No abstract

This report, the first in a series on diet-dependent changes in the serum metabolome (metabolic serotype), describes validation of the use of high performance liquid chromatography (HPLC) separations coupled with Coulometric array detectors to characterize changes in the metabolome. The long-term aim of these studies is to improve understanding of the effects of significant variation in nutritive status on physiology and on disease processes. Initial studies focus on identifying the effects of dietary (or caloric) restriction on the redox-active components of rat serum. Identification of compounds of interest is being carried out using HPLC separations coupled with coulometric array analysis, an approach allowing simultaneous examination of nearly 1200 serum compounds. The technical and practical issues discussed in this report are related to both analytical validity (HPLC running conditions, computer-automated peak identification, mathematical compensation for chromatographic drift, etc.) and biological variability (individual variability, cohort-cohort variability, outliers). Attention to these issues suggests approximately 250 compounds in serum are sufficiently reliable, both analytically and biologically, for potential use in building mathematical models of serotype.

Our research seeks to identify serum profiles, or serotypes, that reflect substantial changes in food intake in both male and female rats. This report validates previously defined subsets of redox-active low-molecular-weight metabolites using independent cohorts of ad libitum consumption (AL) and energy or dietary restricted (DR) 6-mo-old male and female rats. In the male study, both hierarchical cluster analysis (HCA) and principal component analysis (PCA) distinguished the dietary groups of origin in the second male cohort with >85% accuracy using 56 analytically and biologically valid metabolites. Further analysis revealed that 29 metabolites (nine previously unidentified metabolites + 20 chosen from the 56 metabolites) enabled HCA to distinguish dietary groups at 100% efficacy. In the female study, the 63 previously identified serum metabolites were sufficiently robust to enable classification of the dietary intake of two female cohorts (cohorts 2 and 3) that were independent of the cohort on which these markers were initially identified (cohort 1). Classification accuracy was 94 and 100% using HCA and PCA, respectively, in the female cohort 2. HCA and PCA revealed that the 63-metabolite profile distinguished AL and DR samples at 91 and 100% accuracy in the female cohort 3, establishing the 63-metabolite dataset as our baseline profile. These studies used independent cohorts to validate and potentially improve upon previously defined metabolic serotype in male and female rats and set the stage for pattern recognition-based approaches to establish metabolome-based categorical separations.

The aeration-dependent changes in content of various quinones in Escherichia coli were found to be unaffected by a prokaryotic translation inhibitor chloramphenicol. In addition, this process was shown to be completely intact in cells with mutated fnr, arc and appY loci. It is assumed that E. coli possesses a special system of oxygen-dependent post-transcriptional regulation of the quinone biosynthetic pathways.

Induction of the mitochondrial permeability transition (PT) has been proposed to contribute to neuronal cell death. Nearly all studies of the biochemistry of PT induction, however, have been conducted in isolated liver mitochondria. To better understand PT induction in brain mitochondria, we used Ficoll gradients to purify nonsynaptosomal mitochondria from the forebrains of male Fischer 344 rats. Incubation of these mitochondria with Ca²⁺ was associated with a loss of absorbance. Inorganic phosphate enhanced this loss of absorbance, and the PT inhibitor cyclosporin A reduced it, especially in conjunction with ADP. These findings suggest that Ca²⁺-mediated loss of absorbance resulted from PT induction. Na⁺, which enhances mitochondrial Ca²⁺ efflux, but stimulates mitochondrial free radical production, had no effect on PT induction. These data confirm the existence of tissue-specific differences in the nature of PT induction.