Bovine hepatic and adipose retinol-binding protein gene expression and relationship with tumor necrosis factor-α

Retinol-binding protein (RBP) is the main transport system for retinol in circulation, is a relatively small protein with one binding site for retinol in the all-trans form, and is bound to transthyretin. The objectives of this study were to characterize the temporal pattern of bovine hepatic mRNA expression of RBP during the periparturient period and to determine if a relationship exists between the expression of RBP and that of tumor necrosis factor (TNF)-α in dairy cows. In experiment 1, we assessed hepatic mRNA expression of RBP during the periparturient period. Liver tissues were sampled from periparturient dairy cows (n=9) at -21, -4, +1, +7, and +21 relative to parturition and frozen in liquid N(2). Total RNA was extracted from each tissue sample and cDNA was generated. Gene expressions of RBP and β-actin (as a housekeeping gene) were measured as relative quantity using reverse transcription-PCR. Data were analyzed using cycle threshold values, adjusted to β-actin, and significance was determined at P<0.05. Serum samples (-21, -4, +1, +7, and +21 relative to parturition) were analyzed for retinol concentration using a standard HPLC-based method. Cows had variable expression of hepatic RBP and serum retinol over the transition period, with a decline near parturition and a rebound toward prepartum levels later in lactation. In experiment 2, liver and visceral (intestinal) adipose tissues were sampled from dairy cows (n=28) at slaughter. Expression of RBP and TNF-α was detected in all samples and variations among cows were highly significant for both genes. Across tissues, expression of RBP was positively correlated with that of TNF-α (r=0.60). Within adipose tissue, expression of RBP and TNF-α was weakly correlated (r=0.23), whereas in hepatic tissue, expression was strongly correlated (r=0.62). In experiment 3, late-lactation dairy Holstein cows were blocked by parity and feed intake, and randomly assigned to control, recombinant bovine (rb)TNF challenge, or pair-fed control treatment (n=5/treatment). Cows were injected with either rbTNF (subcutaneous injection of 2 µg/kg of body weight in saline) or sterile saline (control and pair-fed control animals) once daily for 7d. Liver biopsy was performed on d 7 and samples were processed for expression of RBP and TNF-α. Although TNF challenge caused an upregulation of hepatic TNF-α expression, as expected, it did not alter hepatic RBP expression. Overall, the temporal pattern of hepatic RBP gene expression during the periparturient period followed, to a great extent, that of plasma retinol. Although a strong positive correlation was previously detected between bovine hepatic RBP and TNF-α transcripts, rbTNF challenge did not cause alter RBP expression. These observations collectively imply that regulation of RBP at the transcription level is influenced by physiological state but may be independent from that of transthyretin, which is altered by proinflammatory stimuli (such as TNF-α) via induction of transcription factor nuclear factor-interleukin 6.