Effects of Changes in Phosphate Availability on Glutamatergic Transmission: Potential Relevance to Alzheimer's or Age-Associated Cognitive Impairment

This article [Glinn et al., 1997: (http://www.ncbi.nlm.nih.gov/pubmed/9200502)] is really interesting, and the authors found that the ATP levels in cultured neurons correlated positively, up to a point, with the availability of inorganic phosphate (Pi). The concentration-dependences found by the authors for that and other correlations, such as of metabolite concentrations with Pi availability, seem to have the potential to be misleading, because many articles have shown that the intracellular ATP or 2,3-bisphosphoglycerate levels do increase in response to seemingly-insignificant, acute increases in extracellular Pi availability (see past postings). It's probably that the availabilities of energy substrates in the culure medium provide the cells with everything they need, and those conditions are unlikely to prevail in vivo. Also, there's some strange issue with the concentrations of free Pi being found by different groups of researchers. Glinn et al. (1997) and other groups have found concentrations in the 30+ mM range, but others have found that the free Pi levels are between 0.8 and 4 or so mM. Maybe there are differences between cell types, but those seem like awfully large differences. I would think it would be difficult for anyone to determine the percentage that would exist unbound, at any given concentration. I'll have to read up on that.

Glinn et al. (1997) also mention some really interesting research suggesting that low Pi availability to the brain may contribute to cognitive dysfunction in people who go on to develop Alzheimer's (reference 30, cited on page 91). They also discuss research showing that Pi availability may help protect against glutamatergic neurotoxicity (i.e. "excitotoxicity"). They mention, earlier in the article, that Pi can be utilized by 3-phosphoglycerate kinase and pyruvate kinase, and I looked into that topic a little bit. In that context, it's interesting that the glycolytic enzymes glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and 3-phosphoglycerate kinase (PGK) form a (dimeric?) enzyme complex, and the 1,3-bisphosphoglycerate (1,3-BPG) that is formed from glyceraldehyde-3-phosphate (GAP), by GAPDH, is then channeled, apparently, to PGK and converted into 3-phosphoglycerate [Ikemoto et al., 2003: (http://www.jbc.org/cgi/content/full/278/8/5929)(http://www.ncbi.nlm.nih.gov/pubmed/12488440?dopt=Abstract)]. What's interesting is that, essentially, Pi is used to fairly directly (via its incorporation into 1,3-BPG) phosphorylate ADP into ATP, and the ATP formed by the GAPDH-PGK complex is preferentially used to transport glutamate into presynaptic vesicles (exogenous ATP is not as effective in promoting vesicular glutamate transport) (Ikemoto et al., 2003). A lot of researchers refer to that ATP-requiring transport as "glutamate uptake," but that terminology could potentially cause one to confuse the process with synaptic glutamate uptake. Ikemoto et al. (2003) are talking about the vesicular glutamate transport that "loads" glutamate in presynaptic vesicles for release. There's other research showing that mitochondrial glutaminase (and the mitochondria that contain it) is localized at the sites of synaptic glutamate uptake and that Pi availability, especially insofar as its availability is important for the activation of glutaminase in astrocytes, plays a role in maintaining synaptic glutamate uptake. Some of those articles that cite Glinn et al. (1997) look interesting (http://scholar.google.com/scholar?cites=5143060761810849298&hl=en).