logical traits of the same group of animals.
2. MATERIAL AND METHODS
2.1. Animals and Phenotypes
Official milk recording data, spanning over 10 years time, from 95 Holstein cows were used in this study. Cows belonged to four commercial neighboring farms located in Northern Italy, the most vocated Italian area for milk production. DNA was obtained from blood for each recorded animal. The following parameters were considered: fat and protein content at the test day; mature equivalent (ME) milk yield, fat and protein content. For 48 of these cows, the results of the morphological evaluation, performed by the Herdbook experts were available; therefore the following parameters were also included in the statistical analysis: stature, body depth, angularity, rump angle, rump width, rear legs, udder depth.
2.2. SNP Detection and Genotyping
Polymorphism detection and genotyping was performed for all samples by direct sequencing on the 3500 Genetic Analyzers (Applied Biosystems, Foster City, CA, USA). For the LEPR gene, 19 primer pairs were designed on the Bos taurus whole genome shotgun sequence (accession no. NW_001494806) and are summarized in Table 1. The amplified fragments include the whole coding sequence and part of the 5’ flanking region. For the LEP gene, all cows were genotyped at the already reported SNP by [15,16,23]. Primers used and position on the gene are summarized in Table 2. In order to sequence the amplicons Promoter 1 (1141 bp) and Promoter 2 (613 bp) that presented a very high CG content together with several deletions and insertion it was necessary to design four more forward (P1-for1, P1-for2, P1-for3, P2-for) and two more reverse primers (P1-rev, P2-rev), that fell inside the amplicon, so to obtain shorter sequences ensuring a greater analytical specificity, without affecting the efficiency.
Table 1. Primers used for the amplification and sequencing of the bovine LEPR gene.
Table 2. Primers used for the amplification and sequencing of the bovine LEP gene.
2.3. Statistical Analysis
The allele substitution effect of each SNP was estimated, similarly to [24,25], by regressing the number of copies of each allele against each of the traits separately, using the following Linear Mixed Model in SAS :
Yijklm = μ + Bi + + Cj + Dk + Gl + Am + eijklmn
where: Yijklm = Phenotype1; μ = overall mean; Bi = fixed effect of the herd; Cj = fixed effect of the season/year of calving; Dk = fixed effect of the age at calving; Gl = allele substitution effect; Am = random animal effect; eijklmn = residual.
A GLM that included only the allele substitution effect as covariate was used for the ME milk yield, fat and protein content, because these parameters were already corrected for the fixed effects. A GLM that included the year of birth and the age at the evaluation as fixed effects was used for the morphological indexes.
A false discovery rate (FDR) was also calculated using PROC MULTTEST in SAS  to account for multiple testing, whereby significance was achieved at an adjusted FDR of P < 0.10, as suggested in . Linkage disequilibrium (LD) between loci was estimated using PROC ALLELE in SAS  in order to evaluate the pair-wise correlation coefficients for all SNP within each of the two genes, so to avoid false-positive effects and not to ascribe to a SNP the effect of another one, in case the alleles of the second were highly correlated with the first.
2.4. Transcription Factor Binding Sites
The MATCH program  in the TRANSFAC® Professional 10.2 http://www.biobaseinternational.com/  was used to perform the profile search for the transcripttion factor binding sites for the two genes. The vertebrate binding matrices and the only high quality matrix option were used in the profile selection, and the cut-offs for core and matrix similarity were set to 0.999 and 0.7 respectively, with no further change in the other options.
In Tables 3-4, the position in the gene and the minor allele frequency (MAF) of the detected SNP in the analyzed population, for the LEPR and the LEP gene was reported. Sequence analysis of the DNA of the 95 Holstein cows, across the 7.8 kb region of the bovine LEPR gene, that includes part of the 5’ flanking region and the whole coding region, revealed a total of 16 SNP; 9 of them being novel detected SNP, not previously referred in the literature (Table 3). The novel SNP LEPR01 and LEPR02 are located in the 5' flanking region; LEPR04, LEPR10 and LEPR16 are synonymous mutations, located, respectively, in exons 7, 12, and 18; LEPR06 is a missense mutation in exon 8 that causes the amino acid substitution S460G at the position 460 of the protein; LEPR09 is also a missense mutation in exon 11, producing the amino acid substitution R615S at the position 615 of the protein. Because  showed that the region from 428 to 635 of the protein includes the binding sites for the leptin, the two missense mutations may be fundamental for binding and activate the leptin hormone. Finally, LEPR11 and LEPR14 are located in the introns 15 and 17.
