Jacques J. H. Hens

Long-term potentiation and synaptic protein phosphorylation

Long-term potentiation (LTP) is a well known experimental model for studying the activity-dependent enhancement of synaptic plasticity, and because of its long duration and its associative properties, it has been proposed as a system to investigate the molecular mechanisms of memory formation. At present, there are several lines of evidence that indicate that pre- and postsynaptic kinases and their specific substrates are involved in molecular mechanisms underlying LTP. Many studies focus on the involvement of protein kinase C (PKC). One way to investigate the role of PKC in long-term potentiation is to determine the degree of phosphorylation of its substrates after in situ phosphorylation in hippocampal slices. Two possible targets are the presynaptic membrane-associated protein B-50 (a.k.a. GAP 43, neuromodulin and F1), which has been implicated in different forms of synaptical plasticity in the brain such as neurite outgrowth, hippocampal LTP and neurotransmitter release, and the postsynaptic protein neurogranin (a.k.a. RC3, BICKS and p17) which function remains to be determined. This review will focus on the protein kinase C activity in pre- and postsynaptic compartment during the early phase of LTP and the possible involvement of its substrates B-50 and neurogranin.

Protein kinase C in synaptic plasticity: changes in the in situ phosphorylation state of identified pre- and postsynaptic substrates

1. Long-term potentiation and its counterpart long-term depression are two forms of activity dependent synaptic plasticity, in which protein kinases and protein phosphatases are essential. 2. B-50/GAP-43 and RC3/neurogranin are two defined neuronal PKC substrates with different synaptic localization. B-50/GAP-43 is a presynaptic protein and RC3/neurogranin is only found at the postsynaptic site. Measuring their phosphorylation state in hippocampal slices, allows us to simultaneously monitor changes in pre- and postsynaptic PKC mediated phosphorylation. 3. Induction of LTP in the CA1 field of the hippocampus is accompanied with an increase in the in situ phosphorylation of both B-50/GAP-43 and RC3/neurogranin, during narrow, partially overlapping, time windows. 4. Pharmacological data show that mGluR stimulation results in an increase in the in situ phosphorylation of B-50/GAP-43 and RC3/neurogranin.

Depolarization-induced phosphorylation of the protein kinase C substrate B 50 (GAP 43) in rat cortical synaptosomes

Abstract: We studied the molecular events underlying K+‐induced phosphorylation of the neuron‐specific protein kinase C substrate B‐50. Rat cortical synaptosomes were prelabelled with 32P‐labelled orthophosphate. B‐50 phosphorylation was measured by an immunoprecipitation assay. In this system, various phorbol esters, as well as a synthetic diacylglycerol derivative, enhance B‐50 phosphorylation. K+ depolarization induces a transient enhancement of B‐50 phosphorylation, which is totally dependent on extracellular Ca2+. Also, the application of the Ca2+ ionophore A23187 induces B‐50 phosphorylation, but the magnitude and kinetics of A23187‐induced B‐50 phosphorylation differ from those induced by depolarization. The protein kinase inhibitors 1‐(5‐isoquinolinylsulfonyl)‐2‐methylpiperazine (H‐7), N‐(6‐aminohexyl)‐5‐chloro‐1‐naphthalenesulfonamide (W‐7), and staurosporine antagonize K+‐ as well as PDB‐induced B‐50 phosphorylation, whereas trifluoperazine and calmidazolium are ineffective under both conditions. We suggest that elevation of the intracellular Ca2+ level after depolarization is a trigger for activation of protein kinase C, which subsequently phosphorylates its substrate B‐50. This sequence of events could be of importance for the mechanism of depolarization‐induced transmitter release.

N‐Terminal‐Specific Anti‐B‐50 (GAP‐43) Antibodies Inhibit Ca2+‐Induced Noradrenaline Release, B‐50 Phosphorylation and Dephosphorylation, and Calmodulin Binding

Abstract: B‐50 (GAP‐43) is a presynaptic protein kinase C (PKC) substrate implicated in the molecular mechanism of noradrenaline release. To evaluate the importance of the PKC phosphorylation site and calmodulin‐binding domain of B‐50 in the regulation of neurotransmitter release, we introduced two monoclonal antibodies to B‐50 into streptolysin O‐permeated synaptosomes isolated from rat cerebral cortex. NM2 antibodies directed to the N‐terminal residues 39–43 of rat B‐50 dose‐dependently inhibited Ca2+‐induced radiolabeled and endogenous noradrenaline release from permeated synaptosomes. NM6 C‐terminal‐directed (residues 132–213) anti‐B‐50 antibodies were without effect in the same dose range. NM2 inhibited PKC‐mediated B‐50 phosphorylation at Ser41 in synaptosomal plasma membranes and permeated synaptosomes, inhibited 32P‐B‐50 dephosphorylation by endogenous synaptosomal phosphatases, and inhibited the binding of calmodulin to synaptosomal B‐50 in the absence of Ca2+. Similar concentrations of NM6 did not affect B‐50 phosphorylation or dephosphorylation or B‐50/calmodulin binding. We conclude that the N‐terminal residues 39–43 of the rat B‐50 protein play an important role in the process of Ca2+‐induced noradrenaline release, presumably by serving as a local calmodulin store that is regulated in a Ca2+‐ and phosphorylation‐dependent fashion.

