Using Capillary Electrophoresis to Elucidate the Interaction Of Sphingomyelin, Indolicidin, and Peptide 45 Undergraduate Research Experience Award Program Final Report By Julie McNutt Supervisors: Heidi Huttunen-Hennelly & Kingsley Donkor Submission Date: Friday September 15th, 2017 Table of Contents Introduction ......................................................................................................... 3 Antimicrobial Peptides ................................................................................................................... 3 Sphingomyelin ................................................................................................................................... 4 Capillary Electrophoresis .............................................................................................................. 5 Affinity Capillary Electrophoresis (ACE)............................................................................................. 7 Frontal Analysis Capillary Electrophoresis (FACE) ...................................................................... 9 Methodology ..................................................................................................... 10 Reagents............................................................................................................................................. 10 BGE Preparation.............................................................................................................................. 11 Sample Preparation ....................................................................................................................... 11 Instrument Conditions .................................................................................................................. 12 Results & Discussion ........................................................................................ 14 Solubility of Sphingomyelin ........................................................................................................ 14 Preliminary ACE Results .............................................................................................................. 14 FACE Study Recommendations .................................................................................................. 14 Neutral Marker Comparison ....................................................................................................... 14 Effect of pH ........................................................................................................................................ 16 Analytical Performance and Troubleshooting ..................................................................... 17 Conclusion......................................................................................................... 18 Future Research ............................................................................................... 18 Acknowledgements .......................................................................................... 19 References ........................................................................................................ 19 Introduction In recent years, antibiotic resistance has become increasingly prevalent in society due to the over usage of antibiotics [1-3]. Some pathogens are even resistant to entire classes of drugs, which has fuelled the need to design new compounds with novel mechanisms of action that demonstrate antimicrobial properties [4]. One particularly promising drug class that has the potential to combat this resistance is antimicrobial peptides (AMPs) [2-3, 5]. Because they attack their targets using several, diverse mechanisms, there is less of a chance that pathogens will develop a resistance to them [5-6]. Antimicrobial Peptides Indolicidin is one type of AMP that has demonstrated effectiveness against gram positive and gram-negative bacteria, fungi and protozoa [1, 6-8]. This has been achieved with capillary electrophoresis to determine the binding between indolicidin and lipopolysaccharide, a major component of the plasma membrane of gram-positive and gram-negative bacteria [7, 9]. Indolicidin has 5 tryptophan residues that contribute to its hydrophobic nature [1, 3, 6, 10]. Recent studies have created derivatives of indolicidin in an attempt to reduce the hemolytic activity without sacrificing antimicrobial activity. This was achieved by replacing incremental amounts of tryptophan with