Blood cells are receiving an increasing attention as an easily accessible source of transcript-based biomarkers. We studied the feasibility of using mouse whole blood RNA in this context. Several paradigms were studied: (i) metabolism-related transcripts known to be affected in rat tissues and peripheral blood mononuclear cells (PBMC) by fasting and upon the development of high fat diet (HFD)-induced overweight were assessed in whole blood RNA of fasted rats and mice and of HFD-fed mice; (ii) retinoic acid (RA)-responsive genes in tissues were assessed in whole blood RNA of control and RA-treated mice; (iii) lipid metabolism-related transcripts previously identified in PBMC as potential biomarkers of metabolic health in a rat model were assessed in whole blood in an independent model, namely retinoblastoma haploinsufficient (Rb+/-) mice. Blood was collected and stored in RNAlater® at -80°C until analysis of selected transcripts by real-time RT-PCR. Comparable changes with fasting were detected in the expression of lipid metabolism-related genes when RNA from either PBMC or whole blood of rats or mice was used. HFD-induced excess body weight and fat mass associated with expected changes in the expression of metabolism-related genes in whole blood of mice. Changes in gene expression in whole blood of RA-treated mice reproduced known transcriptional actions of RA in hepatocytes and adipocytes. Reduced expression of Fasn, Lrp1, Rxrb and Sorl1 could be validated as early biomarkers of metabolic health in young Rb+/- mice using whole blood RNA. Altogether, these results support the use of whole blood RNA in studies aimed at identifying blood transcript-based biomarkers of nutritional/metabolic status or metabolic health. Results also support reduced expression of Fasn, Lrp1, Rxrb and Sorl1 in blood cells at young age as potential biomarkers of metabolic robustness.