Accepted Manuscript Title: Charged Phospholipid Effects on AAPH Oxidation Assay as Determined Using Liposomes Authors: Kervin O. Evans, David L. Compton, Sanghoon Kim, Michael Appell PII: DOI: Reference: S0009-3084(18)30180-4 https://doi.org/10.1016/j.chemphyslip.2019.02.004 CPL 4739 To appear in: Chemistry and Physics of Lipids Received date: Revised date: Accepted date: 26 September 2018 21 December 2018 19 February 2019 Please cite this article as: Evans KO, Compton DL, Kim S, Appell M, Charged Phospholipid Effects on AAPH Oxidation Assay as Determined Using Liposomes, Chemistry and Physics of Lipids (2019), https://doi.org/10.1016/j.chemphyslip.2019.02.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Charged Phospholipid Effects on AAPH Oxidation Assay as PT Determined Using Liposomes Renewable Products Technology Research Unit, ‡Plant Polymer Research N § U SC RI Kervin O. Evans,*,§ David L. Compton§, Sanghoon Kim‡, Michael Appell† A Unit, USDA, †Mycotoxin Prevention and Applied Microbiology Research Unit, M National Center for Agricultural Utilization Research, 1815 N. University TE D Street, Peoria, IL 61604 * CC EP  Kervin O. Evans A Kervin.Evans@ars.usda.gov Highlights  Anionic lipids within liposomes increase oxidation rate of a fluorescent probe.  Cationic lipids can retard oxidation rate of a fluorescent probe in lipid bilayers. 1  Saturated lipids in liquid or gel phase have greater effect in oxidation retardation in the presence of anionic lipids than cationic lipids. PT Abstract RI The capacity of molecules to inhibit oxidation is widely tested using SC liposomes as host matrices of the antioxidant molecule of interest. Spectroscopic assays are readily used for this purpose, specifically assays U using 2,2′-azobis(2-methylpropionamidine) dihydrochloride (AAPH). In this 4,4-difluoro-5-(4-phenyl-1,3-butadienyl)-4-bora-3a,4a-diaza-s- M using A N work the effect that charged lipids have on an AAPH antioxidation assay indacene-3-undecanoic acid (C11-BODIPY® 581/591) as the reporter TE D molecule was investigated. We measured the diameter, zeta potential and spectroscopic rate of decay and area-under-the-curve (AUC) associated with CC EP liposomes containing C11-BODIPY® 581/591 at varying molar percentages (0 to 10 mol%) of charged (cationic or anionic) lipids and compared the results. We showed that although increasing amounts of cationic or anionic A lipids did change the diameter of the liposomes, size had little to no effect on the area-under-the-curve or decay rate of fluorescence. Increased (more positive) or decreased (more negative) zeta potentials did, on the other hand, affect the spectroscopic decay rates and area-under-the-curve. The results 2 demonstrate the importance of considering the presence of charged lipids in the AAPH antioxidation assay. PT Introduction RI Oxidation causes degradation of lipids and is a major cause of rancidity SC in fats and oils worldwide, especially in the food industry. Several methods, such as encapsulating anitoxidants within liposomes 2-4 and enhancing oils for incorporation within liposomes, have been U with antioxidant properties 1 N employed to combat the loss of quality in oils and fats. It, therefore, is M A important to measure antioxidation capacity of bioactives within liposomes. Typically, antioxidation measurements for bioactives within liposome the use D involve of radical TE systems initiator methylpropionamidine) dihydrochloride (AAPH) like 2,2′-azobis(2- 2, 5-8 . AAPH is an aqueous- CC EP soluble molecule that thermally breaks down into two peroxy radicals capable of oxidizing targeted species 9. Liposomes are versatile and simple versions of cells that, typically, A contain primarily lipids. The fact that most lipids are miscible make liposomes highly versatile and readily tunable for various environments. Past studies have demonstrated that the phase state of liposomes can attenuate the oxidation process 5, 10 , and have also demonstrated that the 3 charge state of lipid matrices like emulsions and micelles can affect the oxidation of lipids 11-12 . This work was conducted to investigate how the charged state of liposomes affects the oxidation process within liposomes. PT This study utilized anionic and cationic phospholipid species incorporated RI within phosphatidylcholine-based liposomes. SC Materials and Methods U 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dimyristoyl(DMPC), 1,2-dipalmitoyl-sn-glycero-3- phosphocholine 1,2-dioleoyl-sn-glycero-3-phospho-(1'-rac- N sn-glycero-3-phosphocholine A (DPPC), 1,2-dipalmitoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DPPG), D (DMPG), M glycerol) (DOPG), 1,2-dimyristoyl-sn-glycero-3-phospho-(1'-rac-glycerol)] TE 1,2-dioleoyl- sn-glycero-3-ethylphosphocholine (DOEPC), 1,2-dimyristoyl(DMEPC), and 1,2-dipalmitoyl-sn- CC EP sn-glycero-3-ethylphosphocholine glycero-3-ethylphosphocholine (DPEPC) were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). 