Preparation of Non-Surface-Active Langmuir Trough

8 hours ago - Read OnlinePDF (1 MB). Supporting Info ... The Supporting Information is available free of charge on the ACS Publications website at DOI...
2 downloads 0 Views 1MB Size
Download PDF
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega XXXX, XXX, XXX−XXX

http://pubs.acs.org/journal/acsodf

Preparation of Non-Surface-Active Langmuir Trough Subphases from Milk Luis Real Hernandez and Rafael Jimenez-Flores* Department of Food Science and Technology, The Ohio State University, Columbus, Ohio 43210, United States

Downloaded via 91.204.15.172 on September 6, 2019 at 15:02:48 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Milk polar lipid interfacial behavior continues to be analyzed using Langmuir trough experiments, but the reported Langmuir trough subphases commonly used are not fully representative of milk. A method to transform liquids of biological origin, such as milk, into appropriate Langmuir trough subphases does not currently exist, which hinders the applicability of Langmuir trough experiments to nature. Here, a procedure to manufacture milkderived Langmuir trough subphases with insignificant amounts of surfactants from bovine milk is presented. Ultrafiltration is used to remove the bulk of protein surfactants from milk followed by the creation of solvent-induced emulsions that remove trace proteins and lipids from collected skim milk permeates. Change in surface tension upon compression of resulting washed permeates from the surfactant removal process was ≤0.1 mN/m, terming the permeates nonsurface-active (NSA). NSA permeates (72.4 ± 0.2 mN/m) had a surface tension similar to that of ultrapure water (72.6 ± 0.1 mN/m), but their pH, conductivity, percent total solids, Brix percentage, alkalinity, and hardness were not the same, with NSA permeates being more compositionally similar to skim milk than water. The lift-off points of milk ganglioside GM3 monolayer surface pressure−area isotherms spread on NSA permeates and ultrapure water subphases were significantly different when compared for the same sample, indicating that surface tension measurements obtained on ultrapure water are not the same as with NSA milk permeate. Overall, surfactants were removed from bovine milk without the addition of exogenous compounds, allowing for the production of a NSA solution derived from milk that can be used in the Langmuir trough experiment to more realistically resemble the natural environment of milk polar lipids. The procedure described here was able to produce NSA solutions for other dairy beverages aside from milk, indicating that it can be applicable to other biological fluids.

■

INTRODUCTION Milk polar lipids mainly reside in the milk fat globule membrane (MFGM).1,2 Because of the extensive variety of lipid species found in the MFGM, analyzing the interfacial behavior of milk polar lipids found in milk directly has remained a challenge. In the past decade, the interfacial behavior of MFGM polar lipids has been studied in model systems using Langmuir troughs.3−6 Milk innately contains both protein and lipid molecules that behave as surfactants at the air−water interface, and these molecules can interfere with the collection of accurate data during Langmuir trough experiments, making pure milk an unsuitable subphase for Langmuir trough experiments. Reported Langmuir trough experiments on milk polar lipids over the past decade have used subphases that are not fully representative of milk. Subphases that have been used in milk polar lipid Langmuir trough experiments are ultrapure water, 1,4-piperazinediethane sulfonic acid (PIPES) buffer solutions that may include 50 mM NaCl and 2−5 mM CaCl2, and a saline citrate−phosphate buffer solution with 0.16 M NaCl.4,5,7,8 Ultrapure water (18.0 MΩ·cm, pH = ∼5.7 when exposed to air) can be readily obtained from a water ultrapurification system, but it is not representative of milk because it is more acidic than milk and lacks many of the ions present in milk.9 PIPES buffer solutions can have a pH (6.7) © XXXX American Chemical Society

and ionic strength similar to that of milk, but they lack the presence of lactose, magnesium, potassium, phosphate, and citrate found in milk.6,8,10 Furthermore, the NaCl concentration of 50 mM used in PIPES buffer solutions is above the average concentration of sodium and chloride in both bovine and human milk.10 Saline citrate−phosphate buffer solutions with 150 mM NaCl can have their pH adjusted so that it is comparable to that of milk, but these buffers still lack many of the ions found in milk.7,11 There is currently no reported systematic method of removing surfactants from biological liquids so that they can be used as Langmuir trough subphases. Use of biologically relevant subphases in Langmuir trough experiments is known to affect the results obtained.12 In this work, a procedure to produce virtually surfactant-free solutions from milk is presented as a method to feasibly manufacture realistic subphases from milk for Langmuir trough experiments. A method describing the transformation of milk into a suitable Langmuir trough subphase is not only necessary for improving the accuracy of Langmuir trough experiments on MFGM lipids but it can also become an important methodology for Received: June 6, 2019 Accepted: August 6, 2019

