Post-translational Modification of κ-Casein in the Golgi Apparatus of Mid-lactating Mammary Glands from Rat and Cow

Bovine κ-casein, a phosphoglycoprotein, has mucin-type carbohydrate chains. Subcellular distribution of enzymes that take part in the post-translational modification of κ-casein was examined. In lactating mammary glands from rats and cows, N-acetyl-galactosaminyl transferase, galactosyl transferase, sialyl transferase, and casein kinase were localized specifically in the Golgi apparatus.

The substrate specificities indicate that these enzymes are actually responsible for the processing of κ-casein.

The presence of a phosphate group attached to κ-casein did not affect the rate of glycosylation by N-acetyl-galactosaminyl transferase, while the presence of carbohydrate chains attached to κ- casein strongly reduced the rate of phosphorylation by casein kinase. These results suggest that in the Golgi apparatus, phosphorylation of κ-casein precedes glycosylation.     

Reconstitution of Human Casein Micelle and Its Properties

Human casein micelles were reconstituted from isolated κ- and β-caseins and calcium ions. Micelle formation was recognized in the presence of calcium chloride even at the low concentration of 5mM. At pH levels ranging from 5.5 to 8.0, the re-formed micelles were quite stable so that precipitation of β-casein was not observed. The large micelles were constituted by a higher ratio of β-casein to κ-casein (16:1 by weight) than the small micelles (3: 1). The κ-casein in the small micelles contained carbohydrates to about 43% (w/w) in the molecule, whereas that in the large micelles was only about 25%. When the casein micelles were re-formed from κ-easein and fractionated β-casein components, the extent of phosphorylation of the β-casein component was found to influence the micelle formation; i.e., the β-casein component with no phosphate (the 0-P form) was disadvantageous to form micelles, but the component with 5 phosphates (the 5-P form) formed micelles most easily.     

Attachment Sites of Carbohydrate Portions to Peptide Chain of κ-Casein from Bovine Colostrum

Short glycopeptides were prepared from bovine colostral κ-casein treated with cyanogen bromide and proteases (pronase P and thermolysin), followed by gel filtrations and ion exchange chromatography. It was confirmed by Edman degradation that glycopeptide I among short glycopeptides obtained was homogeneous. From the effect of alkali treatment, it was assumed that three polysaccharide chains of glycopeptide I were attached to the peptide chain through OH groups of threonines. By chemical procedures and carboxypeptidase P treatment, the amino acid sequence of glycopeptide I was established to be Ser-Gly-Glu-Pro-Thr-Ser-Thr-Pro-Thr-Thr-Glu-Ala-Val. Threonine residues of No. 5, 7 and 9 were bound to the carbohydrate chains through galactosamine. The sugar chain bound to the threonine residue at No. 7 contained glucosamine. Glycopeptide I corresponded to residues of No. 127–139 in κ-casein A from normal milk.     

Composition and size distribution of bovine casein micelles

Chromatography of glutaraldehyde-fixed skim-milk on controlled-pore glass (CPG-10, 300 nm) gave three micellar fractions whose averaged diameters, measured by electron microscopy, decreased progressively with increasing elution volume. Casein micelles with diameters up to 680 nm were detected. The casein composition of the same fractions from unfixed skim-milk was determined. As the fraction elution volume increased, κ-casein varied from 7.7 to 11.4% of total casein, giving α_s/κ ratios of 6.1, 4.7 and 3.3.A plot of κ-casein content versus micelle surface-to-volume ratio for skim-milk and the column fractions approximated to a straight line. Re-calculation of the published results from two other studies also gave linear relationships between κ-casein content and surface area for artificial micelles. The three regression lines thus obtained had small intercepts. It was concluded that the data indicated the same fundamental structure for casein micelles, with a pre-dominant surface location for κ-casein, whether the micelles are natural or artificial and whether they are aggregated or by Ca²⁺ alone oy Ca²⁺ together with calcium phosphate-citrate complex.     

Fractionation and Characterization of 9 Subcomponents of Bovine κ-Casein A

κ-casein A was fractionated into 9 subcomponents, all of which were identified as κ-casein from immunological analyses. The microheterogeneity of the subcomponents was explained by stepwise increase of their carbohydrate contents (0~4mol/mol of GalNAc, and 0~8mol/mol of NANA). The micelle-stabilizing ability of κ-casein subcomponents increased with the increase of their carbohydrate contents: the carbohydrate rich subcomponent 7 possessed twice the stabilizing ability of the carbohydrate free subcomponent 1. The sensitivity of synthetic casein micelle composed of κ-casein subcomponents and α_sl-casein to the wheat germ lectin-induced aggregation also increased with the increase of their NANA contents.     