Studies on the Role of B‐50 (GAP‐43) in the Mechanism of Ca2+‐Induced Noradrenaline Release: Lack of Involvement of Protein Kinase C After the Ca2+ Trigger

Abstract: The involvement of B‐50, protein kinase C (PKC), and PKC‐mediated B‐50 phosphorylation in the mechanism of Ca2+‐induced noradrenaline (NA) release was studied in highly purified rat cerebrocortical synaptosomes permeated with streptolysin‐O. Under optimal permeation conditions, 12% of the total NA content (8.9 pmol of NA/mg of synaptosomal protein) was released in a largely (>60%) ATP‐dependent manner as a result of an elevation of the free Ca2+ concentration from 10−8 to 10−5M Ca2+ The Ca2+ sensitivity in the micromolar range is identical for [3H]NA and endogenous NA release, indicating that Ca2+‐induced [3H]NA release originates from vesicular pools in noradrenergic synaptosomes. Ca2+‐induced NA release was inhibited by either N‐ or C‐terminal‐directed anti‐B‐50 antibodies, confirming a role of B‐50 in the process of exocytosis. In addition, both anti‐B‐50 antibodies inhibited PKC‐mediated B‐50 phosphorylation with a similar difference in inhibitory potency as observed for NA release. However, in a number of experiments, evidence was obtained challenging a direct role of PKC and PKC‐mediated B‐50 phosphorylation in Ca2+‐induced NA release. PKC pseudosubstrate PKC19‐36, which inhibited B‐50 phosphorylation (IC50 value, 10−5M), failed to inhibit Ca2+‐induced NA release, even when added before the Ca2+ trigger. Similar results were obtained with PKC inhibitor H‐7, whereas polymyxin B inhibited B‐50 phosphorylation as well as Ca2+‐induced NA release. Concerning the Ca2+ sensitivity, we demonstrate that PKC‐mediated B‐50 phosphorylation is initiated at a slightly higher Ca2+ concentration than NA release. Moreover, phorbol ester‐induced PKC down‐regulation was not paralleled by a decrease in Ca2+‐induced NA release from streptolysin‐O‐permeated synaptosomes. Finally, the Ca2+‐ and phorbol ester‐induced NA release was found to be additive, suggesting that they stimulate release through different mechanisms. In summary, we show that B‐50 is involved in Ca2+‐induced NA release from streptolysin‐O‐permeated synaptosomes. Evidence is presented challenging a role of PKC‐mediated B‐50 phosphorylation in the mechanism of NA exocytosis after Ca2+ influx. An involvement of PKC or PKC‐mediated B‐50 phosphorylation before the Ca2+ trigger is not ruled out. We suggest that the degree of B‐50 phosphorylation, rather than its phosphorylation after PKC activation itself, is important in the molecular cascade after the Ca2+ influx resulting in exocytosis of NA.

B-50/GAP-43 binds to actin filaments without affecting actin polymerization and filament organization.

Abstract: To investigate a possible function of the nervous tissuespecific protein kinase C substrate B‐50/GAP‐43 in regulati of the dynamics of the submembranous cytoskeleton. we studii the interaction between purified 6–50 and actin. Both the phosphorylated and dephosphorylated forms of 8–50 cosedi‐mented with filamentous actin (F‐actin) in a Ca2+‐independent manner. Neither 6–50 nor phospho‐6–50 had any effect on the kinetics of actin polymerization and on the critical concentration at steady state, as measured using pyrenylated actin. tight scattering of F‐actin samples was not increased in the presence of 550, suggesting that 550 does not bundle actin filaments. The number of actin filaments, determined by [3H]cytochalasin B binding, was not affected by either phospho‐ or dephospho‐B‐50, indicating that 550 has neither a severing nor a capping effect. These observations were confirmed by electron microscopic evaluation of negatively stained F‐actin samples, which did not reveal any structural changes in the actin meshwork on addition of 6–50, We conclude that 6–50 is an actin‐binding protein that does not directly affect actin dynamics.

Evidence for a role of protein kinase C substrate B 50 (GAP 43) in Ca2+-induced neuropeptide cholecystokinin-8 release from permeated synaptosomes