alanine [1]. Two derivatives, peptide 45 and peptide 5, produced the most successful results [3]. a) I L P W K W P W W P W R R – NH2 b) I L P W K W P W A P A R R – NH2 c) I L P W K W P W W P A R R – NH2 Figure 1: Amino acid sequence of a) indolicidin, b) peptide 45, where 2 of the tryptophan residues (W) were replaced with alanine (A), and c) peptide 5 where 1 of the tryptophan residues were replaced with alanine [3]. Figure 2: Chemical structure of indolicidin [11]. a) b) Figure 3: Chemical structure of a) tryptophan and b) alanine [12, 13]. Sphingomyelin Sphingomyelin is a sphingolipid that comprises a major portion of the eukaryotic red blood cell (RBC) outer membrane along in addition to phosphatidylcholine [14, 15]. Structurally it is composed of a sphingosphine base, fatty acid tail and a polar head group [16, 17]. The fatty acid tail is typically 18 carbons in length and may be saturated or unsaturated with one double bond [16]. The polar group is usually phosphocholine which is ionized at the amine and phosphate group at a pH greater than 4 [16, 17]. The ionization of the phosphocholine is the reason for the neutral charge on the lipid. Figure 4: The phospholipid bilayer of a RBC and an individual lipid from the bilayer where the polar head group is sphingomyelin or phosphatidylcholine [18]. a) b) Figure 5: Chemical structure of a) sphingosine and b) phosphocholine [19, 20]. The purpose of studying the interaction of sphingomyelin and the two antimicrobial peptides is to determine the safety of these compounds as a therapeutic drug based on calculated binding constants. Furthermore, this study seeks to compare the binding of sphingomyelin and indolicidin versus the binding of sphingomyelin and peptide 45. Capillary Electrophoresis Capillary electrophoresis is a technique based on the principle of applying an electric field to allow a solute move through a background electrolyte (BGE). This ability of a solute is called the electrophoretic mobility. As the sample migrates through the capillary, the components in the sample separate at different times. An electropherogram is generated at the end of each run, which plots the absorbance as a function time. The position of a peak on the x-axis is the migration time of the solute and is the time it took the solute to reach the detector window from the beginning of the capillary. Figure 6: Schematic diagram of a capillary electrophoresis instrument [21]. Equation 1: 𝜇𝑒𝑝 = 𝑈𝑒𝑝 𝐸 μep = electrophoretic mobility Uep = electrophoretic velocity E = electric field Electroosmotic mobility is the movement of the BGE in response to the applied electric field. The silanol groups on the inner capillary wall become ionized in response to NaOH rinses. Cations from the BGE bind to the silanoate ions forming a double layer (fixed and mobile) of cations. In the mobile layer, the cations are solvated and this moves the BGE toward the negatively charged cathode when an electric field is applied. Equation 2: 𝜇𝑒𝑜 = 𝑈𝐸𝑒𝑜 μeo = electroosmotic flow Ueo = electroosmotic velocity E = electric field Since the direction of travel is anode to cathode, small cations are the first to elute. Anions are able to elute as well as a result of electroosmosis. The electroosmotic flow is determined by measuring the migration time of a neutral marker (ie. dimethyl sulphoxide (DMSO), dimethylformamide (DMF), mesityl oxide). The apparent mobility of an ion is the sum of electrophoretic mobility of the ion and the electroosmotic mobility of the BGE. Equation 3: 𝜇! = 𝜇!" + 𝜇!" μe = apparent mobility μep = electrophoretic mobility μeo = electroosmotic flow Several modes of capillary electrophoresis exist such as micellar electrokinetic capillary chromatography, capillary isoelectric focusing, capillary isotachophoresis, and capillary zone electrophoresis. This study utilizes various techniques of capillary zone electrophoresis (CZE) in which analyte separation occurs due to differing electrophoretic mobility. While other techniques have been used to study the mechanism of action of indolicidin such as circular dichroism spectroscopy, transmission electron microscopy, and nuclear magnetic resonance spectroscopy, CE is an efficient technique in that it can detect small quantities of sample and separate and resolve peaks quickly [1, 5, 22-24] Affinity Capillary Electrophoresis (ACE) ACE is one of the most commonly used CE techniques to study the interactions between two biomolecules [26, 27]. In ACE, the binding