A

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

probably a result of a slight dilution of the permeate during the solvent washing steps. Nevertheless, both the crude and solvent-washed permeates had pH values in the pH range commonly accepted for bovine milk (6.6−6.8).21 The electrical conductivity of a solution is related to the ion concentration of the solution. The electrical conductivity of degassed ultrapure water is 1.9 ± 0.1 μS/cm, while the electrical conductivity of bovine milk is reported to be 4.0−5.5 μS/cm.22−24 Electrical conductivity values of analyzed bovine skim milk and associated permeates were within reported literature values. Crude skim milk permeates obtained during ultrafiltration had a significantly lower percentage of total solids and Brix percentage compared to the skim milks used to produce them (Table 1). This was expected because of the removal of proteins during ultrafiltration. No significant difference in the percentage of total solids nor Brix percentage between the crude and solvent washed permeates was detected. The insignificant decrease in percent total solids between solventwashed permeate and CP was due to the removal of trace proteins and lipids adsorbed to the permeate−solvent interface of emulsions separated from the permeate during the solvent washing steps.25 The main solute in solvent-washed permeate is expected to be lactose, which is supported by the fact that the percentage of lactose in milk is between 3.6 and 5.5% and both the percent total solids and Brix percentage lie within this range.21 Ultrafiltration through a 10 kDa membrane should not significantly impact a solution’s concentration of lactose.26 The alkalinity and hardness of bovine skim milk and associated permeates were detected to be over 300 ppm, while both the alkalinity and hardness of ultrapure water were found to be 0 ppm. High alkalinity is indicative of high buffering capacity, while a high hardness value is indicative of a high concentration of divalent ions.27,28 Given that the alkalinity and hardness values of the skim milk permeates did not decrease to zero indicates that they are not the same as ultrapure water, and that ions and small buffering molecules were able to pass through the ultrafiltration membrane. For all parameters tested, the compositional parameters measured for solvent-washed milk permeate were comparable to those of bovine skim milk. It is evident that the solvent-washed milk permeate is not pure water. A more thorough analysis on the concentration of mineral ions, such as calcium, of crude bovine skim milk permeate obtained using an ultrafiltration scheme similar to the one described here has been described before.26 The composition of milk serum can be modeled by simulated milk ultrafiltrate (SMUF), which is predominantly a saturated calcium phosphate solution meant to resemble the calcium concentration in milk for the study of milk proteins.29 While SMUF created from high purity ingredients would theoretically be surfactant-free, it lacks nonprotein nitrogenous compounds present in milk, which have been detected in milk ultrafiltrate by those trying to improve the original SMUF formula.26,30 Because SMUF lacks some chemical species found in milk ultrafiltrate, the solvent-washed skim milk permeate described here would be a more realistic solution of milk serum. For example, monoethanolamide, which has been detected in milk ultrafiltrate, is known to reside at the air− water interface, potentially influencing measured results obtained from Langmuir trough experiments.31 SMUF does not contain monoethanolamide. Furthermore, SMUF as originally described by Jenness and Koops is unstable, and

producing protein-free solutions from milk that can be used to selectively study milk proteins. Formation of suitable subphases from liquids of biological origin will not only improve experiments performed directly on Langmuir troughs, but it can also improve the relevancy of techniques employing a Langmuir trough, such as the creation of Langmuir−Blodgett films.13 It was hypothesized that appropriate subphases could be produced from milk without significantly changing the milk’s composition of nonsurfactant molecules because lipids and proteins can be removed from solutions without the addition of foreign substances. The majority of milk lipids can be removed by passing the milk through a cream separator, and this process is ideal because it is mechanical. Removal of proteins from milk, on the other hand, has traditionally been performed through acid-induced precipitation which can change the concentration of ions in the milk or even contaminate the milk with exogenous chemical species. The procedure described here has two major steps: ultrafiltration to remove most of the proteins in skim milk and then washing of the collected permeate with hexane and chloroform to remove residual polypeptides. Lift-off point analysis of milk ganglioside GM3 surface pressure−area isotherms measured on either washed milk permeate or ultrapure water subphases is presented as an example of the differences that can be observed in milk polar lipid Langmuir trough experiments when a more realistic subphase is used. The objective of this work was to show that there are measurable differences in the results obtained in milk polar lipid Langmuir trough experiments depending on the subphase selected, and not to describe the behavior of specific milk polar lipids. Efficient removal of surfactants from three dairy ingredient-containing beverages aside from skim milk was also achieved, indicating that the procedure described here can transform non-milk dairy liquids, as well as potentially other biological liquids, into suitable Langmuir trough subphases.

■

RESULTS AND DISCUSSION Compositional Parameters of Bovine Skim Milk and Associated Permeates. No significant difference in pH or electrical conductivity was observed between bovine skim milk and its crude permeate (CP) collected from ultrafiltration (Table 1). This was expected because the 10 kDa molecular weight cutoff membrane used for ultrafiltration would allow ions to pass through the membrane. A significant increase in pH was observed in the solvent-washed permeate compared to the pure skim milk used to produce the permeate, but this was Table 1. Compositional Properties of Bovine Skim Milk and Associated Permeatesa sample

pH

conductivity (μS/cm)

total solids (%)

Brix (%)

bovine skim milk crude bovine skim milk permeate solvent washed bovine skim milk permeate

