Pressure-induced unfolding and aggregation of the proteins in whey protein concentrate solutions

Whey protein concentrate solutions (12% w/v, pH 6.65 +/- 0.05) were pressure treated at 800 MPa for 20-120 min and then examined using size exclusion chromatography (SEC), small deformation rheology, transmission electron microscopy, and various types of one-dimensional (1D) and two-dimensional (2D) polyacrylamide gel electrophoresis (PAGE). The pressure-treated samples showed a time-dependent loss of native whey proteins by SEC and 1D PAGE and a corresponding increase in non-native proteins and protein aggregates of different sizes. These aggregates altered the viscosity and opacity of the samples and were shown to be cross-linked by intermolecular disulfide bonds and by noncovalent interactions using 1D PAGE [alkaline (or native), sodium dodecyl sulfate (SDS), and SDS of reduced samples (SDS(R))] and 2D PAGE (native:SDS and SDS:SDS(R)). The sensitivity of the major whey proteins to pressure was in the order beta-lactoglobulin B (beta-LG B) > beta-LG A > bovine serum albumin (BSA) > alpha-lactalbumin (alpha-LA), and the large internal hydrophobic cavity of beta-LG may have been partially responsible for its sensitivity to high-pressure treatments. It seemed likely that, at 800 MPa, the formation of a beta-LG disulfide-bonded network preceded the formation of disulfide bonds between alpha-LA or BSA and beta-LG to form multiprotein aggregates, possibly because the disulfide bonds of alpha-LA and BSA are less exposed than those of beta-LG either during or after pressure treatment. It may be possible that intermolecular disulfide bond formation occurred at high pressure and that hydrophobic association became important after the high-pressure treatment.     

Gelling properties of heat-denatured beta-lactoglobulin aggregates in a high-salt buffer

Thermal denaturation, rheological, and microstructural properties of gels prepared from native beta-lactoglobulin (beta-LG) and preheated or heat-denatured beta-LG (HDLG) aggregates were compared. The HDLG was prepared by heating solutions of 4% beta-LG in deionized water, pH 7.0, at 80 degrees C for 30 min and then diluted to the desired concentration in 0.6 M NaCl and 0.05 M phosphate buffer at pH 6.0, 6.5, and 7.0. When reheated to 71 degrees C, HDLG formed a gel at a concentration of 2% protein. At pH 7.0, 3% HDLG gelled at 52.5 degrees C and had a storage modulus (G') of 2200 Pa after cooling. beta-LG (3%) in 0.6 M NaCl and 0.05 M phosphate buffer, pH 7.0, did not gel when heated to 71 degrees C. The gel point of 3% HDLG decreased by 10.5 degrees C and the G' did not change when the pH was decreased to 6.0. The HDLG gel microstructure was composed of strands and clumps of small globular aggregates in contrast to beta-LG gels, which contained a particulate network of compacted globules. The HDLG formed a gel at a lower concentration and lower temperature than beta-LG in the high-salt buffer, suggesting an application in meat systems or other food products prepared with salt and processed at temperatures of < or =71 degrees C.     

Effects of heat and high hydrostatic pressure treatments on disulfide bonding interchanges among the proteins in skim milk

