Strontium supports synaptic transmission and long-lasting potentiation in the hippocampus

Substitution of extracellular Ca2+ by Sr2+ prolongs inspiratory burst in pre-Bötzinger complex inspiratory neurons

The pre-Bötzinger complex (preBötC) underlies inspiratory rhythm generation. As a result of network interactions, preBötC neurons burst synchronously to produce rhythmic premotor inspiratory activity. Each inspiratory burst consists of action potentials (APs) on top of a 10- to 20-mV synchronous depolarization lasting 0.3-0.8 s known as inspiratory drive potential. The mechanisms underlying the initiation and termination of the inspiratory burst are unclear, and the role of Ca(2+) is a matter of intense debate. To investigate the role of extracellular Ca(2+) in inspiratory burst initiation and termination, we substituted extracellular Ca(2+) with Sr(2+). We found for the first time an ionic manipulation that significantly interferes with burst termination. In a rhythmically active slice, we current-clamped preBötC neurons (Vm ≅ -60 mV) while recording integrated hypoglossal nerve (∫XIIn) activity as motor output. Substitution of extracellular Ca(2+) with either 1.5 or 2.5 mM Sr(2+) significantly prolonged the duration of inspiratory bursts from 653.4 ± 30.7 ms in control conditions to 981.6 ± 78.5 ms in 1.5 mM Sr(2+) and 2,048.2 ± 448.5 ms in 2.5 mM Sr(2+), with a concomitant increase in decay time and area. Substitution of extracellular Ca(2+) by Sr(2+) is a well-established method to desynchronize neurotransmitter release. Our findings suggest that the increase in inspiratory burst duration is determined by a presynaptic mechanism involving desynchronization of glutamate release within the network.

Ionic dependence of Ca2+ channel modulation by syntaxin 1A

Alteration of the kinetic properties of voltage-gated Ca(2+) channels, Ca(v)1.2 (Lc-type), Ca(v)2.2 (N type), and Ca(v)2.3 (R type), by syntaxin 1A (Syn1A) and synaptotagmin could modulate exocytosis. We tested how switching divalent charge carriers from Ca(2+) to Sr(2+) and Ba(2+) affected Syn1A and synaptotagmin modulation of Ca(2+)-channel activation. Syn1A accelerated Ca(v)1.2 activation if Ca(2+) was the charge carrier; and by substituting for Ba(2+), Syn1A slowed Ca(v)1.2 activation. Syn1A also significantly accelerated Ca(v)2.3 activation in Ca(2+) and marginally in Ba(2+). Synaptotagmin, on the other hand, increased the rate of activation of Ca(v)2.3 and Ca(v)2.2 in all permeating ions tested. The Syn1A-channel interaction, unlike the synaptotagmin-channel interaction, proved significantly more sensitive to the type of permeating ion. It is well established that exocytosis is affected by switching the charge carriers. Based on the present results, we suggest that the channel-Syn1A interaction could respond to the conformational changes induced within the channel during membrane depolarization and divalent ion binding. These changes could partially account for the charge specificity of synaptic transmission as well as for the fast signaling between the Ca(2+) source and the fusion apparatus of channel-associated-vesicles (CAV). Furthermore, propagation of conformational changes induced by the divalent ions appear to affect the concerted interaction of the channel with the fusion/docking machinery upstream to free Ca(2+) buildup and/or binding to a cytosolic Ca(2+) sensor. These results raise the intriguing possibility that the channel is the Ca(2+) sensor in the process of fast neurotransmitter release.

Presynaptic facilitation of synaptic transmission in the hippocampus

Biochemical and genetic characterization of proteins in presynaptic axon terminals have led to models of the biochemical pathways underlying synaptic vesicle docking, activation, and fusion. Several studies have attempted recently to assign a precise physiological role to these proteins. This review deals with some of these studies, concentrating on those performed with hippocampal synapses. It is shown that changes in the state of these presynaptic proteins, together with modifications in Ca2+ dynamics in axon terminals, functionally determine the level of basal synaptic transmission, and underlie pharmacologically induced and activity-dependent facilitation of transmitter release in the central nervous system.

