Thrombin receptor peptides induce shape change in neonatal murine astrocytes in culture

Quantitative reverse transcriptase PCR to gauge increased protease-activated receptor 1 (PAR-1) mRNA copy numbers in the Wobbler mutant mouse

Thrombin acts on cells through the surface protease-activated receptor 1 (PAR-1), a G-protein-coupled member of the seven-transmembrane domain superfamily. On neural cells, thrombin has deleterious effects, killing neurons through apoptosis. Consequently, knowledge of PAR-1 expression in the nervous system may help to elucidate the role of thrombin in neurodegenerative disease. We developed a mimic construct to facilitate the highly sensitive technique of quantitative reverse transcriptase to PCR (qRT-PCR) to measure the differential expression of low copy number PAR-1 mRNA in neurodegenerative model systems. In this article, we report our results comparing homozygous wobbler (wr/wr) mice and normal littermates. By optimizing the transcription and quantitative PCR procedures to facilitate rapid copy number determination in small RNA samples, we documented a fivefold greater level of PAR-1 mRNA in the cervical spinal cord of wr/wr.

Involvement of protease-activated receptor-1 in the in vitro development of mesencephalic dopaminergic neurons

In situ hybridization studies have revealed high levels of protease (thrombin)-activated receptor-1 messenger RNA in the mesencephalon of rats, suggesting that dopaminergic neurons are a target for thrombin's actions. We have evaluated the effect of thrombin receptor activation, either by thrombin or by thrombin receptor agonist peptide, a 14 amino acid agonist of protease-activated receptor-1, on tyrosine hydroxylase-positive neurons. Pure cultures of rat mesencephalic neurons or co-cultures of mesencephalic neurons and glial cells were treated with either thrombin or thrombin receptor agonist peptide the day after plating. Tyrosine hydroxylase-positive cell counting, [3H]dopamine uptake and morphometric analysis were performed on day 5. Thrombin and thrombin receptor agonist peptide influenced neurite elongation, branching and the number of primary, secondary and tertiary neurites of tyrosine hydroxylase-positive neurons. In pure cultures, the most significant effects of thrombin and thrombin receptor agonist peptide were to delay branching and to increase the centrifugal growth of neurites without affecting the total neuritic length. Thrombin (up to 10 nM) and thrombin receptor agonist peptide did not affect the number of tyrosine hydroxylase-positive neurons or [3H]dopamine uptake. Neurotrophin-4 also influenced the morphology of tyrosine hydroxylase-positive neurons. The increase of neuritic length initiated by this neurotrophin is complementary to the radial elongation induced by protease-activated receptor-1 activation. When neurons were cultured in the presence of glial cells, the effects of thrombin and thrombin receptor agonist peptide on most of these parameters were larger than those observed with pure cultures. Thus, thrombin is able to initiate a complex remodelling of the architecture of tyrosine hydroxylase-positive neurons through the activation of protease-activated receptor-1. These results provide further support for the involvement of protease-activated receptor-1 activation in the development and differentiation of the central nervous system.

Calcium mobilization and protease-activated receptor cleavage after thrombin stimulation in motor neurons

Thrombin, the ultimate enzyme in the blood coagulation cascade, has prominent actions on various cells, including neurons. As in platelets, thrombin increases [Ca2+]i mobilization in neurons, and also retracts neurites. Both these effects are mediated through a G protein-coupled, proteolytically activated receptor for thrombin (PAR-1). Prolonged exposure to thrombin kills neurons via apoptosis, that may also involve PAR-1 activation. Increased [Ca2+]i has been a unifying mechanism proposed for cell death in several neurodegenerative diseases. Thrombin-elevated calcium levels may activate intracellular cascades in neurons leading to cell death. Since thrombin mediates its diverse effects on cells through both heterotrimeric and monomeric G proteins, we also explored what effect altering differential G protein coupling would have on the neuronal response to thrombin. We studied calcium mobilization by thrombin in a model motor neuronal cell line, NSC19, using fluorescence image analysis. Confirming effects in other neuronal types, thrombin caused dramatic increases in [Ca2+]i levels, both transiently and after prolonged exposure, which involved activation and cleavage of the PAR-1 receptor. Using enzyme linked immunosorbent assay (ELISA) and dot-blot analysis, we found that the N-terminal fragment of PAR-1 was released into the medium after exposure to thrombin. We confirmed that PAR-1 protein and mRNA expression occurred in motor neurons. We found that cholera toxin inhibited thrombin-mediated Ca2+ influx, pertussis toxin did not significantly alter thrombin action, and lovastatin, a small 21-kDa Ras GTPase (Rho) modulator, showed a tendency to reduce the thrombin effect. These data indicate that thrombin-increased [Ca2+]i, sufficient to trigger cell death in motor neurons, might be approached in vivo by modulating thrombin signaling through PAR-1.