Traditionally, milk has been heat treated to control microorganisms and to alter its functionality, for example, to increase its heat stability. Pressure treatment has been considered as a possible alternative for microorganism control, but some of the functionality-related milk protein interactions have not been explored. The present study used two novel two-dimensional polyacrylamide gel electrophoresis (2D PAGE) methods to explore the differences in the irreversible disulfide bond changes among the milk proteins after four common heat treatments and after 30-min pressure treatments of milk at 200, 400, 600, and 800 MPa at ambient temperature (22 degrees C). The pasteurizing heat treatment (72 degrees C for 15 s) denatured and aggregated only a few minor whey proteins, but the high heat treatments (100 degrees C for 120 s, 120 degrees C for 120 s, and 140 degrees C for 5 s) formed disulfide-bonded aggregates that included a high proportion of all of the whey proteins and kappa-casein (kappa-CN) and a proportion of the alpha(s2)-CN. Pressure treatment of milk at 200 MPa caused beta-lactoglobulin (beta-LG) to form disulfide-bonded dimers and incorporated beta-LG into aggregates, probably disulfide-bonded to kappa-CN. The other whey proteins appeared to be less affected at 200 MPa for 30 min. In contrast, pressure treatment at 800 MPa incorporated beta-LG and most of the minor whey proteins, as well as kappa-CN and much of the alpha(s2)-CN, into aggregates. The accessibility of alpha(s2)-CN and formation of complexes involving alpha(s2)-CN, kappa-CN, and whey proteins in the pressure treated milk is an important novel finding. However, only some of the alpha-lactalbumin was denatured or incorporated into the large aggregates. These and other results show that the differences between the stabilities of the proteins and the accessibilities of the disulfide bonds of the proteins at high temperature or pressure affect the formation pathways that give the differences among the resultant aggregates, the sizes of the aggregates, and the product functionalities.     

Hydrophobicity of whey protein concentrates measured by fluorescence quenching and its relation with surface functional properties

Surface hydrophobicity of whey protein concentrate (WPC) under heated (85 degrees C for 5, 10, 20, 30, 40, and 60 min) and unheated conditions was measured using cis-parinaric acid (CPA), 1-anilino-8-naphthalenesulfonate (ANS), and a fluorescence quenching method using acrylamide as a quencher. This last method evaluates the degree of exposure of tryptophanyl residues in proteins to the solvent. The initial slope of Stern-Volmer plots, K(app), was used as an index of protein hydrophobicity. Surface hydrophobicity of WPC exhibited good relation with surface functional properties such as emulsifying and foaming. Analysis of the data obtained in this work showed that the fluorescence quenching method gave results similar to those obtained using CPA and ANS. Therefore, this simple technique is satisfactory in effectively obtaining information about the hydrophobicity of whey proteins.     

Influence of binding of sodium dodecyl sulfate, all-trans-retinol, palmitate, and 8-anilino-1-naphthalenesulfonate on the heat-induced unfolding and aggregation of beta-lactoglobulin B

Heat treatment of bovine beta-lactoglobulin B (beta-LG) causes it to partially unfold and aggregate via hydrophobic association and intra- and interprotein disulfide bonds. The first stage, which involves a "loosening" of the native structure, is influenced by the environmental conditions, such as pressure, pH, and added solutes. In the present study, four potential beta-LG ligands [palmitate, sodium dodecyl sulfate (SDS), 8-anilino-1-naphthalenesulfonate (ANS), and all-trans-retinol (retinol)] were added to beta-LG solutions prior to heat treatment for 12 min at temperatures between 40 and 93 degrees C. The extent of the changes in secondary and tertiary structures, unfolding, and aggregation at 20 degrees C were determined by circular dichroism, fluorescence, and alkaline- and SDS-polyacrylamide gel electrophoresis (PAGE). Both palmitate and SDS stabilized the native structure of beta-LG against heat-induced structural flexibility, subsequent unfolding, and denaturation. Retinol was less effective, probably because of its lower affinity for the calyx-binding site, and ANS did not stabilize beta-LG, suggesting that ANS did not bind strongly in the calyx. It was also noted that holding a beta-LG solution with added SDS or ANS promoted the formation of a hydrophobically associated non-native dimer.     