4,4-difluoro-5-(4-phenyl-1,3-butadienyl)-4-bora- A 3a,4a-diaza-s-indacene-3-undecan oic acid (C11-BODIPY® 581/591) was purchased from methylpropionamidine) ThermoFisher dihydrochloride 2,2′-azobis(2- Scientific. (AAPH) and 2-amino-2- (hydroxymethyl)-1,3-propanediol hydrochloride (TRIS-HCl) were purchased 4 from Sigma-Aldrich (Waltham, MA). Sodium chloride and calcium chloride were purchased from Fisher Scientific (St. Louis, MO). Experimental Procedures PT Liposome Preparation. RI Liposomes were created via the hydration method as previously SC described 13. Lipids in chloroform or 2:1 chloroform:methanol mix were dried U under a gentle stream of argon and subsequently dried under a vacuum for N 3 hr. Lipids were hydrated in the appropriate buffers and vigorously mixed A for at least 30 min. Hydrated lipids were subsequently put through five cycles M of freezing and thawing using dry ice in ethanol and a 60°C water bath, D respectively. Hydrated lipids were finally extruded 11-times through two 100- Canada). TE nm filters using a LiposoFast hand-held extruder (AVESTIN, Inc., Ottawa, All manipulations were done above the phase transition CC EP temperature of the respective phospholipid explored. Particle Size, Zeta Potential. A Dynamic light scattering (DLS) experiments were carried out using a Particle Size Analyzer (Model NanoBrook Omni, Brookhaven Instruments Corporation, Holtsville, NY, USA) equipped with a 658 nm diode laser and an avalanche photodiode detector. All measurements were done at 90° 5 detection angle at 23.0°C. For each sample, ten measurements were conducted and each run lasted 20 s. These data were averaged to obtain the size of particles. All measurements were processed using the software The electrophoretic RI hydrodynamic diameter via a multimodal analysis. PT supplied by the manufacturer (9kpsdw, v.5.31), which provided the mean SC mobilites, and hence the calculated zeta potentials, were determined by electrophoresis and phase analysis light-scattering (PALS) using a Zeta- U PALS function of the aforementioned Particle Size Analyzer. Ten N measurements were carried out for each sample at 23.0°C. All the data were M A taken and processed using the software supplied by the manufacturer (PALS Zeta Potential Analyzer, version 5.73). The average of 10 measurements D and the standard deviation are reported. TE AAPH Antioxidation Assay. CC EP The antioxidation assay used was based on the fact that each molecule of radical initiator AAPH broke down into two radicals that oxidize the hydrophobic reporter molecule C11-Bodipy 581/591 9. The antioxidation A assay in this study was modified from previous work 8. Accordingly, lipids were combined to a total stock concentration of 3.75 mM. C11-Bodipy 581/591, stored in ethanol, was added to give a stock concentration of 3.6 M. The lipid mixture was gently mixed and dried under a gentle argon 6 stream to a thin film. Lipids were further dried for 3 h using a condenser speed vacuum unit and stored under argon at -20°C until needed. Lipids were hydrated in buffer (20 mM Tris-HCl, pH 7.4) and mixed for 30 min prior PT to going through five cycles of freeze/thaw (dry ice in ethanol/60°C RI waterbath). We created liposomes by extruding the hydrated lipids 11-times SC through double-stacked filters with 100-nm pores using a LiposoFast handheld extruder (Avestin, Inc., Ottowa, ON, Canada). Liposomes were stored U at room temperature prior to use. We diluted the lipids and C11-Bodipy A N 581/591 to a final concentration of 0.25 mM and 0.24 M, respectively, in the M cuvette. Liposomes were allowed to equilibrate to 37°C just prior to adding enough AAPH for a final concentration