Preliminary result of a genetic polymorphism of β-lactoglobulin gene and the phylogenetic study of ten balkan and central european indigenous sheep breeds

Genetic polymorphism at the β-lactoglobulin (β-LG) loci in indigenous sheep breeds (Tsigai, Racka, Pramenka) was determined. Altogether 904 sheep were genotyped for the presence of the A, B and C alleles of β-lactoglobulin by PCR-RFLP. The AB genotype was the most common and the β-lactoglobulin A was the most frequent in the Cokanski Tsigai (54%), while the B allele was the most common in the Rusty and the Zomborski Tsigai (59%, 60%). The C allele was found only in one individual from Serbian Cokanski flock. These results differ from those that refer to other native sheep breeds. In the Cokanski Tsigai, deviation from the Hardy-Weinberg equilibrium was detected. Genetic relationship based on β-lactoglobulin polymorphism was the closest between the Rusty and the Cokanski Tsigai among the studied populations and between sheep and goat among the other ruminants. Part of the promoter region (254 bp) of β-LG in studied sheep breeds were sequenced in order to identify polymorphisms, analyze haplotypes, and phylogenetic relationship among them. Sequencing analysis and alignment of the obtained sequences showed one haplotype. Analysis of more samples and longer parts of the promoter region of β-LG are needed to reconstruct a phylogenetic tree.


INTRODUCTION
Milk nutrients are essential to human consumption, especially during the early age of life.It is estimated that 3% of the infants born worldwide are sensitive to milk proteins, and this sensitivity declines to 1% by the time of adulthood.However, the milk protein fractions causing allergic reactions (αs1 casein, αs2 casein, β casein, and β lactoglobulin) consist of numerous variants and not all of the variants trigger such reactions.In order to provide the people manifesting sensitivity to certain milk protein fractions with the nutrients required, a selection method for dairy species should be developed so as to reduce or eliminate the triggering protein fractions causing milk allergies.
β-lactoglobulin is the major whey protein in the milk of ruminants and it also exists in the milk of some nonruminants (pigs, horses, dogs, cats, dolphins, whales).
However, it is absent from the milk of rodents, lagomorphs, camels and humans (Hambling et al., 1992;Kappeler, 1998).This protein was isolated about 60 years ago, but its clear biological function is still not described, in contrast to other main whey proteins, such as α-lactalbumin (Mercier & Vilotte, 1993).The ability of β-LG to bind retinol, fatty acids, and other hydrophobic substances has suggested that its biological function may be the participation in milk fat digestion and transportation of these substances to the newborn (Fogolari et al., 2000).Ovine whey is characterized by a higher proportion of β-LG than caprine or bovine whey (Casper et al., 1998).Aschaffenburg & Drewry (1955) published the first genetic polymorphism data of β-LG.The β-LG encoding gene has been sequenced in sheep, cows and goats (Folch et al., 1994).The β-LG gene is on chromosome 3 in sheep and chromosome 11 in goats and cows (Hayes & Petit, 1993).Allele frequencies of β-lactoglobulin have already been studied in several sheep breeds.The genetic variants A (tyrosine) and B (histidine) differ at the amino