Role of kappa-casein in the association of denatured whey proteins with casein micelles in heated reconstituted skim milk

Reconstituted skim milk at pH from 6.5 to 7.1 was unheated, preheated (68 degrees C/20 min), or heated at 90 degrees C for 20-30 min. On preheating, the size of the casein micelles decreased by about 5-20 nm, with a greater effect at higher pH. The casein micelle size of the heated milk at pH 6.5 increased by about 30 nm when compared to that of the unheated or preheated milk. As the pH was increased before heating, the particle size gradually decreased so that, at pH 7.1, the size was markedly smaller than that for the unheated milk and slightly smaller than that for the preheated milk. High levels (about 85%) of denatured whey protein associated with the casein micelles at pH 6.5, and this level decreased as the pH increased so that, at pH 7.1, low levels (about 15%) were associated with the micelles. Low levels of alphaS-casein and beta-casein were found in the serum regardless of the heat treatment or the pH of the milk. At pH 6.5, low levels (about 10%) of kappa-casein were also found in the milk serum. In the unheated milk, the level of serum kappa-casein increased slightly with increasing pH; in the heated samples, the level of serum kappa-casein increased markedly and linearly with increasing pH so that, at pH 7.1, about 70% of the kappa-casein was in the serum phase. The results of this study indicate that the pH dependence of the levels of serum phase kappa-casein may be responsible for the change in distribution of the whey proteins between the colloidal and serum phases. This is the first report to demonstrate significant levels of dissociation of kappa-casein from the micelles at pH between 6.5 and 6.7, although this dissociation phenomenon is well known on heating milk at high temperatures at pH above 6.7.     

Hydrophobic probe binding of beta-lactoglobulin in the native and molten globule state induced by high pressure as affected by pH,KIO(3) and N-ethylmaleimide

High hydrostatic pressure (HHP) at 500 MPa and 50 degrees C induces beta-LG into the molten globule state. Retinol, cis-parinaric acid (CPA), and 1-anilino-naphthalene-8-sulfonate (ANS) fluorescence from pH 2.5 to 10.5 in the presence of the native and molten globule states of beta-LG indicate that retinol binds to beta-LG in the calyx, CPA at the surface hydrophobic site, and ANS in multiple hydrophobic sites. HHP treatment results in a decrease of beta-LG affinity for retinol and CPA, suggesting conformational changes in the calyx and surface hydrophobic site of beta-LG during HHP treatment. beta-LG treated by HHP in the presence of N-ethylmaleimide (NEM) retains retinol affinity, suggesting that NEM protects the calyx conformation of beta-LG during HHP treatment. HHP treatment of beta-LG in the presence of KIO(3) exhibits a great decrease of CPA affinity compared to HHP-treated beta-LG in the absence of KIO(3), suggesting the formation of non-native disulfide bonding at the CPA binding site.     

木瓜蛋白酶在PEG/PEG-IDA-Fe3+/(NH4)2SO4亲和双水相中的分配系数模型

为预测木瓜蛋白酶在PEG/PEG-IDA-Fe3+/(NH4)2SO4亲和双水相中的分配行为,在298.15K条件下利用浊点法测定并比较了PEG/(NH4)2SO4和PEG/PEG-IDA-Fe3+/(NH4)2SO4的三角相图,建立了木瓜蛋白酶在无亲和配基双水相和含亲和配基双水相中分配模型。考察双水相各组分浓度与木瓜蛋白酶在该体系中分配系数的相关度,基于酶分配系数与双水相中上下相组分浓度差之间较高相关度,提出了木瓜蛋白酶在PEG/(NH4)2SO4双水相中分配系数与上下相组分浓度差的关联模型,模型相对偏差为7.02%。引入亲和配基浓度对酶分配系数的影响因子η,提出了木瓜蛋白酶在PEG/PEG-IDA-Fe3+/(NH4)2SO4亲和双水相中分配系数模型,试验验证,预测值和试验值之间相对误差均在15%以内,能实现亲和双水相中酶的准确预测。研究结果为木瓜蛋白酶在双水相中分配系数的工程计算提供参考。     

