Homocysteine-induced, Metalloproteinase-Mediated, Extracellular-Signal Regulated Kinase Phosphorylation In Cultured Cerebellar Granule Neurons

IntroductionPhosphorylation of extracellular-signal regulated kinase (ERK) 1 and 2 (ERK1/2) generates extracellular-signal regulated kinase 1 and extracellular-signal regulated kinase 2 phosphorylated at serine/threonine residues (p-ERKi/2), the active form of ERK1/2.This activation and ensuing downstream signaling is important for many neuronal functions during development and in the adult brain, where it plays a major role in plasticity and thus in memory formation.Hyperhomocysteinemia is associated with vascular disease and cognitive and neurologic dysfunction. Normally homocysteine is present at plasma concentrations of～10μM, with a slight increase during aging. Homocysteine is formed from the essential amino acid methionine and can be re-converted to methionine by the cobalamine-(vitamin B12) and folate-dependent homocysteine methyltransferase.Exposure to homocysteine at micromolar to millimolar concentrations induces cell death by several different routes in neurons in vivo and in vitro. Robert et al. showed in hippocampal slices that 100μM homocysteine caused phosphorylation of extracellularregulated kinases 1 and 2 (ERK1/2) that was inhibited not only by NMDA or metabotropic glutamate receptor antagonists but also by a non-NMDA receptor antagonist in 2005, and suggested that the response was secondary to glutamate release.In this study micromolar homocysteine-mediated ERK1/2 phosphorylation was replicated in cultured cerebellar granule neurons incubated in tissue culture medium. It involved transactivation, a process where activation of Gi/o or Gq protein-coupled receptors or increase in [Ca2+]i leads to metalloproteinase-catalyzed shedding of an agonist at the epidermal growth factor (EGF) receptor, stimulating EGF receptors, which are widespread in brain. During incubation in amino-acid free saline medium millimolar concentrations of homocysteine were required to obtain ERK1/2 phosphorylation, which was independent of transactivation. Glutamine, an obligatory glutamate precursor, restored the potent response. These findings suggest that this system Xc- may be involved in the micromolar effect of homocysteine.Glutamate/cystine transpoter, designated as system Xc-, transports an anionic form of cystine in exchange for glutamate. The cells expressing system Xc- take up cystine in the medium and reduce it to cysteine, which is in turn used for the synthesis of glutathione and proteins. A part of cysteine is released back into the medium via neutral amino acid transport systems; then cysteine is rapidly oxidized to cystine by oxygen in the medium.Metalloproteinases (MPs) play key roles in the responses of cells to their microenvironment. By effecting proteolytic degradation or activation of cell surface and extracellular matrix (ECM) proteins they can modulate both cell-cell and cell-ECM interactions, which influence cell differentiation, migration, proliferation and survival. Our previous work suggested that metalloproteinases played an important role in transactivation of epidermal growth factor receptors (EGFR) following activation of G-protein coupled receptors in cultured astrocytes. The forms of metalloproteinases have been implicated in transactivation of EGFR including the matrix metalloproteinases (MMPs), the adamalysin-like proteinases with both metalloproteinase and disintegrin-like domains (ADAMs). In the present study we have studied the mechanism of ERK1/2 phosphorylation during homocysteine-induced, metalloproteinases-mediated using 7-8-day-old primary cultures of cerebellar granule neurons. (1) Homocysteine-induced ERK1/2 phosphorylation is transactivation of epidermal growth factor receptors (EGFRs). (2) Homocysteine-induced ERK1/2 phosphorylation is mediated by system Xc-.(3) Expression of metalloproteinases in cultured astrocytes and neurons and the role of it in the EGFR transactivation.MethodsThe primary cerebellar granule neuron culture were the pre-incubate in the corresponding medium without serum at 37℃at different concentration of the homocysteine in the absence or presence of specific inhibitors. The reaction was stopped by washing with ice-cold phosphate-buffered saline (PBS) containing 7.5 mM glucose, and the cells were scraped off the dishes and harvested in 0.5 ml of ice-cold buffer for Western-Blot. The results are analysised with one-way ANOVA by SPSS12.0 software. P＜0.05 indicate the statically significant difference.Results1,Homocysteine-induced ERK1/2 phosphorylation is concentration-and time-dependent.Exposure of cerebellar granule neurons to homocysteine for 20 min during incubation in tissue culture medium (containing essential amino acids,2 mM glutamine and 200μM cystine) caused an increase of p-ERK1/2at homocysteine concentrations at or above 100μM. The course of the effects as a function of the homocysteine concentration suggests a dose-dependent effect. Regardless whether the homocysteine concentration was 100μM or 1 mM maximum response occurred after 20 min, with a beginning increase at 10 min, and followed by a decline towards control values at 40 mm.2,EGF receptor-tyrosine kinase and Zn2+-dependent metalloproteina-se are involved in potent homocysteine-induced ERK1/2 phosphorylation.In the presence of 1μM AG1478, a specific EGF receptor-activated tyrosine kinase inhibitor, or of 10μM GM6001, an inhibitor of metalloproteinases, ERK1/2 phosphorylation after incubation with homocysteine for 20 min was inhibited, indicating EGF transactivation, which depends on metalloproteinase-catalyzed release of an agonist at the EGF receptor.3,NMDA and non-NMDA receptors are involved in potent homocysteine-induced ERK1/2 phosphorylation.MK801, an antagonist of NMDA