A comparison between potencies of external calcium, strontium and barium to support GABAergic synaptic transmission in rat cultured hippocampal neurons

Relative potencies of external Ca2+, Sr2+ and Ba2+ to trigger GABAergic synaptic transmission were evaluated by applying the patch-clamp technique to both presynaptic and postsynaptic hippocampal neurons prepared from neonatal rats. Action potentials were evoked by application of voltage pulses to presynaptic neurons, and GABAergic synaptic currents were recorded in voltage-clamped postsynaptic neurons. No stimuli were delivered during replacement with test solutions and only five pulses were applied to the presynaptic neuron in each test solution. During the five-pulse application, the amplitude of synaptic currents was constant in Ca(2+)-containing solutions, but decreased successively in Ba(2+)- and Sr(2+)-containing solutions without Ca2+. Thus, the amplitude of synaptic currents induced by the first pulse in each ionic condition was used to evaluate the potency of divalent cations. The lowest external concentration required to trigger the transmission was 0.3 mM for Ca2+, 1 mM for Sr2+ and 2 mM for Ba2+, and the concentration required to achieve the same effect as with 2 mM Ca2+ was 6 mM for Sr2+ and 10 mM for Ba2+. These results strongly suggest that Ba2+ as well as Sr2+ can be substituted for Ca2+ in GABAergic synaptic transmission and the order of potency is Ca2+ > Sr2+ > Ba2+.

Adenosine-induced suppression of synaptic responses and the initiation and expression of long-term potentiation in the CA1 region of the hippocampus

The results of several previous studies have suggested that pretreatment with adenosine can block the induction of long-term potentiation (LTP), although other studies have found no effect of adenosine on the induction of LTP. The interaction of adenosine with the induction of LTP in the rat hippocampal slice was investigated. Inhibition of synaptic responses by adenosine either prior to or immediately after high-frequency or theta-burst stimulation did not affect LTP measured after washout of the adenosine. The only conditions under which adenosine blocked the development of LTP was when it was given 3-5 minutes prior to the stimulation train. To understand how it was possible to induce LTP, during the period 1-3 minutes following adenosine when synaptic responses were virtually eliminated, evoked responses during the 100 Hz stimulation train were recorded. Although synaptic responses to low-frequency stimulation were virtually eliminated by adenosine, they reappeared during high-frequency stimulation. These results suggest that although adenosine can depress synaptic responses, an increase in neurotransmission during a high-frequency train can partially overcome this effect of adenosine, and the hypothesis that adenosine can selectively block LTP is not supported.

The EPSP-spike (E-S) component of long-term potentiation in the rat hippocampal slice is modulated by GABAergic but not cholinergic mechanisms

Long-term potentiation of synaptic efficacy (LTP) can be shown to consist of two components: a synaptic and an excitatory postsynaptic potential (EPSP)-spike (E-S) component. The E-S component is expressed as a leftward shift in the curve relating population spike amplitude as a function of EPSP slope. The participation of cholinergic and GABAergic processes in E-S potentiation was studied in field CA1 of rat hippocampal slices. Atropine, a muscarinic antagonist, did not prevent tetanus-induced E-S potentiation. The cholinergic agonist carbachol and the GABAA antagonist picrotoxin produced a leftward shift in the E-S relation; picrotoxin, but not carbachol, prevented the expression of tetanus-induced E-S potentiation. These observations indicate that an increase in the ratio of evoked excitation to inhibition and/or a reduction in tonic inhibition mediated by the activation of GABAA receptors contribute to E-S potentiation produced by high-frequency stimulation.