6.71 ± 0.01a

5.30 ± 0.01a

8.8 ± 0.3a

10.3 ± 0.1a

6.72 ± 0.04a

5.3 ± 0.1a

5.2 ± 0.3b

5.6 ± 0.3b

6.82 ± 0.04b

5.0 ± 0.3a

4.8 ± 0.3b

5.0 ± 0.3b

a

Differing letters indicate significant differences within a parameter (α = 0.05). B

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

Figure 1. Surface activity over time graphs for bovine skim milk, a dairy CIB, a MIF, and a MMR along with their associated permeates. Measurements were started after the surface was fully compressed and the surface was allowed to equilibrate for 1 min. Mean and standard deviation values are given for bovine skim milk and associated permeates, while proof-of-concept measurements are only presented once for nonmilk beverages. For all graphs, values of liquids without modification are plotted with squares, values for the CP of the liquids are plotted as circles, and values for solvent washed permeates (NSAP) are plotted as triangles. For skim milk NSAP, all plotted points had a y-axis standard deviation ≤ 0.06 mN/m.

would need to be prepared by current methods described by Dumpler and collaborators to ensure stability.29,30,32 Surface Activity of Dairy Liquids and Associated Permeates. A surface-active solution is defined here as a solution whose surface tension decreases upon compression. The surface tension of a liquid will only decrease upon compression if surfactants are present at its surface. Bovine skim milk is a surface-active solution given its high protein content (Figure 1). Upon compression, adsorbed proteins at the air−aqueous interface can lower the surface tension more than in the uncompressed surface, causing a measurable difference in surface tension. Permeate from skim milk was also surface-active (Figure 1). This was expected as trace proteins and peptides in the CP could act as surfactants. The change in surface tension upon compression was of higher magnitude for crude skim milk permeates than for pure skim milk probably due to a greater presence of protein at the air−water interface. That being said, surface activity measurements are only used here to determine whether surfactants are present at the air−aqueous interface. Solutions whose surfaces have a decrease in surface tension upon compression without the addition of an external surfactant can be deduced to have endogenous surfactants, but whether the solution has low or high concentration of surfactants is not relevant when the goal is to have no surfactants. Ultrafiltration through a 10 kDa molecular weight cutoff membrane alone was not enough for removing all surfactants from skim milk. A previous study that ultrafiltered bovine skim milk through a 10 kDa molecular weight cutoff membrane found that acidification was necessary to remove proteins in the collected permeate.26 Solvent-washed skim milk permeates had a surface activity ≤ 0.1 mN/m, terming them non-surface-active (NSA) (Figure 1). The NSA nature of solvent-washed milk permeates allows them to be suitable Langmuir trough subphases. Ultrafiltration of milk through a membrane with a lower molecular weight cutoff than 10 kDa might remove more protein, but the resulting permeate might still need to be solvent washed to remove trace amphipathic peptides. It was hypothesized that surfactants can be removed from milk without the need to add substances to the milk. The procedure described here supports this hypothesis as surfactants were unable to be detected after

their removal by ultrafiltration and solvent-induced emulsification, both processes that do not acidify nor purposely add compounds to the milk. To show that the procedure described here is not only applicable to skim milk, NSA permeates from three other dairy-based beverages were produced. The pure and CPs of all three non-milk beverages were surface-active (Figure 1). While MFGM-enriched infant formula (MIF) and milk protein-based meal replacement beverage (MMR) contained protein, cream containing beverage (CIB) had a surface-active CP even though it was not labeled with a significant protein or fat content (Table 2), indicating that trace lipids and proteins can Table 2. Total Fat, Cholesterol, and Protein Content of Analyzed Dairy Liquids dairy liquid bovine skim milk CIB MIF MMR

total fat (g/100 mL)

cholesterol (mg/100 mL)a

protein (g/100 mL)

0

2

3

0.2 3.6 3

0 not given 3

0 1.35 4.6

a

Beverages without cholesterol content in their label were assumed to contain insignificant amounts.

play a role in a solution’s surface activity. All resulting solventwashed permeates from all three non-milk beverages had a surface activity ≤ 0.3 mN/m, essentially terming them NSA. The process described here can produce NSA permeates from multiple dairy liquids, not just milk. With the lack of a method in the literature to transform biological fluids into suitable Langmuir trough subphases, the method described here could be used for that purpose. Surface Tension of Dairy Liquids and Associated Permeates. To compliment surface activity measurements, surface tension measurements were made to observe the relative decrease in surfactants in permeates as surfactants were removed through ultrafiltration and solvent washing of the permeates. The more surfactants a solution has, the lower its surface tension will be compared to its pure solvent. Water is the solvent of milk, so the surface tensions of milk permeates were compared to the surface tension of ultrapure water. C

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

Figure 2. Surface tension over time graphs for bovine skim milk, a dairy CIB, a MIF, and a MMR along with their associated permeates. Mean and standard deviation values are given for bovine skim milk and associated permeate samples, while proof-of-concept measurements are only presented once for non-milk beverages. For all graphs, values for pure liquids are plotted as squares, values for CPs are plotted as circles, and values for NSA permeates are plotted as triangles. For skim milk NSAP, all plotted points had a y-axis standard deviation ≤ 0.3 mN/m. The surface tension of all NSA permeates was comparable to that of ultrapure water, hence the overlap in the plotted points.

