**3. Results and discussion**

#### **3.1. Genetic structure**

The genotype and allele frequencies of the *bPRL* gene studied in four Russian cattle breeds are shown in Table 1.

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and

Designations: *A* and *B,* alleles of the *bPRL* gene; P, Polish; J, Jersey; I, Indian; L, Lithuanian; B&W, Black and White; R, Russian; Holstein Friesian; Mont, Montebeliard; R.R.P., Russian Red Pied; Kostr, Kostroma; Yar, Yaroslavl; Best,

The diagram is based on published data on Polish Jersey [29], Indian Jersey, Indian Holstein Friesian [28], Lithuanian Black & White, Lithuanian Red [30], Russian Black & White [31], Montebeliard [32], and Russian Red Pied [22] cattle. **Figure 3.** Distribution of *bPRL*(*Rsa*I) allele frequencies in the four breeds studied and in other Russian and foreign breeds. The abscissa shows the breeds; the ordinate shows the allele frequencies (in

Designations: *AA*, *AB*, and *BB, bPRL* genotypes. For designations of the breeds, see Figure 2.

**Figure 4.** Distribution of *bPRL* genotype frequencies in the four breeds studied and in other Russian and foreign breeds. The radial axes show the breeds and the corresponding genotype frequencies.

Bestuzhev.

fractions of unity).

Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 41


Notes: *p*± s.e. is the genotype (allele) frequency and its standard error; the number of animals with the given genotype is indicated in parentheses; *n* is the sample size; Hexp ± s.e. is the expected heterozygosity and its standard error.

**Table 1.** The frequencies of the *Rsa*I restriction genotypes and alleles of the *bPRL* gene in four Russian cattle breeds

The G2 test did not show significant differences between the breeds studied with respect to the genotype or allele frequencies for the *bPRL* gene. The genotype frequency distributions in the breeds studied fit the Hardy–Weinberg equilibrium. The *A* allele was prevailing in all breeds studied; its frequency was two to three times higher than the *B* allele frequency. Thus, these breeds had similar genetic structures in terms of the given SNP despite their different origins and directions of artificial selection.

The distribution of the *Rsa*I alleles of the *bPRL* gene is characterized by a higher frequency of the *A* allele in most breeds studied (Figure 3). In Indian Jersey cattle [28], the frequencies of the two alleles were approximately equal. The *A* allele frequency was lower in Polish Jersey cattle [29], which deserves special attention. These differences in allele frequencies may have resulted from different histories of selection for milk yield in different breeds. The breeds of *B. taurus* exhibit a considerable genotypic variation with respect to the *bPRL* gene marker used in this study. This can be seen in the diagram of the genotype frequency distribution (Figure 4). The group of four Russian cattle breeds studied here significantly differed in the genotype frequency distribution from Russian Black & White [31] and Russian Red Pied [22], but not Lithuanian Black & White [30]; (p < 0.006 with the Bonferroni correction).

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 41

40 Prolactin

Yakut (*n* = 41)

Yaroslavl (*n* = 113)

Bestuzhev (*n* = 57)

Kostroma (*n* = 124)

cattle breeds

**3. Results and discussion** 

The genotype and allele frequencies of the *bPRL* gene studied in four Russian cattle breeds

Breed *bPRL*(*Rsa*I) genotypes *Rsa*I alleles of the *bPRL* gene

Hexp± s.e. 0.393±0.120

Hexp± s.e. 0.461±0.066

Hexp± s.e. 0.432±0.085

Hexp± s.e. 0.375±0.133

different origins and directions of artificial selection.

(*n*) *AA* (23) *AB* (14) *BB* (4) *A* (60) *B* (22) *p±s.e.* 0.561±0.078 0.341±0.074 0.098±0.046 0.732±0.098 0.268±0.098

(*n*) *AA* (48) *AB* (50) *BB* (15) *A* (146) *B* (80) *p*±s.e. 0.425±0.047 0.442±0.047 0.133±0.032 0.646±0.064 0.354±0.064

(*n*) *AA* (27) *AB* (24) *BB* (6) *A* (78) *B* (36) *p*±s.e. 0.474±0.066 0.421±0.065 0.105±0.041 0.684±0.087 0.316±0.087

(*n*) *AA* (69) *AB* (48) *BB* (7) *A* (186) *B* (62) *p*±s.e. 0.556±0.045 0.387±0.044 0.056±0.021 0.750±0.055 0.250±0.055

Notes: *p*± s.e. is the genotype (allele) frequency and its standard error; the number of animals with the given genotype is indicated in parentheses; *n* is the sample size; Hexp ± s.e. is the expected heterozygosity and its standard error. **Table 1.** The frequencies of the *Rsa*I restriction genotypes and alleles of the *bPRL* gene in four Russian

The G2 test did not show significant differences between the breeds studied with respect to the genotype or allele frequencies for the *bPRL* gene. The genotype frequency distributions in the breeds studied fit the Hardy–Weinberg equilibrium. The *A* allele was prevailing in all breeds studied; its frequency was two to three times higher than the *B* allele frequency. Thus, these breeds had similar genetic structures in terms of the given SNP despite their

