Metabolomic Assays of Postmortem Brain Extracts: Pitfalls in Extrapolation of Concentrations of Glucose and Amino Acids to Metabolic Dysregulation In Vivo in Neurological Diseases.

Abstract

Glucose utilization is reduced in vulnerable brain regions affected by neurological disorders, especially Alzheimer's disease (AD), but the basis for abnormal glucose homeostasis is unknown. Studies of brain-bank human tissue have made major contributions to understanding complex aspects of neurological, psychiatric, and neurodegenerative diseases, but they are not appropriate for metabolomic analysis of labile metabolites because postmortem intervals between death and tissue freezing are much too long. Recent reports of postmortem brain glucose levels led to suggestions that AD patients may be hyperglycemic and that elevated brain glucose levels along with reduced glycolytic activity reveal abnormal glucose homeostasis before clinical symptoms become manifest. These conclusions are, however, questioned because virtually all brain glucose is consumed within minutes after death, followed by progressive increases in glucose and amino acid levels, presumably due to autolytic changes. To illustrate pitfalls in use of autopsy material for metabolomic assays of labile metabolites, data from living human brain are compared with those from autopsy samples, and metabolism at the onset of postmortem ischemia is compared with calculated glycolytic enzyme activities. Postmortem glucose levels range from extremely low to unrealistically high, precluding their extrapolation to living brain. Indirect evaluation of glycolytic enzyme activities in postmortem AD brain is not valid because the glucose and amino acid concentrations used in the calculations are not stable after death, and reported values are unrealistically high. Specific recommendations are provided for non-invasive longitudinal monitoring of brain metabolism and metabolite levels in patients with neurological diseases.