of 4 mM. Triplicate to sextuplicate TE here. D measurements were conducted; the average for each series is reported CC EP Results and Discussion Effect of Anionic Lipids on AAPH Antioxidation Assay. The effect of anionic lipids on oxidation rates of the reporter molecule Bodipy™ 581/591 A was determined in DOPC liposomes containing DOPG from 0 to 10 mol%. We chose DOPC and DOPG because each lipid had the same acyl chain length and same degree of acyl chain saturation (18:1) which resulted in all lipids having nearly the same phase transition temperatures (-20°C and 7 18°C for DOPC and DOPG, respectively). This ensures that all lipids are in the same phase state at the experimental temperature (37°C), and there were negligible effects due to lipid phase10. This also made negligible, or PT eliminated, any effects due to varying acyl chain lengths. Figure 1 displays RI the results of the AAPH antioxidation assay as a function of increased DOPG SC within the liposomes. The data show that generated AAPH radicals that oxidized the reporter Bodipy molecule caused a loss of fluorescence signal U more rapidly as DOPG concentration increased (figure 1a) within liposomes. N This loss in the fluorescence signal due to increasing DOPG concentration M A also correlated to declining area-under-the-curve (AUC) values and increasing decay rates (Figure 1b). Figure 1b shows that the AUC values D decreased linearly with respect to DOPG concentration. All decay curves fit TE well to a single-exponential decay (Aoe-rt where Ao is the initial concentration, CC EP r is the decay rate constant and t is time in min), which suggests that C 11Bodipy 581/591 fluorescence loss is well described by a first-order reaction. Analysis of the curves’ decay rates (figure 1b) showed that they were of the A order (from slowest to fastest): 100:0 > 99:1 > 98:2 > 95:5 > 90:10, DOPC:DOPG mol% ratios. Analysis also showed that the decay rates increased nearly linearly up to 5 mol% DOPG. All of this suggests that lipids 8 oxidize faster when negatively charged lipids are present; this agreed well A CC EP TE D M A N U SC RI PT with previous studies that showed lipid oxidation occurred much faster 9 RI 0.8 PT DOPC:DOPG 100:0 DOPC:DOPG 99:1 DOPC:DOPG 98:2 DOPC:DOPG 95:5 DOPC:DOPG 90:10 1.0 SC 0.6 0.4 0.2 U Normalized Fluorescence Intensity 1.2 20 40 60 80 100 120 140 160 180 200 M 0 A N 0.0 D Time (min) A CC EP TE Figure 1a. The kinetics of the normalized fluorescence intensity for Bodipy 581/591 as a function of DOPG presence in DOPC liposomes. Time course for the oxidation of Bodipy 581/591 in DOPC liposome with varying concentrations of DOPG at either 0, 1, 2, 5 or 10 mole percent; oxidation was due to the thermal degradation of AAPH at 37°C. Data represents the average of 3 to 6 experiments. 10 30 0.22 AUC decay rate 0.20 0.16 -1 20 Decay Rates (s ) PT 0.18 RI 0.14 0.12 0.10 SC 15 10 0.08 0.06 U Area Under the Curve (AUC) 25 0.04 0 1 2 3 4 5 6 7 0.02 8 9 10 11 M 0 A N 5 DOPG Mole Percent (%) CC EP TE D Figure 1b. Area-under-the curve and decay rate plot as a function of DOPG. Values for the area-under-the-curve (left) and decay rates for the normalized fluorescence signal of Bodipy 581/591 plotted as a function of DOPG mole percent presence in DOPC liposomes. The error bars shown are the calculated standard deviations. when positively charged ions (Fe3+ and Cu2+) were in the presence of both negatively charged emulsions 12 and in systems composed of negatively A charged fatty acid micelles 14. Comparison was also conducted for liposomes containing phospholipids that have saturated acyl chains but were in a fluid state at 37°C. DMPC and DMPG both have a phase transition temperature of 23°C, 11 exist in a complete fluid state at experimental temperature (data not shown), and were zwitterionic and anionic, respectively, under the experimental conditions. Figure 2a shows that the fluorescence signal was lost more PT rapidly as the concentration of DMPG was increased within DMPC RI liposomes, similar to experiments for DOPC:DOPG mixtures; figure 2b SC shows corresponding decreasing AUC values and increasing decay rates with respect to increasing DMPG concentrations, same as was shown for U DOPC and DOPG mixtures. Analysis also shows that the AUC decreased N and fluorescence decay rates increased, both