The distribution of the *Rsa*I alleles of the *bPRL* gene is characterized by a higher frequency of the *A* allele in most breeds studied (Figure 3). In Indian Jersey cattle [28], the frequencies of the two alleles were approximately equal. The *A* allele frequency was lower in Polish Jersey cattle [29], which deserves special attention. These differences in allele frequencies may have resulted from different histories of selection for milk yield in different breeds. The breeds of *B. taurus* exhibit a considerable genotypic variation with respect to the *bPRL* gene marker used in this study. This can be seen in the diagram of the genotype frequency distribution (Figure 4). The group of four Russian cattle breeds studied here significantly differed in the genotype frequency distribution from Russian Black & White [31] and Russian Red Pied

[22], but not Lithuanian Black & White [30]; (p < 0.006 with the Bonferroni correction).

**3.1. Genetic structure** 

are shown in Table 1.

Designations: *A* and *B,* alleles of the *bPRL* gene; P, Polish; J, Jersey; I, Indian; L, Lithuanian; B&W, Black and White; R, Russian; Holstein Friesian; Mont, Montebeliard; R.R.P., Russian Red Pied; Kostr, Kostroma; Yar, Yaroslavl; Best, Bestuzhev.

The diagram is based on published data on Polish Jersey [29], Indian Jersey, Indian Holstein Friesian [28], Lithuanian Black & White, Lithuanian Red [30], Russian Black & White [31], Montebeliard [32], and Russian Red Pied [22] cattle.

**Figure 3.** Distribution of *bPRL*(*Rsa*I) allele frequencies in the four breeds studied and in other Russian and foreign breeds. The abscissa shows the breeds; the ordinate shows the allele frequencies (in fractions of unity).

Designations: *AA*, *AB*, and *BB, bPRL* genotypes. For designations of the breeds, see Figure 2.

**Figure 4.** Distribution of *bPRL* genotype frequencies in the four breeds studied and in other Russian and foreign breeds. The radial axes show the breeds and the corresponding genotype frequencies.

#### **3.2. Genetic variation**

The observed (Hobs) and expected (Hexp) heterozygosities (Table 1) for the *bPRL* gene did not differ significantly from each other in any breed studied; nor did any two breeds differ in these parameters. This was unexpected, considering the substantial differences between these breeds in the *Hinf*I site of the *bPit-1* transcription factor gene (exon 6) and the *Alu*I site of the *bGH* growth hormone gene (exon 5) (our unpublished data), which are also associated with milk productivity.

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and

the breeds studied. However, since artificial selection alters genetic variation, regular

Advances in molecular genetic analysis based on PCR have given rise to novel marker systems for direct genotyping at the DNA level that are independent of the animal's sex and

The *Rsa*I polymorphism of the *bPRL* gene (Table 1), as well as the *Alu*I polymorphism in exon 5 of the *bGH* gene, was used to search for associations of these genes with milk production traits, including the milk yield and milk fat and protein contents, in Yaroslavl cattle. We found the following frequency distribution of genotypes: *LL* (*n* = 32), 0.283 ± 0.042; *LV* (*n* = 63), 0.558 ± 0.047; *VV* (*n* = 18), 0.159 ± 0.034. The genotype frequency distribution in these breeds fit the Hardy–Weinberg equilibrium. Table 1 shows the frequency distribution of *bPRL* genotypes. *AAVL* (*n* = 29) and *ABVL* (*n* = 27) are the most frequent combined genotypes of these genes; the *AALL* and *ABLL* genotypes are somewhat rarer (*n* = 12 and *n* =

We used ANOVA to estimate the isolated effect of the *bPRL* gene and the combined effect of the *bPRL* and *bGH* genes on milk production traits (the milk yield in kilograms and the percentage fat and protein contents for the first three lactations) in Yaroslavl cattle. Table 3 shows the dependence of these parameters for the third lactation on the *bPRL*(*Rsa*I) polymorphism as estimated by one-way ANOVA. The *bPRL* gene was found to affect the fat

Lactation 3

content of milk (*F*(2;63) = 3.18, *p* = 0.048) but not the milk yield or protein content.

Error 63 2.021 0.032 0.0232 0.0004

Protein (%)

lactation on the *PRL*(*Rsa*I) polymorphism in Yaroslavl cattle

Error 63 0.0073 0.0001

probability; Error, residual variance.

 SS MS F P *Rsa*1(*PRL*) 2 0.0005 0.0003 2.34 0.104

significantly higher fat content of milk compared to *BB* cows (*p* = 0.037).