Abstract: To study the involvement of the protein kinase C (PKC) substrate B‐50 [also known as growth‐associated protein‐43 (GAP‐43), neuromodulin, and F1] in presynaptic cholecystokinin‐8 (CCK‐8) release, highly purified synaptosomes from rat cerebral cortex were permeated with the bacterial toxin streptolysin O (SL‐O). CCK‐8 release from permeated synaptosomes, determined quantitatively by radioimmunoassay, could be induced by Ca2+ in a concentration‐dependent manner (EC50 of ∼10‐5M). Ca2+‐induced CCK‐8 release was maximal at 104M Ca2+, amounting to ∼10% of the initial 6,000 ± 550 fmol of CCK‐8 content/mg of synaptosomal protein. Only 30% of the Caa+‐induced CCK‐8 release was dependent on the presence of exogenously added ATP. Two different monoclonal anti‐B‐50 antibodies were introduced into permeated synaptosomes to study their effect on Ca2+‐induced CCK‐8 release. The N‐terminally directed antibodies (NM2), which inhibited PKC‐mediated B‐50 phosphorylation, inhibited Ca2+‐induced CCK‐8 release in a dose‐dependent manner, whereas the C‐terminally directed antibodies (NM6) affected neither B‐50 phosphorylation nor CCK‐8 release. The PKC inhibitors PKC19–36 and 1 −(5‐isoquinolinylsulfonyl)‐2‐methylpiperazine (H‐7), which inhibited B‐50 phosphorylation in permeated synaptosomes, had no effect on Ca2+‐induced CCK‐8 release. Our data strongly indicate that B‐50 is involved in the mechanism of presynaptic CCK‐8 release, at a step downstream of the Ca2+ trigger. As CCK‐8 is stored in large densecored vesicles, we conclude that B‐50 is an essential factor in the exocytosis from this type of neuropeptide‐containing vesicle. The differential effects of the monoclonal antibodies indicate that this B‐50 property is localized in the N‐terminal region of the B‐50 molecule, which contains the PKC phosphorylation site and calmodulin‐binding domain.

Ba2+ replaces Ca2+/calmodulin in the activation of protein phosphatases and in exocytosis of all major transmitters

Exocytosis from nerve terminals is triggered by depolarization-evoked Ca2+ entry, which also activates calmodulin and stimulates protein phosphorylation. Ba2+ is believed to replace Ca2+ in triggering exocytosis without activation of calmodulin and can therefore be used to unravel aspects of presynaptic function. We have analysed the cellular actions of Ba2+ in relation to its effect on transmitter release from isolated nerve terminals. Barium evoked specific release of amino acid transmitters, catecholamines and neuropeptides (EC50 0.2-0.5 mM), similar to K-/Ca(2+)-evoked release both in extent and kinetics. Ba(2+)-and Ca(2+)-evoked release were not additive. In contrast to Ca2+, Ba2+ triggered release which was insensitive to trifluoperizine and hardly stimulated protein phosphorylation. These observations are in accordance with the ability of Ba2+ to replace Ca2+ in exocytosis without activating calmodulin. Nevertheless, calmodulin appears to be essential for regular (Ca(2+)-triggered) exocytosis, given its sensitivity to trifluoperizine. Both Ba(2+)-and Ca(2+)-evoked release were blocked by okadaic acid. Furthermore, anti-calcineurin antibodies decreased Ba(2+)-evoked release. In conclusion, Ba2+ replaces Ca2+/calmodulin in the release of the same transmitter pool. Calmodulin-dependent phosphorylation appears not to be essential for transmitter release. Instead, our data implicate both Ca(2+)-dependent and -independent dephosphorylation in the events prior to neurotransmitter exocytosis.

Noradrenaline release from permeabilized synaptosomes is inhibited by the light chain of tetanus toxin

Noradrenaline release from rat brain cortical synaptosomes permeabilized with streptolysin O can be triggered by μM concentrations of free Ca2+. This process was inhibited within minutes by tetanus toxin and its isolated light chain, but not by its heavy chain. The data demonstrate that the effect of tetanus toxin on NA release from purified synaptosomes is caused by the intraterminal action of its light chain.

Monoclonal Antibody NM2 Recognizes the Protein Kinase C Phosphorylation Site in B-50 (GAP-43) and in Neurogranin (BICKS)

Abstract: Mouse monoclonal B‐50 antibodies (Mabs) were screened to select a Mab that may interfere with suggested functions of B‐50 (GAP‐43), such as involvement in neurotransmitter release. Because the Mab NM2 reacted with peptide fragments of rat B‐50 containing the unique protein kinase C (PKC) phosphorylation site at serine‐41, it was selected and characterized in comparison with another Mab NM6 unreactive with these fragments. NM2, but not NM6, recognized neurogranin (BICKS), another PKC substrate, containing a homologous sequence to rat B‐50 (34–52). To narrow down the epitope domain, synthetic B‐50 peptides were tested in ELISAs. In contrast to NM6, NM2 immunoreacted with B‐50 (39–51) peptide, but not with B‐50 (43–51) peptide or a C‐terminal B‐50 peptide. Preabsorption by B‐50 (39–51) peptide of NM2 inhibited the binding of NM2 to rat B‐50 in contrast to NM6. NM2 selectively inhibited phosphorylation of B‐50 during endogenous phosphorylation of synaptosomal plasma membrane proteins. Preabsorption of NM2 by B‐50 (39–51) peptide abolished this inhibition. In conclusion, NM2 recognizes the QASFR peptide in B‐50 and neurogranin. Therefore, NM2 may be a useful tool in physiological studies of the role of PKC‐mediated phosphorylation and calmodulin binding of B‐50 and neurogranin.