Introduction
The "glutamate hypothesis" of schizophrenia suggests that the disorder is driven by the underactivity of N-methyl-D-aspartate (NMDA) receptors. A critical component of this pathway is D-serine, a co-agonist necessary for NMDA receptor function. The enzyme D-amino acid oxidase (DAO) is responsible for the degradation of D-serine; consequently, elevated DAO activity can lead to D-serine depletion and subsequent NMDA receptor hypofunction.
The protein DAOA (D-amino acid oxidase activator), also known as G72, was initially identified as a putative activator of DAO. However, scientific literature has been inconsistent, with various studies reporting that DAOA activates, inhibits, or has no effect on DAO activity. This study aimed to resolve these contradictions by examining the interaction across different human cell environments and utilizing molecular modeling to observe protein-protein interactions at an atomic level.
Methodology
Researchers employed a multi-disciplinary approach to investigate the DAOA-DAO relationship:
- Cellular Models: Three human cell lines were used to represent diverse physiological environments: SH-SY5Y (neuron-like), 1321N1 (astrocyte-like), and HEK293 (kidney-like).
- Activity Assays: Following the transient transfection and overexpression of DAO and DAOA, enzyme activity was measured using the Amplex Red reagent to detect hydrogen peroxide ($H_2O_2$), a byproduct of D-serine oxidation.
- Molecular Dynamics (MD) Simulations: Simulations were performed to observe the interaction between DAOA and the human DAO (hDAO) monomer in both its apoprotein (enzyme alone) and holoenzyme (enzyme bound to FAD cofactor) states.
- Electrophysiology: Whole-cell patch-clamp recordings were conducted on HEK293 cells expressing NMDA receptors to determine if DAOA overexpression directly influenced receptor currents.
Results and Key Insights
The study revealed that the effect of DAOA on DAO activity is highly dependent on the cellular context.
Cell-Type Specificity
The most significant finding was that DAOA increased DAO activity exclusively in the kidney-derived HEK293 cells. In contrast, no significant change in activity was observed in the brain-derived neuronal (SH-SY5Y) or astrocyte (1321N1) lines. This suggests that the "activator" function of G72 may not be applicable within the central nervous system.
As shown in Figure 2, the addition of G72 to DAO-expressing 1321N1 cells did not significantly alter DAO activity, regardless of the presence of the FAD cofactor. Cell viability also remained stable across conditions, indicating that the lack of increased activity was not due to cellular toxicity.
Molecular Mechanisms and Expression
Molecular dynamics simulations provided a potential explanation for the inhibitory effects reported in some previous studies. The simulations indicated that when DAOA binds to the hDAO holoenzyme, it increases the enzyme's structural flexibility, potentially leading to misfolding and inactivation. Interestingly, this effect was absent when DAOA interacted with the hDAO apoprotein.
Furthermore, expression analysis showed that endogenous DAO mRNA levels were significantly lower in brain-derived lines compared to HEK293 cells. This suggests that in the human brain, these proteins are subject to much tighter regulatory control than in peripheral tissues.
Conclusion and Future Directions
The study concludes that the interaction between DAO and DAOA is complex and dependent on the specific biochemical characteristics of the cell type. While DAOA acts as an activator in kidney-like cells, it appears to have no effect or potentially a destabilizing effect in brain-like environments. Notably, the study found no direct link between DAOA overexpression and changes in NMDA receptor currents, suggesting the relationship is indirect.
Future research should focus on more physiologically relevant models, such as human induced pluripotent stem cells (hiPSCs) derived from patients, to better understand these interactions within a natural genomic context. The discrepancy between the activating effect in HEK293 cells and the destabilizing effect suggested by MD simulations remains a key area for further investigation.
Summary updated