constant is calculated by comparison of the migration time of the neutral marker to the migration time of the complexed species [27]. In the ACE portion of this study, sphingomyelin is in the BGE in increasing concentrations while indolicidin and peptide 45 are the samples held at constant concentration. This is because of the cationic nature of indolicidin and peptide 45 that leads to protein adsorption on the capillary. The disadvantage is the mass of sphingomyelin is unknown introducing error into the binding constant calculation. By this model, indolicidin and peptide 45 are the receptors and sphingomyelin is the ligand. In addition, there are viscosity changes as a result of sphingomyelin in increasing concentrations in the BGE. Pre-incubation ACE (PI-ACE) is a technique similar to ACE used when binding between the two species is slow and consequently increases the analysis time [27]. Instead of sphingomyelin in the BGE, it is placed directly with the indolicidin and peptide 45 as samples. Here, indolicidin and peptide 45 are the ligands in increasing concentrations, which eliminates the error that results from lack of a known mass of sphingomyelin. The effect of sphingomyelin on BGE viscosity is also avoided. Equation 4: x-reciprocal plot [25] (µ e − µ o ) = −K •(µe − µo ) + K(µc − µo ) [ligand] K = −slope = −(−K ) Equation 5: y-reciprocal [25] [ligand] 1 1 = •[ligand]+ (µ e − µ o ) (µ c − µ o ) (µc − µo )K K = slope y − int Equation 6: double-reciprocal [25] 1 1 1 1 = • + (µe − µo ) (µc − µo )K [ligand] (µc − µo ) K = y − int slope where μe = apparent electrophoretic mobility μo = electrophoretic mobility of the free ligand μc = electrophoretic mobility of the complex [ligand] = concentration of indol or peptide 45 in PI-ACE K = binding constant To calculate the binding constants, there are different methods of linear regression models that can be used. With ACE and PI-ACE, there is an assumption that binding is 1:1. Plots that show linearity support this assumption. If the assumption is not supported, x-reciprocal and non-linear regression models are best at showing multiple binding sites. Frontal Analysis Capillary Electrophoresis (FACE) FACE is a useful method for the determination of binding constants because the assumption of 1:1 binding is no longer required. Instead of injecting the sample for 5 seconds, the sample is treated as a rinse step in which it is injected as a large plug for 120 seconds. Consequently, plateaus are formed instead of peaks. Figure 7: Peak (left) generated from PI-ACE or ACE and plateau (right) generated from FACE. Similar to PI-ACE, the ligand and receptors are pre-incubated together where the receptor concentration is held constant and the ligand concentration is varied. The requirement with FACE is that the mobility of the free ligands, indolicidin and peptide 45, must be sufficiently different from the mobility of the complex that they form with the receptor, sphingomyelin. Additionally, the mobility of the free receptor is assumed to be similar to the mobility of the complex. Equation 7: nonlinear regression [25] (µe − µo ) = (µc − µo ) K[ligand] 1+ K[ligand] K = slope where μe = apparent electrophoretic mobility μo = electrophoretic mobility of the free ligand μc = electrophoretic mobility of the complex [ligand] = concentration of indol or peptide 45 in FACE K= binding constant First a calibration curve is generated for the ligand, and then the amount of complexed molecules are plotted as a function of the free ligand. The fraction of the free ligand is determined by comparison to the calibration curves [7]. With FACE, non-linear regression must be used for the determination of the binding constant [7, 25]. Methodology Reagents Sphingomyelin, isolated from chicken egg yolk, was purchased from SigmaAldrich. Monobasic sodium phosphate monohydrate was purchased from Sigma-Aldrich while dibasic sodium phosphate was purchased from Caledon Laboratories. All water used to prepare solutions was 18 MΩ water. Remaining reagents used were of analytical grade and filtered using 0.45 μm Nylon syringe filters. Later in the experiment, 0.20 μm Cellulose Acetate syringe filters were used to filter sphingomyelin, indolicidin, and peptide 45 stock solutions to reduce sample loss. BGE Preparation Dibasic sodium phosphate and monobasic sodium phosphate were each dissolved in 18 MΩ water to concentrations of 100 mM. Two methods were used to prepare the BGEs. In one case, the BGEs were prepared by