Ethanol effect on the structure of beta-lactoglobulin b and its ligand binding

The changes of structure and ligand binding properties of beta-LG B have been studied by fluorescence and circular dichroism spectroscopy in ethanolic solutions. Fluorescence measurements of retinol/beta-LG interactions at 480 nm in various ethanol concentrations show that the maximal fluorescence intensity induced by this interaction between retinol and beta-LG is observed around 20% v/v of ethanol. It is reduced to zero at 40% and 50% of ethanol. These results suggest that there are two distinct structural changes in beta-LG occurring between 20% and 30% and around 40% of ethanol. The first transition, which increases affinity and the apparent number of binding sites for retinol, may be related or similar to the Tanford transition. The strong quenching of retinol emission at 480 nm in 40% of ethanol indicates the radical transformation of beta-LG tertiary structure and the release of retinol. CD spectra at the aromatic region show that secondary and tertiary structures of beta-LG are not significantly affected between 0% and 20% of ethanol. In 30% of ethanol, beta-sheet percentage of beta-LG decreases with respect to native beta-LG (from 55% to 46%). beta-Sheet percentage in beta-LG increases in 40% and 50% alcohol (51% and 53%) relative to 30% of ethanol, which also indicates the strong rearrangement of the secondary structure of beta-LG, while its tertiary structure and beta-LG interactions are radically changed.     

Beta-lactoglobulin structure and retinol binding changes in presence of anionic and neutral detergents

Bovine beta-lactoglobulin (beta-LG) in vivo (in milks) has been found in complexes with lipids such as butyric and oleic acids. To elucidate the still unknown structure-function relationship in this protein, the structural changes of beta-lactoglobulin variant A (beta-LG A) in the presence of anionic surfactant such as sodium n-dodecyl sulfate (SDS) and in the presence of nonionic surfactant such as Triton X-100 have been investigated. Subsequently, the retinol binding by beta-LG has been investigated in the presence of various amounts of these surfactants as its binding indicator. The results of UV-vis and fluorescence studies show a higher denaturating effect of SDS at acid pH that can be due to greater positive charges of beta-LG at this pH indicating also the nonspecific hydrophobic interactions of Triton X-100 with beta-LG at all studied pHs. Isothermal titration calorimetry (ITC) measurements indicate the endothermic nature of beta-LG/SDS interactions and the exothermic nature of Triton X-100/beta-LG interactions. The analysis of the binding data demonstrates the absence of considerable changes in retinol binding properties of beta-LG in the presence of various amounts of these surfactants. This implies that surfactant binding does not change the conformation of beta-LG in the regions defining the retinol-binding site.     

Effects of high-pressure processing at low temperature on the molecular structure and surface properties of beta-lactoglobulin

High-pressure processing (HPP) was utilized to induce unfolding of beta-lactoglobulin (beta-LG). beta-Lactoglobulin solutions at concentrations of 0.5 mg/mL, in pH 7.5 phosphate buffer, were pressure treated at 510 MPa for 10 min at either 8 or 24 degrees C. The secondary structure, as determined by circular dichroism (CD), of beta-LG processed at 8 degrees C appeared to be unchanged, whereas beta-LG processed at 24 degrees C lost alpha-helix structure. Tertiary structures for beta-LG, as determined by near-UV CD, intrinsic protein fluorescence spectroscopy, hydrophobic fluorescent probe binding, and thiol group reactivity, were changed following processing at either temperature. The largest changes to tertiary structure were observed for the samples processed at 24 degrees C. Model solutions containing the pressure-treated beta-LG showed significant decreases in surface tension at liquid-air interfaces with values of 54.00 and 51.69 mN/m for the samples treated at 24 and 8 degrees C, respectively. In comparison, the surface tension for model solutions containing the untreated control was 60.60 mN/m. Changes in protein structure during frozen and freeze-dried storage were also monitored, and some renaturation was observed for both storage conditions. Significantly, the sample pressure-treated at 8 degrees C continued to display the lowest surface tension.     