Sequence analysis of the DNA of the 95 Holstein cows, across the 2.6 kb region of the bovine LEP gene, revealed a total of 26 SNP (24 in the promoter and 2 in the exons). Of the 26 detected SNP, 19 had already been reported by  (accession no. AJ571671); three SNP were reported in the dbSNP (rs29004171; rs29004172; rs29004173); one SNP (accession no. AB070368) had been previously reported by  and two SNP (rs29004488 and rs29004508) had been reported by . Finally, a novel SNP, LEP05, was detected in the present study (g.2003435T>C; accession no. NW_001494939).
Table 3. Analyzed SNP in the LEPR gene.
Table 4. Analyzed SNP in the LEP gene.
In this study, the in silico analysis with the MATCH software was performed to verify whether the SNP that resulted significantly associated to productive traits, were located in putative binding sites of transcription factors. The results of this analysis indicated that LEP14 is located at the putative binding site for v-Maf and NF-E2 transcription factors and that the A allele does not activate the binding sites of either of the two.
In the association analysis, of the 16 SNP of the LEPR gene, the following 6 were used: LEPR01 and LEPR02 because located in the promoter; LEPR03, LEPR09 and LEP15, because they encode a missense mutation; LEPR16 because of its short distance (30 bp) from the 3’UTR. Similarly, of the 26 SNP of the LEP gene, the following 4 were considered in the association analysis: LEP01 because of the significant associations reported by  on growth and fertility; LEP14, because of its position within the binding site of transcription factors (v-Maf and NF-E2), LEP25 and LEP26 because they encode a missense mutations.
In Table 5 and Table 6, the results of the association analysis are presented only for those SNP of LEP and LEPR that showed statistically significant effects, reporting also the effects of the same SNP on other traits, if trending to statistical significance (P < 0.09). Table 5 describes the association of 5 SNP with milk traits. Mean values and standard deviation for each trait are reported in the table, so to allow a direct evaluation of the extent of the allelic substitution effect. All these values fall in the range of the average production traits of Holstein cows in Italy.
Three SNP of the LEPR gene (LEPR03, LEPR09, and LEPR16) influenced ME fat content, with substitution effects ranging from 0.11 (P = 0.04) to 0.26 (P = 0.007); LEPR16 influenced also the test day fat content (0.38; P = 0.02). One SNP of LEP gene (LEP14) influenced the test day fat content (0.19; P = 0.02), and also showed a trend to influence ME fat content (0.10; P = 0.09). Furthermore, three SNP of the LEPR gene (LEPR01, LEPR09 and LEPR16) influenced protein content, with substitution effects ranging from 0.08 (P = 0.04) to 0.11 (P = 0.03) for ME protein content, and from 0.10 (P = 0.02) to 0.11 (P = 0.07) for the test day protein content.
Table 5. Allele substitution effect of the SNP in the LEP and LEPR genes on milk production traits: mature equivalent (ME) and test day.
Table 6. Allele substitution effect of the SNP in the LEP and LEPR genes on morphological indexes expressed in points (pts). The effect refers to the allele of column 2.
In Table 6, the results of the association analysis between 5 SNP of the LEP and LEPR genes and morphological indexes, expressed in points (pts) are shown. These traits were included in this study under the hypothesis that the genes that regulate the fat storage physiology, like the LEP and LEPR genes, might affect some morphological traits; similarly, Waters  quantified the associations of some SNP in the genes of the somatotrophic axis with milk performance and conformation traits in Holstein cows. Three SNP of the LEPR gene (LEPR01, LEPR03, and LEPR16) showed a significant effect on either the angularity or the stature or the rear legs (Table 6). Only one SNP of the LEP gene (LEP01) affected rump width and rear legs with values respectively of 4.3 pts and 5.3 pts. None of the analyzed SNP affected significantly either the rump angle or the udder depth. In Table 7 and Table 8, separately for the SNP of the LEPR and the LEP genes, in order not to ascribe to a SNP the effect of another one, for the SNP that showed some effect on any of the traits, pair-wise allelic correlations were reported.