receptors, and CNQX, an antagonist of non-NMDA ionotropic glutamate receptors individually abolished the stimulation by 100μM homocysteine. Again, the effect of the two inhibitors seemed less complete at 1 mM of homocysteine, suggesting that mechanisms independent of ionotropic glutamate receptors and of transactivation may also contribute to ERK1/2 phosphorylation in the presence of 1 mM homocysteine.4,Metabotropic glutamate receptors are involved in ERK1/2 phosphorylation by millimolar homocysteine concentrations in amino-acid free PBS.All experiments shown up until now were carried out in tissue culture medium containing essential amino acids, glutamine (2 mM), and cystine (200μM). However, when a dose-response study was carried out in PBS without any amino acids, one mM homocysteine had no significant effect on ERK1/2 phosphorylation, two mM homocysteine had a statistically significant effect and a maximum stimulation required homocysteine concentrations of at least 10-15 mM. Moreover, AG1478, GM6001, MK801 and CNQX can not inhibited ERK1/2 phosphorylation. However, ERK phosphorylation induced by 10 mM homocysteine was substantially reduced by either the mGluRl inhibitor LY367385 or the mGluR5 inhibitor MPEP.5,Potent homocysteine-induced ERK1/2 phosphorylation is inhibited by DIDS and SPAP.To substantiate the notion that system Xc- is essential for the potent response to homocysteine its potential inhibition was tested in the presence of the system Xc-inhibitors,100μM DIDS, or 300μM SPAP. Either inhibitor abolished ERK1/2 phosphorylation, establishing the dependence of potent ERK phosphorylation on system Xc-.6,Compared with brain tissue, astrocytes expressed less, MMP15, MMP24 and MMP25 and neurons had less MMP9.We examined the mRNA expression of these MMPs and ADAMs in primary cultures of mouse astrocytes and cerebellar granule neurons by reverse transcriptase PCR (RT-PCR). We found that all of 15 MMPs and ADAMs expressed in brain in vivo were also expressed in cultured astrocytes and neurons. As compared with brain tissue, astrocytes expressed less MMP15, MMP24 and MMP25 and neurons had less MMP9.Discussion1,Micromolar versus millimolar homocysteine effectsThe complete inhibition by either antagonist, at least at 100μM homocysteine, suggests a sequential stimulation of non-NMDA and NMDA receptors, the response required the presence of glutamine in the medium. In the absence of glutamine 10-fold higher concentrations of homocysteine were required. Under these conditions ERK1/2 phosphorylation was transactivation-independent, and antagonists of ionotropic receptors had no inhibitory effect, whereas inhibitors of metabotropic glutamate receptors of group I (mGluRl and mGluR5) greatly reduced the response.2,Dependence on glutamine and the role of system Xc-The dependence of micromolar homocysteine-induced potent ERK1/2 phosphorylation on the presence of glutamine, as a precursor of glutamate, suggested the involvement of system Xc-, a conclusion, which was supported by the inhibitory effect of both DIDS and SPAP. System Xc- functions physiologically as an exchanger between intracellular glutamate and extracellular cystine. Quisqualate is both a system Xc- substrate and an agonist at non-NMDA receptors, moreover, it can "self-sensitize". Since homocysteine is both a glutamate receptor agonist and interacts with system Xc-, a similar"self-sensitization"can be expected and can explain the present results.The dependence on glutamine, which is gradually metabolized to glutamate by a high glutaminase activity in cerebellar granule neurons, suggests that release of intracellularly accumulated homocysteine by system Xc-, is triggered by extracellular availability of glutamate after the exposure to extracellular glutamine.3,Neurotoxicity of homocysteineActivation of ionotropic receptors by micromolar concentrations of homocysteine are likely to contribute to neurotoxic effects of homocysteine in vivo. The present study, pointing out the crucial role of cystine-glutamate exchanger-mediated "self-sensitization"for neuronal effects of clinically relevant homocysteine concentrations, calls for additional studies whether or not pharmacological inhibition of this system. In addition, Ferrer et al. found a neuroprotective effect of ERK1/2 phosphorylation against excitotoxicity.4,Expression of metalloproteinases in cultured astrocytes and neurons and the role of it in the EGFR transactivationSomeones used an RNase protection assay to show that among MMP-2,-9,-11,-12 and-14 were expressed, and mostly at rather low levels, in the mouse brain. But we found in the mouse brain MMP2 and MMP9 mRNA are particularly abundant. The presence of some ADAM mRNAs in the brain of a variety of animal species, demonstrated elsewhere, was confirmed for the mouse brain by Iivari Karkkainen. We found that all of 15 Mps expressed in brain in vivo were also expressed in cultured astrocytes and neurons. As compared with brain tissue, astrocytes expressed less MMP15, MMP24 and MMP25 and neurons had less MMP9. Tissue inhibitors of metalloproteinases-2(TIMP-2) can inhibit active forms of all MMPs studied to date, although TIMP-1 is a poor inhibitor of a number of the MT-MMPs. To detect whether the MT-MMP is essential for the potent shedding, we can use TIMP-1 and TIMP-2. A great deal of data has been collected regarding ectodomain shedding is predominantly mediated by ADAM. Intrerestly, we found ERK1/2 phosphorylation induced by KCl requires ADAM17 function in astrocytes, but does not require ADAM17 function in cerebellar granule neurons.Conclusion1,Potent homocysteine-induced ERK1/2 phosphorylation in cultured neurons via EGFR transactivation and depends on system Xc-self-sensitization.2,We found 15 MMPs and ADAMs expressed in brain in vivo were also expressed in cultured astrocytes and neurons.