Melittin enhances amino acid and free fatty acid release from the in vivo cerebral cortex

Applications and evolution of melittin, the quintessential membrane active peptide

Melittin, the main venom component of the European Honeybee, is a cationic linear peptide-amide of 26 amino acid residues with the sequence: GIGAVLKVLTTGLPALISWIKRKRQQ-NH₂. Melittin binds to lipid bilayer membranes, folds into amphipathic α-helical secondary structure and disrupts the permeability barrier. Since melittin was first described, a remarkable array of activities and potential applications in biology and medicine have been described. Melittin is also a favorite model system for biophysicists to study the structure, folding and function of peptides and proteins in membranes. Melittin has also been used as a template for the evolution of new activities in membranes. Here we overview the rich history of scientific research into the many activities of melittin and outline exciting future applications.

Activation of central orexin/hypocretin neurons by dietary amino acids

Hypothalamic orexin/hypocretin (orx/hcrt) neurons regulate energy balance, wakefulness, and reward; their loss produces narcolepsy and weight gain. Glucose can lower the activity of orx/hcrt cells, but whether other dietary macronutrients have similar effects is unclear. We show that orx/hcrt cells are stimulated by nutritionally relevant mixtures of amino acids (AAs), both in brain slice patch-clamp experiments, and in c-Fos expression assays following central or peripheral administration of AAs to mice in vivo. Physiological mixtures of AAs electrically excited orx/hcrt cells through a dual mechanism involving inhibition of K(ATP) channels and activation of system-A amino acid transporters. Nonessential AAs were more potent in activating orx/hcrt cells than essential AAs. Moreover, the presence of physiological concentrations of AAs suppressed the glucose responses of orx/hcrt cells. These results suggest a new mechanism of hypothalamic integration of macronutrient signals and imply that orx/hcrt cells sense macronutrient balance, rather than net energy value, in extracellular fluid.

The role of the central arachidonic acid-thromboxane A2 cascade in cardiovascular regulation during hemorrhagic shock in rats

The aim of the current study was to elucidate the underlying central mechanism(s) of the cardiovascular effects evoked by centrally injected melittin and arachidonic acid (AA) in hemorrhaged hypotensive condition, specifically, from central AA release from the cell membrane under the influence of phospholipase A(2) (PLA(2)) to central thromboxane A(2) (TXA(2)) signaling via the cyclooxygenase (COX) pathway. As the main control of the study, melittin (3 μg) or AA (150 μg) was injected intracerebroventricularly (i.c.v.) after the hemorrhage procedure, which was performed by withdrawing a total volume of 2.2 ml of blood/100g body weight over a period of 10 min. Both treatments generated a pressor response and abolished the hypotension-induced hemorrhage. Pretreatment with the PLA(2) inhibitor mepacrine (500 μg; i.c.v.) completely blocked the pressor response to melittin in the hemorrhagic hypotensive state. Pretreatments with the nonselective COX inhibitor indomethacin (200 μg; i.c.v.) or the TXA(2) synthesis inhibitor furegrelate (250 or 500 μg; i.c.v.) were made to test the role of central COX activity and, subsequently, the TXA(2) signaling pathway in the melittin- or AA-mediated reversal of hemorrhagic hypotension. Indomethacin completely prevented the pressor response to melittin and AA in the hemorrhaged, hypotensive state, but furegrelate did so only partially. In conclusion, these findings suggest that central COX activity and, subsequently, the central TXA(2) signaling pathway, are, at least in part, involved in the melittin- or AA-induced reversal effect during hemorrhagic shock.

Extracellular signal-regulated kinases mediate melittin-induced hypersensitivity of spinal neurons to chemical and thermal but not mechanical stimuli

Subcutaneous melittin injection causes central plasticity at the spinal level in wide-dynamic-range (WDR) neurons, which are hypersensitive to various nociceptive stimuli. Previous behavioral studies demonstrated that the mitogen-activated protein kinases (MAPKs) extracellular signal-regulated kinase 1/2(ERK1/2), p38 MAPK, and c-Jun N-terminal kinase are involved in both peripheral and spinal processing of melittin-induced nociception and hypersensitivity. Yet the functional roles of the three MAPKs vary among different stimulus modalities, and must be further studied at the cellular level in vivo. In this report, extracellular single unit recordings were performed to investigate whether activation of ERK1/2 in the primary injury site of melittin is essential to the establishment of a spinally sensitized state. Localized peripheral administration of a single dose of the MEK inhibitor U0126 (1 μg/10 μl) significantly suppressed neuronal hyper-responsiveness to thermal stimulus and chemical (melittin)-induced tonic firing of WDR neurons after full establishment of a spinally sensitized state. However, U0126 failed to affect mechanical hypersensitivity to both noxious and non-noxious stimuli. Melittin-induced enhancement of thermal hypersensitivity was also greatly inhibited by a single dose of capsazepine, a thermal nociceptor (TRPV1) blocker. These results suggest that activation of the ERK signaling pathway in the periphery is likely necessary for maintenance of a spinally sensitized state; activation of ERK1/2 in the primary injury site may regulate TRPV1, leading to dorsal horn hypersensitivity to thermal and chemical stimuli. ERK signaling pathways are not likely to be associated with melittin-induced dorsal horn hypersensitivity to mechanical stimuli.