acid position 20 (Moioli et al., 1998;2007).The genetic variant C -a sub-type of variant A -was identified by Erhardt (1989).It is an arginine -glutamine substitution at position 148.Variants A and B were detected in all studied breeds, while variant C was found only in a few breeds.Frequencies of β-LG A were usually higher than that of β-LG B (Ivankovic & Dovc, 2004;Moioli et al., 2007;Elmaci et al., 2006;Dario et al., 2008); however, in some breeds, the B allele was more frequent (Anton et al., 1999;Di Stasio et al., 1997;Pietrola et al., 2000).Variant C was either absent in most of the studied breeds or its frequency was the lowest among β-LG alleles (Erhardt, 1989;Recio et al., 1997;Meie et al., 2006;Elmaci et al., 2006).
During the last decades, genetic polymorphism studies of milk protein genes in different species gained the interest of the animal breeding and dairy industries (Elmaci et al., 2006).One prominent reason for this interest is the increase in the number of people who are suffering from allergies which are mainly caused by the proteins in milk.It has been known for years that α S1 -casein, α-lactalbumin and β-LG arecow's milk allergens.Therefore, manufacturers want to prove the presence or absence of these proteins (Sieber, 2000).Since β-LG is absent from the milk of humans and camels, it is supposed to be the most responsible for milk allergies related to the consumption of cow milk (Kappeler, 1998).However, there is still no clear opinion on whether a product made from cow milk could be substituted by the one that contains the milk of other animal species.Since in camel milk there is a similar amount of essential amino acid content to the levels found in cow milk, camel milk could be a new protein source for children who are allergic to cow milk (Kappeler, 1998;El-Agamy et al., 1997).Buergin-Wolff et al. (1980) and Taylor (1986) have not found any significant difference in the allergenic effect of β-LG and casein of cow milk.Milk allergies cannot be cured while taking medicines at the same time, however they are endurable, conditioned for the affected person to maintain a strict diet.Given that the different variants/alleles can be detected by the use of DNA tests, the use of genomic selection in dairy species represents a viable solution for producing allergen free milk.
Zackel and Tsigai strains are the two most widespread indigenous sheep breed groups found in Eastern and Southern Europe, currently being reared in 14 countries from this regions (Draganescu, 2007).There are historical evidences for the presence of similar types of sheep in the ancient Egypt, from where the migrations at different periods of time took these breeds to the Middle East and Europe.The exact origins of breeds are still unknown.
To the best of our knowledge, no other study concerning the β-lactoglobulin genetic polymorphism in Tsigai and Zackel breed groups exists up to this moment.However, the importance of indigenous breeds as gene reservoirs is recognized worldwide following the large scale crossbreeding's with exotic dairy or meat specialized breeds during the last decades.
In this study, our aim was to determine the allele and genotype frequencies of the β-LG gene in ten different Balkan and Central European indigenous sheep breeds and to study their genetic differences based on their β-LG polymorphism and promoter region.1).