Bovine skim milk had a significantly lower surface tension (42.9 ± 0.7 mN/m) than ultrapure water (72.6 ± 0.1 mN/m) given its high protein content (Figure 2). After ultrafiltration, crude bovine skim milk permeate had a higher surface tension (53 ± 2 mN/m) than pure bovine skim milk because of the removal protein surfactants (Figure 2). NSA skim milk permeate (72.4 ± 0.2 mN/m) had a surface tension similar to ultrapure water. While NSA milk permeate and ultrapure water have similar surface tensions, they are not the same substance. NSA permeate was shown to contain solutes which ultrapure water does not have.26 The similarity in surface tensions could be based on the opposing effects the ions and acids in NSA milk permeate would have on surface tension.33 Lactose is not expected to significantly influence surface tension measurements.34 The solvent of all three non-milk beverages was also water (Table S1). All three non-milk beverages had surface tensions below that of ultrapure water (Figure 2). All CPs collected from the beverages had higher surface tensions than the starting beverages, as seen for skim milk (Figure 2). Solvent washing the CPs caused their surface tension to be comparable to that of ultrapure water (Figure 2). The results found for milk and non-milk beverages followed a similar pattern, indicating that the surfactant removal process described here is not only applicable to milk. NSA permeates of all dairy liquids had a similar surface tension to that of ultrapure water, indicating that NSA permeate produced as described here can replace ultrapure water as a subphase in the production of surface pressure−area isotherms. Milk Ganglioside GM3 Surface Pressure−Area Isotherms. Controlling for the sample analyzed, the lift-off points of milk GM3 surface pressure−area isotherms measured on NSA skim milk permeate subphases were significantly higher than those measured on ultrapure water subphases (Figure 3). This exemplifies how the selection of a biological subphase can influence the data collected during a Langmuir trough experiment. GM3 monolayers at the air−water interface have been shown to expand under the presence of polystyrenebound lactose in the subphase.35 Given that lactose is expected to be the major solute present in NSA skim milk permeate, the expansion of the lift-off point observed for the milk GM3 surface pressure−area isotherms obtained on a NSA skim milk

Figure 3. Representative mean surface pressure−area isotherms for milk ganglioside GM3 monolayers from one sample spread on either ultrapure water or NSA bovine skim milk permeate. Standard deviation bars are shown for each plotted point.

subphase compared to ultrapure water might be because of the interaction of the GM3 molecules with lactose. The interaction between the headgroup of GM3 molecules and lactose might be mediated by calcium ions, but this has not been confirmed for pure GM3 monolayers, such as the ones studied here.36 The presence of ions in Langmuir trough subphases have been reported to influence the interfacial behavior of polar lipids that can be found in milk.37,38 Milk GM3 was analyzed for its reproducible results for the ranges analyzed (see rational in Experimental Section), and only the lift-off point was analyzed as it could be feasibly obtained without having to speculate about possible lipid oxidation of the spread GM3 monolayers. Furthermore, reproducibility of milk GM3 surface pressure−area isotherms at high surface pressures was unobtainable for the conditions analyzed, making it unfeasible to provide a definitive, complete surface pressure−area isotherm for milk GM3. NSA milk permeate is more reminiscent of milk, making the data collected on NSA milk permeate more applicable to milk than the data collected on ultrapure water. It should be noted that D

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

stir-bar revolving at 600 rpm for at least 10 min. The mixture was separated in a separatory funnel after the mixture was left to equilibrate for 2 min. The washed permeate (lower phase) was collected, while the permeate-in-hexane emulsion (upper phase) was discarded. To remove particles that can contaminate the Langmuir trough, a similar process as was performed with hexane was performed using chloroform. HPLC grade chloroform was added to the washed permeate (1:10 v/v) and the mixture was stirred at 600 rpm for at least 10 min. After allowing the permeate−chloroform mixture to equilibrate for 2 min in a separatory funnel, the permeate-inchloroform emulsion (lower phase) was discarded and the washed permeate (upper phase) was decanted into clean glass bottles. Twice washed permeates were vacuum filtered through a 1 mm layer of activated carbon that rested on top of a fiberglass sample pad cut to fit inside a porcelain Büchner funnel. Filtered Permeates were washed sequentially twice more with HPLC grade hexane as was done before. Permeate collected after the fourth solvent wash was again filtered through a 1 mm activated carbon layer supported by a fiberglass sample pad inside a Büchner funnel to remove trace peptides still left in the permeate. After the second round of carbon filtration, permeates were stored at 4 °C until further use. The permeates were not treated further after this point. Analysis of Bovine Skim Milk and Associated Permeates. An Expert Pro-ISM pH probe (30014096, Mettler Toledo) was used to measure the pH. Conductivity (mS/cm) was measured using an H1763100 digital conductivity electrode (Hanna Instruments). Percent total solids (w/w) was measured using a Smart 6 microwave and infrared moisture analyzer (CEM Corporation, Matthews, NC). Brix percentage was measured using a digital pocket refractometer (PAL-α model, ATAGO, Japan) tared using ultrapure water. Alkalinity and hardness were measured (ppm) simultaneously, colorimetrically, and semiquantitatively using EasyStrips (Tetra Holding Inc, Blacksburg, VA). All tests were performed in triplicate, and were used as they are feasible to check the integrity of a solution right before it is used as a subphase. Only the composition of bovine skim milk and associated permeates was analyzed. Surface Activity Measurements. Surface activity was measured on a Teflon-coated KSV NIMA Langmuir trough (Biolin Scientific, Finland) with a working area of 26 700 mm2. The Langmuir trough was housed in a homemade plexiglass box covered during the recording of any measurement. For each experiment, the trough was rinsed five times with ultrapure water after being cleaned with acetone and ethanol. After being warmed to room temperature (21 °C), ∼350 mL of a dairy liquid or permeate was added to the trough, and 2 ethanol-cleaned Delrin barriers were placed on the surface to establish the working area. A 39.24 mm wide Wilhelmy platinum plate was flamed and placed on a Wilhelmy balance so that a third of the plate from the bottom was submerged in the subphase. The balance was allowed to equilibrate for 2 min, after which, the surface tension measured by the Wilhelmy plate balance was tared. The barriers of the trough were compressed so that the working area decreased by 74.1% to the minimum working area the barriers could be compressed to without touching the Wilhelmy plate. The barriers were compressed to the maximum distance allowed, so that all surfactants at the surface could be accounted for. The surface was allowed to equilibrate for 1 min after compression,