These results showed that fat content is the most affected trait by the LEPR and LEP gene polymorphisms, while protein content is significantly affected only by EPR gene polymorphisms. As regards to the LEPR gene, two SNP, LEPR09 and LEPR16, influence both fat and protein (Table 5). LEPR09 encodes a missense mutation in exon 11 causing the amino acids substitution R615S at residue 615. The Arginine is a polar and basic amino acid while the Serine is polar and neutral; moreover, the substitution falls in the extracellular domain, five residuals before the starting of the second W-S-X-W-S motif. Reference  reported that the W-S-X-W-S motifs are involved in the ligand binding. It is then possible to justify the role played by LEPR09 in the variation of fat and protein content. Also LEPR16 affects both parameters, but because it encodes a synonymous mutation, it is likely that such effect be due to the high pair-wise correlations with LEPR09 (0.87; P < 0.0001; Table 7). LEPR03 influenced only fat content; this SNP encodes the missense mutation, in exon 2, causing the amino acid substitution L35V at residue 35, this substitution falls in the extracellular region of LEPR molecule very near the putative signal peptide region. The evidence that the considered SNP affect particularly milk fat is corroborated by the function of the two genes in lipid regulation; however, because the considered animal sample was relatively small, the results should be regarded as an indication of the possible association of these SNP with the considered traits, that need to be further confirmed in independent dairy populations.
The results of this work indicated that also LEPR01 affected significantly protein content but the effect, very similar to the effect of LEPR09, is likely due to the partial and highly significant correlation between the two variants (Table 7).
It is interesting to note that the majority of the authors, that have performed association studies on the LEPR gene, have analyzed only the SNP LEPR15 (rs133672995) which encodes the missense mutation Threonine to Metionine at the 944 residue of the protein sequence (accession no. NP_001012285). Reference  found that this SNP significantly influenced blood concentration of circulating leptin; however, no significant association of this SNP with milk-related traits was found [8,32], in agreement with our results.
Table 7. Allelic correlations between the pairs of SNP of the LEPR gene that were included in the association analysis.
Table 8. Allelic correlations between the pairs of SNP of the LEP gene that were included in the association analysis.
Of the SNP of the LEP gene, only LEP14 showed significant effects on some of the considered traits. On the contrary, a significant effect of LEP25 on total milk yield was proposed by , while  found a significant effect of LEP09 on fat yield, and of LEP11 on protein content. Reference  showed that LEP09 and LEP25 were significantly associated with milk fat and protein content, while LEP26 was associated only with the survival in the herd. Reference  reported a significant effect of LEP01 and LEP09 with milk protein content. Finally,  investigated the associations between the SNP in the LEP and LEPR genes, with growth, milk production, and fertility traits and found that LEP01 was significantly associated with height at withers and with the total number of AI services in heifers, while LEP26 was significantly associated with crown rump length, age at first service, at daily milk yield and 305-d milk yield.
It is interesting to note that LEP14 falls in the core sequence of an important binding site of transcription factors: NF-E2 (gGCTGActttc). The NF-E2 is involved in the regulation of globin gene transcription  and shows a high degree of homology to the Nrf2 transcription factor  that plays a role in lipid metabolism [36, 37]. In fact,  reported that the expression of many genes, that are involved in the synthesis and catabolism of cholesterol and fatty acids, is inversely related to the amount of Nrf2. The Nrf2 might also modulate leptin expression, by interfering with lipogenic pathways, so corroborating the positive effect of allele G of LEP14 on milk fat content.
As regards to the association between the SNP and the morphology, some SNP of the LEPR gene seemed to influence both milk and morphological traits. In fact, the trait angularity, that summarizes the ideal dairy cow type, is positively influenced by the A allele of LEPR16, that showed a positive effect on both fat and protein content (Table 5), confirming that milk traits are positively correlated with the morphological parameters desired for the dairy cow type.
As regards to the LEP gene, we noted that the desired morphological parameters for the dairy cow (body depth, rump width and rear legs) are positively influenced by LEP01 and LEP25 (Table 6), and these SNP are also highly correlated (Table 8).
In this study, for the first time, the bovine LEPR gene was fully characterized by sequencing the whole coding region and part of 5’flanking region and several novel SNP were detected, some of which encode missense mutations; moreover, a novel SNP was detected in the promoter of the LEP gene. In this study, we described 26 SNP of the LEP gene and 16 SNP of the LEPR gene and we reported significant associations of some of them with important milk production and morphological traits. Previous authors had aimed to estimate the LEP and LEPR gene effects on milk production traits, but because only one or few SNP were considered in each study, the results were contrasting and difficult to be extended to other cow populations. To our knowledge, this is the first study that evaluated the effect of many SNP of the LEPR gene on dairy traits. Due to relatively small size of the considered animal sample, the detected effects need to be further confirmed in independent dairy populations; however, because the allele substitution effects of the SNP of both the genes were evident, in most cases, on fat content, the results of this study confirm the role that both genes play in influencing this trait.
This study is part of the GENZOOT research program, funded by the Italian Ministry of Agriculture (Rome, Italy).
1Test day milk fat percentage, test day milk protein percentage.