Nervous system-specific expression of a novel serine protease: regulation in the adult rat spinal cord by excitotoxic injury

A full-length cDNA clone of a previously unidentified serine protease, myelencephalon-specific protease (MSP), has been isolated by using a PCR cloning strategy and has been shown to be expressed in a nervous system and spinal cord-specific pattern. Sequence analysis demonstrated that MSP is most similar in sequence to neuropsin, trypsin, and tissue kallikrein and is predicted to have trypsin-like substrate specificity. MSP mRNA was found to be approximately 10-fold greater in the CNS of the rat and human, as compared with most peripheral tissues, and within the CNS was found to be highest by a factor of four in the medulla oblongata and spinal cord. Levels of mRNA encoding tissue plasminogen activator (tPA) also were elevated in the spinal cord but were more widespread in peripheral tissues as compared with MSP. In the adult rat lumbosacral spinal cord, in situ localization of MSP mRNA demonstrated 2-fold higher levels in the white, as compared with the gray, matter. MSP mRNA expression was shown to increase 3-fold in the white matter and 1.5-fold in the gray laminae at 72 hr after intraperitoneal injection of the AMPA/kainate glutamate receptor-specific agonist, kainic acid (KA). MSP mRNA remained elevated in the ventral gray matter, including expression associated with the motor neurons of lamina IX, at 7 d after the initial excitotoxic insult. Together, these observations indicate that MSP is in a position to play a fundamental role in normal homeostasis and in the response of the spinal cord to injury.

Astrocyte spreading in response to thrombin and lysophosphatidic acid is dependent on the Rho GTPase

Astrocytes are typically star shaped cells playing diverse roles in the function of the nervous system. In astrocyte cultures established from the cerebral hemispheres of newborn rats, the cells have generally a polygonal fibroblast-like morphology, but acquire a stellate shape upon serum removal. When the serine protease thrombin or the bioactive lipid lysophosphatidic acid is added, the stellate cells revert to the flat morphology. Here we show that the effect of these agents is mediated via activation of the small GTP-binding protein Rho. Neither thrombin nor lysophosphatidic acid induced spreading of astrocytes microinjected with C3 transferase, an exoenzyme which ADP-ribosylates and thereby inactivates Rho. In contrast, the response of cells injected with a dominant negative form of Rac was unaffected. In addition, the injection of active Rho into stellate astrocytes mimicked the effect of thrombin and lysophosphatidic acid and an injection of C3 into flat cells grown in serum induced stellation. The conversion from a stellate to a spread morphology upon activation of Rho resulted in the formation of stress fibers and focal adhesions which most probably are key events in establishing and stabilizing the altered cytoarchitecture. These results suggest that Rho plays a crucial role in determining the shape of astrocytes and thereby may modulate their interaction with neurons in vivo.

The role of thrombin-like (serine) proteases in the development, plasticity and pathology of the nervous system

There is increasing evidence suggesting that members of the serine protease family, including thrombin, chymotrypsin, urokinase plasminogen activator, and kallikrein, may play a role in normal development and/or pathology of the nervous system. Serine proteases and their cognate inhibitors have been shown to be increased in the neural parenchyma and cerebrospinal fluid following injury to the blood brain barrier. Zymogen precursors of thrombin and thrombin-like proteases as well as their receptors have also been localized in several distinct regions of the developing or adult brain. Thrombin-like proteases have been shown to exert deleterious effects, including neurite retraction and death, on different neuronal and non-neuronal cell populations in vitro. These effects appear to be mediated through cell surface receptors and can be prevented or reversed with specific serine protease inhibitors (serpins). Furthermore, we have recently shown that treatment with protease nexin-1 (a serpin that inhibits thrombin-like proteases) promotes the survival and growth of spinal motoneurons during the period of programmed cell death and following injury. Taken together, these observations suggest that thrombin-like proteases play a deleterious role, whereas serpins promote the development and maintenance of neuronal cells. Thus, changes in the balance between serine proteases and their cognate inhibitors may lead to pathological states similar to those associated with some neurodegenerative diseases such as Alzheimer's disease. The present review summarizes the current state of research involving such serine proteases and speculates on the possible role of these thrombin-like proteases in the development, plasticity and pathology of the nervous system.

