The action of acetyl-L-carnitine on the neurotoxicity evoked by amyloid fragments and peroxide on primary rat cortical neurones.

The amyloid beta-peptides have been implicated in the excitotoxic mechanism of neuronal injury in the pathogenesis of Alzheimer's disease. In this paper we examine the effect of different amyloid fragments (beta A1-40, A1-28, and A25-35), as well as potential neuroprotective compounds on rat cortical neuron viability. Exposure of neurones to beta A25-35 or A1-40 at concentrations as low as 1 microgram/ml inhibited, significantly, the MTT response and this level of inhibition was similar after 24-h or three-day exposure. Furthermore, the level of inhibition was not affected by the presence or absence of 5% horse serum in the medium. Preexposure (10 min) of neurones to ALC at concentrations of 0.1, 1, 5, and 10 mM attenuated the inhibition of the MTT response caused by beta A25-35 (50 micrograms/ml) in serum free medium for 24 h. The treatment of cells with vitamin E (100 microM), catalase (4 mg/ml), NGF (0.1 and 10 ng/ml), or cycloheximide (0.1 microgram/ml) significantly restored the MTT response that was inhibited by beta A25-35. The mechanism for the protective actions of these compounds against beta A25-35 toxicity is not clear but may involve free radical scavenger action and preservation of energy production, although other mechanisms, especially for ALC, such as a direct effect on A-beta interaction with charged anionic phospholipids and/or stabilizing action on membranes, are also possible.

Structural, metabolic and ionic requirements for the uptake of L-carnitine by primary rat cortical cells.

L-Carnitine (L-C) is involved in the transport of acyl groups into mitochondria for beta-oxidation, although its role in the adult brain is still uncertain. We have shown before that the uptake of L-carnitine into cultured rat cortical neurones was dependent on temperature as well as the Na gradient and is inhibited by compounds resembling its structure, like gamma-aminobutyric acid (GABA), but most potently by specific GABA uptake blockers. In this study we have characterised this uptake process further. We have shown that the uptake of L-carnitine may be dependent on Cl ions, in addition to Na ions, but non on Ca ions. The L-C uptake was inhibited by substituent anions in the order gluconate (83%) > isethionate (32%), with propionate being ineffective, whereas GABA uptake was inhibited most potently by propionate substitution (79%) and equally by isethionate and gluconate (67%). This L-C uptake process was not affected by the amino acids, glutamine or lysine, up to 1 mM concentration, although beta-alanine at 500 microM caused a 38% inhibition. The uptake of L-C was also significantly inhibited by structurally-related compounds, with a carbon chain length of three to six atoms, possessing an amine group and/or a carboxyl group. At a concentration of 500 microM, 3-aminopropane sulphonic acid (53%), gamma-butyrobetaine (31%), gamma-hydroxybutyric acid (34%) and 4 methylaminobutyric acid (33%). Other compounds were effective only at the lower concentration of 10 microM, such as butyric acid (25%), nicotinic acid (26%), isonicotinic acid (26%), hexanoic acid (23%) and at 100 microM, like 6-aminocapric acid (22%). Drugs suggested to affect membrane properties, such as chlorpromazine, was without effect at 1 or 10 microM, whereas flunarizine (FLU) at 1 microM inhibited both L-C (24%) and GABA uptake (17%). Other drugs like the cholinesterase inhibitors, tacrine and eserine, also had a small inhibitory effect on L-C uptake, reducing it at 1 microM by 22 and 21% respectively, although higher concentrations were toxic (> 100 microM). Pretreatment of the cells with neuraminidase (50 U ml-1, 10 min) reduced the subsequent uptake of both L-C (18%) and GABA (42%). Hypoxia (3 h) also significantly attenuated L-C uptake (42%), however part of these effects were related to the loss of cell viability. In summary, L-C uptake occurs by a complex mechanism which at least in part may occur by a Na/Cl cotransport mechanism, which could be similar, to that of GABA or may even in part occur via the GABA transporter.

Protective actions of L-carnitine and acetyl-L-carnitine on the neurotoxicity evoked by mitochondrial uncoupling or inhibitors.