 Milk Yield (kg) Fat (%) Factor D. F. SS MS F P SS MS F P *PRL*(*Rsa*1) 2 0.03 0.015 0.5 0.629 0.002 0.001 3.19 **0.0478** 

Designations: d.f., number of degrees of freedom; SS, sum of squares; MS, mean sum of squares, *F,* Fisher's test; *p,*

**Table 2.** Results of one-way ANOVA showing the dependence of milk production traits for the third

Figures 6a–6c show the dependences of three milk production traits on the *bPRL* genotypes in Yaroslavl cattle. As can be seen in Figure 6a, cows with the *AA* genotype exhibited a

**3.3. Search for associations of SNPs of the prolactin and somatotropin genes** 

monitoring of cattle breeds for the given *Rsa*I genetic marker is advisable.

age. This makes the procedures less time-consuming and more accurate.

**with milk production traits in Yaroslavl cattle** 

16, respectively).

Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 43

In order to compare the genetic diversity in the Russian cattle breeds studied here with those in other Russian and foreign breeds (Figure 5), we calculated the Hexp for all breeds analyzed. The observed and expected heterozygosities for the *bPRL* gene did not differ significantly in most breeds (Figure 5). The Indian Jersey breed was an exception (χ2 = 13.77, *ν* = 1, *p* = 0.00021); the excess of heterozygotes in these cattle apparently resulted from artificial selection.

Designations: Hobs and Hexp, the observed and expected heterozygosities, respectively. For designations of the breeds, see Figure 2. The diagram is based on the expected heterozygosity levels calculated from published data on Montebeliard [32], Russian Red Pied [22], Polish Jersey [29], Lithuanian Black & White, Lithuanian Red [30], Indian Jersey, and Indian Holstein Friesian [28] cattle.

**Figure 5.** The observed and expected heterozygosities for the *bPRL* gene markers in Russian and foreign cattle breeds. The abscissa shows the breeds; the ordinate shows the heterozygosity coefficients (in fractions of unity).

Four Russian breeds studied did not differ from one another in the genetic variation (Hexp). At the same time, as can be seen in Figure 5, this group had significantly higher Hexp values compared to Russian Red Pied [22] Indian Holstein Friesian [28], and Montebeliard [32], cattle (G2 =36.13, *ν* = 4, *p* =3.0∙10–7; G2 =39.19, *ν* = 4, *p* = 1.0∙10–7; G2 =16.72, *ν* = 4, *p* = 0.0022; and G2 =19.41, *ν* = 4, *p* = 0.0007, respectively; after the Bonferroni correction).

Thus, the heterozygosity levels in the Kostroma, Bestuzhev, Yakut, and Yaroslavl breeds are significantly higher than in the breeds used for comparison. This indicates a stable state of the breeds studied. However, since artificial selection alters genetic variation, regular monitoring of cattle breeds for the given *Rsa*I genetic marker is advisable.

#### **3.3. Search for associations of SNPs of the prolactin and somatotropin genes with milk production traits in Yaroslavl cattle**

42 Prolactin

**3.2. Genetic variation** 

with milk productivity.

artificial selection.

Jersey, and Indian Holstein Friesian [28] cattle.

(in fractions of unity).

The observed (Hobs) and expected (Hexp) heterozygosities (Table 1) for the *bPRL* gene did not differ significantly from each other in any breed studied; nor did any two breeds differ in these parameters. This was unexpected, considering the substantial differences between these breeds in the *Hinf*I site of the *bPit-1* transcription factor gene (exon 6) and the *Alu*I site of the *bGH* growth hormone gene (exon 5) (our unpublished data), which are also associated

In order to compare the genetic diversity in the Russian cattle breeds studied here with those in other Russian and foreign breeds (Figure 5), we calculated the Hexp for all breeds analyzed. The observed and expected heterozygosities for the *bPRL* gene did not differ significantly in most breeds (Figure 5). The Indian Jersey breed was an exception (χ2 = 13.77, *ν* = 1, *p* = 0.00021); the excess of heterozygotes in these cattle apparently resulted from

Designations: Hobs and Hexp, the observed and expected heterozygosities, respectively. For designations of the breeds, see Figure 2. The diagram is based on the expected heterozygosity levels calculated from published data on Montebeliard [32], Russian Red Pied [22], Polish Jersey [29], Lithuanian Black & White, Lithuanian Red [30], Indian

Four Russian breeds studied did not differ from one another in the genetic variation (Hexp). At the same time, as can be seen in Figure 5, this group had significantly higher Hexp values compared to Russian Red Pied [22] Indian Holstein Friesian [28], and Montebeliard [32], cattle (G2 =36.13, *ν* = 4, *p* =3.0∙10–7; G2 =39.19, *ν* = 4, *p* = 1.0∙10–7; G2 =16.72, *ν* = 4, *p* = 0.0022; and

Thus, the heterozygosity levels in the Kostroma, Bestuzhev, Yakut, and Yaroslavl breeds are significantly higher than in the breeds used for comparison. This indicates a stable state of

**Figure 5.** The observed and expected heterozygosities for the *bPRL* gene markers in Russian and foreign cattle breeds. The abscissa shows the breeds; the ordinate shows the heterozygosity coefficients

G2 =19.41, *ν* = 4, *p* = 0.0007, respectively; after the Bonferroni correction).