Quantitative PCR analysis reveals novel expression of prothrombin mRNA and regulation of its levels in developing mouse muscle

Precise determination of mRNA levels is an essential element in any investigation of complex regulatory systems. Classical methodologies such as Northern hybridization suffer from requirements for significant samples of material and also a degree of nonspecificity. Recently, quantitative techniques involving PCR amplification have been devised. We have developed and applied such procedures to the determination of prothrombin messages in skeletal muscle cells during development. In addition to its role in the blood coagulation cascade, the serine protease thrombin has been shown to participate in several signaling events in the neuromuscular system. The inactive precursor, prothrombin, primarily produced in the liver, has also been shown to be synthesized and developmentally-regulated in the brain. In skeletal muscle, thrombin is a mediator of activity-dependent polyneuronal synapse elimination (ADPSE) which occurs in early postnatal development. Recent experiments showing that thrombin is released from myotubes in culture under the influence of acetylcholine suggest that locally-synthesized prothrombin may be the source of this Hebbian synaptic interaction. We have determined that prothrombin is expressed in skeletal muscle, as the likely source of thrombin involved in ADPSE, and the current results show the quantitative expression of muscle prothrombin during this time of intense synapse remodeling.

Thrombin receptor on rat primary hippocampal neurons: coupled calcium and cAMP responses

We have tested the hypothesis that hippocampal neurons respond to thrombin via a neuronal thrombin receptor. A human neuroblastoma cell line, SK-N-SH, known to be thrombin responsive morphologically, responded both to thrombin and thrombin receptor agonist peptide (TRAP 42-55) with elevation of intracellular calcium. In Western blots of membranes from SK-N-SH cells and cultured rat hippocampal neurons using an antibody against the N-terminal peptide of the human thrombin receptor, putative receptor proteins of 66 and 47 kDa were detected in both cells. Neurons were treated with thrombin and TRAP 42-55 (TRAP-14) to determine their effects on intracellular levels of calcium and cAMP. Only 10% of the neurons showed a rapid response to thrombin, but most responded rapidly to agonist peptide with a prolonged elevation of intracellular free calcium. Neuronal cAMP levels were decreased by 40% after 24 h thrombin treatment. This decrease in cAMP level could be blocked by both the Gi-protein inhibitor, pertussis toxin, and the thrombin inhibitor, hirudin, suggesting a possible involvement of Gi-protein-coupled receptor activation. Furthermore, rapid calcium and cAMP responses were apparently induced by pre-treatment of neurons with thrombin for 24 h and subsequent washout. In summary, these data indicate that rat primary hippocampal neurons have thrombin receptors whose responses to thrombin apparently are up-regulated by 24 h thrombin pre-treatment. These results may have implications for synaptic remodeling, learning and memory.

Cholinergic stimulation increases thrombin activity and gene expression in cultured mouse muscle