The mechanism for the pathological increase in cell death in various disease states e.g. HIV immunodefficiency or even ageing or Alzheimer's disease, occurs by complex and as yet undefined mechanism(s) related to immunological, virological or biochemical disturbances (i.e. energy depletion, oxidative stress, increased protein degradation). We have studied mitochondrial uncoupling or inhibitor toxicity on neurones at the cellular level and at the mitochondrial level using rhodamine (Rh123) and 10-nonylacridine orange (NAO) fluorescence with confocal microscopy. Blockade of the mitochondrial chain complexes at various points was studied. The possible protective effects of the compound L-carnitine, which plays a central role in mitochondrial function, was tested in this form of neurotoxicity. It appears that L-carnitine and its acetylated form, acetyl-L-carnitine, can attenuate the cell damage, as assessed by lactate dehydrogenase (LDH) release, evoked by the uncoupler, p-(trifluoromethoxy)phenylhdyrazone (FCCP), or by the inhibitors, 3-nitropropionic acid (3-NPA) or rotenone. Further, the FCCP-induced inhibition of Rh123 uptake was antagonized by the preincubation of cells with L-carnitine. Since such neurotoxic mechanisms may be operating in the various pathological forms of myotoxicity and neurotoxicity, these observations suggest potential for a therapeutic approach.

Inhibition of L-carnitine uptake into primary rat cortical cell cultures by GABA and GABA uptake blockers.

L-carnitine plays a central role in mitochondrial function and is found to be differentially distributed in the brain. We have shown before that the uptake of L-carnitine into cultured rat cortical neurones was temperature-dependent, as well as potently inhibited by factors affecting the sodium gradient as well as by molecules resembling its structure, e.g. D-carnitine, acetyl-L-carnitine and gamma-aminobutyric acid (GABA). GABA was the most potent inhibitor of L-carnitine uptake. In the present study we have found that specific GABA uptake blockers, nipecotic acid, cis-4-hydroxynipecotic (HNA), guvacine, 2,4-diaminobutyric acid (DABA) and NO 711 inhibit L-carnitine uptake even more potently than GABA. However, apart from NO 711, they caused about the same maximal inhibition, 67.4% at 50 microM for guvacine, compared to 60.5% by GABA. NO 711 was extremely potent and blocked 80.5% of the L-carnitine uptake. In contrast, the GABAA receptor agonists, isonipecotic acid and isoguvacine, or the antagonist bicuculline, at similar concentrations (50 microM), did not significantly inhibit the uptake of the L-carnitine. However, bicuculline at relatively high concentration (500 microM) was inhibitory (38%). The GABAB receptor agonist, baclofen, or antagonist, phaclofen, were ineffective, although 5-aminovaleric acid did significantly inhibit uptake at both 50 and 500 microM, causing 22 and 48% inhibition respectively. Like bicuculline, it was not as effective as GABA or the specific GABA uptake blockers. The results indicate that the uptake of L-carnitine by rat cortical neurones occurs in part by a process that can be potently inhibited by GABA and GABA uptake blockers.

L-carnitine uptake into primary rat cortical cultures: interaction with GABA.

The ability of the primary rat cortical cells to take up L-carnitine increased with the age of the cultures and plateaued at around day 11 up to 25 days in vitro (DIV) when a slight decline was evident and by 32 DIV there was a major decrease in L-carnitine uptake. The uptake of L-carnitine displayed complex components. Elimination of mitochondrial energy supply by NaCN (1 mM), rotenone (1.25 microM) and DNP (50 microM), caused a small but significant decrease in the uptake (21, 11 and 16%, respectively). The uptake was highly dependent on the Na gradient, since ouabain (0.5 mM) and Na free buffer (replaced by 250 mM sucrose), reduced uptake by 54 and 63%, respectively. There was competition of L-carnitine uptake by molecules resembling its structure, e.g. gamma-aminobutyric acid (GABA), acetyl-L-carnitine (ALC), D-carnitine, L-aminocarnitine and L-choline, with GABA being the most potent inhibitor (57% at 50 microM) and L-choline not being significantly active. The Na-dependent uptake of L-carnitine was saturable with a high Km (692 microM) and Vmax (839 pmol/min/mg). This Na-dependent component was not further additive with the GABA (500 microM) or the DNP (50 microM) inhibitable component, suggesting that it represented the same phenomenon, probably the Na gradient dependent transport of L-carnitine. The results indicate that the uptake of L-carnitine occurs by Na-dependent saturable process as well as non-saturable, Na-independent processes. At least the former uptake mechanism is potently inhibited by GABA.

Stimulation of gonadotropin-releasing hormone secretion by acetyl-L-carnitine in hypothalamic neurons and GT1 neuronal cells.