Advances in molecular genetic analysis based on PCR have given rise to novel marker systems for direct genotyping at the DNA level that are independent of the animal's sex and age. This makes the procedures less time-consuming and more accurate.

The *Rsa*I polymorphism of the *bPRL* gene (Table 1), as well as the *Alu*I polymorphism in exon 5 of the *bGH* gene, was used to search for associations of these genes with milk production traits, including the milk yield and milk fat and protein contents, in Yaroslavl cattle. We found the following frequency distribution of genotypes: *LL* (*n* = 32), 0.283 ± 0.042; *LV* (*n* = 63), 0.558 ± 0.047; *VV* (*n* = 18), 0.159 ± 0.034. The genotype frequency distribution in these breeds fit the Hardy–Weinberg equilibrium. Table 1 shows the frequency distribution of *bPRL* genotypes. *AAVL* (*n* = 29) and *ABVL* (*n* = 27) are the most frequent combined genotypes of these genes; the *AALL* and *ABLL* genotypes are somewhat rarer (*n* = 12 and *n* = 16, respectively).

We used ANOVA to estimate the isolated effect of the *bPRL* gene and the combined effect of the *bPRL* and *bGH* genes on milk production traits (the milk yield in kilograms and the percentage fat and protein contents for the first three lactations) in Yaroslavl cattle. Table 3 shows the dependence of these parameters for the third lactation on the *bPRL*(*Rsa*I) polymorphism as estimated by one-way ANOVA. The *bPRL* gene was found to affect the fat content of milk (*F*(2;63) = 3.18, *p* = 0.048) but not the milk yield or protein content.


Designations: d.f., number of degrees of freedom; SS, sum of squares; MS, mean sum of squares, *F,* Fisher's test; *p,* probability; Error, residual variance.

**Table 2.** Results of one-way ANOVA showing the dependence of milk production traits for the third lactation on the *PRL*(*Rsa*I) polymorphism in Yaroslavl cattle

Figures 6a–6c show the dependences of three milk production traits on the *bPRL* genotypes in Yaroslavl cattle. As can be seen in Figure 6a, cows with the *AA* genotype exhibited a significantly higher fat content of milk compared to *BB* cows (*p* = 0.037).

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and

genotype significantly differed from those with the *AAVL, AALL, ABVV,* and *BBVL* genotypes; and cows with the *BBVV* genotype significantly differed from those with the *ABVV* and *BBVL* genotypes. In terms of the milk protein content, *AALL* cows significantly differed from *AAVV, ABVL, ABLL,* and *BBLL* cows; and *BBVV* cows, from *BBLL* ones.

Trait Milk Yield Fat (%)

*PRL*(*Rsa*I) 2 0.006 0.003 0.2 0.858 4.0E-05 2.0E-05 0.060 0.939 *GH*(*Alu*I) 2 0.084 0.042 2.1 0.123 0.001 2.6E-04 0.920 0.401

**Table 3.** Results of two-way ANOVA showing the dependence of the milk production parameters on

Figure 7 graphically shows the combined effects of the *bPRL* and *bGH* genes. The milk yield (in kilograms) and the fat and protein contents of milk of the first lactation in Yaroslavl cows are plotted in Figures 7a, 7b, and 7c, respectively, against the *bPRL* genotypes (*AA, AB,* and *BB*). The *bGH* genotypes are indicated by dots of different colors. The points corresponding to the same *bGH* genotype but different *bPRL* genotypes are connected by dotted lines. As evident from Figure 7, there was no consistent dependence of the fat or protein content on the dose of any allele of any gene. Let us consider how the mean milk fat content depended on the *bPRL* genotypes in combination with different *bGH* genotypes (Figure 7b). The combinations of the *AA* genotype of the *bPRL* gene with different *bGH* genotypes did not differ significantly from one another, and neither did the combinations of the *BB* genotype of the *bPRL* gene with different *bGH* genotypes. In contrast, the mean percentage fat content of milk of *AB* cows was significantly lower if they had the *LL* genotype of the *bGH* gene than if they had the *VV* genotype. This indicates gene interaction, which should be taken into account because otherwise the effects of individual genes on the formation of these traits

Error 104 2.049 0.020 0.029 2.8E-04

Lactation 1

4 0.001 3.7E-04 2.930 **0.024**

Trait Protein (%)

Error 104 0.013 1.3E-04

Designations are the same as in Table 2.

