Five SNPs on BTA14 were identified as associated with BW in Nellore cattle (P < 1.15 × 10-7), whose surrounding region has been shown to contain many QTLs, genes and variants affecting stature-related traits in cattle by several independent studies [12–15, 21–23]. More particularly, the genes PLAG1, CHCHD7, RDHE2, MOS, RPS20, LYN and PENK have been found to influence both human and cattle height [22–28].
The BTA14 region pointed out by the present study has also been shown to be associated with reproductive traits. Cole et al.  reported a QTL on BTA14 associated with stillbirth, which also has been associated with body size in dairy cattle [8, 30], but found no effect on stature or other conformation traits on that chromosome. The region also associates with many fertility and growth-related traits in the indicine breed Brahman, for example scrotal circumference [31, 32], age at the first corpus luteum[31, 33], blood levels of insulin-like growth factor 1 (IGF1) [32, 33] and hip height .
A significant SNP was found within intron 2 of the XKR4 gene in the present study. Lindholm-Perry et al.  identified five SNPs near XKR4 associated with feed intake and gain in crossbred steers. Bolormaa et al.  found five SNPs in a narrow region of BTA14 encompassing XKR4 associated with rump fat thickness measured at the P8 position (CHILLP8) in seven breeds of cattle, including taurine, indicine and composite breeds. The authors found that four of these SNPs were also associated with CHILLP8 in a confirmatory sample of 1,338 animals, including Angus, Hereford and Brahman cattle. Furthermore, Porto Neto et al.  performed a replication study using samples of Belmont Red, Santa Gertrudis and Brahman animals genotyped for SNPs within XKR4 and found that although the SNP effect may vary depending on the breed, the variant rs42646708 (BTA14:24573257) explain around 1.3% of CHILLP8 variance in cattle. This SNP is also located within intron 2 of XKR4, only 17.6 kb apart from the intronic SNP detected in the present study, which strongly suggests XKR4 as a candidate gene for being further explored in future studies of weight and carcass traits in Nellore cattle.
The most significant SNP (rs133012258, P
GC = 1.34 × 10-9) was found to explain 4.62% of the variance in sires EBVs, with a 95% CI of 2.12-8.09%. One hundred and eighty loci associated with human adult height explain only 10% of the phenotypic variance together, while individual loci account for 0.4% or less . SNPs analyzed by  within a nearby BTA14 region explain from 0.29 to 2.53% of the bovine stature variability, and the quantitative trait nucleotides (QTN) spanning MOS, CHCHD7 and PLAG1 described by  explain from 1.10 to 3.50% of height in Jersey and Holstein breeds. Furthermore, the genome-wide survey performed by  provided strong evidence for two QTL on BTA14 and BTA21 that together explain at least 10% of the variation of EBVs for calving ease in the German Fleckvieh.
Considering that multiple stature-related traits are governed by variants with small effects, and that the genomic region identified in this study has been previously found to be associated with several of these traits, the putative SNP detected in the present analysis can be considered as a marker in linkage disequilibrium (LD) with major untyped (i.e., not probed by the SNP assay used) causative variants affecting BW and other height-associated traits in Nellore cattle, and further studies would be needed to determine if the QTNs reported by  are also segregating in the Nellore population. Also, future investigations are needed to better characterize the effect of nearby SNPs on other weight and carcass traits in Nellore cattle, as it is not clear yet how the putative pleiotropic effect of these variants would be used towards balancing conflicting selection goals for birth, weaning and yearling weights. Although we cannot confirm that the allele substitution effects of these SNPs work in the same direction for all three traits, because only birth weight was analyzed here, these findings suggest that the SNPs identified would be key polymorphisms to be monitored over time. In a scenario where the SNP effects have the same direction in all three traits, one strategy could be avoiding strong positive selection or drifting of the allele that contributes to higher BW EBVs, and identify and promote positive selection of other variants that have effects on weaning and yearling weights only.
The high identity found in the alignment of this BTA14 region against other mammalian species genomes suggests that these orthologous genes are located in a conserved syntenic block which may have arisen and been maintained after speciation from a common ancestor of the mammal clade. Moreover, the evidence for variants associated with growth and stature within this BTA14 region in both taurine and zebu cattle raises two hypotheses: 1) these variants have been introgressed into Nellore via historical admixture with taurine Creole cattle in the maternal line, and was maintained in the breed in spite of several generations of backcrossing; 2) these are ancient polymorphisms, probably already segregating in the founder population of wild Aurochs (Bos primigenius) before subspecies formation.
Regarding functional meaning, the set of genes reported participate in diverse growth and tumor development mechanisms. Among these genes, PLAG1 is the most appealing functional candidate. It is an oncogene that encodes a transcription factor broadly expressed during fetal development, but is down-regulated at birth . It interacts with several growth factors controlling body size, including IGF2. In addition, PLAG1 knock-out mice have been shown to have marked growth retardation and reduced fertility . In a replication study,  confirmed the findings reported by , demonstrating association of growth rate and early life and peripubertal body weight with PLAG1 polymorphisms, supporting its status as a key regulator of mammalian growth.
The lack of significant association between BW and SNPs within other previously described weight- and height-related chromosome regions in the present study should not be interpreted as a lack of existence of true association, but rather it might be due to limitations specific to this study. Firstly, because complex trait mapping requires large sample sizes and only 649 bulls were analyzed here. Secondly, the significance level adopted was highly stringent, which may have caused inflation of type II errors. In spite of these limitations, it was possible to demonstrate that a well-characterized chromosome region affecting human and taurine cattle stature also associates with BW in a zebu breed. The release of a Bos primigenius indicus reference genome assembly, as well as the application of re-sequencing and replication studies would help improve resolution to narrow down the genomic region as close as possible to the true causative variants.