exponentially in relationship M A to DMPG concentration within DMPC liposomes. The fact that the AUC values and decay rates for DOPC:DOPG mixtures are, respectively, at least D three-times smaller and three-times faster, respectively, than those for TE DMPC:DMPG mixtures suggests that the presence of saturatured lipids, CC EP even in a fluid state, can impede lipid oxidation. This agrees well with findings of Bricarello et al, 2012 10 who demonstrated that the physical state of lipids hindered lipid oxidation. Evidence presented here also suggests A that the physical state of the lipids also to some extent counteracts the ability of negatively charged lipids to speed up oxidation. 12 DMPC:DMPG 100:0 DMPC:DMPG 99:1 DMPC:DMPG 98:2 DMPC:DMPG 95:5 DMPC:DMPG 90:10 0.8 RI 0.6 PT 1.0 0.4 SC Normalized Fluorescence Intensity 1.2 0.2 20 40 60 80 100 120 140 160 180 200 A 0 N U 0.0 M Time (min) A CC EP TE D Figure 2a. The kinetics of the normalized fluorescence intensity for Bodipy 581/591 is shown as a function of DMPG presence in DMPC liposomes. Time course for the oxidation of Bodipy 581/591 in DMPC liposome with varying concentrations of DMPG at either 0, 1, 2, 5 or 10 mole percent; oxidation was due to the thermal degradation of AAPH at 37°C. The data displayed is the average of 3 to 6 experiments. 13 0.024 75 AUC decay rate 0.022 0.018 RI 55 PT 60 -1 0.020 Decay Rates (s ) 65 0.016 50 SC Area Under the Curve (AUC) 70 0.014 40 1 2 3 4 5 6 7 9 10 0.012 11 A DMPG Mole Percent (%) 8 N 0 U 45 D M Figure 2b. Area-under-the-curve and decay rates as a function of DMPG. Plot of the area-under-the-curve (left) and decay rates for the normalized fluorescence of Bodipy 581/591 as a function of DMPG concentration in DMPC liposomes. The error bars shown are the calculated standard deviations. TE Comparison of liposomes still in the gel state under the current CC EP conditions was done using a mixture of DPPC (zwitterionic) and DPPG (anionic) lipids where both have a phase transition temperature of 41°C. Figure 3a shows the results of the AAPH antioxidation assays as a function A of DPPG concentration. Similar to the two previous measurements using phosphatidylglycerol lipids, the loss of fluorescence was faster as more anionic lipid (DPPG, in this case) was present within the liposomes. It was also noticeable that AUC values decreased exponentially and decay 14 DPPC:DPPG 100:0 DPPC:DPPG 99:1 DPPC:DPPG 98:2 DPPC:DPPG 95:5 DPPC:DPPG 90:10 1.0 0.8 PT 0.6 RI 0.4 0.2 0.0 20 40 60 80 100 120 140 160 180 200 U 0 SC Normalized Fluorescence Intensity 1.2 N Time (min) A CC EP TE D M A Figure 3a. The kinetics of the normalized fluorescence intensity for Bodipy 581/591 as a function of DPPG. Time course for the oxidation of Bodipy 581/591 in DPPC liposome with varying concentrations of DPPG at either 0, 1, 2, 5 or 10 mole percent; oxidation was due to the thermal degradation of AAPH at 37°C. (0, 1, 2, 5, or 10 mole percent) within DPPC liposomes. Displayed data is the average of 3 to 6 experiments. 15 80 0.026 75 0.024 AUC decay rates 0.018 55 0.016 50 0.014 45 PT 60 RI 0.020 -1 65 Decay Rates (s ) 0.022 SC AUC 70 0.012 40 0.010 1 2 3 4 5 6 7 8 9 10 11 U 0 DPPG Mole Percent (%) D M A N Figure 3b. Area-under-the-curve and decay rate plot as a function of DPPG. Values for the area-under-the-curve (left) and decay rates for the normalized fluorescence signal of Bodipy 581/591 plotted as a function of DPPG mole percent presence in DPPC liposomes. The error bars shown are the calculated standard deviations. CC EP (figure 3b). TE rates increased nearly exponentially as a function of DPPG concentration Effect of Cationic Lipids on AAPH. Comparison of the effects of cationic (positively) charged lipids (ethylphosphatidylcholine - EPC) was also A investigated. Lipids of comparable saturation and acyl chain length (i.e. DOPC:DOEPC, DMPC:DMEPC, and DPPC:DPEPC) were again matched together and employed to minimize or eliminate any effects that could be due to varying lipid phases throughout the membrane. 