Phospholipase A2 Activation Enhances Inhibitory Synaptic Transmission in Rat Substantia Gelatinosa Neurons

Phospholipase A2 (PLA2) activation enhances glutamatergic excitatory synaptic transmission in substantia gelatinosa (SG) neurons, which play a pivotal role in regulating nociceptive transmission in the spinal cord. By using melittin as a tool to activate PLA2, we examined the effect of PLA2 activation on spontaneous inhibitory postsynaptic currents (sIPSCs) recorded at 0 mV in SG neurons of adult rat spinal cord slices by use of the whole cell patch-clamp technique. Melittin enhanced the frequency and amplitude of GABAergic and glycinergic sIPSCs. The enhancement of GABAergic but not glycinergic transmission was largely depressed by Na+ channel blocker tetrodotoxin or glutamate-receptor antagonists (6-cyano-7-nitroquinoxaline-2,3-dione and/or dl-2-amino-5-phosphonovaleric acid) and also in a Ca2+-free Krebs solution. The effects of melittin on glycinergic sIPSC frequency and amplitude were dose-dependent with an effective concentration of ∼0.7 μM for half-maximal effect and were depressed by PLA2 inhibitor 4-bromophenacyl bromide or aristolochic acid. The melittin-induced enhancement of glycinergic transmission was depressed by lipoxygenase inhibitor nordihydroguaiaretic acid but not cyclooxygenase inhibitor indomethacin. These results indicate that the activation of PLA2 in the SG enhances GABAergic and glycinergic inhibitory transmission in SG neurons. The former action is mediated by glutamate-receptor activation and neuronal activity increase, possibly the facilitatory effect of PLA2 activation on excitatory transmission, whereas the latter action is due to PLA2 and subsequent lipoxygenase activation and is independent of extracellular Ca2+. It is suggested that PLA2 activation in the SG could enhance not only excitatory but also inhibitory transmission, resulting in the modulation of nociception.

Phospholipase A2-independent Ca2+ entry and subsequent apoptosis induced by melittin in human MG63 osteosarcoma cells

Melittin, a peptide from bee venom, is thought to be a phospholipase A(2) activator and Ca(2+) influx inducer that can evoke cell death in different cell types. However, the effect of melittin on cytosolic free Ca(2+) concentration ([Ca(2+)](i)) and viability has not been explored in human osteoblast-like cells. This study examined whether melittin altered [Ca(2+)](i) and killed cells in MG63 human osteosarcoma cells. [Ca(2+)](i) changes and cell viability were measured by using the fluorescent dyes fura-2 and WST-1, respectively. Melittin at concentrations above 0.075 microM increased [Ca(2+)](i) in a concentration-dependent manner. The Ca(2+) signal was abolished by removing extracellular Ca(2+). Melittin-induced Ca(2+) entry was confirmed by Mn(2+) quenching of fura-2 fluorescence at 360 nm excitation wavelength which was Ca(2+)-insensitive. The melittin-induced Ca(2+) influx was unchanged by modulation of protein kinase-C activity with phorbol 12-myristate 13-acetate (PMA) and GF 109203X, or inhibition of phospholipase A(2) with AACOCF(3) and aristolochic acid; but was substantially inhibited by blocking L-type Ca(2+) channels. At concentrations of 0.5 microM and 1 microM, melittin killed 33% and 45% of cells, respectively, via inducing apoptosis. Lower concentrations of melittin failed to kill cells. The cytotoxic effect of 1 microM melittin was completely reversed by pre-chelating cytosolic Ca(2+) with BAPTA. Taken together, these data showed that in MG63 cells, melittin induced a [Ca(2+)](i) increase by causing Ca(2+) entry through L-type Ca(2+) channels in a manner independent of protein kinase-C and phospholipase A(2) activity; and this [Ca(2+)](i) increase subsequently caused apoptosis.

