The GH and IGF-I allele frequencies in this study are in agreement with those reported previously in Holstein-Friesian for GH [18, 19, 27] and in Holstein for IGF-I [29–31].
There was no significant association between GH genotype and productive parameters (milk, 4%FCM and total solid yields) in accordance with Yao et al.  in Holstein bulls and Balogh et al.  in Holstein-Frisian cows. However, a trend was found for the interaction of GH genotype and dpp on FCM yield in farm 2 (multiparous cows), as LL cows produced more than LV cows during early lactation. Similarly, Shariflou et al.  suggested that the L allele appeared to have an additive effect on milk production only at the beginning of the lactation. Besides, Lucy et al.  reported that cows carrying LL genotype yielded more milk, fat, and protein than LV cows. No effect of the interaction between GH genotype and dpp was found in primiparous cows (farm 1), and this could be associated with the level of production and/or a differential role of GH genotype in growing animals. In contrast, Dybus et al.  determined an effect of GH genotype on milk, fat and protein yield in primiparous but not in multiparous cows, and they suggested that the observed differences could have resulted from another source of variation (e.g. effects of herd, sires) not considered in the study.
The IGF-I genotype tended to affect FCM yield in farm 2 (multiparous cows), and the interaction between IGF-I genotype and dpp tended to affect FCM yield in farm 1 (primiparous cows), as AB cows yielded more FCM than BB (farm 1) or AA (farm 2) cows. Similarly, Siadkowska et al.  reported that AB cows yielded more FCM than AA and BB cows. In contrast, Hines et al.  did not find any effect of IGF-I genotype on productive parameters. In our study, cows were under grazing conditions and were average producing cows (17 L/day in farm 1 and 22 L/day in farm 2). Previous studies that could not find any effect of the genotype on milk production stated that genotype differences might not be expressed at this level of production [38, 27]. In addition to this, Chilibroste et al.  and Kolver and Muller  reported that DM intake is not enough to achieve the genetic potential on grazing milk production systems.
In our study there was no effect of GH genotype on reproductive parameters in none of the farms. Balogh et al.  found no effect of the GH genotype on the time of the first pospartum ovulation. Lechniak et al.  reported a tendency for greater non-return rates of VV beef bulls at 60 dpp and Lechniak et al.  found no effect of the GH genotype on oocyte number. No data as such has been found for the relationship between IGF-I genotype and reproduction in dairy cows. For IGF-I genotype there was a significant effect on calving-first service interval only in farm 1, as BB cows had a longer interval than AB cows.
We found only one report regarding the effects of GH genotype  and none of IGF-I genotype on metabolic and/or endocrine profiles in the transition dairy cow. Balogh et al.  did not found either an effect of GH genotype on plasma BHB, insulin, and IGF-I concentrations, but they performed only one postpartum determination (4 to 13 dpp). In the present study we have included pre and postpartum determinations which in our understanding, allowed a better comprehension of the metabolic endocrinology during the peripartum period.
In farm 1 (primiparous cows), NEFA and IGF-I concentrations were affected by GH genotype, as LL cows had lower NEFA and greater IGF-I concentrations than LV cows. Since NEFA and IGF-I are both indicators of the metabolic status , these data suggest that LL cows presented a better energy status than LV cows. It is supposed that bovine GH with Leu127 stimulate the release of IGF-I more than other variants of bGH  which is consistent with the results found in the present study. In contrast, Schlee et al.  observed that Simmental LV bulls presented greater IGF-I concentrations. This differential metabolic/endocrine environment was not reflected on productive/reproductive traits, which could be due to the level of production of primiparous cows as discussed before and/or to the extra energy demands for growth in these cows. In farm 2 (multiparous cows) there was no effect of GH genotype on any of the metabolites and hormones studied. Although a reduced number of animals were included in this farm, there are more metabolic/endocrine time measurements which allowed a better metabolic description of the NEB. Indeed, NEFA and β-hydroxibutirate concentrations increased around calving reflecting fat mobilization as reported before [1–3]. As expected, insulin concentrations decreased around calving as has previously been observed . This decrease in plasma insulin is a metabolic adaptation to cope with the energy demands of lactation as reported earlier [42, 43], since low insulin concentrations favours gluconeogenesis and lipolysis  (e.g. homeorhetical effect). The decrease in IGF-I concentrations at calving confirmed the uncoupled somatotropic axis (GH-IGF-I), which mediates nutrient partitioning for lactogenesis . We have no obvious explanation for the differential effect of GH genotype on metabolic/endocrine profiles found in primiparous cows (farm 1) vs. multiparous cows (farm 2), but as previously suggested it could be due to the differential role of GH during growth and development. Indeed, primiparous cows present greater insulin and IGF-I concentrations than multiparous cows .
Insulin-like growth factor-I genotype affected NEFA and BHB concentrations in farm 1 (primiparous cows); as BB cows had lower concentrations than AA and AB cows. In accordance to the better energy balance in BB cows reflected by these metabolites, these cows presented greater insulin concentrations at 30 dpp. Unexpectedly, IGF-I concentrations were not affected by IGF-I genotype. Ge et al.  and Maj et al.  reported lower or greater IGF-I blood concentrations in BB young Angus cattle or BB Holstein -Friesian young bulls and heifers, respectively. Since this polymorphism is located in the promoter region of the IGF-I gene, a variety of responses in gene expression may result depending on the physiological and/or nutritional status of the animal. This differential energy balance is not consistent with the calving first service interval, since it is known that cows in a better energy balance have also better reproductive performance , and in this study BB cows presented longer calving-first service interval than AB cows (103 vs 73 days, respectively). Unfortunately, we do not have the endocrine/metabolic profiles at the time of the initiation of the services which could clarify these contradictory results; indeed the endocrine system changes dynamically according to the nutritional and productive status. Moreover, BB cows had not only a reduced reproductive performance but they tended to yield less FCM at 120 and 210 dpp than AB cows. Staples and Thatcher  reported that cows with more DM intake, present not only greater milk production, but also better reproductive performance. In farm 2 (multiparous cows), there was no effect of IGF-I genotype on plasma NEFA, BHB, and insulin concentrations. On the other hand, IGF-I profiles suggest a greater uncoupling of the somatotropic axis in AB and BB cows than AA cows which is in accordance with the greater FCM yield of AB than AA cows.