Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Glutamate and aspartate are the major excitatory neurotransmitters acting in the CNS (Monaghan et al., 1989). Their interactions with specific membrane receptors, which are currently divided into NMDA, kainate, AMPA and metabotropic receptors, are responsible for many neurological functions such as cognition, learning, memory and sensation (Gasic and Hollmann, 1992). In addition, NMDA receptors are involved in the developmental plasticity of synaptic connections in the nervous system (Lipton and Kater, 1989). However, excessive glutamate release and consequent overstimulation of EAA receptors can lead to excitotoxicity. The latter has been associated with a wide range of neurological disorders, including trauma and stroke damage, epilepsy and Alzheimer's and Huntington's diseases (Lipton and Rosenberg, 1994).

Activation of central EAA receptors releases both ADO and NE (Hoehn et al., 1990; Craig and White, 1991). In fact, activation of non-NMDA receptors itself releases ADO, whereas activation of NMDA receptors releases a nucleotide that is later metabolized extracellularly to adenosine (Craig and White, 1993). NMDA receptor-evoked release is Ca⁺⁺-dependent, whereas non-NMDA receptor-evoked release is not (Craig and White, 1993). NMDA is 33 times more potent in releasing ADO than in releasing NE (Hoehn et al., 1990). Moreover, partial noncompetitive antagonism of NMDA-evoked ADO release by Mg⁺⁺, MK801 or the glycine site antagonist 7-chlorokynurenic acid can be overcome by high concentrations of NMDA (Hoehn et al., 1990; Craig and White, 1991). These results support the idea that spare receptors exist for NMDA-evoked ADO release but not for NE release. Previous studies in our laboratory showed that the ADO released during low-level NMDA receptor activation provides an inhibitory threshold against NMDA-mediated neurotransmission such as the release of NE (Craig and White, 1992; White et al., 1993). This inhibitory threshold must be overcome in order for NMDA-mediated responses to proceed maximally (White et al., 1993), and this provides selectivity for critical functions, such as learning, memory and synaptic plasticity in the cortex. Indeed, results from electrophysiological studies indicate that ADO, released as a result of low-level NMDA receptor activation during glutamatergic transmission, acts presynaptically to decrease the release of glutamate and depress excitation in the CA1 region of the hippocampus (Mitchell et al., 1993; Manzoni et al., 1994).

PKC is a Ca⁺⁺- and phospholipid-dependent enzyme that is highly concentrated in the brain (Nishizuka, 1986). Its activity is very important in mediating neurotransmitter release and synaptic plasticity (Hollingsworth et al., 1985; Harvey and Collingridge, 1993; Ohtani et al., 1995). The facilitatory effects of metabotropic glutamate receptors on NMDA-evoked responses have been reported to be mediated by the activation of PKC (Aniksztejn et al., 1992; Kelso et al., 1992). PKC activation by phorbol esters has also been shown to enhance NMDA currents in various systems, including rat trigeminal neurons (Chen and Huang, 1991) and oocytes expressing total rat brain mRNA (Kelso et al., 1992). In addition, NMDA-evoked ADO release from rat cortical slices is potentiated when M₃ muscarinic receptors are activated (Semba and White, 1997). M₃ muscarinic receptors are G protein-coupled receptors that activate PKC. This raises the possibility that the potentiating effect of M₃ agonists on NMDA-evoked ADO release might be mediated by the activation of PKC.

The present study was undertaken to investigate the possible role of PKC in NMDA-evoked ADO and NE release from rat cortical slices and to determine whether PKC activation affects NMDA-evoked ADO and NE release in the cortex.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Preparation of slices. Male Sprague-Dawley rats weighing 250 to 350 g (Charles River Canada, St Constant, Quebec, Canada) were decapitated, and their brains were removed rapidly to ice-cold Krebs-Henseleit bicarbonate medium containing (mM) NaCl, 111; NaHCO₃, 26.2; NaH₂PO₄, 1.2; KCl, 4.7; CaCl₂, 1.8; MgCl₂, 1.2 and glucose, 11, gassed with 95% O₂-5% CO₂ to maintain a pH of 7.4. The lateral 1 to 1.5-mm portion of parietal cortex was removed from both hemispheres of the brain with a recessed tissue slicer. Coronal slices (0.4 mm) of parietal cortex were cut with a McIlwain tissue chopper. Adjacent slices were placed alternately into each of two tissue baths so that each bath contained six slices from both sides of the brain, weighing a total of about 80 mg. This procedure made it possible to use paired statistical analyses of the data obtained from the slices in the two baths.

