**2.6. Statistical analysis**

Extrasynthèse SA (Genay, France). Soybean lecithin Beakin LV3 was a gift from Archer Daniels Midland Co (Decatur, IL, USA). Ethanol, thimerosal, phenylmethanesulfonyl fluoride, and monobasic potassium phosphate were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dimethyl sulfoxide (DMSO) was purchased from Fisher Scientific (Pittsburg, PA, USA). Deionized water was prepared by passing distilled water over a mixed bed of cation-

Whole caseins were prepared by isoelectric precipitation of the individual skimmed milk samples obtained from a Jersey cow and French-Alpine goats. The precipitate was dissolved by the addition of NaOH to yield a solution of pH 7.0. The casein was re-precipitated, washed, and then re-suspended. The sodium caseinate was subsequently cooled to 4°C and centrifuged at 100,000 × g for 30 min to remove residual fat using a Beckman Optima XL-A (Beckman Instruments Inc., Palo Alto, CA, USA) analytical ultracentrifuge. Finally, the suspension was dialyzed against cold deionized water at 4°C for 72 h with three changes and then lyophilized. The integrity of the samples was confirmed by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and the percentages of the various component caseins estimated by densitometry as previously described [22]. The casein composition of the samples is as follows: αs2-casein (bovine casein 12.1%; caprine αs1-I-casein 9.2%; caprine αs1-II-casein 5.3%), αs1-casein (type I): (bovine casein 0%; caprine αs1-I-casein 4.0%; caprine αs1-II-casein 0%); αs1 casein (type II): (bovine casein 39.5%; caprine αs1-I-casein 21.1%; caprine αs1-II-casein 25.6%); β-casein: (bovine casein 37.2%; caprine αs1-I-casein 51.6%; caprine αs1-II-casein 60.6%); and κ-casein: (bovine casein 11.2%; caprine αs1-I-casein 13.8%; caprine αs1-II-casein 9.6%) [20].

A corn oil-in-water emulsion was prepared as follows: an organic phase was prepared by diluting 2.5% (wt/wt) of the commercial lutein in corn oil. Soybean lecithin 0.5% (wt/wt) was dissolved in the corn oil. An aqueous phase was prepared by dispersing 1.0% (wt/ wt) of lyophilized bovine casein or caprine casein into the aqueous buffer solution (5 mM phosphate, pH 7.0). A coarse emulsion of oil-in-water was prepared by mixing the organic phase (10%, wt/wt) and the aqueous phase (90%, wt/wt) using a hand-held homogenizer (Biospec Products, Inc., Bartlesville, OK, USA) at low speed. The coarse emulsion was then homogenized five times at 82.74 MPa (12,000 psi) through a high-pressure TC5 homogenizer (Stansted Fluid Power, Harlow, UK). The fine emulsion produced was then diluted (1:1, v/v) with buffer solution containing an antimicrobial agent [5 mM phosphate buffer at pH 7.0 and 1 mM (wt/v) thimerosal]. The final diluted emulsions that were used for the stability studies contained 5% (wt/wt) oil phase and 250 mg per liter of lutein. Resulting emulsions were

The particle size of the oil droplets in the lutein-enriched emulsions was measured at day 0 and day 7 after homogenization with a SALD-2101 laser diffraction particle analyzer (Shimadzu Corporation, Columbia, MD, USA). The charge of the oil droplets in the lutein-enriched

anion exchanger and was used throughout this study.

**2.2. Source of caseins**

164 Progress in Carotenoid Research

**2.3. Emulsion preparation**

stored in the dark at 5 and 15°C for 7 days.

**2.4. Physical characterization of the emulsion**

In each experiment, the results of triplicate analyses were used to test experimental variables. The data were analyzed by ANOVA using PRO GLM procedure of SAS (version 8.2, SAS Institute, Cary, NC, USA). The least significant test was used to determine significant differences among means at p < 0.05.
