This study was performed to investigate the hypothesis that the onset of lactation in dairy cows leads to an up-regulation of genes involved in hepatic carnitine synthesis and uptake of carnitine. As expected, the transition from late pregnancy to early lactation was associated with a strong negative energy balance resulting in increased concentrations of NEFA and BHBA in plasma and an increase in hepatic TAG concentration. Similar metabolic changes during the periparturient period in cows have been observed in many other studies [e.g. [16, 18, 29]]. In agreement with a recent study , we observed that the negative energy balance occurring at early lactation was associated with an increased expression of several PPARA target genes involved in fatty acid uptake, mitochondrial and peroxisomal fatty acid oxidation and ketogenesis. Although we were not able to give direct proof of PPARA activation due to small liver sample amount available, an up-regulation of various PPARA target genes is indicative of an activation of that transcription factor in the liver. While there is less research about activation of PPARA in cattle, studies in other species such as rodents or pigs have clearly shown that increased plasma NEFA concentrations, induced by energy deprivation, are leading to an activation of PPARA in liver and other tissues [12, 13, 30]. As long-chain fatty acids are also acting as agonists of PPARA in bovine cells , it seems justified to speculate that high plasma NEFA concentrations in dairy cows during early lactation were causing an activation of PPARA in the liver. It should be noted, however, that in opposite to our study and the study of Loor et al. , there are also studies which did not observe an up-regulation of PPARA and PPARA target genes in the liver of dairy cattle during early lactation, particularly when cows had prepartum a caloric intake in excess of 100% of their energy requirement [31, 32]. The lack of up-regulation of PPARA during early lactation in these studies has been explained by a hepatic inflammatory response, induced by an excessive prepartum caloric intake, which decreased pre- and postpartum hepatic expression of expression of PPARA .
In accordance with the hypothesis of this study, we observed for the first time that the transition from late pregnancy to early lactation leads to an up-regulation of various genes involved in carnitine synthesis (ALDH9A1, TMLHE, BBOX1) and carnitine uptake (SLC22A5) in the liver of cows at 1 wk postpartum. As all these genes are PPARA target genes with functional PPREs identified in their promoters or first introns [32–35], we assume that the up-regulation of these genes in the liver of dairy cows at 1 wk postpartum was caused by a potential activation of PPARA. The finding of positive correlations between plasma concentrations of NEFA, which might be regarded as natural agonists, and hepatic carnitine concentrations during lactation supports a role of PPARA in the regulation of genes of carnitine synthesis and uptake.
The present study confirms previous studies in showing that liver carnitine concentration is increasing during the transition from late pregnancy to lactation and is thereafter continuously decreasing to values similar or even below those observed in pregnancy [18, 19]. In the study of Carlson et al. , hepatic total carnitine concentrations were around 1.6-fold higher on 2 d of lactation compared to 3 wk prepartum, while values at d 28 were even lower than those 3 wk prepartum. In the study of Grum et al. , concentrations of acid-soluble carnitine (free carnitine plus short-chain acyl carnitine) sharply increased from 3 wk prepartum to 1 d postpartum and returned to prepartum values at d 21 postpartum. As carnitine is synthesized from trimethyllysine released primarily from turnover of skeletal muscle proteins, Grum et al.  suggested that an increased hepatic carnitine concentration at 1 d postpartum might be due to an enhanced catabolism of muscle protein in this early stage of lactation. Although we have not directly measured carnitine synthesis and uptake into cells, the findings of increased mRNA expression of genes of carnitine synthesis and uptake suggest that increased hepatic carnitine concentrations at early lactation could also be due, at least in part, to an increased carnitine synthesis in the liver and an increased uptake of carnitine from blood into the liver. This suggestion is supported by studies in rodents and pigs which found that an up-regulation of enzymes of carnitine synthesis in the liver, caused either by treatment with PPARA agonists or by energy deprivation, leads to an increased hepatic carnitine concentration without changing the concentrations of BB or TML, the precursors of carnitine synthesis [7, 8, 36, 37]. The finding that the concentration of free carnitine in plasma is strongly decreasing from 3 wk prepartum to 1 wk postpartum fits into this suggestion. Studies in rodents have shown that an up-regulation of SLC22A5 by treatment with PPARA agonists or by energy deprivation leads to a reduction of plasma carnitine concentration, due to an increased transport of carnitine from plasma into tissues [6, 8, 36]. The decrease in plasma carnitine concentrations during early lactation might be in part explained by the transfer of carnitine from plasma into the milk in mammary gland. Nevertheless, with respect to the findings in rodents, we assume that reduced plasma carnitine concentrations in early lactation could also be caused by an increased uptake of free carnitine from plasma into tissues, including the liver. In accordance with present study, Carlson et al.  also observed a strong reduction of plasma carnitine concentration from 21 d prepartum to 9 or 27 d postpartum. Those authors  also found that carnitine concentration in skeletal muscle is not changing significantly during transition from pregnancy into lactation. This finding also agrees with studies in rodents which show that energy deprivation does not influence muscle carnitine concentrations [7, 36].
