PROTEIN ACYL-THIOESTERASE ENZYME ACTIVITY IN THE POSTMORTEM DORSOLATERAL PREFRONTAL CORTEX IN SCHIZOPHRENIA

Abstract Background The role protein trafficking and localization is a recent target of investigation in schizophrenia pathophysiology. An important mediator of protein trafficking is S-acylation, also known as S-palmitoylation, which is the reversible attachment of long chain fatty acids to cysteine residues. S-acylation is a dynamic post-translational modification that modulates hydrophobicity of proteins, regulating their membrane association and subcellular localization. Notably, we have previously reported a proteome-wide decrease in S-acylated protein levels in the dorsolateral prefrontal cortex (DLPFC) of subjects with schizophrenia. One potential mechanism of decreased S-acylation is increased removal of acyl groups from proteins by protein acyl-thioesterase enzymes (PATs). Here we describe the optimization of an assay to measure the activity of the PAT family of enzymes in human postmortem cortical tissue and use the assay to address our hypothesis that PAT activity is increased in the DLPFC of subjects with schizophrenia. Methods To determine PAT activity, tissue homogenate was incubated with 4-methylumbelliferyl-6-thio-palmitate-β-D-glucopyranoside (4MU-Gluc-Palm) and 1U of exogenous β-glucosidase (to hydrolyze the 4MU-Gluc reaction intermediary). Released 4MU was excited at 360 ± 40 nm and fluorescent emission was measured, per minute, at 460 ± 40 nm. To determine the relationship between initial reaction rate and amount of enzyme, the initial reaction rate using 300 µM 4MU-Gluc-Palm was measured in homogenate containing 1 – 10 µg of total protein from the DLPFC of a subject with no history of psychiatric illness. The PAT activity of DLPFC homogenate boiled for 30 min and total protein homogenate from lymphocytes were measured as negative and positive control reactions, respectively. To estimate the maximum reaction rate (Vmax) and the concentration of 4MU-Gluc-Palm which achieved ½ Vmax (Km; a measure of enzyme-substrate affinity) the initial reaction rate was calculated in the presence of 0 – 200 µM 4MU-Gluc-Palm and the Michaelis-Menten equation was fit to plots of concentration vs. initial rate. Reactions were performed on 2.5 µg total protein homogenate from the DLPFC of 24 subjects with schizophrenia and 24 non-psychiatrically ill subjects. Results A fluorescent signal, which increases with time to a plateau upon substrate depletion, is detectable in total protein homogenate from DLPFC and lymphocytes, but not boiled DLPFC homogenate. In the DLPFC the initial reaction rate is linear with total protein amount [r2 = .99; p = .007], demonstrating that the reaction is sensitive to varying amounts of enzyme in a 10-fold range. When compared between schizophrenia and control subjects, neither Vmax [t(46) = 0.756; p = .45] nor Km [t(46) = 0.780; p = .44] were statistically significantly different. Discussion Here we have demonstrated that PAT activity is measurable in human cortical tissue homogenate. Additionally, we have found no difference in the Vmax or Km of the combined PAT enzyme group in schizophrenia, providing no evidence to support our hypothesis that total PAT activity is increased in subjects with schizophrenia. This suggests that the proteome-wide decrease in S-acylated proteins in schizophrenia is caused by another mechanism, possibly increased expression or function of one or more of the specific PATs, leading to substrate specific changes in S-acylation, or a decrease in activity the acyl protein transferase enzymes, which attach acyl groups to proteins.