Effect D.f. SS MS F P *PRL*(*Rsa*I) 2 2.4E-05 1.2E-05 0.100 0.908 *GH*(*Alu*I) 2 2.3E-04 1.1E-04 0.890 0.412

the *bPRL* and *bGH* genes and their interaction in Yaroslavl cattle

D.f. SS MS F P SS MS F P

4 0.064 0.016 0.8 0.520 0.003 0.001 2.590 **0.041** 

Effect Lactation 1

*PRL*(*Rsa*I) *\*GH*(*Alu*I)

*PRL*(*Rsa*I) *\*GH*(*Alu*I)

Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 45

**Figure 6.** Dependence of (a) the milk yield (in kilograms), (b) the fat content, and (c) the protein content of milk of the third lactation on the *bPRL* genotypes in Yaroslavl cattle. The abscissas show the *bPRL* genotypes; the ordinates show the mean values of the traits; vertical bars are 0.95 confidence intervals.

Considering that quantitative traits, including milk production traits, are each determined by a number of genes, we also estimated the possible combined effect of two factors, the *bPRL* and *bGH* genes, on the milk production traits using two-way ANOVA. The analysis did not reveal a significant isolated effect of any one-locus genotype for any lactation. Still it showed that the *bPRL* genotype tended to affect the third-lactation milk fat content (*p* = 0.064), which was in contrast to the results of one-way ANOVA showing its significant effect (Table 2). Like one-way ANOVA, the two-way analysis demonstrated that *AA* cows were characterized by a higher fat content of milk than *BB* cows. However, the main result concerning the relationship between the *bPRL* and *bGH* gene polymorphisms and milk production was that we found combined effects of these two genes on the fat and protein contents of first-lactation milk in Yaroslavl cattle (*F*(4;104) = 2.59, *p* = 0.041 and *F*(4;104) = 2.93, *p* = 0.024, respectively) (Table 3).

In order to identify the genotypes whose carriers significantly differed in the mean percentage fat and protein contents of milk, we performed post-hoc pairwise comparisons of these traits in cows with different combined genotypes of the *bPRL* and *bGH* genes. Table 4 shows the results of this analysis. In terms of the milk fat content, cows with the *ABLL*

genotype significantly differed from those with the *AAVL, AALL, ABVV,* and *BBVL* genotypes; and cows with the *BBVV* genotype significantly differed from those with the *ABVV* and *BBVL* genotypes. In terms of the milk protein content, *AALL* cows significantly differed from *AAVV, ABVL, ABLL,* and *BBLL* cows; and *BBVV* cows, from *BBLL* ones.


Designations are the same as in Table 2.

44 Prolactin

confidence intervals.

0.024, respectively) (Table 3).

**Figure 6.** Dependence of (a) the milk yield (in kilograms), (b) the fat content, and (c) the protein content of milk of the third lactation on the *bPRL* genotypes in Yaroslavl cattle. The abscissas show the *bPRL* genotypes; the ordinates show the mean values of the traits; vertical bars are 0.95

Considering that quantitative traits, including milk production traits, are each determined by a number of genes, we also estimated the possible combined effect of two factors, the *bPRL* and *bGH* genes, on the milk production traits using two-way ANOVA. The analysis did not reveal a significant isolated effect of any one-locus genotype for any lactation. Still it showed that the *bPRL* genotype tended to affect the third-lactation milk fat content (*p* = 0.064), which was in contrast to the results of one-way ANOVA showing its significant effect (Table 2). Like one-way ANOVA, the two-way analysis demonstrated that *AA* cows were characterized by a higher fat content of milk than *BB* cows. However, the main result concerning the relationship between the *bPRL* and *bGH* gene polymorphisms and milk production was that we found combined effects of these two genes on the fat and protein contents of first-lactation milk in Yaroslavl cattle (*F*(4;104) = 2.59, *p* = 0.041 and *F*(4;104) = 2.93, *p* =

In order to identify the genotypes whose carriers significantly differed in the mean percentage fat and protein contents of milk, we performed post-hoc pairwise comparisons of these traits in cows with different combined genotypes of the *bPRL* and *bGH* genes. Table 4 shows the results of this analysis. In terms of the milk fat content, cows with the *ABLL*

**Table 3.** Results of two-way ANOVA showing the dependence of the milk production parameters on the *bPRL* and *bGH* genes and their interaction in Yaroslavl cattle

Figure 7 graphically shows the combined effects of the *bPRL* and *bGH* genes. The milk yield (in kilograms) and the fat and protein contents of milk of the first lactation in Yaroslavl cows are plotted in Figures 7a, 7b, and 7c, respectively, against the *bPRL* genotypes (*AA, AB,* and *BB*). The *bGH* genotypes are indicated by dots of different colors. The points corresponding to the same *bGH* genotype but different *bPRL* genotypes are connected by dotted lines. As evident from Figure 7, there was no consistent dependence of the fat or protein content on the dose of any allele of any gene. Let us consider how the mean milk fat content depended on the *bPRL* genotypes in combination with different *bGH* genotypes (Figure 7b). The combinations of the *AA* genotype of the *bPRL* gene with different *bGH* genotypes did not differ significantly from one another, and neither did the combinations of the *BB* genotype of the *bPRL* gene with different *bGH* genotypes. In contrast, the mean percentage fat content of milk of *AB* cows was significantly lower if they had the *LL* genotype of the *bGH* gene than if they had the *VV* genotype. This indicates gene interaction, which should be taken into account because otherwise the effects of individual genes on the formation of these traits would be incorrectly estimated. Note that, among the combined genotypes of the *bPRL* and *bGH* genes shown in Figure 7, the *AALL, AAVL, ABVV,* and *BBVL* genotypes were characterized by a significantly higher milk fat content of milk compared to the *ABLL* gene (Figure 7b).