mixing specific volumes of dibasic and monobasic together to obtain a desired pH. In another case, monobasic or dibasic was used and the pH was adjusted with 0.10 M NaOH and 0.10 M of HCl, respectively. BGE pH was determined by using a Mettler Toledo pH meter. While a physiological pH of 7.40 ± 0.10 for the BGE was originally desired, this proved to be challenging in some experiments. Consequently, the pH of the BGE was lowered to 4.5 ± 0.10 for the method optimization process and it was increased as seen fit. After desired pH was obtained, the BGEs were sonicated and filtered using 0.45 μm Nylon syringe filters before. BGEs were remade as needed and were not stored for longer than 30 days. Sample Preparation The mass of sphingomyelin was obtained by weighing by difference on an analytical balance. Wax weigh paper was used instead of weigh boats to minimize loss from static cling. In a volumetric flask, sphingomyelin was dissolved in 50 mM phosphate BGE with 50 v/v% of 1-propanol to facilitate dissolution [16]. The stock solution was prepared to concentrations varying between 100 and 400 ppm. The stock solution was sonicated and filtered using 0.20 μm Cellulose Acetate syringe filters. It was stored in the fridge in a clear glass vial for approximately 30 years or until evidence of degradation. Indolicidin and peptide 45 were prepared in the same manner with the exception of the addition of 1-propanol. The stock solutions were prepared to a concentration of approximately 150 ppm. DMF was prepared in accordance with Darlington [16] in which it was dissolved first in 1-propanol to make a 5 % DMF solution. An aliquot of this was added to 50 mM phosphate BGE to make a 0.1% DMF solution. This solution was used to add to the samples as the neutral marker in a concentration of 0.01% DMF. DMSO was prepared directly in the BGE because of its polar nature. It was prepared such that the concentration of DMSO in the sample would be 0.01% DMSO. Polyethylene oxide (PEO) was prepared in 1 M HCl to a concentration of 0.2% PEO. HCl and NaOH were prepared as needed at concentrations of 0.1 M and 1.0 M. These reagents and the neutral markers were filtered with 0.45 μm Nylon syringe filters [16, 27]. Instrument Conditions Data collection was carried out using Beckman P/ACE system MDQ. The ultraviolet detector was set at 214 nm and direct absorbance was used for most runs, while indirect absorbance was during attempts of FACE calibration curves. The silica capillary, from Polymicro Technologies, had an outer diameter of 366.0 μm and an inner diameter of 50.4 μm. The length of the capillary was approximately 58 cm with an effective length of 10 cm less. To remove heat and prevent Joule heating, a fluorocarbon coolant was used to maintain a cartridge temperature of 25 °C. Run time typically was set at 20 minutes, with the exception of certain ACE experiments when a longer run time was needed. Separation of the samples was completed at voltage of 10-20 kV and normal polarity. New capillaries were conditioned with 1.0 M NaOH for 1 hour followed by 0.1 M NaOH for 30 minutes and finally with methanol for 5 minutes, each at 20 psi. Prior to the beginning of a sequence, the capillaries were rinsed with water, 0.1 M NaOH, water again, and the first BGE in the sequence each for 10 minutes at 20 psi. Initially, rinse steps were completed in between each run in the sequence according to table 1. Later when a polyethylene coating was deemed necessary, the conditions in table 2 were used and followed for the remainder of the study. Table 1: CE conditions without coating (only used until coating proved to be necessary) [27]. Event Duration Value Rinse - 0.1 M NaOH 4 min 20 psi Rinse - H2O 2 min 20 psi Rinse - BGE 4 min 20 psi Injection 5.0 sec 5 psi 30 min 10 kV Separate Voltage Ramp 0.17 min, normal polarity Table 2: CE conditions with PEO coating. The injection event would have been changed to a rinse step for 2 minutes for FACE studies [16]. Event Duration Value Rinse - MeOH 2 min 20 psi Rinse - H2O 1 min 20 psi Rinse - 1 M HCl 4 min 20 psi Rinse - 0.2% PEO 8 min 20 psi Rinse - BGE 5 min 20 psi Injection 5.0 sec 5 psi Separate Voltage Ramp 0.17 min, normal 30 - 60 polarity min 20 kV Results & Discussion Solubility of Sphingomyelin Working with sphingomyelin proved to be a difficult task. Due to the non-polar nature of the lipid, it was not readily soluble in the phosphate BGE. This required addition of 1-propanol at 50 % v/v to facilitate