16 DOPC liposomes containing DOEPC exhibited increased ability to retain the fluorescence signal of C11-Bodipy 581/591 as the amount of DOEPC present in the lipid PT membrane went from 0 to 10 mol%, as evidenced in figure 4a. SC 0.8 RI DOPC:DOEPC 100:0 DOPC:DOEPC 99:1 DOPC:DOEPC 98:2 DOPC:DOEPC 95:5 DOPC:DOEPC 90:10 1.0 0.6 U 0.4 0.2 N Normalized Fluorescence Intensity 1.2 20 40 60 80 100 Time (min) 120 140 160 M 0 A 0.0 180 200 A CC EP TE D Figure 4a. The kinetics of the normalized fluorescence intensity for Bodipy 581/591 as a function of DOEPC. Time course for the oxidation of Bodipy 581/591 in DOPC liposome containing varying concentrations of DOEPG at either 0, 1, 2, 5 or 10 mole percent; oxidation was due to the thermal degradation of AAPH at 37°C. Data displayed is the average of 3 to 6 experiments. 17 1.0 0.8 100:0 99:1 98:2 95:5 90:10 PT DMPC:DMEPC DMPC:DMEPC DMPC:DMEPC DMPC:DMEPC DMPC:DMEPC RI 0.6 SC 0.4 0.2 0.0 0 20 40 60 80 100 120 160 180 200 A N Time (min) 140 U Normalized Fluorescence Intensity 1.2 CC EP TE D M Figure 4b. Kinetics of Bodipy 581/591 normalized fluorescence intensity as a function of DMEPC in DMPC liposomes. Time course for the oxidation of Bodipy 581/591 as a function of DMEPC (0, 1, 2, 5, or 10 mole percent) within DMPC liposomes is displayed; oxidation was due to the thermal degradation of AAPH at 37°C. Data displayed is the average of 3 to 6 experiments. The increased retention of fluorescence signal also correlated into the AUC A values and decay rates, respectively, increasing and decreasing linearly as a function of DOEPC concentration (supplemental figure 1). Liposomes containing DMPC with increasing amounts of DMEPC, on the other hand, exhibited little to no change in fluorescence signal from 0 to 2 mol% DMEPC, 18 but did show an increasing fluorescence signal from 2- mol% DMEPC to 10 mol% DMEPC. Analysis of the AUC values for DMEPC-containing liposomes reveals a similar trend in that AUC values changed little as well PT over DMEPC concentration of 0 to 2 mol% and but are best described as RI sigmoidal in nature. The decay rates for liposomes with DMEPC also stayed SC relatively the same from 0 to 2 mol% DMEPC, but decreased beyond 2-mole percent DMEPC presence (supplemental figure 2). DPEPC-containing U liposomes exhibited a similar trend to DMEPC-containing liposomes where N there was little to no change in fluorescence intensity for 0, 1 and 2 M A mol%DPEPC, but greater retention of fluorescence signal for DPPC liposomes containing 2, 5, and 10 mol%DPEPC (figure 4c). This, too, D translated into both AUC values that increased and decay rates that TE decreased in a sigmoidal fashion as a function of DPEPC concentration CC EP (supplemental figure 3). For all instances studied, the fact that the fluorescence signal was lost more rapidly in the presence of increased anionic lipid concentration and was A retained more readily with increased cationic lipid concentration clearly suggests that electrostatic interactions play an important in role in the interactions between the reporter molecule C11-Bodipy 581/591 and the radical initiator AAPH. The fact that the fluorescence decay rate of C1119 Bodipy 581/591 correlates to its rate of oxidation by AAPH radicals15 and that C11-Bodipy 581/591 rate of oxidation was either enhanced or retarded by PT anionic and cationic lipids, respectively, further suggests that the rate of RI DPPC:DPEPC: 100:0 DPPC:DPEPC 99:1 DPPC:DPEPC 98:2 DPPC:DPEPC 95:5 DPPC:DPEPC 90:10 1.0 SC 0.8 0.6 U 0.4 0.2 0.0 20 40 60 80 100 120 140 160 180 M Time (min) 200 A 0 N Normalized Fluorescence Intensity 1.2 A CC EP TE D Figure 4c. Kinetics of Bodipy 581/591 normalized fluorescence intensity as a function of DPEPC in DPPC liposomes. Time course for the oxidation of Bodipy 581/591 as a function of DPEPC (0, 1, 2, 5, or 10 mole percent) within DPPC liposomes is displayed; oxidation was due to the thermal degradation of AAPH at 37°C. Data displayed is the average of 3 to 6 experiments. 