Activity-dependent synapse reduction is a major determinant of neuromuscular innervation. Previous research has shown that nanomolar concentrations of hirudin, a specific thrombin antagonist, significantly attenuates this reduction, and protease nexin 1 (PN1), an endogenous thrombin inhibitor closely localized to the neuromuscular synapse, can inhibit synapse reduction at similar concentrations. Protease inhibitors which do not inhibit thrombin, including cystatin and aprotinin, had no effect on synapse reduction. We present a series of experiments examining whether prothrombin and/or PN1 gene expression, as well as thrombin activity, are regulated in muscle cultures by acetylcholine (ACh) receptor activation. We also studied the effect of exogenous thrombin on synapse elimination in co-cultures of muscle and cholinergic neurons. Cultured muscle cells were electrically blocked with tetrodotoxin (TTX), or co-treated with ACh in order to isolate ACh receptor activation. Electrical blockade resulted in a decrease in thrombin release to about two-thirds of control values. The application of ACh to electrically blocked muscle cultures resulted in a 2.5-fold increase in thrombin activity released into the medium and a 2-fold increase in prothrombin gene expression. In contrast, ACh treatment in the presence of TTX had no effect on PN1 gene expression compared to treatment with TTX alone. In addition, exogenous thrombin significantly increased synapse elimination in unstimulated muscle/cholinergic neuron co-cultures. These results suggest that thrombin or a thrombin-like molecule released from muscle is required for activity-dependent synapse elimination and is regulated by neuromuscular activity.

Nip and tuck at the neuromuscular junction: a role for proteases in developmental synapse elimination

During late embryonic and early postnatal development, synaptic connections are extensively modified so that some functional connections are weakened and eliminated from a neural circuit while others are strengthened and maintained. The mechanisms that underlie synapse elimination are beginning to be understood from studies of the neuromuscular junction. A recent paper provides some intriguing insights into the role proteases may play in the developmental disassembly of neuromuscular synapses.

Pharmacological characterization of protease-activated receptor (PAR-1) in rat astrocytes

The proteolytic action of thrombin on its receptor (protease-activated receptor-1 or PAR-1) results in a conformational change in which the new N-terminal sequence auto-activates the receptor. Peptide analogs of this N-terminal sequence (TRAPs) are able to mimic the effect of thrombin and an extensive search has led to the definition of the structural requirement for the agonist and antagonist activity on thrombin receptors in several peripheral systems. Thrombin plays an important role in central and peripheral nervous system development and PAR-1 is present in neurons and astrocytes. We have now characterized thrombin receptors pharmacologically in cultured rat astrocytes by using [3H]thymidine incorporation and reversal of stellation induced by Bt2cAMP as end-points. Thrombin increased [3H]thymidine incorporation into DNA with an EC50 of 1 nM and induced a complete reversion of cell stellation. The effects of thrombin on [3H]thymidine incorporation were mimicked by TRAP-14 (EC50 = 3 microM) and a peptide containing non-natural amino acids Ala-Phe(p-F)-Arg-Cha-HArg-Tyr-NH2 (A6Y; EC50 = 0.8 microM). Similarly, these two peptides reversed Bt2cAMP-induced stellation. The effect of thrombin, TRAP-14 and A6Y on [3H]thymidine incorporation into DNA was significantly prevented by L9R, a 9-amino-acid peptide (Leu-Val-Arg-D-Cys-Gly-Lys-His-Ser-Arg; IC50 = 180 microM against thrombin and TRAP-14 and 800 microM against A6Y) previously described as an antagonist in human platelet aggregation. L9R antagonized also thrombin effects on astrocyte morphology. These results demonstrate that rat astrocytes express PAR-1 receptors which are pharmacologically similar to those previously characterized in human platelets.

Trypsin induced von Willebrand factor release from human endothelial cells in mediated by PAR-2 activation

Nystedt and co-workers cloned in 1994 a second protease activatable receptor (PAR-2) that could be activated by trypsin but not by thrombin (1). In this study, we investigated whether trypsin induced stimulation of endothelial cells is linked to PAR-2 activation. We have found by mRNA analysis that endothelial cells of venous and arterial origin express both protease activatable receptors. The functional thrombin receptor and the protease activated receptor-2 (PAR-2) mediate apparently the same effects in human vascular endothelial cells. Both, the activation of the thrombin receptor with thrombin or SFLLRN and the activation of the PAR-2 with trypsin or SLIGRL induced intracellular calcium mobilisation and a subsequent release of von Willebrand factor (vWf) from Weibel-Palade bodies. As a consequence, it can be concluded that endothelial cells have two different receptors mediating the same cellular responses after activation.