Abnormal aluminium, cobalt, manganese, strontium and zinc concentrations in untreated epilepsy

The concentration of 38 trace and bulk elements in the serum from 19 patients with recent onset of epilepsy and 20 age- and sex-matched controls was estimated by neutron activation analysis or inductively coupled plasma source by mass spectrometry. The concentrations of aluminium, strontium and zinc were significantly higher and the concentrations of cobalt and manganese were significantly lower than controls. Low concentrations of manganese and high concentrations of zinc in epilepsy have been previously reported but the abnormalities of aluminium, cobalt and strontium are new findings. The possible significance of these results in the pathogenesis of epilepsy is discussed.

Long-term potentiation involves enhanced synaptic excitation relative to synaptic inhibition in guinea-pig hippocampus

1. Tetanization of hippocampal pyramidal cell afferents travelling in stratum radiatum of area CA1 induces both long-term potentiation (l.t.p.) of extracellularly recorded excitatory postsynaptic potentials (e.p.s.p.s), and an increase in the number of cells firing, as measured by the extracellular population spike, for a given sized field e.p.s.p. The mechanism of this latter change, known as e.p.s.p.-spike (E-S) potentiation, was investigated in the guinea-pig hippocampal slice preparation. 2. Plots of the E-S relation before and after tetanization were constructed from measures taken over a series of stimulus strengths. Tetanization of afferents in stratum radiatum decreased the spike threshold by 24%, while the gamma-aminobutyric acid antagonist picrotoxin (PTX) decreased spike threshold by 72%. Sequential administration of PTX and tetanization, in either order, resulted in no more change in the E-S threshold than did PTX application alone. 3. Extracellular synaptic potentials, matched for initial slope before and after tetanization by adjusting the stimulus strength, showed an increased peak amplitude and increased peak latency following tetanization. PTX produced similar but larger percentage changes. Tetanization in the presence of PTX, however, did not alter the field potential wave shape. 4. Intracellular postsynaptic potentials (p.s.p.s) were also matched for initial slope before and after tetanization. Tetanization induced p.s.p. shape changes similar to those observed extracellularly, i.e. in the direction of less inhibition. Such changes did not occur in the presence of PTX. 5. Inhibitory p.s.p.s (i.p.s.p.s) were studied in depolarized pyramidal cells with microelectrodes filled with QX-314. Tetanization of afferents in stratum radiatum produced i.p.s.p. increases in eight of nineteen cells. These increases were generally attributable to an increased activity in the recurrent inhibitory pathway. Tetanization of the alveus failed to produce any lasting increases in the i.p.s.p. amplitude. 6. Tetanization of afferents in stratum radiatum decreased the ratio of the intracellular i.p.s.p. to field e.p.s.p. over stimulus strengths below population spike threshold. Above population spike threshold, the ratio tended towards its pretetanization level. 7. The results indicate that E-S potentiation results from an increase in the level of depolarization reached by a synaptic potential of given initial slope. These findings support the hypothesis that tetanization induces greater l.t.p. of excitatory inputs onto pyramidal cells than of inputs onto feed-forward inhibitory interneurones.

Heterosynaptic changes accompany long-term but not short-term potentiation of the perforant path in the anaesthetized rat

Brief high-frequency trains of electrical stimulation delivered to the perforant path result in long-term potentiation (l.t.p.) of field potentials recorded extracellularly from granule cells of the dentate gyrus. L.t.p. of the population spike is often disproportionately greater than l.t.p. of the population excitatory post-synaptic potential (e.p.s.p.). We have investigated the basis of this effect in rats anaesthetized with sodium pentobarbitone. A series of graded stimuli were given before and after tetanization of the perforant path. From data obtained in this way, we plotted stimulus-response curves, and the relation (E-S curve) between the slope of the population e.p.s.p. (E) and the amplitude of the population spike (S). Curves relating spike onset latency to the slope of the e.p.s.p. were also constructed. Tetanization of the combined medial and lateral components of the perforant path led to long-term changes in the relation between the e.p.s.p. and the population spike. For a given e.p.s.p., the corresponding population spike was of greater amplitude and earlier onset. This E-S potentiation was marked by a shift to the left of the E-S amplitude curve and a downward displacement of the E-S latency curve. Tetanization of the lateral component of the perforant path had two long-term effects on responses evoked by test stimuli to the untetanized medial component: (1) long-term depression of the medial e.p.s.p. and (2) long-term E-S potentiation. The net result of these two heterosynaptically induced effects was to leave unaltered information transfer across medial perforant path-granule cell synapses; for a given test volley the e.p.s.p. was smaller, but because of E-S potentiation the population spike remained relatively unaffected. Short-term potentiation, which has a time course of only a few minutes and is presumed to be mediated by presynaptic mechanisms, was not accompanied by E-S potentiation or by corresponding changes in spike latency. Possible mechanisms of long-term heterosynaptic depression of the e.p.s.p. and of homo- and heterosynaptic E-S potentiation, are discussed. We conclude that although these effects probably reflect a generalized post-synaptic change, this change is unlikely to be a prolonged reduction in the membrane potential of granule cells.