Heat-induced redistribution of disulfide bonds in milk proteins. 1. Bovine beta-lactoglobulin

Changes in the structure and chemistry of beta-lactoglobulin (beta-LG) play an important role in the processing and functionality of milk products. In model beta-LG systems, there is evidence that the aggregates of heated beta-LG are held together by a mixture of intermolecular non-covalent association and heat-induced non-native disulfide bonds. Although a number of non-native disulfide bonds have been identified, little is known about the initial inter- and intramolecular disulfide bond rearrangements that occur as a result of heating. These interchange reactions were explored by examining the products of heat treatment to determine the novel disulfide bonds that form in the heated beta-LG aggregates. The native protein and heat-induced aggregates were hydrolyzed by trypsin, and the resulting peptides, before and after reduction with dithiothreitol, were separated by high-performance liquid chromatography and their identities confirmed by electrospray ionization mass spectrometry. Comparisons of these peptide patterns showed that some of the Cys160 was in the reduced form in heated beta-LG aggregates, indicating that the Cys160-Cys66 disulfide bond had been broken during heating. This finding suggests that disulfide bond interchange reactions between beta-LG non-native monomers, or polymers, and other proteins could occur largely via Cys160.     

On-line MS/MS monitoring of acrylamide generation in potato- and cereal-based systems

An on-line MS/MS technique was used to study the generation of acrylamide in rye-, wheat-, and potato-based systems during cooking. Acrylamide release to the gas phase was monitored continuously and was correlated with the acrylamide content of samples using a calibration based upon the partition of [1,2,3-(13C3)]acrylamide. On-line results at 180 degrees C were compared with data relating to the same systems obtained through GC-MS analysis. Agreement between the two techniques was notable, both in terms of the temporal profiles of acrylamide generation and when comparing the relative magnitudes of results for potato, wheat, and rye determined by each method. The effects of pH (citric acid) and added amino acids (soy protein hydrolysate) on the generation of acrylamide in hydrated potato flake were modeled at 180 degrees C. It was concluded that a combined treatment of low levels of each additive could result in significant reductions in acrylamide, although the effects of such treatments on sensory properties such as color and flavor remain to be evaluated.     

Effect of pH on the association of denatured whey proteins with casein micelles in heated reconstituted skim milk

Skim milk was adjusted to pH values between 6.5 and 6.7 and heated (80, 90, and 100 degrees C) for up to 60 min. Changes in casein micelle size, level of whey protein denaturation, and level of whey protein association with the micelles were monitored for each milk sample. Changes in casein micelle size were markedly affected by the pH at heating. At low pH (6.5-6.55), the casein micelle size increased markedly during the early stages of heating, and the size plateaued on prolonged heating. The maximum increase in size was approximately 30-35 nm. In contrast, at high pH (6.7), much smaller changes in size were observed on heating and the maximum increase in size was only approximately 10 nm. An intermediate behavior was observed at pH values between these two extremes. The rate of denaturation of the major whey proteins, alpha-lactalbumin and beta-lactoglobulin, was essentially unaffected by the pH at heating for the small pH changes involved in this study, and the changes in casein micelle size were poorly related to the level of whey protein denaturation. In contrast, the level of denatured whey proteins associating with the micelles was markedly dependent on the pH at heating, with high levels of association at pH 6.5-6.55 and low levels of association at pH 6.7. Changes in casein micelle size were related to the levels of denatured whey proteins that were associated with the casein micelles, although there was a small deviation from linearity at low levels of association (<15%). Further studies on reconstituted and fresh milk samples at smaller pH steps confirmed that the association of whey proteins with the casein micelles was markedly affected by the pH at heating. These results indicate that the changes in casein micelle size induced by the heat treatment of skim milk were a consequence of the whey proteins associating with the casein micelles and that the level of association was markedly influenced by small pH changes of the milk. It was not possible to determine whether the association itself influenced the casein micelle size or whether parallel reactions involving micellar aggregation caused the increase in micelle size as whey protein association progressed.     