Effects of Oleic Acid on Distinct Populations of Neurons in the Hypothalamic Arcuate Nucleus Are Dependent on Extracellular Glucose Levels

Pharmacological manipulation of fatty acid metabolism in the hypothalamic arcuate nucleus (ARC) alters energy balance and glucose homeostasis. Thus, we tested the hypotheses that distinctive populations of ARC neurons are oleic acid (OA) sensors that exhibit a glucose dependency, independent of whether some of these OA sensors are also glucose-sensing neurons. We used patch-clamp recordings to investigate the effects of OA on ARC neurons in brain slices from 14- to 21-day-old Sprague–Dawley (SD) rats. Additionally, we recorded spontaneous discharge rate in ARC neurons in 8-wk-old fed and fasted SD rats in vivo. Patch-clamp studies showed that in 2.5 mM glucose 12 of 94 (13%) ARC neurons were excited by 2 μM OA (OA-excited or OAE neurons), whereas six of 94 (6%) were inhibited (OA-inhibited2.5or OAI2.5neurons). In contrast, in 0.1 mM glucose, OA inhibited six of 20 (30%) ARC neurons (OAI0.1neurons); none was excited. None of the OAI0.1neurons responded to OA in 2.5 mM glucose. Thus OAI2.5and OAI0.1neurons are distinct. Similarly, in seven of 20 fed rats (35%) the overall response was OAE-like, whereas in three of 20 (15%) it was OAI-like. In contrast, in fasted rats only OAI-like response were observed (three of 15; 20%). There was minimal overlap between OA-sensing neurons and glucose-sensing neurons. In conclusion, OA regulated three distinct subpopulations of ARC neurons in a glucose-dependent fashion. These data suggest that an interaction between glucose and fatty acids regulates OA sensing in ARC neurons.

Phospholipase A2 activation by melittin enhances spontaneous glutamatergic excitatory transmission in rat substantia gelatinosa neurons

In order to know a role of phospholipase A2 in modulating nociceptive transmission, the effect of a secreted phospholipase A2 activator melittin on spontaneous glutamatergic excitatory transmission was investigated in substantia gelatinosa neurons of an adult rat spinal cord slice by using the whole-cell patch-clamp technique. Bath-applied melittin at concentrations higher than 0.5 microM increased both the amplitude and the frequency of spontaneous excitatory postsynaptic current in a manner independent of tetrodotoxin; the latter effect of which was examined in detail. In 80% of the neurons examined (n = 64), melittin superfused for 3 min gradually increased spontaneous excitatory postsynaptic current frequency (by 65+/-6% at 1 microM; n = 51) in a dose-dependent manner (effective concentration for half-maximal effect = 1.1 microM). This effect subsided within 3 min after washout. The spontaneous excitatory postsynaptic current frequency increase produced by melittin was reduced by the phospholipase A2 inhibitor 4-bromophenacryl bromide (10 microM) while being unaffected by the cyclooxygenase inhibitor indomethacin (100 microM) and the lipoxygenase inhibitor nordihydroguaiaretic acid (100 microM). A similar increase in spontaneous excitatory postsynaptic current frequency was produced by exogenous arachidonic acid (50 microM); this effect was also unaffected by the cyclooxygenase or lipoxygenase inhibitor. Melittin failed to increase spontaneous excitatory postsynaptic current frequency in a nominally Ca2+-free or La3+-containing Krebs solution. We conclude that melittin increases the spontaneous release of L-glutamate to substantia gelatinosa neurons by activating secreted phospholipase A2 and increasing Ca2+ influx through voltage-gated Ca2+ channels in nerve terminals, probably with an involvement of arachidonic acid but not its metabolites produced by cyclooxygenase and lipoxygenase. Considering that the substantia gelatinosa plays an important role in regulating nociceptive transmission, it is suggested that this transmission may be positively modulated by secreted phospholipase A2 activation in the substantia gelatinosa.