Purification and Characterization of UDP-N-Acetylgalactosamine: κ-Casein Polypeptide N-Acetylgalactosaminyltransferase from Mammary Gland of Lactating Cow

UDP-N-acetyl-d-galactosamine: κ-casein polypeptide N-acetylgalactosaminyltransferase was purified from a crude Golgi apparatus of lactating bovine mammary gland after solubilization with Triton X-100. Through chromatography on DEAE-Sephadex A-50, apomucin-Sepharose 4B, FPLC mono S, and Sephacryl S-200, and then electrofocusing, the enzyme was purified up to 7500-fold from the homogenate.

The molecular weight of the enzyme was estimated at 200,000 from gel filtration. The pI value of the enzyme was 6.4 on electrofocusing. The purified enzyme transferred GalNAc from UDP-GalNAc, not to the carbohydrate chains but to the polypeptide chains of the substrates, κ-casein and mucin. The enzyme required Mn²⁺, DTT, and Triton X-100 for maximal activity. The Km value for UDP-GalNAc was 16.2μm. Km values for K-subcomponents 1 and 7, and apomucin were 1.15, 5.10, and 0.192mg/ml, and V_max values were 254, 259, and 581 nmol/hr/mg, respectively. Thermal stability and the effects of pH, milk components, lectins, and nucleotides were examined.

α_s1-Casein strongly inhibited GalNAc transfer to κ-casein. The inhibitory effect of α_s1-casein was canceled by the addition of Ca²⁺, which causes casein micelle formation. This means that the glycosylation of κ-casein occurs after casein micelle formation triggered by the accumulation of Ca²⁺ in vivo.     

Interaction of Calcium Ion with Bovine Caseins

In order to clarify the interaction of calcium ion with casein, the volume change associated with the interaction was measured by dilatometric procedures. When CaCl₂ was added to the casein solutions at neutral pH, a volume increase occurred and reached a constant saturated value of about 700 ml per 10⁶ g protein with increasing CaCl₂ concentrations for whole-, α_s- and β-casein solutions, but there was no volume change for κ-casein solution. On the other hand, the binding of calcium ion to the casein fractions was determined by a gel filtration procedure at pH 6.0 to 9.0. The number of Ca²⁺ ions bound to the caseins increased with the CaCl₂ concentration and pH value, and the relative order of binding capacities for the caseins was: α_s-casein > whole-casein > β-casein > κ-casein.

It was found that the volume changes obtained by the dilatometry were smaller than the calculated volume increases based on the assumption that these are caused by the binding of Ca²⁺ ion to the caseins. Therefore it is necessary to introduce another factor which reduces the volume increase due to the Ca²⁺ ion binding in order to reasonably explain the measured volume changes. At present it is presumed that there occurs the unfolding of peptide chain of casein molecule on Ca²⁺ ion binding, which has been known to decrease the volume of the protein solution.     

Identification of the milk fat globule membrane proteins: I. Isolation and partial characterization of glycoprotein B

The salt soluble proteins from the fat globule membrane of cow's milk were resolved into three fractions by Sephadex column chromatography in sodium dodecyl sulfate. One of the fractions, termed glycoprotein B, was purified by rechromatography to essentially one band on sodium dodecyl sulfate gel electrophoresis. It was found to contain 14% carbohydrate including sialic acid, mannose, galactose, glucose, glucosamine and galactosamine. The amino acid composition of glycoprotein B was determined; it has amino terminal serine and carboxyl terminal leucine. The molecular weight of this glycoprotein as estimated by sodium dodecyl sulfate gel electrophoresis is 49 500.     

