D‐Amino acids in the brain: pyridoxal phosphate‐dependent amino acid racemases and the physiology of d‐serine

Abstract

Many millions of years of evolution have resulted in the virtually complete homochirality of amino acids in mammals. Not only are the amino acids that form the building blocks almost exclusively in their l-isomeric form, but free amino acids seldom appear as d-isomers. However, it has been known for a long time that d-amino acids, principally d-alanine and d-glutamic acid, are present in bacteria where they are used as building blocks for the peptidoglycan of the cell wall. Almost 60 years ago, in 1948, Sir Hans Krebs published the discovery of an enzyme that he had somewhat accidentally isolated from kidney that specifically degraded d-amino acids (but not their l counterparts). He termed this enzyme d-amino acid oxidase. However, the real function of this enzyme in vivo remained unclear. Some years later, another flavoenzyme, d-aspartate oxidase, was also discovered in mammals, and its specificity was shown to be focused on dicarboxylic d-amino acids. The identification of free d-amino acids in several tissues and in the blood of mammals finally provided a rationale for the widespread occurrence of these oxidases. Initially, the mammalian d-amino acids were thought to arise from endogenous microbial flora, from ingestion in the diet, or from spontaneous racemization of l-amino acids incorporated into polypeptides during aging. With the improvement in HPLC techniques, the unambiguous detection of d-amino acids has been possible. The three most abundant d-amino acids in mammalian brain are d-alanine, d-serine and d-aspartic acid. For instance, d-alanine has been found in rat brain as well as in the pituitary gland. d-Serine is also present in the nervous system in high concentrations, mostly located in the grey matter, in the hippocampus, in the anterior olfactory nucleus and in the amygdala. High levels of d-aspartic acid also occur in the brain and endocrine glands, such as pineal, adrenal and pituitary. The most important breakthrough in the field is that of Wolosker, Snyder and coworkers who, using dozens of rat brains, were able to isolate an enzymatic activity that could convert l-serine into d-serine. Elegant experiments allowed them to clone the structural gene and demonstrate that the enzyme was a pyridoxal phosphate (PLP)-containing racemase. This mammalian serine racemase could increase the intersynaptic levels of d-serine, a molecule that used the so-called glycine site within the N-methyl d-aspartate (NMDA) receptor. Less is known about the enzyme(s) responsible for the synthesis of d-aspartic acid or d-alanine. The observation that amino-oxyacetic acid, a potent inhibitor of PLP-dependent enzymes, could abrogate the release of d-aspartic acid in primary neuronal cultures indicates that aspartate racemase might also be a PLP-containing enzyme. So far the identification of the racemases responsible for the synthesis of d-aspartic acid and d-alanine has been elusive. In these three reviews we initially analyze the physiology of d-serine in the brain, notably the localization and regulation of brain serine racemase, both in neurotransmission and in neurodegeneration. In the second article we describe how PLP-dependent enzymes perform the racemization reaction, both in microorganisms and in higher organisms. Finally, in a third review, the regulation of serine racemase by divalent cations, nucleotides and interacting proteins is also addressed.