20 lipid oxidation can be enhanced or limited if anionic or cationic lipids, respectively, are present Validation of the surface charge for each liposome series was PT conducted by measuring the zeta potential of each sample. Results in figure RI 5 (lower panel) show that, indeed, increasing the amount of anionic lipids SC within the liposomes did increase the negative surface charge, and that increasing the cationic lipid present in the liposome increased the positive N U surface charge. We also measured liposome diameter to determine if there A was size differences as function of charged lipid present. Liposomes M containing DOPG were roughly 109 to 112 nm in diameter (figure 5, top D panel). Liposomes containing DMPG had a slightly wider size range, from TE 109 to 118 nm. Both set of liposomes containing DOPG and DMPG exhibit no distinct size trend as a function of anionic lipids. Liposomes containing CC EP DOEPC, DMEPC and DPPG decreased in size as their respective concentrations increased. Liposomes containing DOEPC were roughly 112 nm with 0 mol%DOEPC present but were roughly 105 nm in diameter at 2 A and 5 mol% and about 101 nm in diameter at 10 mol%DOEPC. Liposomes containing DMEPC were about 125 nm in diameter at 0 mol% DMEPC, ~95 nm at 1 mol%, 104 to 101 nm in diameter at 2 and 5 mol%, respectively, and 107 nm at 10 mol%. Liposomes with 21 240 200 PT 180 160 RI 140 120 SC Liposome Diameter (nm) 220 100 U 80 50 N 40 DOPG DOEPC DMPG DMEPC DPPG DPEPC A 20 10 M Potential (mV) 30 0 -10 D -20 -40 -50 -60 1 2 CC EP 0 TE -30 3 4 5 6 7 8 9 10 11 Lipid Mole% A Figure 5. Liposome diameter and -potential as a function of charged lipid. The average liposome diameter (top panel) and zeta potential (bottom panel) as a function of charged lipid concentration within liposomes are displayed. The base phospholipid in liposome for DOPG and DOEPC was DOPC, for DMPG and DMEPC was DMPC, and for DPPG and DPEPC was DPPC. Error bars are ± 1 standard deviation (n = 10). 22 DPPG initially were ~210 nm in diameter at 0 mol%, ~168 nm at 1 mol% and finally around 155 nm at DPPG concentration above 1 mol% DOPG. Liposomes containing DPEPC exhibited the greatest size fluctuation in that PT they alternated between 210 and 170 nm in size (figure 5, top panel) over RI the entire mol% range explored. It would appear possible that size may SC affect the oxidation rate in the assay as smaller liposomes have greater curvature, which translates into tighter lipid packing and subsequently more 16 U defects that allow increased access inside the lipid membrane . DOPC N liposomes, however, extruded at sizes ranging from 30 to 200 nm (64 to 164 M A nm actual diameter) exhibited little to no difference in decay rates of TE (supplemental figure 4). D fluorescence signal as data overlay one another without distinction It is, however, unclear how the increased surface charge allows for CC EP enhanced or limited lipid (or probe in this case) oxidation to occur. Apak et al, 2016 describe antioxidant activity measuring assays like the one employed herein as occurring through the transfer of a hydrogen atom A transfer 15 . Typically, this hydrogen atom transfer occurs between radicals generated by AAPH thermal decomposition and an antioxidant which would compete with the fluorescent probe (Bodipy 581/591). The experiments herein were minus any antioxidant. Therefore, there was no competitive 23 kinetics occurring; instead, the AAPH radicals have direct access to the fluorescent probe for completing the one-electron oxidation. The current work shows that the oxidation of the fluorescent probe occurred faster in the PT presence of negatively charged lipids and slower in the presence of positively RI charged lipids. This may be due to electrostatic interactions where the AAPH SC radicals are more positively charged and are attracted greater to an increasingly negatively charged surface (thus a faster rate of oxidation) or U are repelled greater by an increasingly positively charged surface (thus a N slower rate of oxidation). There is also the possibility that the negatively M A charged lipids create an environment within the lipid bilayer where the fluorescent probe is an energy state where it is more easily oxidized and the D positively charged lipids create an environment where the opposite is true. TE It is unclear, however, which is more likely. CC EP Further analysis of the decay rates for loss in fluorescence (figure 6) reveals that DOPG lipids induced a decay rate that was an order of magnitude faster than that induced by DMPG or DPPG, and that DOEPC A lipids had decay rates that were 3 or 4 times faster than DMEPC or DPEPC lipids. These were not a surprising results as lipid oxidation is fundamentally expected to proceed more rapidly in unsaturated lipids than saturated lipids. 