Pulsatile gonadotropin-releasing hormone (GnRH) secretion from perifused hypothalamic cells and GT1-1 neuronal cells was significantly increased after culture in medium containing 100 microM acetyl-L-carnitine (ALC). This action of ALC was largely due to an increase in the spike amplitude of GnRH release. In addition, the receptor-mediated release of GnRH by N-methyl-D-aspartic acid and endothelin was significantly increased in perifused cells cultured in ALC-enriched medium. Stimulatory effects of ALC on basal, high K(+)- and agonist-induced GnRH release were also observed during long-term culture of primary hypothalamic neurons. Similar effects of ALC were evident in cultured GT1-1 cells and were accompanied by a significant increase in cell number. These observations in normal and transformed GnRH neurons demonstrate that ALC promotes the growth and secretory activity of neuropeptide-producing cells of the hypothalamus.

Actions of acetyl-L-carnitine on the hypothalamo-pituitary-gonadal system in female rats.

Acetyl-L-carnitine (ALC) is known to affect several aspects of neuronal activity. To evaluate the neuroendocrine actions of this compound, several endocrinological parameters were followed in ALC-treated and control animals during recovery from dark-induced anestrus. In treated animals, serum luteinizing hormone (LH) and prolactin levels were higher than those of controls during the proestrous and estrous phases of the cycle, and serum estradiol levels were higher during estrus. No significant changes were observed in serum levels of follicle-stimulating hormone and progesterone. Uterine weight was increased in ALC-treated rats during proestrus and estrus, but not in diestrus. The basal release of gonadotropin-releasing hormone (GnRH) from perifused hypothalamic slices of ALC-treated animals was elevated at proestrus and diestrus, and GnRH release elicited by high K+ was higher during all three phases of the cycle. The basal release of LH from perifused pituitaries of treated animals was elevated in diestrus, and the LH response to GnRH was higher in estrus and diestrus I. Depolarization with K+ caused increased LH secretion during proestrus and estrus in treated animals. In contrast to these effects of ALC treatment in vivo, no direct effects of ALC were observed during short- or long-term treatment of cultured pituitary cells. These results indicate that ALC treatment influences hypothalamo-pituitary function in a cycle stage-dependent manner, and increases the secretory activity of gonadotrophs and lactotrophs. Since no effects of ALC on basal and agonist-induced secretory responses of gonadotrophs were observed in vitro, it is probable that its effects on gonadotropin release are related to enhancement of GnRH neuronal function in the hypothalamus.

Calcium signaling and episodic secretion of gonadotropin-releasing hormone in hypothalamic neurons.

Gonadotropin-releasing hormone (GnRH) is released episodically into the pituitary portal vessels and from hypothalamic tissue of male and female rats in vitro. Perifused primary cultures of rat hypothalamic neurons, as well as the GT1-1 GnRH neuronal cell line, spontaneously exhibited episodic GnRH secretion of comparable frequency to that observed with perifused hypothalami. Such pulsatile GnRH release from GT1 cells indicates that GnRH neurons generate rhythmic secretory activity in the absence of input from other cell types. In primary hypothalamic cultures, the frequency of GnRH pulses increased with the duration of culture. The spontaneous pulsatility in GnRH release was abolished in Ca(2+)-deficient medium and was markedly attenuated in the presence of nifedipine, an antagonist of voltage-sensitive Ca2+ channels. The basal intracellular Ca2+ level of perifused GT1-1 cells cultured on coverslips was also dose-dependently reduced by nifedipine. Conversely, depolarization with high K+ increased intracellular Ca2+ and GnRH release in an extracellular Ca(2+)-dependent and nifedipine-sensitive manner. The dihydropyridine Ca2+ channel agonist Bay K 8644 increased basal and K(+)-induced elevations of intracellular Ca2+ concentration and GnRH secretion. These findings demonstrate that pulsatile neuropeptide secretion is an intrinsic property of GnRH neuronal networks and is dependent on voltage-sensitive Ca2+ influx for its maintenance.

Dependence of hormone secretion on activation-inactivation kinetics of voltage-sensitive Ca2+ channels in pituitary gonadotrophs.

