Different rates of synthesis of whey protein and casein by alleles of the β-lactoglobulin and αs1-casein locus in cattle

de Bavière; 6353 échantillons de lait ont été analysés pour la caséine ce, 1 et 5355 pour la ,0-lactoglobuline. Les individus homozygotes aa -Cn° produisent significativement plus de caséine que les individus homozygotes a. 1-Cn D. L’expression de l’allèle (3-Lg A est supérieure à celle de l’allèle (3-Lg B chez les individus hétérozygotes ou homozygotes. Chez les hétérozygotes, l’allèle (3-LgA a une production de protéine supérieure d’environ 50%


INTRODUCTION
In cattle rather few loci have been identified and efforts to link them to quantitative traits have not been very successful. Milk protein genes, however, are associated with the quantitative variation of the proteins for which the codominant alleles are coding. Moustgaard et al. (1960), Golikova and Panin (1972), Michalak (1973), Cerbulis and Farrell (1975), Komatsu et al. (1977), Mariani et al. (1979), McLean et al. (1984), Ng-Kwai-Hang et al. (1987) and Aaltonen and Antila (1987) demonstrated that the !3-Lg genotype AA produce more ,0-lactoglobulin than genotypes BB or AB. Also McLean et al. (1984) on cattle and Boulanger et al. (1984) and Grosclaude et al. (1987) for goats showed that a sl -Cn genotypes influence the production of a,,-casein.
Although ,!-Lg and a sl -Cn genotypes show a different rate of protein synthesis, there is little known about the expression of the alleles in heterozygotes. However, the haemoglobin of sickle-cell heterozygote is composed of more than 60% haemoglobin A and less than 40% of haemoglobin S (Wellis and Itano, 1951;Wrightstone and Huisman, 1968). Such different rates of expression of globin genes appear to be even more marked in Hb-C heterozygotes (Boyer et al., 1963;Itano, 1965) and in thallasemias (Na-Nakorn and Wasi, 1970;Huisman et al., 1972). Here we report on differences in the concentration of a,,-caseins and #-lactoglobulins coded by the different alleles of heterozygotes and homozygotes of the Bavarian Simmental and Bavarian Brown Alpine cattle.

MATERIALS AND METHODS
The data are based on casein resp. whey protein analysis of 6353 resp. 5355 milk samples from 2059 Simmental and 1809 Brown Alpine cows. Simrrzental cows were sampled twice, Brown Alpine cows once. The statistical analysis of Simmental data was based on a model with effects of herd, year-season, stage and number of lactation and cows; that of the Brown Alpine herd, year-season, stage and number of lactation, sire of the cow and genotypes at 3 loci (in the case of the a,,-Cn expression, the 3 -Cn, x-Cn and ,0-Lg locus; in the case of the 3 -Lg expression, the a si -Cn, /!-Cn and x-Cn locus). The different mean expression of the alleles of heterozygous genotypes was tested by a simple t-test; those of the homozygous genotypes by the Student-Newman-Keuls test.
In Sirrcmental cows 2 samples were analysed from nearly every cow. This permitted estimation of the repeatability of the ratio of the proteins in the heterozygotes (a sl -Cn B /a s1 -Cn c resp. !3-LgA/,Q-LgB).
The milk protein content was measured by the amido-black method, the proportion of the a si -casein B resp. C and 0 -lactoglobulin A resp. B by quantitative photometric determination from cellogel electropherograms (Kirchmeier, 1975;personal communication, 1988), where the optical density of the bands was measured by a photodensitometer. The area under the respective peaks was recorded and the integral area computed. This corresponds to the relative quantity of the protein, provided that the specific affinity to bind the dye is taken into consideration. !3-lactoglobulin was isolated from whey proteins after removal of a-lactalbumin (Sluyterman and Elgersma, 1978). The separation of the two genetic variants was achieved by chromatofocusing (Sluyterman and Wijdenes, 1978). Purity and homogeneity was checked by Page electrophoresis (Raymond and Weintraub, 1959).
For determination of the specific dye binding affinity, known quantities of !3-lactoglobulins were electrophorized, the bands coloured by amido-black and measured densitometrically. In comparison with the standard !3-lactoglobulin A, ,Q-lactoglobulin B had a dye-binding activity of 1.05, similar to published results (Reimerdes and Mehrens, 1978;Krause, personal communication, 1988). The analogous coefficient for a sl -casein B relative to a sl -casein C was taken as 1.06, as published previously by McLean et al. (1982).

RESULTS
The average differences between the expression of a sl -casein B and C alleles in heterozygotes were insignificant (Table I). However, homozygous a,,-Cn cc cows had a higher a sl -casein content than the alternative BB homozygote. As shown in Figure 1, the degree of activity of the alleles in the heterozygote varied considerably and its distribution approached that of a normal curve.
The two alleles of !3-lactoglobulin heterozygote ,B_LgAB differed significantly in their activity. 3 -Lg A produced about 50% more lactoglobulin A than /3-Lg B did lactoglobulin B. This difference is paralleled by the difference between alternative homozygotes. The distribution (Fig. 1) indicates considerable variability and a leptocurtosis.
In Figs. 2 to 4, the course over seasons in 2 years of the ratio between the proteins produced by the alleles of the respective a,,-Cn and 3 -Lg heterozygotes and the expression of the alleles in homozygotes is shown. The difference between the whey proteins of the #-Lg heterozygotes remains nearly stable during the 2 years of the investigation (Fig. 4). In contrast, the B allele of the a sl -Cn heterozygote shows significantly more synthetic activity during the springsummer seasons than the Callele (Fig. 2). Even in homozygous genotypes, a Sl -Cn B shows more activity in this period (Fig. 3). In general, !i-LgB and a sl -Cn C show a more constant expression in heterozygous genotypes than the resp. homologous alleles.
For the ratio of Qs1 -caseins in heterozygotes, repeatability was estimated as 18%, and as about 50% for the ,Q-lactoglobulins. This indicates that this ratio reflects to a considerable degree an innate property of cows which probably is inherited to a large extent. However, even for whey protein, a large proportion of the variability is due to factors not accounted for in the model. The lower repeatability of the ratio between the caseins may reflect inter alia the interaction between the allelic activity and seasonal influences.

DISCUSSION
The two breeds Bavarian Simmental and Bavarian Brown Alpine are located in different regions and the analysis of the milk samples was performed at different times. The differences between genotypes in both breeds are similar (Table I), as are the distributions and the seasonal changes. As to seasonal effects on the ratio of caseins, we can only speculate at this time. During spring-summer seasons, cows are either on pasture or zero-grazing and receive fresh grass which contains steroids which, in turn, may activate the different alleles to different degrees.
The above average expression of the B allele in a sl -Cn heterozygotes could, to some degree, be a product of a so -caseins of C-a s1 protein co-migrating with the B-a sl protein. For the B protein, the contribution of the a so -casein is evident in the electrophoregram and has been considered in estimating the B-fraction. Also the area in the case of homozygotes was corrected for where indeed CC genotypes produce significantly more casein than the BB genotypes. Therefore, the above average expression of the B alleles in heterozygotes during spring-summer could be influenced also by differences in phosphokinase activity. However, the significant increase in the expression of BB homozygotes in the spring-summer season cannot be accounted for by such an influence.