the GM3 samples used were composed from multiple chemical species, creating vast variability between the samples analyzed.

■

EXPERIMENTAL SECTION Materials. Bovine skim milk, a dairy CIB, a MIF, and a MMR were all purchased locally from a major supermarket. Table 2 reports the fat, cholesterol, and protein content of all four dairy liquids analyzed (see Table S1 for ingredient information). Skim milk instead of whole milk was used because it is commercially available and has most lipids already removed. MIF was prepared following manufacturer directions using ultrapure water with a resistivity of 18.2 MΩ·cm and a pH of 5.7. Ultrapure water was obtained from a Thermo Scientific Barnstead Smart2Pure ultrapure water purification system (50129870) equipped with an ultrafiltration module (50133981). HPLC grade chloroform and hexane and ACS grade acetone were purchased from Fisher Scientific (Fair Lawn, NJ). Particle size −100 activated carbon was purchased from Sigma-Aldrich (St. Louis, MO). CEM fiberglass 10 × 10 cm sample pads were purchased from CEM Corporation (Matthews, NC). HPLC grade methanol was purchased from EDM Millipore Corporation (Billerica, MA). USP Spec grade ethanol was purchased from Decon Labs Inc (King of Prussia, PA). Bovine milk ganglioside GM3 (>99%) was purchased from Avanti Polar Lipids (Alabaster, AL), and stored without further purification at −20 °C. GM3 is an anionic ganglioside that is present in whole bovine milk at a concentration of 0.15 ppm (Figure 4).14 The ceramide backbone of bovine milk ganglioside GM3 is mainly composed of long-chain (C22:0, C23:0, C24:0) fatty acids.15

Figure 4. Chemical structure of ganglioside GM3. Fatty acid is variable.

Ultrafiltration. Bovine skim milk and dairy beverages were ultrafiltered through a 10 kDa molecular weight cut-off 0.57 m2 Biomax membrane (P3B010A05) encased in a stainless steel Pellicon 3 cassette ultrafiltration module (MilliporeSigma, USA) by recirculating mode. Sample was fed into the module by an easy-load MasterFlex pump (Millipore, USA) at a flow rate of 3.0 ± 0.5 L/min in accordance with recommended feed flow rate parameters for the membrane used.16 Feed inlet pressure (186 ± 6 kPa) and temperature (22 ± 2 °C) for all materials was kept constant throughout filtration, except for MMR, in which the feed inlet pressure was maintained at 207 ± 7 kPa to obtain the desired feed flow rate. All samples were concentrated by a volumetric factor of 2. Permeate was collected and stored at 4 °C until further use. Permeate collected at this stage was termed crude as it had no further purification. Solvent Removal of Permeate Proteins. HPLC grade hexane was added to collected permeate (1:10 v/v). The permeate−hexane mixture was stirred using a Teflon-coated E

