Aroma compounds-proteins interaction using headspace techniques.

Interactions between volatile compounds and proteins in aqueous solution, were studied using static and dynamic headspace techniques. The decrease of aroma compounds concentration in the headspace determined in static mode indicates a retention by beta-lactoglobulin, 2.4% (w/w), pH 3.4, 25 degrees C, varying from 8 to 60% for most of the compounds tested. However 2-hexanone and isoamyl acetate are not significantly retained by the protein and a salting-out effect is noticed for limonene. On the basis of these preliminary results the intensity of interactions of methyl ketones (C7 to C9), ethyl esters (C6 to C9), limonene and mgamma rcene and beta-lactoglobulin for different pH values (2.0 to 11.0) were estimated by the determination of the infinite dilution activity coefficients (gamma i) by exponential dilution. For a constant pH value, the relative activity coefficient in the presence of beta-lactoglobulin (gamma r) of methyl ketones decreases significantly with the hydrophobicity of the volatiles whereas the relative activity coefficient value reaches a minimum for ethyl octanoate in the ester series. For limonene and myrcene an increase of gamma r is generally noticed whereas a decrease occurs in the presence of sodium caseinate and bovine serum albumin. The variations of relative activity coefficient according to the pH of the medium can be related to structural modifications of the beta-lactoglobulin.

Determination of apparent binding constants for aroma compounds with beta-lactoglobulin by dynamic coupled column liquid chromatography.

Apparent binding constants of aroma compounds limonene, alpha- and beta-ionone, and terpenyl acetate, with beta-lactoglobulin (BLG), were determined, using dynamic coupled column liquid chromatography, for pH values varying from 3 to 11. K(a) values varied from 2.61 to 3.21 x 10(3) M(-1) for limonene, indicating a strong interaction with BLG. Similarly, significant and close apparent binding constants were obtained for alpha- and beta-ionone, 1.7 x 10(2) and 4.5 to 5.4 x 10(2) M(-1), respectively. These data indicated that a similar mechanism is involved for the binding of these two molecules. The weaker values obtained at low pH, for alpha-ionone relative to beta-ionone, can be explained by the existence of steric hindrance. An increase of the apparent binding constant was observed, for all the compounds studied, when the pH was increased from 3 to 9. At this pH, an apparent binding constant was obtained for terpenyl acetate (1.04 x 10(2) M(-1)), whereas this determination was not possible at pH 3 and 6. The apparent binding constant increase was in agreement with the decrease of aroma compound relative activity coefficient in the presence of BLG, previously observed at this pH. It indicated a best accessibility to the same binding site. The binding constants of all the aroma compounds studied decreased at pH 11 as a result of the important release of the BLG structure previously reported.

Effect of pH on retention of aroma compounds by beta-lactoglobulin.

Interactions between volatile compounds and BLG in aqueous solution were studied using static and dynamic headspace techniques (exponential dilution). The intensity of interactions between methyl ketones (C7-C9), ethyl esters (C6-C9), limonene, myrcene, and beta-lactoglobulin (BLG) were estimated by determination of the relative infinite dilution activity coefficients (gamma(r)). For a constant pH value, the methyl ketones retention by BLG increased significantly with the hydrophobicity of the volatiles, whereas the retention reached a maximum for ethyl octanoate in the ester series, indicating a possible steric hindrance. For limonene and myrcene an unexpected increase in headspace concentration or "salting out" effect was noticed for acid pH. The variations of the retention according to the pH increase of the medium from pH 3 to pH 11 could be related to structural modifications of the BLG. The retention increase observed between pH 3 and pH 9 resulted from the flexibility modification of the protein, allowing better accessibility to the primary or the secondary hydrophobic sites, whereas the dramatic decrease observed at pH 11 was the consequence of the alkaline denaturation of BLG. Electrostatic interactions occurring at pH 7.5 could also explain the observed retention increase.