dissolution. Consequently, this made the system less biologically relevant. This was done in accordance with Darlington [16], however, they were successful at lower concentrations of 1propanol. When this was emulated, precipitation of sphingomyelin occurred after a few days of storage. Tests should be done to determine the time it takes for precipitation to occur when lower concentrations of 1-propanol are used. FACE Study Recommendations FACE was the first method attempted during this study with the first set of CE conditions. With little success, the conditions were changed and ACE was attempted instead. FACE should still be completed because of it does not require the 1:1 binding assumption once the ACE study is completed. The calibration curves and the binding should be performed on the same day as per the suggestion of Dufort-Lefrancois [27]. Preliminary ACE Results After FACE was attempted, it was decided that ACE may be an easier starting point since plateaus can be difficult to produce. First indolicidin, peptide 45, and sphingomyelin were run as samples at a pH of 4.5 as per the conditions of Table 2 to ensure peaks were visible. They were run at two different concentrations to ensure growth of the peaks. The migration times of indolicidin and peptide 45 were found to be 14.3 minutes and 14.2 minutes, respectively (figure 8). Sphingomyelin did not produce a peak because the method used [16] was for proteins and to visualize the binding between the proteins and sphingomyelin but not sphingomyelin itself. An ACE test was run with singlet samples of indolicidin and peptide 45. A change in migration time was observed but PI-ACE was desired because the binding was very slow. This was not achieved due to time constraints because lots of effort was spent on trouble shooting when certain results were unable to be replicated. UV - 214nm JM-070717-004.dat 0.015 UV - 214nm JM-070717-005.dat 0.010 0.010 0.005 0.005 0.000 0.000 -0.005 -0.005 -0.010 -0.010 -0.015 -0.015 -0.020 -0.020 -0.025 AU AU 0.015 -0.025 0 2 4 6 8 10 12 14 16 18 20 Minutes Figure 8: Electropherogram of indol at 50 ppm (black) and 90 ppm (blue). The 100mM phosphate BGE had a pH of 4.5. The small peak at 1 minute is DMF. 0.025 UV - 214nm JM-070717-006.dat 0.025 UV - 214nm JM-070717-007.dat 0.015 0.015 0.010 0.010 0.005 0.005 0.000 0.000 -0.005 -0.005 AU 0.020 AU 0.020 -0.010 -0.010 0 2 4 6 8 10 Minutes 12 14 16 18 20 Figure 9: Electropherogram of peptide 45 at 50 ppm (black) and 90 ppm (blue). The 100mM phosphate BGE had a pH of 4.5. The small peak at 1 minute is DMF. Neutral Marker Comparison A good neutral marker for ACE studies should remain uncharged, it should be detectable at the chosen wavelength, pure in form, it should be soluble in the BGE, not interact with the wall of the capillary. DMF and DMSO were both tested as neutral markers under the same conditions. DMF proved to be more reliable from the beginning, however, poor performance from DMSO was likely attributed to light degradation. Since DMF was promising and validated by other experiments, the study moved forward with this a neutral marker. To ensure the correct peak was located for DMF and determine its migration time, a sample was run with DMF in the BGE followed by a spiked sample of DMF in the BGE. The migration time was found to be approximately 1 minute. Effect of pH In order to mimic the conditions of a biological system, the pH of the BGE should be 7.40 ± 0.10. The method used in the beginning of the study met this requirement, however, this method was developed to study the interaction between indolicidin and lipopolysaccharide not sphingomyelin [27]. After little success, a new method was sought that had been used to study a protein interaction with sphingomyelin [16]. This method utilized the polyethylene oxide coating of the inner capillary wall. The caveat was that the pH of the BGE in this method was 4.5 ± 0.10. This proved to be more successful and was maintained throughout the remaining experiments. The method was tested and adjusted to ensure optimal conditions and then the pH was increased. The increases in pH were done with the DMF, indolicidin, and peptide 45. Mixed results were observed and this may have been due to the preparation of the BGEs. As described earlier, BGEs were prepared by two different methods. When attempting the pH tests, the BGEs that were prepared by addition of NaOH and HCl. To achieve the proper pH, significant amounts of acid or base had to be added which