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

Afterward, the barriers on trough were compressed at a rate of 5 mm/min/barrier until a surface pressure of 2 mN/m was reached. Surface pressure−area isotherms were recorded to a target surface pressure of 2 mN/m as this was enough to determine an extrapolated lift-off point of each isotherm. The lift-off point of each isotherm was calculated by extrapolating the mean isotherm data at the surface pressure points recorded at 0.5, 1.0, and 1.5 to 0 mN/m using OriginPro 2018 (OriginLab Corporation). Using extrapolation to determine the lift-off point of an isotherm does not necessary allow for the calculation of the empirical lift-off point, but using extrapolation has the advantage of being systematic, so that multiple isotherms can be compared in the same manner, and for the lift-off point calculated to not be heavily influenced by noise because Wilhelmy plate balance measurements have an error of at least 0.1 mN/m.19 All isotherm data were collected at 21 ± 1 °C and a relative humidity of 55 ± 4%. Individual samples were measured at least once in each subphase, producing at least three isotherms for each subphase from three different samples. Milk GM3 was analyzed as it was found to have reproduceable surface pressure−area isotherm results for the surface pressure range analyzed for each individual sample tested. GM3 isotherms on an ultrapure water subphase for each individual sample had measured surface pressure and mean molecular area standard deviations of ±0.5 mN/m and ±1 Å2, respectively, which were similar to measured standard deviations of single lipid species isotherms. 20 Surface pressure−area isotherms on an ultrapure water subphase were attempted for other milk polar lipids extracted from milk aside from GM3, but the variability of obtained results for individual samples of those lipids was too high to consider the results reproduceable (data not shown). The variability in the results obtained was due to the heterogeneity of lipid species within the samples as the samples were extracted from milk and composed of multiple lipid species. Statistics. All statistics were computed on OriginPro 2018. Permeate composition measurements were analyzed using oneway ANOVA tests followed by Tukey post-hoc tests. Milk ganglioside GM3 monolayer lift-off point measurements were compared between subphases by a paired sample t-test. α was set at 0.05 for all statistical tests.

and then surface activity was recorded for 10 min. All measurements were taken at 21 ± 1 °C and at a relative humidity of 55 ± 6%. For bovine skim milk and associated permeates, measurements were performed in triplicate. Only one representative measurement for non-milk beverages and associated permeates is shown as these measurements were only used to show proof-of-concept. Surface Tension Measurements. Inside a plexiglass box, ∼50 mL of dairy liquid or permeate was added to a clean glass dish on a height-adjustable stage placed below a Wilhelmy plate balance. A 39.24 mm wide platinum plate was flamed and then placed on the Wilhelmy plate balance without touching the dish or its contents. The stage was slowly raised so that the liquid on the dish covered half of the Wilhelmy plate, and then lowered, so that the Wilhelmy plate was close to the surface of the liquid on the dish, but without touching it. After 2 min of letting the plate stabilize, the mass of the plate was tared. Very slowly, the stage was raised until the plate made first contact with the surface of the liquid on the dish. A cover was placed on the plexiglass box, and the entire system was left to equilibrate for 3 min. The surface tension was measured for 40 min as this is around the maximum amount of time that a Langmuir trough experiment analyzing unsaturated lipids can be performed for before the unsaturated lipids undergo significant oxidation.17,18 All surface tension measurements were performed at 21 ± 1 °C and a relative humidity of 51 ± 7%. For bovine skim milk and associated permeates, measurements were performed in triplicate. As for surface activity measurements, only one representative measurement for nonmilk beverages and associated permeates is shown as these measurements were only used to show proof-of-concept. Milk Ganglioside GM3 Monolayer Surface Pressure− Area Isotherms. Milk ganglioside GM3 samples were warmed up to room temperature (21 °C) for 1 h while sealed in nitrogen gas. Afterward, each GM3 sample, as received, was dissolved in a chloroform−methanol (2:1 v/v) mixture to reach a final GM3 concentration of 1.4 mM. The average molar mass (1268.653 g/mol) stated for each GM3 sample was used in calculating the final concentration of each solution. For each experiment, ∼350 mL of desired subphase, either ultrapure water or NSA bovine skim milk permeate, was added to a clean Teflon-coated KSV NIMA Langmuir Trough with a working area of 26 700 mm2. The Langmuir trough was housed in a homemade plexiglass box, which was covered during the recording of every measurement. Two Delrin barriers were placed on the surface of the subphase to form the working area. A 39.24 mm wide platinum plate was flamed and then placed on a Wilhelmy plate balance so that a third of the plate from the bottom was submerged in the subphase. The barriers were compressed to decrease the working area by 74.1% to make sure that the change in surface tension due to compression was ≤0.2 mN/m, indicating a clean subphase. The surface pressure of NSA skim milk permeate remained as ≤0.2 mN/m when the working area was compressed to its maximum point and then fully decompressed, making NSA skim milk permeate an ideal subphase for Langmuir trough experiments. Using a gas-tight syringe and with the barriers fully expanded, 24 μL of prepared GM3 solution was spread dropwise on the working area. A light-blocking curtain was placed on top of the plexiglass box to block the spread GM3 monolayer from direct light. The solvent used to spread the GM3 monolayer was allowed to evaporate for 10 min.