Effects of calcium, strontium, and barium ions on phosphorylation of hippocampal proteins in vitro

Calcium ion alone or in the presence of added calmodulin stimulated in vitro transfer of 32P from [gamma 32P]ATP into several proteins of mitochondrial and synaptosomal particulate fractions from rat brain. Strontium ion was capable of substituting for calcium ion in this stimulation, but barium ion lacked this capacity. These results bring into question the hypothesis that calcium-dependent protein phosphorylation of synaptic proteins is intrinsic to neurotransmitter release during neurotransmission, but they do not rule out that possibility.

Long-term potentiation in the hippocampus

Long-term potentiation of field and single neuronal responses recorded in various hippocampal fields is described on the basis of author's and literary data. Most of intrahippocampal and extrinsic connections in both in vivo and in vitro hippocampal preparations show this phenomenon after one or several conditioning trains of comparatively short duration (20 s or less) at various frequencies (from 10 to 400 Hz). Properties of hippocampal potentiation are described. The properties include long term persistence (hours and days) of the potentiated response, its low frequency depression, self-restoration after the depression, specificity of the potentiation for the tetanized pathway, necessity of activation of a sufficient number of neuronal elements ('cooperativity') to produce the potentiation, possible involvement of 'reinforcing' brain structures during conditioning tetanization. These properties are distinct from those of 'usual' short-term post-tetanic potentiation and lead to the suggestion that the neuronal mechanisms underlying long-term post-tetanic are similar to those underlying memory and behavioral-conditioned reflex. Neurophysiological mechanisms of long-term potentiation are discussed. The main mechanism consists in an increase in efficacy of excitatory synapses as shown by various methods including intracellular recording and quantal analysis. The latter favours presynaptic localization of changes of synaptic efficacy showing increase in the number of transmitter quanta released per presynaptic impulse. However, changes in the number of subsynaptic receptors or localized changes in dendritic postsynaptic membrane are not excluded. Biochemical studies indicate the increase in transmitter release and calcium-dependent phosphorylation of pyruvate dehydrogenase after tetanization. Instances of persistent response facilitations at other levels of the vertebrate central nervous system (especially at neocortical level) are considered and compared with hippocampal long-term potentiation. It is suggested that modifiable excitatory synapses necessary for learning have been identified in studies of long-term potentiation. These synapses are presumably modified as a result of close sequential activation of the following three structures: excitatory presynaptic fibers, the postsynaptic neuron and a 'reinforcing' brain system.

Asymmetric relationships between homosynaptic long-term potentiation and heterosynaptic long-term depression