likely diluted the BGE from its concentration of 100 mM and changed the ionic strength. Indolicidin was visible on the electropherogram until a pH of 5.0 while peptide 45 was visible until a pH of 5.5. With increasing pH the intensity of the peaks decreased. In terms of DMF it appeared to elute at higher pHs but this varied between runs. The intensity of the DMF peaks started to change so spiking was attempted again to determine where the peak was located. Due to time constraints, the other method for BGE preparation was not used to attempt the pH tests. However, this should be done to determine if this is the problem as there are other studies that have successfully identified indolicidin at physiologic pH on an electropherogram. Once the peptides and DMF are consistently producing peaks at physiological pH then the pH testing should be repeated with sphingomyelin present. This way it will be clear that any issues at physiologic pH upon the addition of sphingomyelin, must be due to sphingomyelin and not the peptides or DMF. Analytical Performance and Troubleshooting CE is a great tool to elucidate the binding interactions between two biomolecule, however, there are some disadvantages of only study one ligand and one receptor as this is not representative of biological systems. In actuality, many other factors come into play such as the potential of the membranes, proteins in the membrane, and pH differences between the extracellular and intracellular matrices [27]. The inner diameter of the capillary was chosen to be 50 μm because this size minimizes Joule heating in comparison to larger capillaries. In addition, narrow bore capillaries were used which helps to dissipate heat while the coolant removes the heat generated. Part of the reason I was unable to get the results I had hoped was because of encountering different problems with the instrument. As I became more used to working with the instrument, I became more proficient at trouble shooting errors. One error that I encountered was an autozero fail in which the entire sequence would fail after the rinse steps and before the sample was injected. The solution to this was to clean off the detector probe with a damp kim wipe and to check the surface for deposits with a magnifying glass. Additionally, it was prudent to ensure that the cartridge was setup correctly and the capillary did not appear to be broken. Conclusion In this study, I was able to work on CE method optimization to study the interaction of sphingomyelin, indolicidin, and peptide 45. In the preliminary ACE study, it was clear that there was some degree of interaction, however, there was difficulties replicating this. While there are little results from the work I was able to complete, this research was still a valuable stepping-stone in determining the safety of antimicrobial peptides as an antibiotic. Future Research While I continued to build upon my technical skills throughout this project, I also learned to be patient, to maintain focus, and developed ways to trouble shoot. Despite being unable to accomplish certain research milestones, what I learned from the process was equally as valuable. In the future, I hope to obtain the data needed to calculate the binding constants between sphingomyelin and the two antimicrobial peptides: indolicidin and peptide 45. Students who continue with this project in the future should start by attempting to replicate the study with peptide 5. This should be relatively easy as no new methodology would be required. Following this, the binding between all three antimicrobial peptides and phosphatidylcholine should be determined. This may require adjustments to the method due to the different structure of phosphatidylcholine. Finally, the pH study should be continued and a temperature study should be employed to maintain the biological relevance. Acknowledgements I would like to thank my supervisors Dr. Heidi Huttunen Hennelly and Dr. Kingsley Donkor for their ongoing support, guidance, and knowledge for the duration of this project. Additionally I would like to thank Tallon Milne for providing training on the capillary electrophoresis and for being available for assistance. Finally, I would like to thank the Undergraduate Research Experience Award Program for providing me with the funds to further this research throughout the summer. References 1. Jones DSA, Huttunen-Hennelly HEK, et al. 2010. The design and synthesis of alanine-based indolicidin derivatives with identical physicochemical properties and their separation using capillary electrophoresis. Anal Bioanal Chem. 398(7-8):3073-79. 