■

CONCLUSIONS Ultrafiltration and solvent-induced emulsions can be used to turn milk into an appropriate Langmuir trough subphase. NSA milk permeates can be used to more accurately resemble the environment milk polar lipids are exposed to in nature during Langmuir trough experiments. This is important in increasing the relevancy of these experiments to the behavior of MFGM lipids in milk and validating previous studies performed with simple subphases such as ultrapure water. Surfactants were removed from milk without the addition of foreign substances, supporting the hypothesis that surfactants in dairy liquids can be removed without having to add exogenous compounds that could change the composition of the liquids. Milk was not the only dairy liquid that could be transformed into a NSA solution, indicating that the procedure for producing Langmuir subphases described here could be applied to other biological fluids. Transformation of liquids of biological or natural origin into ideal Langmuir trough subphases is necessary for advancing the relevancy of techniques that employ a Langmuir trough to natural systems, such as foods. F

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

■

Article

Lipid-Protein Interactions in the Milk Fat Globule Membrane. Int. Dairy J. 2012, 22, 58−65. (10) Fox, P. F.; Uniacke-Lowe, T.; McSweeney, P. L. H.; O’Mahony, J. A. Salts of Milk. Dairy Chemistry and Biochemistry; Springer International Publishing: Cham, 2015; pp 241−270. (11) Gaucheron, F. The Minerals of Milk. Reprod., Nutr., Dev. 2005, 45, 473−483. (12) Selladurai, S. L.; Miclette Lamarche, R.; Schmidt, R.; Dewolf, C. E. Model Lung Surfactant Films: Why Composition Matters. Langmuir 2016, 32, 10767−10775. (13) Li, T.; Lilja, K.; Morris, R. J.; Brandani, G. B. Langmuir− Blodgett technique for anisotropic colloids: Young investigator perspective. J. Colloid Interface Sci. 2019, 540, 420−438. (14) Lee, H.; German, J. B.; Kjelden, R.; Lebrilla, C. B.; Barile, D. Quantitative Analysis of Gangliosides in Bovine Milk and ColostrumBased Dairy Products by Ultrahigh Performance Liquid Chromatography-Tandem Mass Spectrometry. J. Agric. Food Chem. 2013, 61, 9689−9696. (15) Bode, L.; Beermann, C.; Mank, M.; Kohn, G.; Boehm, G. Human and Bovine Milk Gangliosides Differ in Their Fatty Acid Composition. J. Nutr. 2004, 134, 3016−3020. (16) MilliporeSigma. Pellicon 3 Cassettes with Biomax Membrane Data Sheet, 2017. (17) Liljeblad, J. F. D.; Bulone, V.; Tyrode, E.; Rutland, M. W.; Johnson, C. M. Phospholipid Monolayers Probed by Vibrational Sum Frequency Spectroscopy: Instability of Unsaturated Phospholipids. Biophys. J. 2010, 98, L50−L52. (18) Telesford, D.-M.; Verreault, D.; Reick-Mitrisin, V.; Allen, H. C. Reduced Condensing and Ordering Effects by 7-Ketocholesterol and 5β,6β-Epoxycholesterol on DPPC Monolayers. Langmuir 2015, 31, 9859−9869. (19) Gaines, G. L. Insoluble Monolayers at Liquid-Gas Interfaces; Interscience Publishers: New York, 1966. (20) Adams, E. M.; Verreault, D.; Jayarathne, T.; Cochran, R. E.; Stone, E. A.; Allen, H. C. Surface organization of a DPPC monolayer on concentrated SrCl2 and ZnCl2 solutions. Phys. Chem. Chem. Phys. 2016, 18, 32345−32357. (21) Bylund, G.; Malmgren, B.; Holanowski, A.; Hellman, M.; Mattsson, G.; Svensson, B.; Pålsson, H.; Bronsveld, E.; Adamson, N. Dairy Processing Handbook; Tetra Pak Processing Systems, 2015. (22) Francis, M. J. Effect of Degassing on the Electrical Conductivity of Pure Water and Potassium Chloride Solutions. J. Phys. Chem. C 2008, 112, 14563−14569. (23) Sastry, S. K. Electrical Conductivity of Foods. Engineering Properties of Foods; CRC Press, 2005; pp 461−500. (24) Henningsson, M.; Ö stergren, K.; Dejmek, P. The Electrical Conductivity of Milk-The Effect of Dilution and Temperature. Int. J. Food Prop. 2005, 8, 15−22. (25) Kotsmar, C.; Aksenenko, E. V.; Fainerman, V. B.; Pradines, V.; Krägel, J.; Miller, R. Equilibrium and dynamics of adsorption of mixed β-casein/surfactant solutions at the water/hexane interface. Colloids Surf., A 2010, 354, 210−217. (26) Dumpler, J.; Kieferle, I.; Wohlschläger, H.; Kulozik, U. Milk Ultrafiltrate Analysis by Ion Chromatography and Calcium Activity for SMUF Preparation for Different Scientific Purposes and Prediction of Its Supersaturation. Int. Dairy J. 2017, 68, 60−69. (27) Thermo Scientific. Technical Bulletin 517Total Alkalinity Measurement in Natural Waters, 2008. (28) Fisher Scientific. APHA Method 2340 Hardness, 2000. (29) Jenness, R.; Koops, J. Preparation and Properties of a Salt Solution Which Simulates Milk Ultrafiltrate. Ned. Melk- Zuiveltijdschr. 1962, 16, 153−164. (30) Dumpler, J.; Kieferle, I.; Wohlschläger, H.; Kulozik, U. Preparation of Simulated Milk Ultrafiltrate. Int. Dairy J. 2017, 68, 60−69. (31) McWilliams, L. E.; Valley, N. A.; Wren, S. N.; Richmond, G. L. A Means to an Interface: Investigating Monoethanolamine Behavior at an Aqueous Surface. Phys. Chem. Chem. Phys. 2015, 17, 21458− 21469.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.9b01659.