Superfusion of slices. The slices rested on a nylon mesh screen in tissue baths adjusted to 0.5 ml as described previously (Hoehn et al., 1990) and were immersed in a circulating water bath at 36°C. Two baths were run in parallel and were assigned in alternate experiments to either "test" or "control" observations. After an initial 5-min superfusion period, slices were labeled with [³H]NE by superfusion for 10 min with oxygenated Krebs-Henseleit bicarbonate medium containing freshly prepared 10⁷ M [³H]NE (specific activity, 13.1 Ci/mmol) at 36°C. Superfusion was continued with Krebs-Henseleit bicarbonate medium for a further 65 min before collection of 10 serial 2.5-min fractions. After collection of three samples to determine basal release, the superfusing medium was switched for 10 min to medium containing NMDA, after which the superfusing buffer was switched back to Krebs-Henseleit buffer for the final three fractions. PMA or GFX was introduced into the superfusing buffer 10 min before collection of the first fraction and continued until the end of the experiment. In experiments conducted in "0 Mg⁺⁺", slices were superfused for 65 min before sample collection with medium from which MgCl₂ had been omitted.

Determination of ADO release. Samples of superfusate were deproteinated with Ba(OH)₂ and ZnSO₄ and then reacted with chloroacetaldehyde to form 1-N⁶-ethenoadenosine, which was assayed using HPLC with fluorescence detection essentially as described previously (Hoehn and White, 1990; White, 1996). ADO standards in Krebs-Henseleit medium were treated identically to the samples, and the amount of ADO in the samples was quantitated by comparison of peak heights with the standards.

Determination of [³H]NE release. After removal of 0.5 ml of the superfusate for determination of ADO, 1 ml was placed into scintillation vials containing 10 ml of Aquasol-2 and the disintegrations per minute of [³H]NE released were determined with a Beckman model LS5801 scintillation counter (Hoehn et al., 1990). The slices were weighed and then solubilized in 1 ml of Protosol. Tissue [³H]NE contents were determined by scintillation spectrometry in 14 ml of Econofluor. The rate of [³H]NE release was standardized as the percentage of total tissue [³H]NE content at the beginning of the sample collection period. The rate of evoked [³H]NE release was obtained by subtracting, from every other sample, the percentage of release per minute in the sample immediately preceding exposure to the releasing agent; it was expressed as percentage of content. Total evoked [³H]NE release was determined as the percentage of [³H]NE content released during the entire 17.5-min period after exposure to the releasing agent.

Statistical analysis. Paired Student's t tests were conducted on the total ADO release and [³H]NE release from the two groups of slices.

Materials. The following drugs and chemicals were used in their study: ADO, NMDA, 1-octanesulfonic acid (Sigma Chemical Co., St. Louis, MO), acetonitrile (BDH, Dartmouth, Nova Scotia, Canada), chloroacetaldehyde (ICN, Plainview, NY), PMA, GF109,203X, GFX (Research Biochemicals Inc., Natick, MA) and L-[7-³H]NE, Protosol, Aquasol-2 and Econofluor (Du Pont-New England Nuclear Canada Inc., Markham, Ontario, Canada). All other chemicals were obtained from commercial sources.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of PKC inhibition with GFX on NMDA-evoked ADO and [³H]NE release in the absence of Mg⁺⁺. GFX is a cell-permeable protein kinase inhibitor that is structurally similar to staurosporin and inhibits PKC by acting as a competitive inhibitor of the ATP-binding site of PKC. GFX is a highly selective and potent inhibitor of PKC compared with staurosporin (Toullec et al., 1991; Davis et al., 1992).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effect of PKC inhibition with 1 µM GFX on 100 µM NMDA-evoked ADO (panel A) and [3H]NE release in the absence of Mg++ (panel B). GFX was present from 15 min before exposure to NMDA until the end of the experiment. NMDA was present from 0 to 10 min. Values are means ± S.E.M. from six experiments. Histograms represent the total amount of ADO or [3H]NE released.

Effects of PKC activation with PMA on NMDA-evoked ADO and [³H]NE release in the absence of Mg⁺⁺. PMA (1 µM) by itself did not release either ADO or [³H]NE (data not shown), which indicates that activation of PKC by itself does not promote the release of either ADO or [³H]NE from rat cortical slices. In the absence of Mg⁺⁺, 1 µM PMA increased 2-fold the maximal rate of ADO release evoked by 20 µM NMDA (fig. 2A). The total amount of NMDA-evoked ADO release was also increased 2-fold by PMA (fig. 2A). This low concentration of NMDA (20 µM) did not release [³H]NE (fig. 2B).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Effect of PKC activation with 1 µM PMA on 20 µM NMDA-evoked ADO (panel A) and [3H]NE (panel B) release in the absence of Mg++. PMA was present from 15 min before exposure to NMDA until the end of the experiment. NMDA was present from 0 to 10 min. Values are means ± S.E.M. from five experiments. Histograms represent the total amount of ADO or [3H]NE released. * Significantly different from control (P < .05, paired t test).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Effect of PKC activation with 1 µM PMA on 100 µM NMDA-evoked ADO (panel A) and [3H]NE (panel B) release in the absence of Mg++. PMA was present from 15 min before exposure to NMDA until the end of the experiment. NMDA was present from 0 to 10 min. Values are means ± S.E.M. from eight experiments. Histograms represent the total amount of ADO and [3H]NE released. ** Significantly different from control (P < .01, paired t test).