Signaling effects of alpha-thrombin and SFLLRN in rat glioma C6 cells

Effects of thrombin on brain cells, including change of neurite outgrowth and astrocyte shape, are described, but the molecular mechanisms are unclear. We investigated the effects of human alpha-thrombin and a six amino acid thrombin receptor activating peptide (TRAP-6, SFLLRN) on [Ca2+]i, phosphoinositide hydrolysis, and protein kinase C in rat glioma C6 cells. Stimulation of C6 cells with both alpha-thrombin and TRAP-6 resulted in [Ca2+]i mobilization, [3H]Inositol phosphate response, and enhanced immunoreactivity of the protein kinase C (PKC) isoenzymes alpha, beta, gamma, delta, and epsilon. Results suggest that alpha-thrombin and TRAP-6 activate at least partially the same intracellular signaling pathways in rat glioma C6 cells, which is evidence for involvement of "tethered ligand" receptor in thrombin induced signaling in glioma C6 cells.

The thrombin receptor is present in myoblasts and its expression is repressed upon fusion

Cultured myoblasts derived from limb muscle of newborn rats express thrombin receptor immunoreactivity on their surface. Receptor expression is repressed upon myoblast fusion. This is due at least in part to a decrease in the amount of the thrombin receptor mRNA. Addition of thrombin triggers calcium transients only in mono- but not multinucleated muscle cells. Furthermore, thrombin increases the rate of myoblast proliferation that coincides with an activation of mitogen-activated protein kinase. Northern analysis of thrombin receptor mRNA expression in skeletal muscle showed that the transcript is present at a relatively high level at birth, but is almost undetectable in the adult. By in situ hybridization, the mRNA at birth appeared to be present mostly in mononucleated cells grouped in clusters, but not in muscle fibers. Very few nuclei surrounded by a mRNA signal were present on muscle sections of rats 24 days postnatally. These results suggest that the thrombin receptor plays a role in muscle development.

The serine protease granzyme A does not induce platelet aggregation but inhibits responses triggered by thrombin

Granzyme A is a serine protease stored in cytoplasmic granules of cytotoxic and helper T lymphocytes. This protease seems to elicit thrombin receptor-mediated responses in neural cells, thereby triggering neurite retraction and reversal of astrocyte stellation. Here we report that granzyme A does not cause platelet aggregation even at concentrations that are more than two orders of magnitude higher than the EC50 for granzyme A in causing morphological changes in neural cells. However, granzyme A blocks thrombin-induced platelet aggregation in a dose-dependent manner without affecting the response to either ADP or to the peptide agonist of the thrombin receptor SFLLRN that corresponds in sequence to the tethered ligand domain. The inability of granzyme A to cause aggregation and its inhibition of thrombin-induced aggregation were seen in platelets from man, rat and mouse. Granzyme A does not affect the catalytic activity of thrombin in cleaving a chromogenic substrate or the macromolecular substrate fibrinogen. However, granzyme A does seem to cleave the thrombin receptor on platelets to produce a weak Ca2+ signal and reduce the response to subsequent challenge with thrombin, but does not induce a signal in thrombin-stimulated platelets. It is proposed that granzyme A interacts with the thrombin receptor found on platelets in a manner that is insufficient to cause aggregation, but sufficient to compete with thrombin for the receptor. These results suggest that granzyme A cleaves the thrombin receptor at a rate that is insufficient to cause platelet aggregation but is sufficient to cause morphological changes in neural cells. Furthermore, these observations demonstrate that granzyme A release occurring during immune responses within blood vessels would not directly cause platelet aggregation.

Dual effects of thrombin and a 14-amino acid peptide agonist of the thrombin receptor on septal cholinergic neurons

We have compared the effects of thrombin and of the 14-amino acid peptide agonist (TRAP-14) of the thrombin protease activated receptor (PAR) on cholinergic neurons in pure cultures of rat septal neurons and in co-cultures of septal neurons and glial cells. In pure septal cultures, low concentrations of thrombin (up to 10 nM) did not affect choline acetyltransferase (ChAT) activity, a marker of cholinergic neurons, or 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) reduction, an index of cell viability. However, 100 nM thrombin decreased ChAT activity and MTT reduction by 44 and 17%, respectively. In co-cultures, a low concentration of thrombin (1 nM) increased ChAT activity (+75%), whereas a high concentration (100 nM) decreased it (-83%). At this high concentration, thrombin was neurotoxic, as indicated by a large decrease in MTT reduction (-80%). Thrombin effects on ChAT activity were mimicked by TRAP-14 both in pure septal cultures (no effect at 0.1 microM and -63% at 100 microM) and in co-cultures (+25% at 0.1 microM and -28% at 100 microM). In contrast, this peptide did not affect MTT reduction. These dual effects of thrombin and TRAP-14 on ChAT activity in co-cultures, were also observed on pure cultures of septal cells supplied with NGF. The activation and inhibition by TRAP-14 of the expression of ChAT activity in septal neuron/glial cell cultures were inhibited by a 9-amino acid peptide antagonist of thrombin PAR. Thus, the effects of thrombin on cholinergic neurons seem to be mainly mediated by thrombin PAR and glial cells seem to play a major role in these thrombin actions.