Materials
Methods.Genomic DNA was extracted from total blood samples using standard protocol (Zsolnai & Orbán, 1999).PCR amplification was carried out in a 25 µl reaction mixture containing 100-150 ng DNA.β-LG A and B alleles were determined, as previously reported by Feligini et al. (1998).β-LG C allele was characterized from β-LG A, as described by Anton et al. (1999).Promoter region of β-LG gene of five animals per breed was sequenced by Mastrangelo et al. (2012) procedure.

RESULTS AND DISCUSSION
At the β-LG locus, 3 alleles were characterized.The 904 sheep, belonging to three breeds (Zomborski, Rusty, Cokanski), were genotyped for the presence of the A, B and C alleles.The method of Feligini et al. (1998) was used to differentiate between A+C and B alleles, while since C is a subtype of allele A, a second genotyping was needed to differentiate between A and C alleles using the method of Anton et al. (1999).Genotype and allele frequency data of the β-LG obtained by PCR-RFLP of all three studied breeds are presented in Table 1.Three genotypes (AA, AB, BB) and two alleles (A, B) were presented because C allele was detected only in one individual (Cokanski Tsigai) in heterozygote form.The most frequent was AB in all three breeds, just like in Pag ewes, Polish Merino and crosses derived from Finn, Romanov and Booroola rams (Cubrik-Curik et al., 2002;Piwczynski et al., 2002).BB genotype was in the Russian Karakul breed, while AB genotype in Indian Jalauni sheep was not detected (Nassiry et al., 2002;Arora et al., 2010).Frequency of genotypes of AA, AB, BB were observed at 18%, 72%, 9% in the Cokanski, at 11%, 59%, 30% in the Rusty Tsigai and at 13%, 54%, 33% in the Zomborski Tsigai, respectively.Except for the Cokanski Tsigai, the other breeds were in Hardy-Weinberg equilibrium at the β-LG locus (Table 1).The frequency of A, B and C β-LG alleles were estimated to be 0.54, 0.46 and 0.005 in the Cokanski, 0.41, 0.59 and 0.00 in the Rusty, as well as 0.40, 0.60 and 0.00 in the Zomborski Tsigai.Results revealed that the β-LG A allele was more frequent than the B allele only in the Cokanski breed from among the studied breeds; however, based on other publications, β-LG A seems to be more frequent than the B allele in the Iranian Karakul, the Finnish Landrace, the Russian Karakul, the Ghezel, the Hungarian Racka, the Pag island sheep, the Manchega and the Hyfer Border Leicester breeds (Mohammadi et al., 2006;Thomas et al., 1989;Rampilli et al., 1997;Lopez-Galvez et al., 1994;Ivankovic & Dovc, 2004;Kerekes et al., 2008;Amigo et al., 2000).The β-LG B allele was found to be more fre-quent than A in the Romanov (Macha & Novackova, 1974), Tajik (Aliev & Koloteva, 1975), Sarda (Bolla et al., 1989), Lacha (Recio et al., 1997), and Chios (Macha & Novackova, 1974) breeds.Our results showed that the frequency of β-LG B allele in the studied breeds was higher than in the breeds studied previously (Mohammadi et al., 2006;Amigo et al., 2000).A relatively low incidence of the C allele was found in the studied breeds, as previously reported in the Merino, Massese, Merinoland, Kivircik, Gokceada and Sakiz breeds by Recio et al. (1997), Erhardt (1989), Meie et al. (2006), Elmaci et al. (2006).Amigo et al. (2000) stated that β-LG A could be the original form of ovine β-LG.
Genetic relationships between the studied breeds based on β-LG polymorphism are presented in Fig. 1.The closest relationship was found between the Rusty and the Cokanski Tsigai, while the Zomborski Tsigai was clustered into another group.Since there is a similarity between caprine, ovine and bovine milk proteins, genetic relationship between ovine, bovine and caprine was also studied based on the β-LG nucleotide sequence (Fig. 2).The smallest genetic distance was observed between ovine and caprine.
After alignment and reduction (254 bp) of the sequences of promoter region of β-LG only one haplotype was detected.
Regrettably, within the sampling farms, no available data existed regarding the milk chemical composition for the studied populations.As a result, the comparison between milk content and individual genotypes, even though it could had given interesting results, was not feasible in the current study.Kawecka & Radko (2011) have not found statistical differences between β-lactoglobulin genotype and milk yield composition in the Polish Merino.Conversely to results published by Mroczkowski et al. (2004) for the same breed.Yousefi et al. (2013), Ramos et al. (2009) and Dario et al. (2008) revealed significant associations between AB genotype of β-LG and high percentages of fat and milk lactose in different dairy breeds, respectively.
For the future, we would like to characterize all known alleles of ovine milk protein genes in the studied breeds, sequence the whole promoter region of β-LG and determine genetic variability within breeds.Our plan is also to extend the study by examining the association between milk composition and β-LG polymorphism.
Hereby, we took the first steps forward towards allergy-free products based on genotyped animals from indigenous sheep breeds reared in Central-, Eastern-and Southern-Europe.

Figure 1 .
Figure 1.Genetic distance between the studied indigenous breeds based on β-LG genotypes

Table 1 . Distribution of β-lactoglobulin genotype and allele frequencies
*A+C alleles together.a Test of Hardy-Weinberg equilibrium; NS, not significant.b Test of Hardy-Weinberg equilibrium; p<0.001 .c Test of Hardy-Weinberg equilibrium; p<0.01.d Upper row: observed value; Lower row: expected value.