24 Unsaturated lipids possess one or more alkene functional groups that increase potential sites for oxidative transformations. What 0-mole percent 1-mole percent 2-mole percent 5-mole percent 10-mole percent RI -1 Decay Rates (s ) 0.20 PT 0.25 U SC 0.15 N 0.10 M A 0.05 0.00 DMPG DPPG DOEPC D DOPG DMEPC DPEPC A CC EP TE Figure 6. Fluorescence decay rates as a function of charged lipids within liposomes. Decay rates from time course measurements of Bodipy 581/591 are display in relationship to each charged lipid investigated. Error bars are reported as the calculated standard deviation with n = 3 – 6. 25 was surprising was that the physical state of the lipids had little or no effect on decay rates. Liposomes containing the anionic lipids DMPG or DPPG (which were in a fluid and gel state, respectively) had nearly the same decay PT rates and so did liposomes containing the cationic counterpart DMEPC or RI DPEPC (again, fluid state and gel state, respectively). This suggests that SC the physical state of the lipids (fluid or gel state) matters little when all lipids are saturated. This is different from the findings of Bricarello et al, 2012 who U showed that the gel state of lipids further impeded fluorescence decay in N liposomes 10. It should be noted that the lipids Bricarello et al, 2012 employed systems examined herein. M A have a larger difference in phase temperatures (-1°C/41°C) than the lipid The lipids respective phase transition D temperatures can be explained by the difference in acyl chain lengths (12- TE carbons versus 16-carbons, respectively), whereas the ones employed in CC EP this study have phase temperature of 23°C and 41°C due to their respective acyl chain lengths of 14 and 16 carbons. A CONCLUSIONS Various forms of the anionic lipid phosphatidylglycerol were shown to increase the oxidation of C11-Bodipy 581/591 within phosphatidylcholinebased liposomes; several forms of the cationic lipid ethylphosphatidylcholine were found to retard oxidation. Saturated lipids limited enhanced oxidation 26 induced by phosphatidylglycerol lipids but added little to the retardation ability of positively charged lipids. Overall, implications are that positively charged moieties, specifically lipids, may be used to limit oxidation of other PT lipids for such uses as increasing the shelf-life of edible oils, improving the RI useful lifetime of lubrication oils, limiting oxidation in fats used in food and A CC EP TE D M A N U SC beverage processes, or increasing shelf life of paints and pigments. 27 Acknowledgements The authors greatly appreciate and recognize the professional technical assistance provided by Ms. Leslie Smith. Mention of trade names PT or commercial products in this publication is solely for the purpose of RI providing specific information and does not imply recommendation or SC endorsement by the United States Department of Agriculture (USDA). A CC EP TE D M A N U USDA is an equal opportunity provider and employer. 28 References 1. Balanč, B. D.; Ota, A.; Djordjević, V. B.; Šentjurc, M.; Nedović, V. A.; Bugarski, B. M.; Ulrih, N. P., Resveratrol-loaded liposomes: Interaction of PT resveratrol with phospholipids. European Journal of Lipid Science and Laszlo, J. A.; Cermak, S. C.; Evans, K. O.; Compton, D. L.; SC 2. RI Technology 2015, 117 (10), 1615-1626. Evangelista, R.; Berhow, M. A., Medium-chain alkyl esters of tyrosol and U hydroxytyrosol antioxidants by cuphea oil transesterification. European A Evans, K. O.; Compton, D. L., Phosphatidyl-hydroxytyrosol and M 3. N Journal of Lipid Science and Technology 2013, 115, 363-371. 2017, 202, 69-76. Compton, D. L.; Goodell, J. R.; Berhow, M. A.; Kenar, J. A.; Cermak, TE 4. D phosphatidyl-tyrosol bilayer properties. Chemistry and Physics of Lipids CC EP S. C.; Evans, K. O., Feruloylated Products from Coconut Oil and Shea Butter. JAOCS, Journal of the American Oil Chemists' Society 2017, 94 (3), 397411. Tikekar, R. V.; Nitin, N., Effect of physical state (solid vs. liquid) of lipid A 5. core on the rate of transport of oxygen and free radicals in solid lipid nanoparticles and emulsion. Soft Matter 2011, 7 (18), 8149-8157. 29 6. Mosca, M.; Ceglie, A.; Ambrosone, L., Effect of membrane composition on lipid oxidation in liposomes. Chemistry and Physics of Lipids 2011, 164 (2), 158-165. An, C.-B.; Li, D.; Liang, R.; Bu, Y.-Z.; Wang, S.; Zhang, E.-H.; Wang, PT 7. RI P.; Ai, X.-C.; Zhang, J.