High-level, stage- and mammary-tissue-specific expression of a caprine kappa-casein-encoding minigene driven by a beta-casein promoter in transgenic mice.

A 5' truncated caprine (ca) kappa-casein-encoding gene (kappa Cas) was fused to the 3' end of a 3' truncated ca beta Cas. The kappa Cas form comprised the 0.8-kb 3' end of intron 2, the remaining part of the transcription unit containing codons -2 to stop 172, and 0.43 kb of the 3' flanking region. The beta Cas form comprised a 3-kb 5' flanking region and the 5' end of the transcription unit terminating 69 bp downstream from exon 2 which encodes the 15-amino-acid (aa) signal peptide and the first 2 aa of mature beta Cas. The resulting hybrid gene driven by the beta Cas promoter was expressed in all eight lines of transgenic mice investigated, although at different levels. In two lines, the yield of recombinant (re-) kappa Cas was > or = 3 mg/ml of milk. The stage- and mammary tissue-specific expression was similar to that of endogenous beta Cas. The re-kappa Cas differed from its goat milk counterpart by the occurrence of four extra aa at the N-terminal end, indicating that the signal peptidase released the beta Cas signal peptide. According to sedimentation analyses of murine milk containing > or = 3 mg re-kappa Cas/ml, the latter essentially occurred in micelles. Preliminary comparative assays of the behavior of ca alpha s1Cas-kappa Cas and alpha s1Cas-re-kappa Cas mixtures upon incremental addition of Ca2+ showed that re-kappa Cas had the capacity to protect alpha s1Cas against Ca(2+)-induced precipitation in forming stable micelles.(ABSTRACT TRUNCATED AT 250 WORDS)

New data on the proteins of rabbit (Oryctolagus cuniculus) milk.

The main rabbit milk proteins have previously been prepared by reversed-phase HPLC of the acid-precipitated material ('whole casein') and of its supernatant (acid whey). Most of them were nearly homogeneous on SDS-PAGE. Among those isolated from whole casein, alpha s1-, beta- and kappa-caseins, as well as whey acidic protein (WAP) were identified by N-terminal sequencing. After further internal sequencing, two unknown proteins were found to be the putative products, alpha s2a- and alpha s2b-caseins of two recently sequenced transcripts from rabbit mammary gland. Each whole casein component gave several bands on IEF. For kappa-casein, this was probably due to uneven glycosylation as in all kappa-caseins studied so far. For the other whole casein components, including WAP, the number of bands roughly reflected the number of potential phosphorylation sites predicted from the sequences. For alpha s1- and alpha s2-caseins polymorphism could be detected. From acid whey, in addition to WAP, which was a minor component, reversed phase HPLC separated three proteins. These were alpha-lactalbumin, transferrin and serum albumin, on the basis of their apparent molecular weights deduced from SDS-PAGE. WAP was a major component of the native whey obtained by ultracentrifugation of rabbit milk. It was found to consist of two identical subunits linked by at least one disulfide bridge.

Biochemical and genetic analysis of variant C of caprine alpha s2-casein (Capra hircus).

Two alleles, A and B, were previously described at the goat alpha s2-casein locus. Isoelectric focusing allowed us to subdivide the former one in two new alleles, called A and C. Although alpha s2-casein C cannot actually be distinguished from its A counterpart by starch or polyacrylamide gel electrophoresis, it differs from the previous allele by a single substitution Lys (A)/Ile (C) at position 167, which was confirmed at the nucleotide level. The frequencies of the three alpha s2-casein alleles A, B and C were estimated to be 0.85, 0.04 and 0.11 in the French dairy breeds 'Alpine' and 'Saanen'.

Primary structure of bovine lactoperoxidase, a fourth member of a mammalian heme peroxidase family.