Studies on κ-Casein of Bovine Milk

κ-Caseins were prepared by the calciurn-ethanol method, the Sephadex method and the urea-sulfuric acid method. Some important properties of κ-caseins were investigated using isoelectric focusing, starch gel electrophoresis, ultracentrifugation, chemical analysis, stabilization test of α_s-casein, and rennin treatment. Isoelectric focusing established that κ-casein had its isoelectric point near pH 6.0 in 6 m urea, usually accompanied by a second peak around pH 5.6. Ultracentrifugation, however, showed a single peak having a s_20,w value of 2.6 ~ 3.8 in the presence of 6 m urea and of 14.4 in the absence of such dispersing reagents. Normal contents of hexose, sialic acid, phosphorus, and nitrogen were about 1.5, 0.8, 0.2, and 14%, respectively. Relative patterns of amino acid composition were similar in all of the κ-caseins. In addition, amino acid composition in intact κ-casein and in the further purified κ-casein which formed the second peak in DEAE cellulose chromatography were almost identical, indicating that the κ-casein of the first peak is not an impurity but is one of the components which formed the original κ-casein complexes. The ability of κ-caseins to stabilize α_s-casein in the presence of calcium increased when purified by DEAE cellulose chromatography.     

Phosphorylation of casein by cyclic AMP-dependent protein kinase.

The catalytic subunit of rabbit muscle cyclic AMP-dependent protein kinase (EC 2.7.1.37; ATP:protein transferase) has been tested on a variety of caseins. The B variant of β-casein was phosphorylated at a much greater rate than other β-caseins, α_s1-caseins, and κ-caseins. Whole casein homozygous for β-casein B was phosphorylated at 2.5 times the rate of commercial whole casein. Gel electrophoresis experiments indicate that β-casein is the predominant component phosphorylated in commerical casein. It is therefore suggested that phosphorylation of whole casein depends on its content of the specific genetic variant, β-casein B.     

Photolysis of Aryl Glycosides with Ultraviolet Irradiation

Bovine casein components (α_sl-, β-, and κ-caseins) were chemically phosphorylated and the properties of the modified components were compared with those of the native to clarify the function of the intrinsic phosphate groups of casein components in casein micelle formation. The calcium binding ability of casein components increased after chemical phosphorylation. The concentrations of calcium chloride required to precipitate modified α_sl- and β-caseins were higher than those for native components. However, phosphorylation of α_sl- and β-caseins did not affect their properties of forming micelles through interaction with κ-casein. The stabilizing ability of κ-casein for α_sl- and β caseins was impaired by its phosphorylation, but the stability was recovered by treating phosphorylated κ-casein with phosphoprotein phosphatase. The results show that the phosphate content of κ-casein must be low to form a stable casein micelle. The results also explain why the specific phosphorylation of casein components in the mammary gland is required.     

Amino acids content and electrophoretic profile of camel milk casein from different camel breeds in Saudi Arabia

This study aimed to evaluate amino acids content and the electrophoretic profile of camel milk casein from different camel breeds. Milk from three different camel breeds (Majaheim, Wadah and Safrah) as well as cow milk were used in this study.Results showed that ash and moisture contents were significantly higher in camel milk casein of all breeds compared to that of cow milk. On the other hand, casein protein of cow milk was significantly higher compared to that of all camel milk breeds. Molecular weights of casein patterns of camel milk breeds were higher compared to that of cow milk.Essential (Phe, Lys and His) and non-essential amino acids content was significantly higher in cow milk casein compared to the casein of all camel milk breeds. However, there was no significant difference for the other essential amino acids between cow casein and the casein of Safrah breed and their quantities in cow and Safrah casein were significantly higher compared to the other two breeds. Non-essential amino acids except Arg and the essential amino acids (Met, Ile, Lue and Phe) were also significantly higher in cow milk α-casein compared to α-casein from all camel breeds. Moreover, essential amino acids (Val, Phe and His) and the non-essential amino acids (Gly and Ser) content was significantly higher in cow milk β-casein compared to the β-casein of all camel milk breeds and the opposite was true for Lys, Thr, Met and Ile. However, Met, Ile, Phe and His were significantly higher for β-casein of Majaheim compared to the other two milk breeds. The non-essential amino acids (Gly, Tyr, Ala and Asp) and the essential amino acids (Thr, Val and Ile) were significantly higher in cow milk κ-casein compared to that for all camel milk breeds. There was no significant difference among all camel milk breeds in their κ-casein content of most essential amino acids.Relative migration of casein bands of camel milk casein was not identical. The relative migration of α_s-, β- and κ-casein of camel casein was slower than those of cow casein. The molecular weights of α_s-, β- and κ-casein of camel caseins were 27.6, 23.8 and 22.4 KDa, respectively. More studies are needed to elucidate the structure of camel milk.     