-P.; Skibsted, L. H., Chain Length Effects in SC Isoflavonoid Daidzein Alkoxy Derivatives as Antioxidants: A Quantum Mechanical Approach. Journal of agricultural and food chemistry 2011, 59 Laszlo, J. A.; Evans, K. O.; Vermillion, K. E.; Appell, M., Feruloyl N 8. U (23), 12652-12657. M A Dioleoylglycerol Antioxidant Capacity in Phospholipid Vesicles. Journal of agricultural and food chemistry 2010, 58 (9), 5842-5850. Zhang, J.; Stanley, R. A.; Melton, L. D., Lipid peroxidation inhibition D 9. TE capacity assay for antioxidants based on liposomal membranes. Molecular CC EP Nutrition & Food Research 2006, 50 (8). 10. Bricarello, D. A.; Prada, M. J.; Nitin, N., Physical and chemical modifications of lipid structures to inhibit permeation of free radicals in a A supported lipid membrane model. Soft Matter 2012, 8 (43), 11144-11151. 11. Mei, L.; McClements, D. J.; Decker, E. A., Lipid Oxidation in Emulsions As Affected by Charge Status of Antioxidants and Emulsion Droplets. Journal of agricultural and food chemistry 1999, 47 (6), 2267-2273. 30 12. Mei, L.; McClements, D. J.; Wu, J.; Decker, E. A., Iron-catalyzed lipid oxidation in emulsions as affected by surfactant, pH and NaCl. Food Chemistry 1998, 61 (3), 307-312. Evans, K. O., Room-temperature ionic liquid cations act as short-chain PT 13. RI surfactants and disintegrate a phospholipid bilayer. Colloids and Surfaces A: 14. SC Physicochemical and Engineering Aspects 2006, 274 (1-3), 11-17. Yoshida, Y.; Niki, E., Oxidation of methyl linoleate in aqueous U dispersions induced by copper and iron. Arch Biochem Biophys 1992, 295 A Apak, R.; Özyürek, M.; Güçlü, K.; Çapanoğlu, E., Antioxidant M 15. N (1), 107-114. Activity/Capacity Measurement. 2. Hydrogen Atom Transfer (HAT)-Based, D Mixed-Mode (Electron Transfer (ET)/HAT), and Lipid Peroxidation Assays. Talbot, W. A.; Zheng, L. X.; Lentz, B. R., Acyl chain unsaturation and CC EP 16. TE Journal of agricultural and food chemistry 2016, 64 (5), 1028-1045. vesicle curvature alter outer leaflet packing and promote poly(ethylene A glycol)-mediated membrane fusion. Biochemistry 1997, 36 (19), 5827-36. 31 50 0.050 AUC Decay Rate 0.035 SC 35 30 0.030 0.025 20 1 2 3 4 5 6 M 0 A N 25 7 -1 0.040 RI 40 Decay Rates (s ) PT 0.045 U Area Under the Curve (AUC) 45 0.020 8 9 10 11 DOEPC Mole Percent (%) A CC EP TE D Supplemental Figure 1. Area-Under-the-Curve and decay rates for DOPC liposomes as function of DOEPC concentration. Values for the areaunder-the-curve (left) and decay rates for the normalized fluorescence signal of Bodipy 581/591 plotted as a function of DOEPC mole percent presence in DOPC liposomes. The error bars shown are the calculated standard deviations. 32 105 0.016 AUC Decay rate 0.014 -1 Decay Rates (s ) PT 95 90 RI 0.012 85 0.010 SC 80 75 0.008 U Area Under the Curve (AUC) 100 65 0 1 2 3 4 A N 70 5 6 7 0.006 8 9 10 11 M DMEPC Mole Percent (%) A CC EP TE D Supplemental Figure 2. Area-Under-the-Curve and decay rates for DMEPC liposomes as function of DMPC concentration. Values for the area-under-the-curve (left) and decay rates for the normalized fluorescence signal of Bodipy 581/591 plotted as a function of positively charged DMEPC mole percent presence in DMPC liposomes. The error bars shown are the calculated standard deviations. 33 0.013 120 0.012 AUC Decay Rate 110 RI AUC 0.009 0.008 SC 90 0.007 U 80 0.006 N 0 1 2 3 4 A 70 5 6 7 -1 0.010 100 Decay Rates (s ) PT 0.011 0.005 8 9 10 11 M DPEPC Mole Percent (%) A CC EP TE D Supplemental Figure 3. Area-Under-the-Curve and decay rates for DPPG liposomes as function of DPPC concentration. Values for the areaunder-the-curve (left) and decay rates for the normalized fluorescence signal of Bodipy 581/591 plotted as a function of positively charged DPPG mole percent presence in DPPC liposomes. The error bars shown are the calculated standard deviations. 34 1.2 30-nm 50-nm 100-nm 200-nm PT 0.8 RI 0.6 0.4 SC Normalized Fluorescence Intensity F/Fo 1.0 0.2 20 40 60 80 100 120 140 160 180 200 A 0 N U 0.0 M Time (min) A CC EP TE D Supplemental Figure 4. AAPH oxidation assay as a function of DOPC liposome extrusion pore size. Time-course kinetics for the oxidation of Bodipy 581/591 inside of DOPC liposomes extruded at varying sizes is shown; oxidation was caused by the thermal degradation of AAPH at 37°C. Data displayed is the average of 3 to 6 experiments. 35