Long-term potentiation in the hippocampal slice: possible involvement of pyruvate dehydrogenase

Oxygen sensitivity of mitochondrial redox status and evoked potential recovery early during reperfusion in post-ischemic rat brain

Inspired oxygen (FiO2) was manipulated during the early reperfusion period after global cerebral ischemia (four-vessel occlusion of 20 or 30 min duration) in anesthetized rats. The goal was to determine whether oxygen availability during the early reperfusion period alters recovery of mitochondrial redox state and evoked electrical activity. The effectiveness of post-ischemic oxygen treatment was monitored at the tissue level with oxygen sensitive microelectrodes, and at the mitochondrial level by reflection spectrophotometry of the redox state of cytochrome oxidase. Transiently decreasing FiO2 from 0.3 to 0.15 limited reperfusion-induced hyperoxygenation and post-ischemic mitochondrial hyperoxidation (PIMHo). Evoked potential recovery was improved by this treatment after 20 min ischemia but not after 30 min ischemia. Increasing FiO2 from 0.3 to 1.0 exacerbated PIMHo and tissue hyperoxygenation. Transient elevation of tissue oxygen tension after 30 min of global ischemia inhibited recovery of evoked potentials. These data suggest that a period of heightened vulnerability to oxidative stress occurs within the first 10 min of reperfusion after global ischemia. This period is characterized by tissue hyperoxygenation and mitochondrial hyperoxidation. Limiting oxygen availability during this period may improve the outcome while conversely elevating oxygenation may be detrimental.

Release of proteins during long-term potentiation in the hippocampus of the anaesthetized rat

Using a push-pull device, the release of endogenous proteins in the extracellular space was investigated in the CA1 region of the hippocampus of anaesthetized rats. With low-frequency stimulation of the Schaffer collaterals, there was a relatively stable release of 5 proteins (64, 54, 48, 45 and 16 kDa). A train of high-frequency stimulation produced a long-lasting enhancement of the negative field EPSP and a delayed (90-120 min) enhancement of the release of these proteins. An additional 19 kDa protein was present only 90 min after the train. These observations raise the possibility that release of proteins might be involved in the maintenance of LTP.

Synapsin I, a phosphoprotein associated with synaptic vesicles: possible role in regulation of neurotransmitter release

The data presented here provide evidence that the study of neuronal phosphoproteins can lead to the identification of previously unknown proteins and that these proteins may play important roles in neuronal communication. Specifically, in the case of synapsin I, direct evidence has been obtained that this phosphoprotein is involved in regulating neurotransmitter release. A tentative explanation of the results obtained in the micro-injection studies is as follows: synapsin I, in the dephosphostate, is bound to the cytoplasmic surface of synaptic vesicles and inhibits the ability of the vesicle to interact with the plasma membrane; increases in intracellular calcium activate calmodulin kinase II which in turn phosphorylates synapsin I and the phosphorylated synapsin I dissociates from the synaptic vesicle thus removing a constraint on the release of neurotransmitter. Clearly, more studies need to be done to rigorously test this hypothesis. Nevertheless these studies of synapsin I suggest that the study of previously unknown phosphoproteins will lead to the elucidation of previously unknown regulatory processes in neurons.

Low frequency perforant path stimulation as a conditioned stimulus demonstrates correlations between long-term synaptic potentiation and learning

Stimulation of the perforant path with impulse trains of 15 cps and 670 msec duration was used as a conditioned stimulus in a two-way shuttle box avoidance on rats. Field potentials in the dentate area evoked by test stimuli were measured after the training sessions until the 7th day. Foot-shock and unconditioned escape elicited only a transient slight depression of the population spike amplitude (P) and increased also slightly the slope function (SF) of the population EPSP of the evoked test potentials. The control stimulation of the perforant path without pairing with foot-shock as in conditioning did only slightly increase SF of test potentials, but produced a strong transient inhibition followed by a long lasting moderate depression of P. After conditioning, all animals exhibited the same initial inhibition of P as shown in control stimulation of the perforant path. However during the following 4 hours, good learners with a relearning index greater than 30% developed a significant potentiation of P lasting until the second training session 24 hours later, which resulted in a further enhancement. SF of the evoked test potentials increased in good learners with a similar time course after conditioning but without initial depression. After 7 days P showed still enhanced but non-significant values. Poor learners with a relearning index less than 10% did not develop a potentiation of P after conditioning and initial inhibition, but a long-term depression. Also SF of test potentials decreased in poor learners during 4 hours after conditioning and returned almost to baseline until the following day. After 7 days, P and SF did not differ from baseline. The analysis of the observed synaptic changes by E-S curves demonstrated the post-tetanic LTP seems to differ in some ways from post-conditioning LTP in good learners. The latter exhibits a clear tendency of a right shift contrary to the left shift commonly occurring after tetanization. Furthermore poor learners do not only fail to produce long-term potentiation, but fail to show a change in the opposite direction with a left shift of the E-S curves. The observed correlation of LTP in the conditioning pathway with the learning ability suggests an involvement of LTP at least in the acquisition and early retention of this learned behavior. The results do however not finally clarify the role of LTP in long-term retention.

Protein phosphorylation and neuronal function

Studies in the past several years have provided direct evidence that protein phosphorylation is involved in the regulation of neuronal function. Electrophysiological experiments have demonstrated that three distinct classes of protein kinases, i.e., cyclic AMP-dependent protein kinase, protein kinase C, and CaM kinase II, modulate physiological processes in neurons. Cyclic AMP-dependent protein kinase and kinase C have been shown to modify potassium and calcium channels, and CaM kinase II has been shown to enhance neurotransmitter release. A large number of substrates for these protein kinases have been found in neurons. In some cases (e.g., tyrosine hydroxylase, acetylcholine receptor, sodium channel) these proteins have a known function, whereas most of these proteins (e.g., synapsin I) had no known function when they were first identified as phosphoproteins. In the case of synapsin I, evidence now suggests that it regulates neurotransmitter release. These studies of synapsin I suggest that the characterization of previously unknown neuronal phosphoproteins will lead to the elucidation of previously unknown regulatory processes in neurons.

Factors affecting the phosphorylation of a 41,000 dalton protein of hippocampal subcellular fractions

The phosphorylation of the alpha-subunit of mitochondrial pyruvate dehydrogenase may be involved in the development of long-term potentiation in the hippocampus. Study of this hypothesis is hampered by variability in the incorporation of 32P into pyruvate dehydrogenase of hippocampal subcellular preparations, in vitro. 32P from [gamma-32P]ATP was incorporated into pyruvate dehydrogenase present in mitochondria and in a membrane-enriched synaptic particulate fraction from hippocampus. However, the presence of the synaptic fraction decreased isotopic labeling of the mitochondrial protein. This effect was not due to inhibition of the protein kinase or activation of a protein phosphatase, but the rate of ATP hydrolysis was found to be higher in the synaptic fraction than in the mitochondria (34 nmol/mg protein/min vs 14 nmol/mg protein/min). These data raise a variety of questions about the interpretation of the in vitro phosphorylation assay. It is concluded that variability in in vitro labeling can be minimized if the effect of ATP hydrolysis is diminished by use of a higher concentration of ATP. In addition, these data indicate that quantitative comparisons of the in vitro phosphorylation of diverse subcellular preparations must take into account differential rates of ATP hydrolysis.