L.; Glonek, T. P-31 NMR Analysis of Phospholipids from Cultured Human Corneal Epithelial, Fibroblast and Endothelial Cells. Curr. Eye Res. 1990, 9, 1167−1176. (32) Toptygin, D.; Brand, L. Determination of DPH Order Parameters in Unoriented Vesicles. J. Fluoresc. 1995, 5, 39−50. (33) Lentz, B. R. Use of Fluorescent Probes to Monitor Molecular Order and Motions Within Liposome Bilayers. Chem. Phys. Lipids 1993, 64, 99−116. (34) Lentz, B. R.; Barenholz, Y.; Thompson, T. E. Fluorescence Depolarization Studies of Phase Transitions and Fluidity in Phospholipid Bilayers. 1. Single Component Phosphatidylcholine Liposomes. Biochemistry 1976, 15, 4521−4528. (35) Lentz, B. R.; Barenholz, Y.; Thompson, T. E. Fluorescence Depolarization Studies of Phase Transitions and Fluidity in Phospholipid Bilayers. 2. Two-Component Phosphatidylcholine Liposomes. Biochemistry 1976, 15, 4529−4537. (36) Wang, S.; Beechem, J. M.; Gratton, E.; Glaser, M. Orientational Distribution of 1,6-Diphenyl-1,3,5-hexatriene in Phospholipid Vesicles as Determined by Global Analysis of Frequency Domain Fluorometry Data. Biochemistry 1991, 30, 5565−5572. (37) Jahnig, F. Structural Order of Lipids and Proteins in Membranes: Evaluation of Fluorescence Anisotropy Data. Proc. Natl. Acad. Sci. U. S. A. 1979, 76, 6361−6365. (38) Esquembre, R.; Ferrer, M. L.; Gutierrez, M. C.; Mallavia, R.; Mateo, C. R. Fluorescence Study of the Fluidity and Cooperativity of the Phase Transitions of Zwitterionic and Anionic Liposomes Confined in Sol-Gel Glasses. J. Phys. Chem. B 2007, 111, 3665−3673. (39) Mykytczuk, N. C. S.; Trevors, J. T.; Leduc, L. G.; Ferroni, G. D. Fluorescence Polarization in Studies of Bacterial Cytoplasmic Membrane Fluidity Under Environmental Stress. Prog. Biophys. Mol. Biol. 2007, 95, 60−82. (40) Parasassi, T.; Krasnowska, E. K.; Bagatolli, L.; Gratton, E. Laurdan and Prodan as Polarity-Sensitive Fluorescent Membrane Probes. J. Fluoresc. 1998, 8, 365−373. (41) Parasassi, T.; Gratton, E.; Yu, W. M.; Wilson, P.; Levi, M. TwoPhoton Fluorescence Microscopy of Laurdan Generalized Polarization Domains in Model and Natural Membranes. Biophys. J. 1997, 72, 2413−2429. (42) Parasassi, T.; Di Stefano, M.; Ravagnan, G.; Sapora, O.; Gratton, E. Membrane Aging During Cell Growth Ascertained by Laurdan Generalized Polarization. Exp. Cell Res. 1992, 202, 432−439. (43) Parasassi, T.; De Stasio, G.; Ravagnan, G.; Rusch, R. M.; Gratton, E. Quantitation of Lipid Phases in Phospholipid Vesicles by the Generalized Polarization of Laurdan Fluorescence. Biophys. J. 1991, 60, 179−189. 10541