Exogenous thrombin inhibits neuritogenesis in cultured neuroblastoma cells but not in rat hippocampal neurons

Thrombin has been reported to inhibit neurite outgrowth from neuroblastoma cells grown in serum-containing medium after switching to serum-free medium. A test of the serum and substrate-dependence of this inhibition became possible with the development of Neurobasal/B27 serum-free medium. Inhibition of sprouting of Nb2a neuroblastoma cells by thrombin occurred from the substrate where it was bound to material adsorbed from serum. Neuritogenesis of primary hippocampal neurons was unaffected by exogenous thrombin on polylysine substrates with or without serum treatment. However, sprouting of hippocampal neurons was stimulated by treating the substrate with hirudin, a highly specific thrombin inhibitor. This suggests that hippocampal neurons are not directly responsive to added thrombin, perhaps because they produce their own thrombin.

Glial development in primary cultures established from normal and X-irradiated neonatal spinal cord

The glial population of the lumbosacral spinal cord of the rat can be markedly depleted by exposure to ionizing radiation during the first postnatal week. Identification of specific cell populations which survive the exposure to radiation is difficult in situ; therefore, the present investigation used in vitro approaches to address issues related to specific phenotypes and maturational states of glia in cultures derived from non-irradiated (control) and irradiated (experimental) lumbosacral spinal cords of 3-day-old rats. Cultures were established from the spinal cords 2 to 4 hours following irradiation and were compared to cultures from non-irradiated, littermate controls. By 4 days in vitro (DIV) the numbers of cells in experimental cultures were profoundly reduced when compared to controls, and this reduction persisted through the termination of the study (8 DIV). In addition to reduction in numbers, astrocyte phenotypes were altered in experimental cultures, with greater proportions of the astrocyte population being constituted by the flat angular, large angular, and pancake types and a lesser proportion by stellate cells. The non-astrocytic cell types were dramatically reduced as evidenced by the paucity of oligodendrocytes immunoreactive for galactocerebroside and of small, non-process bearing cells binding the lectin, Griffonia (Bandeiraea) simplicifolia, a marker for microglia. Experimental cultures contained an increased incidence of binucleate astrocytes, an increase not restricted to a particular astrocyte phenotype. This study established the feasibility of utilizing this combined in vivo/in vitro approach in assessment of glial populations in immature spinal cords, and further investigations are in progress using this model.

Proteolytic action of thrombin is required for electrical activity-dependent synapse reduction

Molecular mechanisms of activity-dependent synapse reduction were studied in an in vitro mammalian neuromuscular preparation. Synapse reduction in this model is activity-dependent and is substantially reduced by the broad-spectrum protease inhibitor, leupeptin, suggesting the role of activity-dependent proteolytic action in the process. Our present experiments show that a potent and specific thrombin inhibitor, hirudin, at nanomolar concentration completely blocked the activity-dependent synapse reduction. Furthermore, a naturally occurring serine protease inhibitor, protease nexin I (PNI), which closely colocalizes with acetylcholine receptors at the neuromuscular junction, inhibited the synapse reduction at the same low concentration. In contrast, neither cystatin, a cysteine protease inhibitor, nor aprotinin, a serine protease inhibitor that does not inhibit thrombin, blocked the synapse reduction. Similarly, neither of the inhibitors of the calcium-activated proteases calpain I and II prevented the reduction of synapses. These results strongly suggest that serine proteolytic action by thrombin or thrombin-like molecules is required for synapse reduction in our in vitro model of the mammalian neuromuscular junction.