2. Matheson SMG, Cheeptham N, et al. 2013. Investigating the effects of hydrophobicity and charge on the therapeutic ability of the antimicrobial histatin 8 peptide for potential use in oral applications. Int J Biol. 5(2):85-91. 3. Podorieszach AP, Huttunen-Hennelly HEK. 2010. The effects of tryptophan and hydrophobicity on the structure of bioactivity of novel derivatives with promising pharmaceutical potential. Org Biomol Chem. 8(7):1679-87. 4. Andersson DI, Hughes D, et al. 2016. Mechanisms and consequences of bacterial resistance to antimicrobial peptides. Drug Resist Updat. 26(1):4357. 5. Azad MAA, Huttunen-Hennelly HEK, et al. 2011. Bioactivity and the first transmission electron microscopy immunogold studies of short de novodesigned antimicrobial peptides. Antimicrob Agents Ch. 55(5):2137-45. 6. Falla TJ, Nedra Karunaratne D, et al. 1996. Mode of action of antimicrobial peptide indolicidin. J Biol Chem. 271(32):19298-303. 7. Dufort-Lefrancois E. Determination of the strength and binding of indolicidin with lipopolysaccharide using capillary electrophoresis. 2014. TRU UREAP Proposal. 8. Subbalakshmi C, Krishnakumari V, et al. 1996. Requirements for antibacterial and hemolytic activities in the bovine neutrophil derived 13 residue peptide indolicidin. FEBS Lett. 395(1):48-52. 9. Olsen J. 2011. Frontal analysis capillary electrophoresis optimization for determining the binding constant between indolicidin and lipopolysaccharide. TRU research report. 10. Ladokhin AS, Selsted ME, et al. 1997. Bilayer interactions of indolicidin, a small antimicrobial peptide rich in tryptophan, proline, and basic amino acids. Biophys J. 72(2):794-805. 11. ChemicalBook [Internet]. c2017. [cited 2017 Sept 28]. Available from: http://www.chemicalbook.com/ChemicalProductProperty_EN_CB1719637.ht m 12. Wikipedia [Internet]. c2017. [cited 2017 Sept 28]. Available from: https://en.wikipedia.org/wiki/Tryptophan 13. Pill Scout [Internet]. c2014. [cited 2017 Sept 28]. Available from: http://www.pillscout.com/supplements/beta-alanine-powerful-pre-workoutamino/ 14. Himbert S, Alsop RJ, Rose M, Hertz L, Dhaliwal A, Moran-Mirabal JM, Verschoor CP, Bowdish DME, Kaestner L, Wagner C, et al. 2016. The molecular structure of human red blood cell membranes from highly oriented, solid supported multi-lamellar membranes. Sci Rep. 7: 1-14. 15. Mohandas N, Gallagher PG. 2008. Red cell membrane: pas, present, and future. Blood. 112(10): 3939-3948. 16. Darlington A. 2011. Application of capillary electrophoresis for lantibioticphospholipid interaction and lantibiotics separation using phospholipid coating. TRU Masters Thesis. 17. PubChem [Internet]. c2004. C18:1 Sphingomyelin. [cited 2017 Sept 28]. Available from: https://pubchem.ncbi.nlm.nih.gov/compound/6443882#section=Top 18. Brown E. [Internet]. Earl Brown. [cited Sept 28 2017]. Available from: http://bauernfaenger.info/brightpcolorsof-phospholipid-bilayer.htm 19. Wikipedia [Internet]. c2017. Sphingosine structure. [cited 2017 Sept 28]. Available from: https://commons.wikimedia.org/wiki/File:Sphingosine_structure.svg 20. Toronto Research Chemicals [Internet]. [cited 28 Sept 2017]. Available from: https://www.trc-canada.com/product-detail/?G598700 21. Wikipedia [Internet]. c2017. Capillary electrophoresis. [cited 29 Sept 2017]. Available from: https://en.wikipedia.org/wiki/Capillary_electrophoresis 22. Friedrich CL, Rozek A, Patrzykat, et al. 2001. Structure and mechanism of action of an indolicidin peptide derivative with improved activity against gram-positive bacteria. J Biol Chem. 276(26):24015-22. 23. Jiang C, Armstrong DW. 2010. Use of CE for the determination of binding constants. Electrophoresis. 31(1):17-27. 24. Ostergaard J, Heegaard NH. 2003. Capillary electrophoresis frontal analysis: principles and applications for the study of drug-plasma protein binding. Electrophoresis. 24(17):2903-13. 25. Donkor K. Application of Capillary electrophoresis for the study of interaction between alpha linolenic acid and epigallocatechin gallate. Powerpoint presentation, TRU. 26. Heegaarda NHH, Nilssonb S, Guzmanc NA. 1998. Affinity capillary electrophoresis: important application and some recent developments. 715(1): 29-54. 27. Dufort-Lefrancois E. Exploring capillary electrophoresis methods for the affinity interaction of indolicidin and lipopolysaccharide. 2015. TRU Masters Thesis.