Isolation and properties of human kappa-casein

Human kappa-casein was isolated from human whole casein by gel filtration with Sephadex G-200 and hydroxylapatite chromatography in the presence of sodium dodecyl sulfate (SDS). The kappa-casein was calcium-insensitive and did stabilize human beta-casein and bovine alpha s1-casein against precipitation by calcium ions. Formation of micelles from human beta- and kappa-caseins, and calcium ions was confirmed by electron microscopic observation. On SDS-polyacrylamide gel electrophoresis (SDS-PAGE), a single band was obtained. The formation of para-kappa-caseins by chymosin was confirmed by SDS-PAGE. Two para-kappa-caseins with apparent molecular weights of 13,000 and 11,000 appeared. The molecular weight of intact human kappa-casein was estimated to be approximately 33,000. The human kappa-casein contained about 40% carbohydrate (15% galactose, 3% fucose, 15% hexosamines, and 5% sialic acid) and 0.10% (1 mol/mol) phosphorus. Its amino acid composition was similar to that of bovine kappa-casein except for serine, glutamic acid, and lysine contents.     

Proteomic analysis of whey and casein proteins in early milk from the marsupial Trichosurus vulpecula,the common brushtail possum

Whey and casein proteins representing the first and second halves of the early lactation phase in the common brushtail possum (Trichosurus vulpecula) have been compared by two dimensional gel electrophoresis. Nine components of whey were differentially expressed during early lactation, including proteins identified as cathepsin B, clusterin, late lactation protein, lysozyme, ganglioside M2 activator and neutrophil gelatinase-associated lipocalin. A major novel protein, termed very early lactation protein (VELP), was identified in whey. Partial amino acid sequence data obtained from VELP did not appear to match any other reported protein sequence. VELP was shown to be an acidic glycoprotein of 20–30 kDa which exists as a homodimer. In the casein fraction, κ-casein appeared to be differentially post-translationally modified during early lactation and fragments of β-casein were relatively more abundant at the earlier lactation stage.     

κ-Casein terminates casein micelle build-up by its “soft” secondary structure

In our previous paper (Nagy et?al. in J Biol Chem 285:38811–38817, 2010) by using a multilayered model system, we showed that, from α-casein, aggregates (similar to natural casein micelles) can be built up step by step if Ca-phosphate nanocluster incorporation is ensured between the protein adsorption steps. It remained, however, an open question whether the growth of the aggregates can be terminated, similarly to in nature with casein micelles. Here, we show that, in the presence of Ca-phosphate nanoclusters, upon adsorbing onto earlier α-casein surfaces, the secondary structure of α-casein remains practically unaffected, but κ-casein exhibits considerable changes in its secondary structure as manifested by a shift toward having more β-structures. In the absence of Ca-phosphate, only κ-casein can still adsorb onto the underlying casein surface; this κ-casein also expresses considerable shift toward β-structures. In addition, this κ-casein cover terminates casein aggregation; no further adsorption of either α- or κ-casein can be achieved. These results, while obtained on a model system, may show that the Ca-insensitive κ-casein can, indeed, be the outer layer of the casein micelles, not only because of its “hairy” extrusion into the water phase, but because of its “softer” secondary structure, which can “occlude” the interacting motifs serving casein aggregation. We think that the revealed nature of the molecular interactions, and the growth mechanism found here, might be useful to understand the aggregation process of casein micelles also in?vivo.     

Membrane interactions and lipid binding of casein oligomers and early aggregates

Caseins constitute the main protein components in mammalian milk and have critical functions in calcium transport and prevention of protein aggregation. Fibrillation and aggregation of κ-casein, a phenomenon which has only recently been detected, might be associated with malfunctions of milk secretion and amyloidosis phenomena in the mammary glands. This study employs a newly-designed chromatic biomimetic vesicle assay to investigate the occurrence and the parameters affecting membrane interactions of casein aggregates and the contribution of individual casein members to membrane binding. We show that physiological casein colloids exhibit membrane activity, as well as early globular aggregates of κ-casein, a prominent casein isoform. Furthermore, inhibition of κ-casein fibrillation through complexation with α_S-casein and β-casein, respectively, was found to go hand in hand with induction of enhanced membrane binding; these data are important in the context of casein biology since in secreted milk κ-casein is found only in assemblies containing also α_S-casein and β-casein. The chromatic experiments, complemented by transmission electron microscopy analysis and fluorescence quenching assays, also revealed significantly higher affinity early spherical aggregates of k-casein to anionic phosphatidylglycerol-lipids, as compared to zwitterionic phospholipids. Overall, this study suggests that lipid interactions play important roles in maintaining the essential physiological functions of caseins in mammalian milk.     