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and

**Figure 7.** Dependence of (a) the milk yield (in kilograms), (b) the fat content, and (c) the protein content of milk of the first lactation on combined genotypes of the *bPRL* and *bGH* genes in Yaroslavl cattle. The abscissas show the *bPRL* genotypes (*AA, AB,* and *BB*); the ordinates show the mean values of the traits; vertical bars are 0.95 confidence intervals. Different *bGH* genotypes are denoted by dots of different colors; the points corresponding to the same *bGH* genotype but different *bPRL* genotypes are connected

by dotted lines.

Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 47


**Table 4.** Results of post-hoc comparisons of the mean first-lactation fat (above the diagonal) and protein (below the diagonal) contents of milk of Yaroslavl cows with different combined *bPRL*–*bGH* genotypes

Regarding the dependence of the mean protein content of milk in Yaroslavl cattle on the *bPRL* genotypes combined with different *bGH* genotypes (Figure 7c), the combined *AALL* genotype differs from *AAVV*, and *BBVV* differed from *BBLL*. In contrast, no combination of the *AB* genotype of *bPRL* with a *bGH* genotype differed from its combination with any other *bGH* genotype in this respect. Note that, in the given sample of Yaroslavl cattle, the *AAVV, ABLL,* and *BBLL* genotypes were preferable over the *AALL* genotype in terms of the percentage protein content of milk, as was the *BBLL* genotype over the *BBVV* genotype.

Thus, in Yaroslavl cattle, the *AA* genotype of the *bPRL* gene was characterized by a significantly higher percentage fat content of milk than the *BB* genotype, in contrast to Russian Red Pied cattle, where *AB* cows had a higher milk fat content than *AA* and *BB* cows (*p* < 0.05) [22]. Our study has been the first to demonstrate the combined effect of the *bPRL* and *bGH* genes (i.e., their combined genotypes) on the milk fat and protein contents. A number of other authors have also studied the relationship of milk production with combined genotypes of SNPs in the same or different genes; however, the possibility of their combined effect has almost never been considered. One exception is the study on the effects of the combined genotypes of the *Alu*I and *Msp*I polymorphic sites in exon 5 and intron 3, respectively, of the growth hormone gene in Polish Black & White cattle [20]. However, these authors studied combined *bGH* genotypes as a single factor; hence, they did not consider the effect of interaction between individual SNPs on the traits studied. In addition, the marker system used by them was hardly suitable for revealing the interaction of these SNPs because they were located in the same gene.

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 47

46 Prolactin

(Figure 7b).

would be incorrectly estimated. Note that, among the combined genotypes of the *bPRL* and *bGH* genes shown in Figure 7, the *AALL, AAVL, ABVV,* and *BBVL* genotypes were characterized by a significantly higher milk fat content of milk compared to the *ABLL* gene

Combined *bPRL* and *bGH* genotypes (*bPRL*(*Rsa*I)–*bGH*(*Alu*I))

*AAVV* 0.933 0.725 0.243 0.945 0.182 0.272 0.267 0.968 *AAVL* 0.102 0.701 0.163 0.809 **0.041** 0.175 0.186 0.984 *AALL* **0.038** 0.378 0.336 0.572 **0.044** 0.139 0.369 0.806 *ABVV* 0.067 0.484 0.986 0.125 **0.008 0.038** 0.955 0.339 *ABVL* 0.432 0.181 **0.058** 0.125 0.069 0.218 0.143 0.919 *ABLL* 0.733 0.086 **0.029** 0.068 0.570 0.882 **0.009** 0.259 *BBVV* 0.058 0.343 0.725 0.736 0.108 0.064 **0.042** 0.312 *BBVL* 0.559 0.368 0.153 0.208 0.960 0.727 0.159 0.36

*BBLL* 0.713 0.086 **0.036** 0.054 0.293 0.491 **0.045** 0.387

protein content of milk, as was the *BBLL* genotype over the *BBVV* genotype.

SNPs because they were located in the same gene.

