Despite massive research efforts, the molecular etiology of bovine polledness and the developmental pathways involved in horn ontogenesis are still poorly understood. In a recent article, we provided evidence for the existence of at least two different alleles at the Polled locus and identified candidate mutations for each of them. None of these mutations was located in known coding or regulatory regions, thus adding to the complexity of understanding the molecular basis of polledness. We confirm previous results here and exhaustively identify the causative mutation for the Celtic allele (PC) and four candidate mutations for the Friesian allele (PF). We describe a previously unreported eyelash-and-eyelid phenotype associated with regular polledness, and present unique histological and gene expression data on bovine horn bud differentiation in fetuses affected by three different horn defect syndromes, as well as in wild-type controls. We propose the ectopic expression of a lincRNA in PC/p horn buds as a probable cause of horn bud agenesis. In addition, we provide evidence for an involvement of OLIG2, FOXL2 and RXFP2 in horn bud differentiation, and draw a first link between bovine, ovine and caprine Polled loci. Our results represent a first and important step in understanding the genetic pathways and key process involved in horn bud differentiation in Bovidae.

The large spectrum of limb morphologies reflects the wide evolutionary diversification of the basic pentadactyl pattern in tetrapods. In even-toed ungulates (artiodactyls, including cattle), limbs are adapted for running as a consequence of progressive reduction of their distal skeleton to symmetrical and elongated middle digits with hoofed phalanges. Here we analyse bovine embryos to establish that polarized gene expression is progressively lost during limb development in comparison to the mouse. Notably, the transcriptional upregulation of the Ptch1 gene, which encodes a Sonic hedgehog (SHH) receptor, is disrupted specifically in the bovine limb bud mesenchyme. This is due to evolutionary alteration of a Ptch1 cis-regulatory module, which no longer responds to graded SHH signalling during bovine handplate development. Our study provides a molecular explanation for the loss of digit asymmetry in bovine limb buds and suggests that modifications affecting the Ptch1 cis-regulatory landscape have contributed to evolutionary diversification of artiodactyl limbs.

In cattle, the embryo production rate after superovulation varies between individuals and is difficult to predict. Recently, we proposed that anti-Müllerian hormone (AMH) plasma levels measured before treatment can help predict superovulatory responses. To establish whether blood measurement of AMH can help predict the number of embryos produced by a given cow after superovulation, data collected over 4 years from 45 dairy cows submitted to repeated embryo production were analysed in a retrospective study. A high within-animal repeatability (0.38 and 0.36) and a strong effect of the father of the donor cow (P < 0.01) were observed for the numbers of collected and transferable embryos, respectively. AMH concentration, measured in the plasma of donor cows during first lactation and several months before the start of the embryo production campaigns, was found to be highly correlated with the maximal number of collected (P < 0.0001) and transferable (P < 0.01) embryos per cow. In conclusion, the capacity of embryo production is a repeatable and probably heritable trait in the cow, and blood measurement of AMH in potential donor cows could be of value in determining a cow's intrinsic capacity to produce transferable embryos.

Reproductive performance has decreased over the last decades in many dairy systems. This study aimed at comparing the effects of a high-and a low-feeding level on reproductive stages (cyclicity, oestrus and fertility) of Holstein and Normande cows in a grass-based winter compact calving system. High-fed cows received a total mixed ration composed of 55% maize silage, 15% dehydrated alfalfa pellets and 30% concentrate in winter and 4 kg/day concentrate supply at grazing. Low-fed cows only received 50% grass silage and 50% haylage in winter and no concentrate at grazing. Low-fed cows produced less milk over 44 weeks, but lost more body condition (BC) than high-fed ones (5207 v. 7457 kg, 21.28 v. 20.96 unit, P , 0.001). Normande cows produced less milk and lost less BC than Holstein ones (5596 v. 7068 kg, 20.89 v. 21.36 unit, P , 0.001). Post-partum ovarian activity was little affected by the feeding level. In both breeds, ovulation detection rate was higher in low-fed cows (74% v. 59%, P , 0.001) and detection relied more frequently on standing behaviour (67% v. 55%, P 5 0.022). In both breeds, the recalving rate following the first and second inseminations was not significantly affected by the feeding level, although more non-fertilisations or early embryo mortalities occurred in low-fed cows (43% v. 26%, P 5 0.004). For the Holstein breed, this was clearly explained by more late embryo mortalities in high-fed cows (30% v. 9%, P 5 0.004). Finally, pregnancy rate by the end of the 13-week breeding period was similar between feeding groups. Conversely, the Normande dual-purpose cows had a higher pregnancy rate by the end of the breeding period than the Holstein dairy cows (72% v. 54%, P 5 0.007), owing to a better ovarian activity (79% v. 54% normal, P , 0.001) and a higher recalving rate following insemination (53% v. 37%, P 5 0.007). In conclusion, this study demonstrates that feeding levels with converse effects on milk yield and BC score also have converse effects on reproductive stages and lead to quite similar final reproductive performance in compact calving systems. Normande cows benefit from an overall better reproductive performance, but do not suit high feeding levels for very compact calving systems owing to depressed oestrous behaviour.

Mammalian pigmentation is controlled by the concerted action of Tyr, Tyrp1 and Dct producing eumelanin and/or pheomelanin in melanocytes. The ratio of these two pigments is determined by the agonist alpha-melanocyte stimulating hormone and the antagonist Agouti protein acting on the Mc1r. Here we show that the Agouti gene is over-expressed in Normande breed compared with Prim'Holstein breed. The Normande cattle have a characteristic coat color phenotype with a variable presence of black (eumelanin) hair over a red/brown background. We have found a previously undescribed full-length L1-BT element inserted in the 5'-genomic sequence of the Agouti gene in Normande cattle which promotes the over-expression of alternative transcripts. The variable expression of the alternative transcript directed by the long interspersed nuclear element promoter may be the origin of the brindle coat color pattern of the Normande breed. This new bovine Agouti allele isolated in Normande breed has been named Abr. Finally, as ectopic over-expression of Agouti in Ay mice is responsible for the obesity syndrome, we discuss the possible consequences of Abr for meat and milk production in cattle.