Effects of GFX on PMA-potentiated release of ADO and [³H]NE evoked by NMDA in the absence of Mg⁺⁺. We tested whether the potentiating effect of PMA on the NMDA-evoked release of both ADO and [³H]NE was mediated by the activation of PKC. At 20 µM NMDA, the potentiating effect of PMA on evoked ADO release was depressed by the PKC inhibitor GFX (1 µM, fig. 4A). Similarly, the facilitatory effect of PMA on 100 µM NMDA-evoked release of [³H]NE was decreased by coadministration with 1 µM GFX (fig. 4B). These findings confirm that the potentiating effects of PMA on the NMDA-evoked release of both ADO and NE are mediated by activation of PKC.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effect of 1 µM GFX on PMA's effects on 20 µM NMDA-evoked ADO release (panel A) and 100 µM NMDA-evoked [3H]NE release (panel B), in the absence of Mg++. GFX and PMA were present from 15 min before exposure to NMDA until the end of the experiment. NMDA was present from 0 to 10 min. Values are means ± S.E.M. from four experiments for ADO release and means ± S.E.M. from 6 experiments for [3H]NE release. Histograms represent the total amount of ADO and [3H]NE released. * Significantly different from control (P < .05, paired t test).

Effects of PKC activation with PMA on NMDA-evoked [³H]NE release in the presence of Mg⁺⁺. Chen and Huang (1992) suggested that PKC activation may decrease the voltage-sensitive Mg⁺⁺ block of NMDA receptors. Thus we determined whether activation of PKC might permit NMDA to release NE in the presence of Mg⁺⁺. However, in the presence of 1.2 mM Mg⁺⁺, 1 µM PMA did not permit 300 µM NMDA to release [³H]NE (data not shown).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Endogenous PKC appears to be required for NMDA-evoked increases in cytosolic Ca⁺⁺ in rat striatal neurons (Murphy et al., 1994). However, the selective PKC inhibitor GFX did not diminish NMDA-evoked ADO and [³H]NE release in the present study, which suggests that endogenous PKC activity does not play an essential role in these release processes. Using the two-electrode voltage-clamp technique, Kelso et al. (1992) showed that protein kinase inhibitors do not depress NMDA-activated currents expressed in Xenopus oocytes, which also suggests that PKC does not play an essential role in NMDA receptor function.

Although PKC does not appear to be involved in the NMDA-evoked release of either ADO or [³H]NE in rat cortical slices, activation of PKC by PMA did potentiate their releases evoked by submaximal concentrations of NMDA, a result that suggests that PKC plays a role in modulating these releasing processes. The potentiating effects of PMA on both ADO and NE release evoked by submaximal concentrations of NMDA appear to be related specifically to activation of PKC, because they were reversed by the specific PKC inhibitor GFX.

In electrophysiological studies, Chen and Huang (1992) proposed that PKC potentiated the NMDA-activated currents mainly by reducing the voltage-dependent Mg⁺⁺ block of the NMDA receptor channels. However, in the present study, PMA did not overcome the Mg⁺⁺-block of NMDA-evoked [³H]NE release from rat cortical slices. Similar conclusions have been reached by other investigators (Murphy et al., 1994; Patel et al., 1995; Wagner and Leonard, 1996).

The results of the present study can be explained if activation of PKC increases the agonist affinity of NMDA receptors. This would increase responses when activation of NMDA receptors is submaximal but would have no effect on responses to maximal NMDA receptor activation. It is also possible that activation of PKC exerts its effects at some site after NMDA receptor activation, perhaps on the transduction mechanisms that promote ADO and NE release. However, these transduction mechanisms cannot include PKC activation, consequent to NMDA receptor activation, because neither the release of ADO nor the release of [³H]NE is diminished by PKC inhibitors.

Our current findings indicate that activation of PKC by the phorbol ester PMA potentiated ADO release at low levels of NMDA receptor activation, whereas at higher levels of NMDA receptor activation, it potentiated [³H]NE release but had no effect on ADO release. This could have important functional implications. ADO, released during low levels of NMDA receptor activation, provides an inhibitory threshold that must be overcome in order for excitatory NMDA-mediated processes to proceed maximally (Craig and White, 1992). When only a few NMDA receptors are activated, PKC increases extracellular ADO and thus elevates the inhibitory threshold against excitatory neurotransmission that might play a role in modulating normal physiological processes such as learning and memory. This would provide even more selectivity for these essential excitatory processes. However, when the presynaptic release of glutamate is very high, large numbers of NMDA receptors become activated, and the inhibitory threshold provided by ADO is overcome. Under these circumstances, activation of PKC promotes the excitatory actions of NMDA receptor activation (e.g., NE release) without producing a corresponding increase in the inhibitory threshold (e.g., adenosine release). This has the effect of accelerating NMDA-mediated excitatory responses and promoting processes such as learning and memory.