**Table 4.** Results of post-hoc comparisons of the mean first-lactation fat (above the diagonal) and protein (below the diagonal) contents of milk of Yaroslavl cows with different combined *bPRL*–*bGH* genotypes

Regarding the dependence of the mean protein content of milk in Yaroslavl cattle on the *bPRL* genotypes combined with different *bGH* genotypes (Figure 7c), the combined *AALL* genotype differs from *AAVV*, and *BBVV* differed from *BBLL*. In contrast, no combination of the *AB* genotype of *bPRL* with a *bGH* genotype differed from its combination with any other *bGH* genotype in this respect. Note that, in the given sample of Yaroslavl cattle, the *AAVV, ABLL,* and *BBLL* genotypes were preferable over the *AALL* genotype in terms of the percentage

Thus, in Yaroslavl cattle, the *AA* genotype of the *bPRL* gene was characterized by a significantly higher percentage fat content of milk than the *BB* genotype, in contrast to Russian Red Pied cattle, where *AB* cows had a higher milk fat content than *AA* and *BB* cows (*p* < 0.05) [22]. Our study has been the first to demonstrate the combined effect of the *bPRL* and *bGH* genes (i.e., their combined genotypes) on the milk fat and protein contents. A number of other authors have also studied the relationship of milk production with combined genotypes of SNPs in the same or different genes; however, the possibility of their combined effect has almost never been considered. One exception is the study on the effects of the combined genotypes of the *Alu*I and *Msp*I polymorphic sites in exon 5 and intron 3, respectively, of the growth hormone gene in Polish Black & White cattle [20]. However, these authors studied combined *bGH* genotypes as a single factor; hence, they did not consider the effect of interaction between individual SNPs on the traits studied. In addition, the marker system used by them was hardly suitable for revealing the interaction of these

*AAVV AAVL AALL ABVV ABVL ABLL BBVV BBVL BBLL* 

**Figure 7.** Dependence of (a) the milk yield (in kilograms), (b) the fat content, and (c) the protein content of milk of the first lactation on combined genotypes of the *bPRL* and *bGH* genes in Yaroslavl cattle. The abscissas show the *bPRL* genotypes (*AA, AB,* and *BB*); the ordinates show the mean values of the traits; vertical bars are 0.95 confidence intervals. Different *bGH* genotypes are denoted by dots of different colors; the points corresponding to the same *bGH* genotype but different *bPRL* genotypes are connected by dotted lines.

### **4. Conclusion**

We have found that four Russian cattle breeds, Yakut, Bestuzhev, Kostroma, and Yaroslavl cattle, are similar in genetic structure. All of them are characterized by a low frequency of the *BB* genotype of the *Rsa*I polymorphic site in the *bPRL* gene (from 0.056 ± 0.021 to 0.133 ± 0.031) and high frequencies of heterozygotes and homozygotes for the *A* allele. The breeds of *B. taurus* exhibit a considerable genotypic variation with respect to the *bPRL* gene marker used in this study. The group of these four Russian cattle breeds significantly differs in the genotype frequency distribution from other breeds, such as Russian Black & White and Russian Red Pied. At the same time, these four breeds do not differ significantly in the observed or expected heterozygosity for the *bPRL* gene either from each other or from other breeds used for comparison.

Use of the Bovine Prolactin Gene (*bPRL*)for Estimating Genetic Variation and

*Department of Genetics, Koltsov Institute of Developmental Biology, Russian Academy of Sciences,* 

*Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow, Russian Federation* 

This study was supported by the Subprogram of the Presidium of the Russian Academy of Sciences "Gene Pools and Gene Diversity," State Contract no. 14.740.11.0164 in the framework of the Federal Target Program "Science and Education Professionals of Innovative Russia," and the Russian Foundation for Basic Research (grant no. 12-04-92214

[1] Bernichtein S., Touraine P., Goffin V. REVIEW New concepts in prolactin biology.

[2] Horseman N.D., Zhao W., Montecino-Rodriguez E., Tanaka M., Nakashima K., Engle S.J., Smith F., Markoff E., Dorshkind K. Defective mammopoiesis, but normal hematopoiesis, in mice with a targeted disruption of the prolactin gene. EMBO Journal

[3] Rischkowsky B., Pilling D., editors. The State of the World's Animal Genetic Resources for Food and Agriculture Food and Agriculture Organization of the United Nations.

[4] Ernst L.K., Dmitriev N.G., Paronyan I.A., editors. Geneticheskie resursy selskokhozyaistvennykh zhivotnykh v Rossii i sopredel'nykh stranakh (Genetic Resources of Farm Animals in Russia and Neighboring Countries). St. Petersburg,

[5] Korotov G.P. Krupnyi Rogatyi Skot Yakutskoi FSSZ i Metody Ego Uluchsheniya (Yakut Cattle and Breeding Methods of Its Improvement). Yakutsk, USSR: Yakutknigoizdat;

[6] Sulimova G.E., Lazebnaja I.V., Perchun A.V., Voronkova V.N., Ruzina M.N., Badin G.A. Uniqueness of Kostroma breed of cattle from a position of molecular genetics.

Advances in science and technology of Agro-Industrial Complex 2011;9 52-54. [7] Weller J.I. Quantitative Trait Loci Analysis in Animals. London: CABI Publishing;

O.E. Lazebny

S.R. Khatami

Mong\_a).

**5. References** 

1997;16(23) 6926–6935.

Russia: VNIIGRZh; 1994.