The relationships between the activation status of voltage-sensitive Ca2+ channels and secretory responses were analyzed in perfused rat gonadotrophs during stimulation by high extracellular K+ concentration ([K+]e) or the physiological agonist, gonadotropin-releasing hormone (GnRH). Increase of [K+]e to 50 mM evokes an on-off secretory response, with a rapid rise in luteinizing hormone (LH) secretion to a peak at 35 sec (on response) followed by an exponential decrease to the steady-state level. Cessation of K+ stimulation elicits a transient (off) response followed by an exponential decrease to the basal level. The LH response to high [K+]e is nifedipine-sensitive and its amplitude depends on membrane potential. There is a close relationship between the LH secretory response to high [K+]e and the amplitude of the inward Ca2+ current measured at 100 msec in whole-cell patch clamp experiments. In addition, the profile of the LH secretory response is similar to that of the response of intracellular Ca2+ concentration ([Ca2+]i) in K(+)-stimulated cells. In Ca2(+)-deficient medium, the effect of high [K+]e is abolished; subsequent elevation of [Ca2+]e during the K+ pulse is followed by restoration of the on response, but with reduced magnitude. Agonist stimulation during the steady-state phase of the [K+]e pulse or after repetitive stimulation by high [K+]e elicited biphasic [Ca2+]i and secretory responses with a significantly reduced plateau phase; conversely, K(+)-induced LH release was reduced in cells treated with desensitizing doses of GnRH. These findings indicate that depolarization-induced changes in the status of voltage-sensitive Ca2+ channels determine the profiles of [Ca2+]i and LH responses to stimulation by high [K+]e; the initial activation of dihydropyridine-sensitive Ca2+ channels is clearly dependent on membrane potential, whereas their subsequent inactivation depends on increased [Ca2+]i. Such inactivation of voltage-sensitive Ca2+ channels also occurs during GnRH action and may represent an additional regulatory mechanism to limit the entry of extracellular Ca2+ during prolonged or frequent agonist stimulation.

Stimulation of luteinizing hormone release by gamma-aminobutyric acid (GABA) agonists: mediation by GABAA-type receptors and activation of chloride and voltage-sensitive calcium channels.

The mechanism by which gamma-aminobutyric acid (GABA) stimulates the release of LH was analyzed in cultured female rat pituitary cells. In 3-h incubations, GABA (1-100 microM) caused a dose-dependent increase in LH release, with the maximal response about 16% of that evoked by 10 nM GnRH. GABA action was independent of the GnRH receptor, since 1 microM GnRH antagonist [( N-acetyl-D-p-Cl-Phe1,2,D-Trp3,D-Lys6,D-Ala10] GnRH), which completely inhibits GnRH action, did not affect the response to GABA. In studies on the effects of GABA receptor agonists and antagonists, 4,5,6,7-tetrahydoisoxazolo-[5,4-c]pyridin-3(2H)-one (THIP) and muscimol (GABAA agonists) gave similar response patterns, with the same maximal stimulation as GABA but much higher potencies. In contrast, the GABAB receptor agonist baclofen did not stimulate LH release. The GABAA receptor antagonist SR95531 caused dose-dependent inhibition of the LH-releasing effects of GABA and muscimol (10 microM), with complete blockade at 10 microM SR95531. T-Butylbicyclophosphorothionate, an inhibitor of the GABAA receptor-associated chloride channel, also dose-dependently reduced the releasing effect of 100 microM GABA. These results indicate that GABA action is mediated by the chloride channel-associated GABAA receptor. However, the other GABAA receptor antagonists, including bicuculline, picrotoxin, and strychnine, did not attenuate the LH-releasing effect of 100 microM GABA in concentrations up to 100 microM, suggesting that GABA action is mediated by nonclassical GABAA receptors. Incubation in the presence of nifedipine (1 microM) or in calcium-free medium inhibited the LH-releasing action of GABA, indicating that calcium influx through voltage-sensitive calcium channels (VSCC) is required for GABA-induced LH release. Such entry of Ca2+ would result from activation of VSCC by depolarization due to the increased Cl- conductance caused by GABAA receptor activation. In cell perfusion studies, the actions of GABA and muscimol were attenuated or abolished after repetitive stimulation, consistent with desensitization of the GABA receptors. These findings have demonstrated that the stimulation of LH release by GABA is independent of GnRH action, occurs via binding to nonclassical GABAA receptors, which rapidly desensitize, and is mediated by the activation of VSCC.

Generation and amplification of the cytosolic calcium signal during secretory responses to gonadotropin-releasing hormone.