Investigation of the lactosylation of whey proteins by liquid chromatography-mass spectrometry

Heat treatment of milk induces a reaction between the milk proteins and lactose, resulting in lactosylated protein species. The lactosylation of the two major whey proteins alpha-lactalbumin and beta-lactoglobulin was investigated by reversed phase liquid chromatography-mass spectrometry (LC-MS). Three sample series, consisting of aqueous model solutions of each whey protein separately and in mixture and whole milk, were heated for different time periods, and the progression of the lactosylation reaction was monitored. The observed degrees of lactosylation and the reaction kinetics showed that the lactosylation of beta-lactoglobulin was not influenced by the presence of other components, whereas the lactosylation of alpha-lactalbumin was enhanced in whole milk compared to the aqueous model systems. An in-depth evaluation of the LC-MS data yielded information regarding changes of physicochemical properties of the whey proteins upon lactosylation. Whereas retention time shifts indicated changes in hydrophobicity for both alpha-lactalbumin and beta-lactoglobulin, changes in the charge state distribution denoting conformational alterations were observed only for beta-lactoglobulin. The analysis of different liquid and solid milk products showed that the lactosylation patterns of the whey proteins can be used as indicators for the extent of heat treatment.     

Physical and chemical interactions in cold gelation of food proteins

pH-Induced cold gelation of whey proteins is a two-step process. After protein aggregates have been prepared by heat treatment, gelation is established at ambient temperature by gradually lowering the pH. To demonstrate the importance of electrostatic interactions between aggregates during this latter process, beta-lactoglobulin aggregates with a decreased iso-electric point were prepared via succinylation of primary amino groups. The kinetics of pH-induced gelation was affected significantly, with the pH gelation curves shifting to lower pH after succinylation. With increasing modification, the pH of gelation decreased to about 2.5. In contrast, unmodified aggregates gel around pH 5. Increasing the iso-electric point of beta-lactoglobulin via methylation of carboxylic acid groups resulted in gelation at more alkaline pH values. Comparable results were obtained with whey protein isolate. At low pH disulfide cross-links between modified aggregates were not formed after gelation and the gels displayed both syneresis and spontaneous gel fracture, in this way resembling the morphology of previously characterized thiol-blocked whey protein isolate gels (Alting, et al., J. Agric. Food Chem. 2000, 48, 5001-5007). Our results clearly demonstrate the importance of the net electric charge of the aggregates during pH-induced gelation. In addition, the absence of disulfide bond formation between aggregates during low-pH gelation was demonstrated with the modified aggregates.     

Interactions between bovine beta-lactoglobulin and peptides under different physicochemical conditions

The aim of this study was to determine if peptides could interact with beta-lactoglobulin (beta-LG) and what the physicochemical conditions promoting their interaction with the protein are. The binding of negatively charged (beta-LG 125-135 and 130-135), positively charged (beta-LG 69-83 and 146-149), and hydrophobic (alphaS1-CN 23-34 and beta-LG 102-105, both bioactive peptides) peptides to bovine beta-LG was determined using an ultrafiltration method under different physicochemical conditions: pH 3.0, 6.8, and 8.0; buffers of 0.05 and 0.1 M; 4, 25, and 40 degrees C; beta-LG/peptide ratios of 1:5 and 1:10. At pH 3.0, none of the peptides interacted with beta-LG at any temperature, buffer molarity, or beta-LG/peptide ratio probably due to electrostatic repulsions between the highly protonated species. At pH 6.8 and 8.0, charged peptides beta-LG 130-135, 69-83, and 146-149 bound to beta-LG under some physicochemical conditions, possibly by nonspecific binding. However, both hydrophobic peptides probably bind to the inner cavity (beta-barrel) of beta-LG, provoking the release of materials absorbing at 214 nm. Given the known biological activities of the hydrophobic peptides used in this study (opioid and ACE-inhibitory activities), their binding to beta-LG may be relevant to a better understanding of the physiological function of the protein.     