Much is known about bovine lactoperoxidase but no data are available on its primary structure. In this work its main active fraction was isolated from cow's milk and sequenced using a conventional strategy. A clear similarity was found with human myeloperoxidase, eosinophil peroxidase and thyroperoxidase, the sequences of which were recently elucidated from those of their cDNAs and/or genes. The single peptide chain of bovine lactoperoxidase contains 612 amino acid residues, including 15 half-cystines and 4 or 5 potential N-glycosylation sites. The corresponding peptide segments of human myeloperoxidase, eosinophil peroxidase and thyroperoxidase display 55%, 54% and 45% identity with bovine lactoperoxidase, respectively, with 14 out of the 15 half-cystines present in each of the four enzymes being located in identical positions. The occurrence of an odd number of half-cystines in bovine lactoperoxidase supports the recent finding of a heme thiol released from this enzyme by a reducing agent, suggesting that the heme is bound to the peptide chain via a disulfide linkage, since the absence of free thiol in the enzyme was reported long ago.

Two of the three genetic variants of goat alpha s1-casein which are synthesized at a reduced level have an internal deletion possibly due to altered RNA splicing.

This paper describes the elucidation of the primary structure of the three genetic variants of goat alpha s1-casein, alpha s1-Cn D, E and F, which have been found to be associated with reduced amounts of alpha s1-casein in milk. Variant E has the same electrophoretic mobility as variant B, but differs from the latter by the substitutions of Arg for Lys and of Thr for Ala at positions 100 and 195. A genetically controlled event which does not affect the amino acid sequence of this variant might be responsible for its lower rate of synthesis compared to that of alpha s1-casein B. The deletion of 11 amino acids at positions 59-69 and of 37 amino acids at positions 59-95 in variant B leads to variants D and F. In both cases the deletions, which start at the same position of the polypeptide chain, include the major phosphorylation site of the protein. On the basis of sequence data for casein genes and cDNAs, it was concluded that the deletions occurring in the D and F variants are due to the exclusion of one and several exons, respectively. The observed deletions in the proteins could thus be the consequence of splice site mutations which would induce altered RNA processing and hence reduce the rate of synthesis of the casein.

Sequence of caprine alpha s1-casein and characterization of those of its genetic variants which are synthesized at a high level, alpha s1-CnA, B and C.

The sequence of caprine alpha s1-casein (199 residues) was established. The peptide chain has the same length as, and shows a 88% degree of identity with, its bovine counterpart. With the ovine alpha s1-casein, the sequence of which was deduced from that of its mRNA, the degree of identity is 97%, counting as one difference a deletion of eight residues in the ovine protein. The differences between the three genetic variants associated with a high alpha s1-casein content in milk are simple substitutions. Variant alpha s1-CnA differs from variant alpha s1-CnB by two substitutions, 16 Leu (A)----Pro (B) and 77 Gln (A)----Glu (B), the latter inducing the appearance of a phosphate group on 75 Ser. Variant alpha s1-CnC differs from alpha s1-CnB by three substitutions, 8 His (B)----Ile (C), 100 Arg (B)----Lys (C) and 195 Thr (B)----Ala (C). The original type of caprine alpha s1-casein could be another hypothetical genetic variant, having the same electrophoretic mobility as alpha s1-CnB.

Preparation and amino acid sequence of human kappa-casein.

Human kappa-casein was prepared from whole casein by successive hydroxyapatite and thiol-Sepharose chromatographies. The primary structure of its 99-residue N-terminal fragment has been determined by sequencing peptides obtained by tryptic and chymotryptic digestions of the whole protein. This fragment overlaps the known sequence of the 65-residue C-terminal fragment. The 158-residue sequence of human kappa-casein was compared to those of goat, ewe, cow and rat kappa-caseins. Only 22% of the residues are identical in homologous positions. The rate of divergence of the 93-residue N-terminal segment (para-kappa-casein) appears to be higher than that of the rest of the molecule.

Quantification of beta-casein in human milk.