Gonadotropin-releasing hormone (GnRH) stimulates characteristic biphasic increases in cytosolic calcium concentration ([Ca2+]i) and in luteinizing hormone (LH) release in cultured gonadotrophs, with an early peak followed by a prolonged plateau in both responses. Analysis of [Ca2+]i by dual-wavelength fluorimetric assay and of LH release at 5-sec intervals in perifused pituitary cells revealed increases in both responses within a few seconds of exposure to GnRH. The maximum elevation of [Ca2+]i occurred within 20 sec, and the peak gonadotropin release in 35 sec; the total duration of the spike phase for both [Ca2+]i and LH release was 2.5 min. Under extracellular Ca2(+)-deficient conditions, the GnRH-induced peak in [Ca2+]i was reduced by about 20% and the plateau phase was abolished. Concomitantly, the magnitude of the acute phase of LH release was reduced by 40% and that of the second phase by about 90%. Recovery of the plateau phase of LH release occurred within 25 sec after addition of 1.25 mM Ca2+ to Ca2(+)-deficient medium. In a dose-dependent manner, the non-selective Ca2+ channel blockers Co2+ and Cd2+ reduced the Ca2+ current measured by whole-cell recording in pituitary gonadotrophs and abolished the extracellular Ca2(+)-dependent component of LH release. The selective calcium channel blocker, nifedipine, decreased the magnitude of the Ca2+ current and reduced the plateau phase of LH release by 50%; conversely, the dihydropyridine agonist methyl, 1,4,dihydro-2,6-dimethyl 3-nitro-4-(2-trifluorome) (Bay K 8644) consistently enhanced the amplitudes of both Ca2+ current and GnRH-induced LH release. These data reveal a close temporal correlation between changes in [Ca2+]i and LH release during GnRH action, with Ca2+ mobilization during the spike phase and Ca2+ influx through dihydropyridine-sensitive and insensitive sets of receptor-operated calcium channels during the spike and plateau phases. In addition, analysis of the magnitudes of the [Ca2+]i and LH responses to a wide range of GnRH concentrations in the presence and absence of extracellular Ca2+ is consistent with amplification of the [Ca2+]i signal in agonist-stimulated gonadotrops.

Efflux of putative transmitters from superfused rat brain slices induced by low chloride ion concentrations.

Slices of rat cerebral cortex, preloaded with [14C]gamma-aminobutyric acid (GABA) and either [3H]5-hydroxytryptamine (5-HT) or [3H]noradrenaline, were superfused with media in which varying concentrations of Cl- had been replaced with other monovalent anions. Rapid reduction of [Cl-], by superfusion with media containing instead the impermeant anions propionate, isethionate, gluconate, or methyl sulphate, caused increases in the efflux of tritiated biogenic amines, but the increase in that of [14C]-GABA was not significant. The increased efflux of [3H]5-HT evoked by superfusion with low Cl- levels when propionate was the replacement anion, was transient and was linearly related to the log[Cl-]-1. It was not affected by removal of Ca2+ or by addition of 10 mM Mg2+ and was delayed but not abolished by tetrodotoxin. The low Cl(-)-evoked efflux of [3H]5-HT was not affected by pretreatment with neuronal reuptake blockers but was inhibited by picrotoxin, strychnine, and 4-acetamido-4-isothiocyanostilbene-2,2-disulphonic acid and was enhanced by glycine. Muscimol and GABA were without effect. These observations are taken to indicate that the efflux of biogenic amines is brought about by terminal depolarisation due to outward movement of Cl- in low chloride-containing media. They are of relevance to other physiological and pharmacological studies in which anion concentrations are manipulated and suggest that the anion-evoked release phenomenon may provide a model for the analysis of Cl(-)-dependent mechanisms in nerve terminals.

Peptidases involved in the catabolism of neurotensin: inhibitor studies using superfused rat hypothalamic slices.

In order to identify which peptidases are involved in the catabolism of neurotensin in the CNS, [3H-Tyr3,11]-neurotensin was superfused over rat hypothalamic slices in the presence and absence of peptidase inhibitors. The degree of degradation of the peptide was determined by reverse phase HPLC separation of 3H-labelled neurotensin from 3H-labelled products. Very little degrading activity was released from the slice into the medium during the superfusion. In the absence of inhibitors, 20 to 50% of 3H-neurotensin was degraded giving mainly 3H-Tyr along with other unidentified 3H-labelled products. Inhibitors of endopeptidase 24.11 (phosphoramidon) and proline endopeptidase (antibody) had no effect on the degradation. Captopril, an inhibitor of angiotensin converting enzyme, had a small inhibitory effect. In contrast, dynorphin(1-13), an inhibitor of a soluble, thiol dependent metallopeptidase which hydrolyses neurotensin at Arg8-Arg9, gave greater than 80% inhibition of 3H-neurotensin degradation in the slice preparation. 1,10-Phenanthroline, an inhibitor of metallopeptidases, was also an effective inhibitor. The dynorphin sequence responsible for the inhibition contains the Arg6-Arg7 bond. Other peptides (bradykinin and angiotensin) which are substrates of the soluble metallopeptidase also inhibited neurotensin breakdown by the slice. This evidence suggests that this thiol dependent metalloendopeptidase is the major neurotensin catabolizing enzyme in hypothalamic slices.