1983.

2001.

*Moscow, Russian Federation* 

**Acknowledgement** 

*Department of Genetics, Faculty of Science, Shahid Chamran University of Ahvaz, Ahvaz, Iran Department of Comparative Genetics of Animals,* 

Journal of Endocrinology 2010;206(1) 1–11.

FAO. Rome. 2007. ISBN 978-92-5-105762-9

Milk Production in Aboriginal Russian Breeds of *Bos taurus* L. 49

We have demonstrated a combined effect of the *bPRL* and *bGH* genes on the percentage protein and fat contents of milk. Each trait has been found to be significantly positively associated with some of the combined genotypes. No genotype has been found to positively affect both traits. At the same time, some genotypes are associated positively with one trait and negatively with the other one: *ABVV* cows are characterized by a high fat content and low protein content of milk, while this is the other way round with *ABLL* cows. This could be used for selecting cattle for high individual productivity traits. Note that only one combined genotype (*BBVV*) is unfavorable in terms of both traits. However, being doubly homozygous, it may serve as a reserve for obtaining genotypes that are valuable in terms of either fat (*ABVV* and *BBVL*) or protein (*AAVV* and *BBLL*) content of milk. Thus, the study of the combined effects of the *bPRL* and *bGH* genes and the breeding practice taking these effects into account allow the cattle productive potential to be analyzed in more detail. In addition, involvement of epistatic gene interaction in the formation of selectively valuable quantitative traits has been further confirmed.

We believe that data on the *bPRL*(*Rsa*I) SNP marker and the heterozygosity estimates calculated from its allele frequencies may be used in programs for conservation of aboriginal breeds while maintaining the optimal balance between the goals of artificial selection and preservation of the genetic diversity, which is necessary for sustained reproduction of cattle breeds with all their unique characters. This is especially important because of the rapid decrease in the stocks of cattle breeds and the related threat of partial loss of the genetic resources for stockbreeding on a global scale. [33].

#### **Author details**

I.V. Lazebnaya and G.E. Sulimova *Department of Comparative Genetics of Animals, Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow, Russian Federation* 

#### O.E. Lazebny

48 Prolactin

**4. Conclusion** 

breeds used for comparison.

quantitative traits has been further confirmed.

resources for stockbreeding on a global scale. [33].

*Department of Comparative Genetics of Animals,* 

*Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow,* 

**Author details** 

*Russian Federation* 

I.V. Lazebnaya and G.E. Sulimova

We have found that four Russian cattle breeds, Yakut, Bestuzhev, Kostroma, and Yaroslavl cattle, are similar in genetic structure. All of them are characterized by a low frequency of the *BB* genotype of the *Rsa*I polymorphic site in the *bPRL* gene (from 0.056 ± 0.021 to 0.133 ± 0.031) and high frequencies of heterozygotes and homozygotes for the *A* allele. The breeds of *B. taurus* exhibit a considerable genotypic variation with respect to the *bPRL* gene marker used in this study. The group of these four Russian cattle breeds significantly differs in the genotype frequency distribution from other breeds, such as Russian Black & White and Russian Red Pied. At the same time, these four breeds do not differ significantly in the observed or expected heterozygosity for the *bPRL* gene either from each other or from other

We have demonstrated a combined effect of the *bPRL* and *bGH* genes on the percentage protein and fat contents of milk. Each trait has been found to be significantly positively associated with some of the combined genotypes. No genotype has been found to positively affect both traits. At the same time, some genotypes are associated positively with one trait and negatively with the other one: *ABVV* cows are characterized by a high fat content and low protein content of milk, while this is the other way round with *ABLL* cows. This could be used for selecting cattle for high individual productivity traits. Note that only one combined genotype (*BBVV*) is unfavorable in terms of both traits. However, being doubly homozygous, it may serve as a reserve for obtaining genotypes that are valuable in terms of either fat (*ABVV* and *BBVL*) or protein (*AAVV* and *BBLL*) content of milk. Thus, the study of the combined effects of the *bPRL* and *bGH* genes and the breeding practice taking these effects into account allow the cattle productive potential to be analyzed in more detail. In addition, involvement of epistatic gene interaction in the formation of selectively valuable

We believe that data on the *bPRL*(*Rsa*I) SNP marker and the heterozygosity estimates calculated from its allele frequencies may be used in programs for conservation of aboriginal breeds while maintaining the optimal balance between the goals of artificial selection and preservation of the genetic diversity, which is necessary for sustained reproduction of cattle breeds with all their unique characters. This is especially important because of the rapid decrease in the stocks of cattle breeds and the related threat of partial loss of the genetic *Department of Genetics, Koltsov Institute of Developmental Biology, Russian Academy of Sciences, Moscow, Russian Federation* 

S.R. Khatami

*Department of Genetics, Faculty of Science, Shahid Chamran University of Ahvaz, Ahvaz, Iran Department of Comparative Genetics of Animals, Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow, Russian Federation* 