Article


The Journal of Physical Chemistry B (44) Parasassi, T.; De Stasio, G.; d’Ubaldo, A.; Gratton, E. Phase Fluctuation in Phospholipid Membranes Revealed by Laurdan Fluorescence. Biophys. J. 1990, 57, 1179−1186. (45) Parasassi, T.; Conti, F.; Gratton, E. Time-Resolved Fluorescence Emission Spectra of Laurdan in Phospholipid Vesicles by Multifrequency Phase and Modulation Fluorometry. Cell. Mol. Biol. 1986, 32, 103−108. (46) Harris, F. M.; Best, K. B.; Bell, J. D. Use of Laurdan Fluorescence Intensity and Polarization to Distinguish Between Changes in Membrane Fluidity and Phospholipid Order. Biochim. Biophys. Acta, Biomembr. 2002, 1565, 123−128. (47) Nyholm, T.; Nylund, M.; Soderholm, A.; Slotte, J. P. Properties of Palmitoyl Phosphatidylcholine, Sphingomyelin, and Dihydrosphingomyelin Bilayer Membranes as Reported by Different Fluorescent Reporter Molecules. Biophys. J. 2003, 84, 987−997. (48) Sanchez, S. A.; Tricerri, M. A.; Gratton, E. Laurdan Generalized Polarization Fluctuations Measures Membrane Packing MicroHeterogeneity In Vivo. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 7314−7319. (49) Zhang, Y.-L.; Frangos, J. A.; Chachisvilis, M. Laurdan Fluorescence Senses Mechanical Strain in the Lipid Bilayer Membrane. Biochem. Biophys. Res. Commun. 2006, 347, 838−841. (50) Shinitzky, M.; Barenholz, Y. Dynamics of the Hydrocarbon Layer in Liposomes of Lecithin and Sphingomyelin Containing Dicetylphosphate. J. Biol. Chem. 1974, 249, 2652−2657. (51) Spencer, R. D.; Weber, G. Influence of Brownian Rotations and Energy Transfer Upon the Measurements of Fluorescence Lifetime. J. Chem. Phys. 1970, 52, 1654−1663. (52) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer: New York, 2006. (53) Marky, L. A.; Breslauer, K. J. Calculating Thermodynamic Data for Transitions of Any Molecularity From Equilibrium Melting Curves. Biopolymers 1987, 26, 1601−1620. (54) De, S.; Ghosh, A. K.; Basu, R.; Nandy, P. Calculation of van’t Hoff Enthalpy Associated with Aspirin-Induced Change in Liposomal Membrane Anisotropy. Phys. Med. Biol. 1996, 41, 383−386. (55) Parasassi, T.; De Stasio, G.; Rusch, R. M.; Gratton, E. A Photophysical Model for Diphenylhexatriene Fluorescence Decay in Solvents and in Phospholipid Vesicles. Biophys. J. 1991, 59, 466−475. (56) Parasassi, T.; Conti, F.; Glaser, M.; Gratton, E. Detection of Phospholipid Phase Separation. A Multifrequency Phase Fluorometry Study of 1,6-Diphenyl-1,3,5-hexatriene Fluorescence. J. Biol. Chem. 1984, 259, 14011−14017. (57) Squier, T. C.; Mahaney, J. E.; Yin, J. J.; Lai, C. S.; Lakowicz, J. R. Resolution of Phospholipid Conformational Heterogeneity in Model Membranes by Spin-Label EPR and Frequency-Domain Fluorescence Spectroscopy. Biophys. J. 1991, 59, 654−669. (58) Kalb, E.; Paltauf, F.; Hermetter, A. Fluorescence Lifetime Distributions of Diphenylhexatriene-Labeled Phosphatidylcholine as a Tool for the Study of Phospholipid-Cholesterol Interactions. Biophys. J. 1989, 56, 1245−1253. (59) Alcala, J. R.; Gratton, E.; Prendergast, F. G. Resolvability of Fluorescence Lifetime Distributions Using Phase Fluorometry. Biophys. J. 1987, 51, 587−596. (60) Choy, C. K. M.; Cho, P.; Boost, M. V.; Benzie, I. F. F. Do Multipurpose Solutions Damage Porcine Corneal Epithelial Cells? Optom. Vis. Sci. 2009, 86, 447−453. (61) Labbe, A.; Pauly, A.; Liang, H.; Brignole-Baudouin, F.; Martin, C.; Warnet, J.-M.; Baudouin, C. Comparison of Toxicological Profiles of Benzalkonium Chloride and Polyquaternium-1: An Experimental Study. J. Ocul. Pharmacol. Ther. 2006, 22, 267−278. (62) Shafirovich, V. Y.; Batova, E. E.; Levin, P. P. Monitoring of Molecular Dynamics in Lipid Bilayers of Small Unilamellar Vesicles by Magnetic-Field-Sensitive Probes. J. Am. Chem. Soc. 1995, 117, 6093− 6100. (63) Andrews, J. R.; Hudson, B. S. Environmental Effects on Radiative Rate Constants with Applications to Linear Polyenes. J. Chem. Phys. 1978, 68, 4587−4594.