■

List of ingredients for bovine skim milk and other dairy beverages analyzed (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: jimenez-fl[email protected]. Phone: (614) 292-1993. ORCID

Rafael Jimenez-Flores: 0000-0003-4905-5021 Funding

This research was supported by the Dairy Parker Chair in Dairy Foods at The Ohio State University, the OSU CFAESOARDC minority scholarship fund, and an OSU Graduate Enrichment Fellowship. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to thank Dr. Heather Allen and her research group at The Ohio State University for providing technical guidance on surface tensiometry measurements. The authors would also like to thank Dr. Derek Gragson for his comments on the experimental design of this study.

■

REFERENCES

(1) Singh, H.; Gallier, S. Nature’s complex emulsion: The fat globules of milk. Food Hydrocolloids 2017, 68, 81−89. (2) Lopez, C.; Ménard, O. Human Milk Fat Globules: Polar Lipid Composition and in Situ Structural Investigations Revealing the Heterogeneous Distribution of Proteins and the Lateral Segregation of Sphingomyelin in the Biological Membrane. Colloids Surf., B 2011, 83, 29−41. (3) Gallier, S.; Gragson, D.; Jiménez-Flores, R.; Everett, D. W. Surface Characterization of Bovine Milk Phospholipid Monolayers by Langmuir Isotherms and Microscopic Techniques. J. Agric. Food Chem. 2010, 58, 12275−12285. (4) Obeid, S.; Guyomarc’h, F.; David-Briand, E.; Gaucheron, F.; Riaublanc, A.; Lopez, C. The Phase and Charge of Milk Polar Lipid Membrane Bilayers Govern Their Selective Interactions with Proteins as Demonstrated with Casein Micelles. J. Colloid Interface Sci. 2019, 534, 279−290. (5) Priyanka, M.; Sabine, D.; Aman, P.; Koen, D.; Christophe, B. The Interfacial Properties of Various Milk Fat Globule Membrane Components Using Langmuir Isotherms. Food Biosci. 2017, 20, 96− 103. (6) Murthy, A. V. R.; Guyomarc’h, F.; Paboeuf, G.; Vié, V.; Lopez, C. Cholesterol Strongly Affects the Organization of Lipid Monolayers Studied as Models of the Milk Fat Globule Membrane: Condensing Effect and Change in the Lipid Domain Morphology. Biochim. Biophys. Acta, Biomembr. 2015, 1848, 2308−2316. (7) Danthine, S.; Blecker, C. Interactions of Lipases with Milk Fat Globule Membrane Monolayers Using a Langmuir Film Balance. Int. Dairy J. 2014, 35, 81−87. (8) Cheng, K.; Ropers, M.-H.; Lopez, C. The Miscibility of Milk Sphingomyelin and Cholesterol Is Affected by Temperature and Surface Pressure in Mixed Langmuir Monolayers. Food Chem. 2017, 224, 114−123. (9) Gallier, S.; Gragson, D.; Jiménez-Flores, R.; Everett, D. W. βCasein-Phospholipid Monolayers as Model Systems to Understand G

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX

ACS Omega

Article

(32) Gao, R.; van Halsema, F. E. D.; Temminghoff, E. J. M.; van Leeuwen, H. P.; van Valenberg, H. J. F.; Eisner, M. D.; Giesbers, M.; van Boekel, M. A. J. S. Modelling Ion Composition in Simulated Milk Ultrafiltrate (SMUF). I: Influence of Calcium Phosphate Precipitation. Food Chem. 2010, 122, 700−709. (33) dos Santos, A. P.; Levin, Y. Surface Tensions and Surface Potentials of Acid Solutions. J. Chem. Phys. 2010, 133, 154107. (34) Adhikari, B.; Howes, T.; Shrestha, A.; Bhandari, B. R. Effect of Surface Tension and Viscosity on the Surface Stickiness of Carbohydrate and Protein Solutions. J. Food Eng. 2007, 79, 1136− 1143. (35) Matsuura, K.; Kobayashi, K. Analysis of GM3-Gg3 Interaction Using Clustered Gycoconjugate Models Constructed from Glycolipid Monolayers and Artificial Glycoconjugate Polymers. Glycoconjugate J. 2004, 21, 139−148. (36) Seah, N.; Santacroce, P. V.; Basu, A. Probing the Lactose·GM3 Carbohydrate−Carbohydrate Interaction with Glycodendrimers. Org. Lett. 2009, 11, 559−562. (37) Baumler, S. M.; Blanchard, G. J. The Influence of Metal Ions on the Dynamics of Supported Phospholipid Langmuir Films. Langmuir 2017, 33, 2986−2992. (38) Adams, E. M.; Casper, C. B.; Allen, H. C. Effect of Cation Enrichment on Dipalmitoylphosphatidylcholine (DPPC) Monolayers at the Air-Water Interface. J. Colloid Interface Sci. 2016, 478, 353−364.

H

DOI: 10.1021/acsomega.9b01659 ACS Omega XXXX, XXX, XXX−XXX