All synaptically-based neuropsychological theories of learning postulate that there are changes resulting from neural activity which are long-lasting and confined to specific sets of synapses. In the past decade a form of synaptic strengthening known as long-term potentiation (LTP) has been found which results from high-frequency neural activity and is of sufficient duration to model as a learning mechanism. Some early tests of the synaptic specificity of LTP in area CA1 of the hippocampus indicated that although LTP was specific to the tetanized pathway, in a converging untetanized pathway it was associated with depression of synaptic transmission lasting for at least 30 min. However, others have found that this heterosynaptic depression more usually decays within 5-15 min post-tetanus despite the maintenance of LTP in the tetanized pathway. Similarly, in the dentate gyrus (DG), LTP of either the lateral (LPP) or medial (MPP) components of the perforant path afferents has been associated with only short-lasting reciprocal heterosynaptic depression. Here, using more detailed measurement of stimulus intensity curves, we report that tetanization of either MPP or LPP reliably depresses synaptic transmission in the other pathway for at least 3 h. This heterosynaptic depression, considerably smaller than the usual magnitude of LTP, was obtained regardless of whether LTP had been produced in the tetanized homosynaptic pathway. Heterosynaptic long-term depression was not observed if the test pathway had been previously tetanized.

Modulation of long-term potentiation: effects of adrenergic and neuroleptic drugs

A variety of drugs which either mimic or antagonize the effects of norepinephrine and dopamine were tested for their ability to modulate long-term potentiation (LTP) in the rat hippocampus in vitro. Neither administration of norepinephrine, amphetamine or adrenergic antagonists, nor pretreatment with reserpine or DSP4 (which selectively destroys noradrenergic afferents to the hippocampus) had any significant effect on the magnitude of LTP. Isoproterenol, a beta-adrenergic receptor agonist, was able to partially block LTP, but this did not appear to be due to a direct action of isoproterenol on LTP. Neuroleptic drugs such as trifluoperazine were able to block LTP almost completely; however, this action was not stereospecific. Other dopamine antagonists such as sulpiride had no effect on LTP. The ability of neuroleptics to antagonize LTP was more closely related to their ability to block calmodulin than to their relative potencies as dopamine antagonists. It would appear that neither norepinephrine nor adrenergic antagonists influence the amount of LTP elicited by repetitive stimulation; however, drugs which have been shown to interfere with calmodulin-mediated cellular processes do antagonize this phenomenon.

Increased excitability of hippocampal unmyelinated fibres following conditioning stimulation

Unmyelinated fibres in the stratum radiatum (area CA1) were electrically stimulated in hippocampal slices from guinea pigs. The size of the evoked fibre volley was increased by a preceding conditioning stimulation to the same fibres, provided that the strength of the conditioning stimulus was larger than that of the test pulse. This facilitation was maximal for 20-30 ms interstimulus intervals and had a duration of about 200 ms. The results are compatible with an increased excitability (reduction in threshold) specific for fibres activated by the conditioning stimulation.

Two types of synaptic facilitation recorded in pyramidal cells of in vitro hippocampal slices from guinea pigs

PyramidaL cells in CA1 hippocampus show two types of increase in the excitatory postsynaptic potential (EPSP) during paired pulse stimulation of the afferent fibres in stratum radiatum. For 15-300 ms interstimulus intervals, both the initial slope and the peak amplitude of the EPSP are increased, which is mainly due to an increased excitatory synaptic current. For 300 ms-2 s intervals, the predominant change is an increase in the peak amplitude. The latter type of facilitation is caused by a decrease in a superimposed inhibitory postsynaptic potential (IPSP).

Hippocampal glutamate receptors

For years, the hippocampus has been the privileged domain of anatomists and electrophysiologists for investigating various neurobiological processes. The present review deals with recent work which shows that this structure is also well suited to study the role of glutamate as a neurotransmitter and more particularly the characteristics of glutamate receptors and their possible involvement in hippocampal function. After a brief description of the main anatomical features of the hippocampus, we attempt a critical evaluation of the electrophysiological studies of hippocampal glutamate receptors. We then describe the properties of Na-independent 3H-glutamate binding sites in hippocampal membranes, and discuss the possibility that these binding sites are related to postsynaptic glutamate receptors. Finally we show that these binding sites are extremely labile and that hippocampal membranes possess various mechanisms which regulate their number. In particular we develop the idea that the calcium-stimulation of 3H-glutamate binding in hippocampal membranes may be the mechanism by which electrical activity regulates the number of glutamate receptors at hippocampal synapses and thus induces long-lasting changes in synaptic transmission.