Transfer of aroma compounds in water-lipid systems: binding tendency of beta-lactoglobulin

Interactions of volatile aroma compounds with protein in aqueous solutions, especially whey proteins, have received significant attention in recent years. This work attempts to improve our understanding of the mass transfer in multiphasic systems, such as emulsions at the lipid-water interface, and to reveal the role of beta-lactoglobulin in the release rate of solutes. For this purpose the rotating diffusion cell has been used. From a practical point of view it enables evaluation of the transfer through the aqueous phase, through the oil and the interfacial transfer. The effect of beta-lactoglobulin, medium pH, and solute concentration has been investigated. Benzaldehyde and 2-nonanone have been studied, and miglyol has been chosen as an oil phase. It has been demonstrated that mass transfer has a rate-limiting step, which depends on physicochemical parameters such as hydrophobicity of the volatile, diffusion and partition coefficients, and rheological properties of the aqueous phase.     

Heat-induced aggregation of whey proteins: comparison of cheese WPC with acid WPC and relevance of mineral composition

Heat-induced aggregation of whey proteins in solutions made from two commercial whey protein concentrates (WPCs), one derived from mineral acid whey (acid WPC) and the other from cheese whey (cheese WPC), was studied using polyacrylamide gel electrophoresis (PAGE), size exclusion chromatography (SEC), and transmission electron microscopy (TEM). Heat treatment (75 degrees C) of acid WPC solutions (12.0%, w/w, pH 6.9) resulted in formation of relatively small "soluble" aggregates that were predominantly disulfide-linked. By contrast, heat treatment of the cheese WPC solutions (under the same conditions) caused formation of relatively large aggregates, containing high proportions of aggregates linked by noncovalent associations. The rate of aggregation of both beta-lactoglobulin and alpha-lactalbumin at 75 degrees C, measured as the loss of native proteins by PAGE, was higher in the cheese WPC solution than in the acid WPC solution. Cross dialysis of the two WPC solutions resulted in alteration of the mineral composition of each WPC solution and reversing their heat-induced aggregation behavior. The results demonstrated that the mineral composition is very important in controlling the aggregation behavior of WPC products.     

Heat-induced redistribution of disulfide bonds in milk proteins. 2. Disulfide bonding patterns between bovine beta-lactoglobulin and kappa-casein

Heat treatment of milk causes the heat-denaturable whey proteins to aggregate with kappa-casein (kappa-CN) via thiol-disulfide bond interchange reactions. The particular disulfide bonds that are important in the aggregates are uncertain, although Cys(121) of beta-lactoglobulin (beta-LG) has been implicated. The reaction at 60 degrees C between beta-LG A and an activated kappa-CN formed small disulfide-bonded aggregates. The tryptic peptides from this model system included a peptide with a disulfide bond between a Cys residue in the triple-Cys peptide [beta-LG(102-124)] and kappa-CN Cys(88) and others between kappa-CN Cys(88) or kappa-CN Cys(11) and beta-LG Cys(160). Only the latter two novel disulfide bonds were identified in heated (90 degrees C/20 min) milk. Application of computational search tools, notably MS2Assign and SearchXLinks, to the mass spectrometry (MS) and collision-induced dissociation (CID)-MS data was very valuable for identifying possible disulfide-bonded peptides. In two instances, peptides with measured masses of 4275.07 and 2312.07 were tentatively assigned to beta-LG(102-135):kappa-CN(11-13) and beta-LG A(61-69):kappa-CN(87-97), respectively. However, sequencing using the CID-MS data demonstrated that they were, in fact, beta-LG(1-40) and beta-LG(41-60), respectively. This study supports the notion that reversible intramolecular disulfide-bond interchange precedes the intermolecular interchange reactions.