RESUMO
Liver performs several important functions; however, predicting its functions is difficult. Methods of analyzing gene expression profiles, for example, microarray, provide functional information of tissues. Liver and peripheral blood leukocytes (PBLs) were collected from Holstein cows subjected to two different physiological conditions (non-pregnant and pregnant), and PBLs were fractionated by gradient cell separation. RNA from PBLs and liver were applied to oligo-DNA microarray and reverse transcription-quantitative polymerase chain reaction (RT-qPCR). It revealed a group of stable bovine liver genes under constant physiological conditions. When they applied to physiological conditions including non-pregnant and pregnant, the profiles of some genes in liver were consistent with those in PBLs. Microarray data subjected to a principal component analysis (PCA) showed that the hepatic gene expression profiles were more consistent with those of granulocytes than mononuclear cells. The relationship of gene profiles in liver with granulocytes was confirmed by RT-qPCR and hierarchical cluster analysis. Gene profiles of granulocytes were more reliable than those of mononuclear cells, which reflected liver functions. These results suggest that the genes expressed in PBLs, particularly granulocytes, may be convenient bioindicators for the diagnosis of clinical disorder and/or detecting aberration of liver functions in cows subjected to different physiological conditions.