A method is described for preparing immunologically homogeneous human milk beta-casein, against which monospecific rabbit antiserum was prepared. The antiserum was used to quantify beta-casein, the major human casein, by rocket immunoelectrophoresis in individual milk samples. However, it was found that in most samples beta-casein occurred together with degradation products originating from its proteolysis by plasmin. Immunological quantification of human beta-casein, treated with plasmin for various time periods, showed that rocket height was not affected by proteolysis up to degradation states clearly more advanced than those observed in all samples of fresh human milk tested. Assays of 150 individual milk samples from 80 women, covering a lactation period of up to 730 d, gave an average concentration of beta-casein (native + degraded) of 4.67 +/- 0.89 standard deviation (g/l); extremes at 2.1 and 7.3 g/l did not vary significantly during the period under study. Comparison of this average value with an accepted casein content of 4.4 g/l (Macy & Kelly, 1961) showed that the casein content of human milk is underestimated when obtained by N determinations on milk and on its supernatants at pH 4.6 (whey). Caseins other than beta-casein occurred only in minute amounts, if at all.

Does beta-lactoglobulin occur in human milk?

Although beta-lactoglobulin (beta-lg) has been considered to be absent from human milk, recent results of other workers, based on immunological reactions between human milk and rabbit antiserum to bovine beta-lg, suggest that this protein may be present. Although our results show similar immunological reactions, we consider that lactoferrin is responsible for these, as it was the only reactive protein species which could be prepared to homogeneity. Indeed two types of antibodies were found by ELISA test in the antisera to bovine beta-lg. One of them would be able to bind loosely to human lactoferrin, but its binding sites would not be antigenic in the rabbit.

[A new method for human milk protein separation]. / A new method for human milk protein separation.

Attempts at isolating individual human milk proteins showed that cross interactions made it difficult to obtain of homogeneous components. A new method was devised, based on complete precipitation of milk proteins with saturated ammonium sulphate and progressive solubilization of the precipitate on a column of Sephadex G10 with a linear gradient of ammonium sulphate (from saturation to water). Three fractions were obtained. The first contained lactoferrin, serum albumin, lysozyme and traces of alpha-lactalbumin. Lysozyme could be obtained free from contaminants by chromatography on Ultrogel AcA 54. Lactoferrin and serum albumin coeluting as a single peak, were separated by a further chromatography on DEAE-cellulose. From the other two fractions recovered on Sephadex G10, it should be possible to prepare immunoglobulins, alpha-lactalbumin and the bulk of caseins. The homogeneity of the preparations of lysozyme, lactoferrin and serum albumin was assessed by SDS polyacrylamide gel electrophoresis, acrylamide agarose electrophoresis and immunoelectrophoresis.

[The primary structure of bovine alpha s2 caseins]. / Premiers elements de structure primaire des caseines alphas2 bovines

The bovine alphas2-, alphas3-, alphas4- and alphas6-caseins [1] were isolated. The 4 proteins had the same amino-acid composition and C-terminal sequence, but different phosphorus contents. From a mixture of these proteins (designated as 'alphas2-complex') and from alphas3-casein a single and identical N-terminal sequence was obtained by Edman degradation. It seems therefore that the 4 proteins have the same peptide chain and only differ in their phosphorus content. For this reason we propose to modify the nomenclature of Annan and Mason [1] and to use in future the single term alphas2 to designate the caseins which have been previously called alphas2,alphas3,alphas4 and alphas6 by these authors. The study of the primary structure of the peptide chain, which has confirmed these results, was undertaken on the S-carboxymethylated alphas2-complex. From a cyanogen bromide digest and from a tryptic hydrolyzate of the alphas2-complex, 5 and 25 peptides were obtained respectively, both sets of peptides accounting for the whole peptide chain. Examination of the tryptic peptides containing methionine combined with the N- and C-terminal sequences of the alphas2-complex and some CNBr peptides, gave the order of the CNBr peptides, H.CN4--CN2--CN5--CN1--CN3.OH, which contain 4, 22, 115, 49 and 17 residues respectively. A partial sequence accounting for half of the peptide chain of the alphas2-complex is given. This peptide chain is likely composed of 207 amino-acid residues.