A potentially critical role of phospholipases in central nervous system ischemic, traumatic, and neurodegenerative disorders

Phospholipases are a diverse group of enzymes whose activation may be responsible for the development of injury following insult to the brain. Amongst the numerous isoforms of phospholipase proteins expressed in mammals are 19 different phospholipase A2's (PLA2s), classified functionally as either secretory, calcium dependent, or calcium independent, 11 isozymes belonging to three structural groups of PLC, and 3 PLD gene products. Many of these phospholipases have been identified in selected brain regions. Under normal conditions, these enzymes regulate the turnover of free fatty acids (FFAs) in membrane phospholipids affecting membrane stability, fluidity, and transport processes. The measurement of free fatty acids thus provides a convenient method to follow phospholipase activity and their regulation. Phospholipase activity is also responsible for the generation of an extensive list of intracellular messengers including arachidonic acid metabolites. Phospholipases are regulated by many factors including selective phosphorylation, intracellular calcium and pH. However, under abnormal conditions, excessive phospholipase activation, along with a decreased ability to resynthesize membrane phospholipids, can lead to the generation of free radicals, excitotoxicity, mitochondrial dysfunction, and apoptosis/necrosis. This review evaluates the critical contribution of the various phospholipases to brain injury following ischemia and trauma and in neurodegenerative diseases.

Characterization of modes of release of amino acids in the ischemic/reperfused rat cerebral cortex

Brain extracellular levels of glutamate, aspartate, GABA and glycine increase rapidly following the onset of ischemia, remain at an elevated level during the ischemia, and then decline over 20-30 min following reperfusion. The elevated levels of the excitotoxic amino acids, glutamate and aspartate, are thought to contribute to ischemia-evoked neuronal injury and death. Calcium-evoked exocytotic release appears to account for the initial (1-2 min) efflux of neurotransmitter-type amino acids following the onset of ischemia, with non-vesicular release responsible for much of the subsequent efflux of these and other amino acids, including taurine and phosphoethanolamine. Extracellular Ca(2+)-independent release is mediated, in part by Na(+)-dependent amino acid transporters in the plasma membrane operating in a reversed mode, and by the opening of swelling-induced chloride channels, which allow the passage of amino acids down their concentration gradients. Experiments on cultured neurons and astrocytes have suggested that it is the astrocytes which make the primary contribution to this amino acid efflux. Inhibition of phospholipase A(2) attenuates ischemia-evoked release of both amino and free fatty acids from the rat cerebral cortex indicating that this group of enzymes is involved in amino acid efflux, and also accounting for the consistent ischemia-evoked release of phosphoethanolamine. It is, therefore, possible that disruption of membrane integrity by phospholipases plays a role in amino acid release. Recovery of amino acid levels to preischemic levels requires their uptake by high affinity Na(+)-dependent transporters, operating in their normal mode, following restoration of energy metabolism, cell resting potentials and ionic gradients.

A melittin-related peptide from the skin of the Japanese frog, Rana tagoi, with antimicrobial and cytolytic properties

Two peptides with antimicrobial and cytolytic properties were purified from an extract of the skin of Tago's brown frog Rana tagoi. The primary structure of one peptide (FLPILGKLLS(10)GIL.NH(2)) identifies it as a member of the temporin family, whereas the second peptide (AIGSILGALA(10)KGLPTLISWI(20)KNR.NH(2)) displays 78% sequence identity to melittin from the venom of the honeybee Apis florea. Compared with melittin, the melittin-related peptide (MRP) was equipotent in inhibiting the growth of the Gram-positive bacterium Staphylococcus aureus, 5-fold less potent against the Gram-negative bacterium Escherichia coli and against the fungal pathogen, Candida albicans. MRP was 13-fold less hemolytic than melittin against human erythrocytes and 4- and 5-fold less cytolytic against mouse EL4 T-lymphoma-derived cells and L929 fibroblasts, respectively. However, at non-cytotoxic concentrations (<or=8 microM), MRP did not protect HeLa cells from cell death produced by human rhinovirus type 2 infection.

Differential effects of phospholipase inhibitors on free fatty acid efflux in rat cerebral cortex during ischemia-reperfusion injury

Free fatty acid (FFA) elevation in the brain has been shown to correlate with the severity of damage in ischemic injury. The etiology of this increase in FFA remains unclear and has been hypothesized to result from phospholipase activation. This study examines the effects of specific phospholipase inhibitors on FFA efflux during ischemia-reperfusion injury. A four-vessel occlusion model of cerebral ischemia was utilized to assess the effects of PLA(2) and PLC inhibitors on FFA efflux from rat cerebral cortex. In addition, FFA efflux from non-ischemic cortices exposed to PLA(2) and PLC was measured. Concentrations of arachidonic, docosahexaenoic, linoleic, myristic, oleic, and palmitic acids in cortical superfusates were determined using high performance liquid chromatography (HPLC). Exposure to the non-selective PLA(2) inhibitor 4-bromophenylacyl bromide (BPB) significantly inhibited FFA efflux during ischemia-reperfusion injury (P<0.01 arachidonic, oleic and palmitic; P<0.05 all others); exposure to the PLC inhibitor U73122 had no observed effect. The effects of the Ca(2+)-dependent PLA(2) inhibitor arachidonyl trifluoromethyl ketone (AACOCF(3)) mirrored the effects of BPB and led to reductions in all FFA levels (P<0.01 arachidonic, oleic and palmitic; P<0.05 all others). Exposure to the secretory PLA(2) inhibitor 3-(3-acetamide-1-benzyl-2-ethyl-indolyl-5-oxy) propane sulfonic acid (LY311727) and to the Ca(2+)-independent PLA(2) inhibitor bromoenol lactone (BEL) had only minimal effects on FFA efflux. Application of both PLA(2) and PLC to non-ischemic cortices resulted in significant increases in efflux of all FFA (P<0.05). The study suggests that FFA efflux during ischemia-reperfusion injury is coupled to activation of Ca(2+)-dependent PLA(2) and provides further evidence of the potential neuroprotective benefit of Ca(2+)-dependent PLA(2) inhibitors in ischemia.