Comparison of casein of cynomolgus monkey (Macaca fascicularis) with human casein

Casein of cynomolgus monkey was compared with those from human and bovine milk. Cynomolgus monkey casein showed similar electrophoretical patterns to those of human casein on Disc- and SDS-electrophoresis. It consisted of beta- and kappa-casein-like components. The component corresponding to bovine alpha s1-casein was not detected. The beta-casein-like fraction of cynomolgus monkey showed 9 bands on Disc-PAGE. These were suggested to be the same protein binding different levels of phosphorus by dephosphorylation experiment using an acid phosphatase. The kappa-casein-like component of cynomolgus monkey was highly glycosylated (about 50% carbohydrate) similarly as human kappa-casein and the constituent carbohydrates were same as those detected in human kappa-casein (galactose, fucose, N-acetylgalactosamine, N-acetylglucosamine, and sialic acid). Amino acid composition of cynomolgus monkey kappa-casein bore a resemblance to those of both human and bovine kappa-caseins. Amino acid composition of cynomolgus monkey beta-casein was also similar to those of human and bovine beta-caseins.     

Enzymatic Cross-Linking of Casein Facilitates Gel Structure Weakening Induced by Overacidification

Casein (α_S1, α_S2, β, κ) is the major protein fraction in milk and, together with heat denatured whey proteins, responsible for gel network formation induced by acidification. Rheological measurements during gelation typically reveal a maximum storage modulus (G') at a pH close to the isoelectric point (pI) of casein (~4.6). With further decreasing pH gel stiffness decreases because of increased electrostatic repulsion, which is referred to as overacidification. In this study we investigated the effect of casein cross-linking with microbial transglutaminase on gel structure weakening induced by acidification to pH below the pI. Although enzymatic cross-linking increased the maximum stiffness (G' _MAX ) of casein gels the reduction of G' during overacidification, expressed as ratio of the plateau value (G' _FINAL ) to G' _MAX , was more pronounced. Almost no soluble protein was detected in the serum of gels from cross-linked casein, whereas considerable amounts of α_S- and κ-casein were released from reference gels below the pI. This suggests that covalent cross-linking of casein retains charged molecules within the gel network and therefore causes a higher reduction of protein-protein interactions because of higher electrostatic repulsion. Furthermore, higher amounts of uncross-linked β-casein, which was the only casein type not found in the serum, resulted in higher G' _FINAL to G' _MAX ratios, underlining the important contribution of β-casein to acid gel formation and prevention of gel structure weakening.     

Bovine colostrum fraction as a serum substitute for the cultivation of mouse hybridomas

Summary Fractions of bovine colostrum were prepared and their ability to support the growth of mouse-mouse hybridomas in culture was tested. Whey was prepared from defatted colostrum by removal of casein using acid precipitation. An ultrafiltrate was obtained from cleared whey by filtration through membranes with a nominal molecular mass cut-off of 100 000 Da. Colostrum ultrafiltrate contained 1.16 g/l protein, 0.24 g/l immunoglobulin G (IgG) and less than 0.24 EU (endotoxin unit)/ml endotoxins. The effect of defatted colostrum, whey and ultrafiltrate as serum substitutes was examined by cultivation of hybridoma cells in minimal essential medium containing different concentrations of the supplements. Under optimal conditions in ultrafiltrate-supplemented medium, the maximal cell concentration was 35–40% of that obtained using 10% foetal bovine serum, and IgG production per cell was equal to that achieved using serum. In 1% defatted colostrum the maximum hybridoma concentration was about 30% of that in 10% serum, but at higher concentrations hybridoma growth was significantly reduced. The growth-promoting activity of whey was low. The results show that bovine colostrum ultrafiltrate provides a very attractive alternate to serum for production of monoclonal antibodies. Correspondence to: R. Pakkanen     

Purification and characterization of the free secretory piece of human colostrum (author's transl)]

The free secretory piece is isolated from human colostrum by gel filtration and ion-exchange chromatography in high yield (200 mg/l colostrum). DEAE-Cellulose chromatography separates the free secretory piece in two fractions which are electrophoretically distinct, but otherwise have the same characteristics, like molecular weight, antigenic determinants, N-terminal sequence, peptide map and amino acid composition. It was therefore concluded that the protein part of the secretory piece is homogenous.