Furthermore, we observed that milk carnitine concentration is highest at 1 wk postpartum and is thereafter decreasing to 5 wk and 14 wk postpartum. This finding agrees with the study of Carlson et al.  which found a strong reduction of milk carnitine concentration from 2 wk to 6 wk postpartum. It is possible that the high carnitine concentration in milk at wk 1 postpartum is due to the strong negative energy balance of the cows. Carlson et al.  found that energy restriction of cows increased milk carnitine concentrations. In the mammary gland, carnitine is secreted into the milk by several transporters (SLC22A4, SLC22A5, OCTN3, SLC6A14, SLC6A10) . Possibly, one or more of these transporters are up-regulated in a state of a negative energy balance. Studies in rats also found a reduction of milk carnitine concentration from early to later stage of lactation, due to a down-regulation of SLC22A5, OCTN3 and SLC6A14 .
Ketosis and fatty liver are two diseases in dairy cows during early lactation which are linked to hepatic fatty acid oxidation . As carnitine is involved in β-oxidation due to its role in the transport of long chain fatty acids into the mitochondrion, it was interesting to explore whether plasma ketone body concentrations or hepatic TAG concentrations are correlated with hepatic carnitine concentrations. The observation that there were no correlations between hepatic carnitine concentration and both, plasma BHBA and hepatic TAG concentrations, at 1, 5 and 14 wk postpartum suggests that the availability of carnitine in the liver had no influence on ketogenesis and TAG accumulation in the liver. The observation that hepatic carnitine concentration does not correlate with BHBA concentration is in accordance with a study showing that carnitine supplementation does not influence plasma BHBA concentration in early lactating dairy cattle . The finding that the activity of CPT1A does not play a primary role in the etiology of ketosis  is another indication that the availability of carnitine, which acts as a cofactor of that enzyme, is not a key factor in the production of ketone bodies. Accumulation of TAG in the liver is explained by the observation that the capacity of bovine liver tissue to convert fatty acids to esterified products is strongly increased during the early postnatal period, whereas fatty acid oxidation is only slightly increased, meaning that NEFA from mobilization are directed towards conversion to TAG . The finding that there was no correlation between hepatic carnitine concentration and hepatic TAG concentration thus indicates that a higher hepatic carnitine concentration did not stimulate hepatic fatty acid oxidation. This indication is, however, in contradiction to some in vitro and in vivo studies in dairy cows. In vitro studies using bovine liver slices have shown that addition of carnitine enhances the oxidation of palmitate [44, 45]. Moreover, postruminal infusion of carnitine enhanced palmitate oxidation and decreased liver lipid accumulation in cows with experimentally induced negative energy balance . These studies suggested that carnitine might be the rate-limiting factor of hepatic β-oxidation in dairy cows during the periparturient period and that carnitine supplementation might prevent the development of a fatty liver.