A new perspective on the importance of glycine N–acyltransferase in the detoxification of benzoic acid
Abstract
Despite being the first biochemical reaction to be discovered, the glycine conjugation pathway remains poorly
characterised. It has generally been assumed that glycine conjugation serves to increase the water solubility of organic acids, such as benzoic acid and isovaleric acid, in order to facilitate urinary excretion of these compounds. However, it was recently suggested that the conjugation of glycine to benzoate should be viewed as a neuroregulatory process that prevents the accumulation of glycine, a neurotransmitter, to toxic levels. The true importance of glycine conjugation in metabolism is therefore not well understood. However, no genetic defect of glycine conjugation has ever been reported. This seems to suggest that glycine conjugation is a fundamentally important metabolic process, whatever its function may be. Therefore, a major objective of this thesis was to develop a deeper understanding of glycine conjugation and its metabolic significance. A review of the literature on GLYAT and glycine conjugation suggested that the primary purpose of glycine conjugation is indeed to detoxify benzoate and other aromatic acids of dietary origin. However, the commonly held assumption, that glycine conjugation increases the water solubility of aromatic acids in order to facilitate urinary excretion, seems to be incorrect. A better explanation for the detoxification of benzoate by means of glycine conjugation is based on hydrophilicity, not water solubility. Because of its lipophilic nature, benzoic acid is capable of passively diffusing across the mitochondrial inner membrane into the matrix space, where it accumulates due to the pH gradient over the inner membrane. Although benzoate can be exported from the matrix by organic anion transporters, this process would likely be futile because benzoic acid can simply diffuse back into the matrix. Hippurate, however, is
significantly less lipophilic and therefore less capable of diffusing into the matrix. It is therefore not transport out of the mitochondrial matrix that is facilitated by glycine conjugation, but rather the ability of the glycine conjugates to re-enter the matrix that is decreased. The conversion of benzoate to hippurate is a two-step process. First, benzoate is activated by an ATP-dependent acid:CoA ligase (ACSM2A) to form the more reactive benzoyl-CoA. Second, glycine N-acyltransferase (GLYAT) catalyses the formation of hippurate and CoASH from benzoyl-CoA and glycine. Another major objective of this thesis was to gain a better understanding of the structure and function of the GLYAT enzyme. While the substrate selectivity and enzyme kinetics of GLYAT have been investigated to some extent, almost nothing has been published on the structure, active site, or catalytic mechanism of GLYAT. Furthermore, while interindividual variation in the rate of glycine conjugation has been reported by several researchers, it is not known if, or how, genetic variation in the human GLYAT gene contributes to this inter individual variation. To address these issues, systems for the bacterial expression of recombinant bovine GLYAT and recombinant human GLYAT were developed. Because no crystal structure of GLYAT has been reported, homology modelling was used to generate a molecular model of bovine GLYAT. By comparing the molecular model to other acyltransferases for which the catalytic residues were known, Glu227 of bovine GLYAT was identified as a potential catalytic residue. Site directed mutagenesis was used to generate an E227Q mutant recombinant bovine GLYAT lacking the proposed catalytic residue. Characterisation of this mutant suggested that Glu227 was indeed the catalytic residue, and the GLYAT catalytic mechanism was elucidated. The molecular model was also used to identify Asn131 of bovine GLYAT as a potential active site residue. Site-directed mutagenesis was used to generate an N131C mutant, which was sensitive to inhibition by the sulfhydryl reagent DTNB. This suggests that the Asn131 residue of bovine GLYAT may
be situated in the active site of bovine GLYAT, but more work is needed to confirm this result. Finally, site-directed mutagenesis was used to generate variants of recombinant human GLYAT corresponding to six of the known SNPs in the human GLYAT gene. Expression and characterisation of the recombinant human GLYAT variants revealed that the enzyme activity and KM (benzoyl-CoA) parameter of the recombinant human GLYAT were influenced by SNPs in the human GLYAT gene. This suggests that genetic variation in the human GLYAT gene could partly explain the inter individual variation in the rate of glycine conjugation observed in humans. Interestingly, the SNPs that negatively influenced enzyme activity also had low allele frequencies, suggesting that there may be some selective advantage to having high GLYAT activity.