(64) Yaroslavov, A. A.; Kiseliova, E. A.; Udalykh, O. Y.; Kabanov, V. A. Integrity of Mixed Liposomes Contacting a Polycation Depends on the Negatively Charged Lipid Content. Langmuir 1998, 14, 5160− 5163. (65) Yaroslavov, A. A.; Kulkov, V. E.; Polinsky, A. S.; Baibakov, B. A.; Kabanov, V. A. A Polycation Causes Migration of Negatively Charged Phospholipids From the Inner to Outer Leaflet of the Liposomal Membrane. FEBS Lett. 1994, 340, 121−123. (66) Kabanov, V. A.; Yaroslavov, A. A. What Happens to Negatively Charged Lipid Vesicles Upon Interacting with Polycation Species? J. Controlled Release 2002, 78, 267−271. (67) Wytrwal, M.; Bednar, J.; Nowakowska, M.; Wydro, P.; Kepczynski, M. Interactions of Serum with Polyelectrolyte-Stabilized Liposomes: Cryo-TEM Studies. Colloids Surf., B 2014, 120, 152−159. (68) Bordi, F.; Sennato, S.; Truzzolillo, D. Polyelectrolyte-Induced Aggregation of Liposomes: A New Cluster Phase with Interesting Applications. J. Phys.: Condens. Matter 2009, 21, 203102. (69) Bordi, F.; Cametti, C.; Sennato, S.; Diociaiuti, M. Direct Evidence of Multicompartment Aggregates in Polyelectrolyte-Charged Liposome Complexes. Biophys. J. 2006, 91, 1513−1520. (70) Bordi, F.; Cametti, C.; Sennato, S. Polyions Act as an Electrostatic Glue for Mesoscopic Particle Aggregates. Chem. Phys. Lett. 2005, 409, 134−138. (71) Bordi, F.; Cametti, C.; Diociaiuti, M.; Sennato, S. Large Equilibrium Clusters in Low-Density Aqueous Suspensions of Polyelectrolyte-Liposome Complexes: A Phenomenological Model. Phys. Rev. E 2005, 71, 050401. (72) Bordi, F.; Cametti, C.; Diociaiuti, M.; Gaudino, D.; Gili, T.; Sennato, S. Complexation of Anionic Polyelectrolytes with Cationic Liposomes: Evidence of Reentrant Condensation and Lipoplex Formation. Langmuir 2004, 20, 5214−5222. (73) Bordi, F.; Cametti, C.; Gili, T.; Gaudino, D.; Sennato, S. Time Evolution of the Formation of Different Size Cationic LiposomePolyelectrolyte Complexes. Bioelectrochemistry 2003, 59, 99−106. (74) Cametti, C.; Truzzolillo, D. Many Facets of the Polyelectrolyte and Oppositely Charged Colloidal Particle Complexation: Counterion Release and Electrical Conductivity Behavior. J. Phys. Chem. B 2011, 115, 7248−7255. (75) Huebner, S.; Battersby, B. J.; Grimm, R.; Cevc, G. Lipid-DNA Complex Formation: Reorganization and Rupture of Lipid Vesicles in the Presence of DNA as Observed by Cryoelectron Microscopy. Biophys. J. 1999, 76, 3158−3166.

10542


Effects of Polyhexamethylene Biguanide and Polyquaternium-1 on

Recommend Documents