Melittin exerts multiple effects on the release of free fatty acids from L1210 cells: lack of selective activation of phospholipase A2 by melittin

Melittin is known as a phospholipase A2 (PLA2) activator, but the selectivity of its effect on PLA2 is uncertain. We examined the selectivity of melittin effect on the release of free fatty acids (FFAs) from L1210 cells using various inhibitors. A systemic lipid analysis by HPLC and GLC revealed that melittin induced release of various FFAs including saturated, monounsaturated, and polyunsaturated FFAs. Various PLA2 inhibitors examined exerted only minimal effects on the melittin-induced arachidonic acid (AA) and palmitic acid (PAL) releases. Specific inhibitors of phosphatidylinositol-phospholipase C (U73122) and diacylglycerol lipase (RHC80267) exerted significant inhibitory effects on both AA and PAL releases. These results suggest that melittin-induced FFA release is most likely due to multiple participations of various types of lipases. Since BAPTA/AM, an intracellular Ca2+ chelator, did not influence the FFA release, the Ca2+ influxed by melittin appeared not to be a key factor for the FFA release. The mimicking of the melittin-induced FFA release by digitonin, a membrane-permeabilizing agent, implies that the membrane-perturbing action of melittin is likely the cause of the FFA release. Melittin also induced release of multiple FFAs from other cell lines including P388D1 and HL60. The rapid melittin-stimulated phospholipase D (PLD) observed in L1210 cells appeared not directly related to the steady release of FFA, as indicated by the fact that the PLD was not blocked by RHC80267. In view of melittin's multiple effects on the composition of cellular lipids, we conclude that melittin does neither exclusively release any single FFA nor selectively activate PLA2 in L1210 cells. The problem of using melittin as a PLA2 activator is discussed.

Inhibition of Na(+)/H(+) exchange by 5-(N-ethyl-N-isopropyl)-amiloride reduces free fatty acid efflux from the ischemic reperfused rat cerebral cortex

Brain tissue acidosis is considered to be a contributor to ischemic brain injury. The deleterious effects of marked acidosis may be associated with reperfusion and an excessive entry of Na(+) into cerebral neurons and glia as intracellular pH is restored by Na(+)/H(+) exchange. Normalization of pH, with activation of many calcium-dependent and other phospholipases and proteases with pH optima in the neutral or alkaline range, could account for the pronounced elevation in extracellular levels of free fatty acids which occurs during reperfusion following cerebral ischemia. In the present investigation we evaluated the effects of inhibition of Na(+)/H(+) exchange with N-(N-ethyl-N-isopropyl)-amiloride (EIPA; 25 microM) applied topically onto the rat cerebral cortex prior to and during ischemia. Free fatty acid levels in cortical superfusates, withdrawn at 10-min intervals from bilateral cortical windows, were analyzed by high pressure liquid chromatography. EIPA application effectively inhibited the increases in arachidonic and linoleic acid release observed in the control rats during reperfusion, and non-significantly depressed that of palmitic and oleic acids. Superfusate levels of glucose, which decline to near zero levels during ischemia and then rebound during reperfusion, were not affected by EIPA administration. Lactate levels in cortical superfusates from EIPA-treated animals rose more rapidly during reperfusion than did those in the control rats and then significantly declined towards basal levels. The data indicate that inhibition of Na(+)/H(+) exchange prevented the activation of phospholipases that usually occurs during reperfusion following a cerebral ischemic episode. These results are the first demonstration of such an effect and may provide an explanation for the cerebroprotective effects that have been observed in stroked animals following administration of Na(+)/H(+) exchange inhibitors.