Search

Neuronal voltage-dependent calcium channels undergo inhibitory modulation by G-protein activation, generally involving both kinetic slowing and steady-state inhibition. We have shown previously that the beta-subunit of neuronal calcium channels plays an important role in this process, because when it is absent, greater receptor-mediated inhibition is observed (). We therefore hypothesized that the calcium channel beta-subunits normally may occlude G-protein-mediated inhibition. Calcium channel beta-subunits bind to the cytoplasmic loop between transmembrane domains I and II of the alpha1-subunits (). We have examined the hypothesis that this loop is involved in G-protein-mediated inhibition by making chimeras containing the I-II loop of alpha1B or alpha1A inserted into alpha1E (alpha1EBE and alpha1EAE, respectively). This strategy was adopted because alpha1B (the molecular counterpart of N-type channels) and, to a lesser extent, alpha1A (P/Q-type) are G-protein-modulated, whereas this has not been observed to any great extent for alpha1E. Although alpha1B, coexpressed with alpha2-delta and beta1b transiently expressed in COS-7 cells, showed both kinetic slowing and steady-state inhibition when recorded with GTPgammaS in the patch pipette, both of which were reversed with a depolarizing prepulse, the chimera alpha1EBE (and, to a smaller extent, alpha1EAE) showed only kinetic slowing in the presence of GTPgammaS, and this also was reversed by a depolarizing prepulse. These results indicate that the I-II loop may be the molecular substrate of kinetic slowing but that the steady-state inhibition shown by alpha1B may involve a separate site on this calcium channel.

The dihydropyridine-sensitive calcium channel agonist (-)-BayK 8644 was found to produce an enhancement of the intrinsic hydrolysis of GTP by Go in rat frontal cortex membranes. An anti-calcium channel beta-subunit antiserum abolished the (-)-BayK 8644-stimulated hydrolysis of GTP by Go and reduced the dihydropyridine binding capacity of the cortical membranes. A peptide which mimics the beta-subunit binding domain of the calcium channel complex, also attenuated (-)-BayK 8644 activation of GTPase. This study suggests that the calcium channel beta-subunit is the principal component of the channel complex involved in linking dihydropyridine agonist binding to enhanced hydrolysis of GTP by Go. This may be a mechanism by which calcium channels can normally act to limit the duration of a G-protein modulatory signal.

P21-ras in combination with ras-GAP has been found to block activation of atrial K+ channels by M2 muscarinic receptors (mAChRs), an effect which was suggested to involve interference with receptor-G-protein coupling [l]. Since both M2 and M4 mAChRs are known to inhibit Ca2' channels directly via the pertussis toxin (PTX)-sensitive G-protein, Go, p21-ras may also be involved in regulation of neuronal Ca2+ channels [2, 31. Here we report the results from a study in which the whole cell patch clamp technique has been used to examine the involvement of endogenous p21-ras in regulation of Ca" channels in primary cultured rat dorsal root ganglion cells (DRGs). In preliminary ex eriments, the presence of the muscarinic agonist, carbachol(10' M) in the external medium caused a 32.0 * 4.0 YO (mean i s.e.m.) inhibition of calcium channel current, IBa. Use of the M 1 M4-selective muscarinic antagonist, pirenzepine ( M), the M4-selective antagonist, himbacine ( M), and PTX (230 ng/ml for 15 h) suggested the involvement of a PTXsensitive M4 receptor in rat DRGs. In DRGs microinjected with the p21-ras neutralising antibody (Y13-259, 0.1 mg/ml, 50 pl/cell) 2 h prior to experiments, a 43.7 f 4.0 % inhibition of IBa was observed in the presence of carbachol. This result was not significantly different from control cells injected with 1 mg/ml of mouse IgG (36.2 5 4.0 %) or from noninjected cells (42.9 * 4.0 %). The voltage-dependence of both activation and inactivation of IB, shifted between 5-10 mV in a depolarising direction in all three treatments following application of carbachol, although these shifts were not significant either within, or, between treatments. Thus, microinjection of Y 13-259 had no effect on the ability of the muscarinic agonist to inhibit I B ~ Importantly, however, there was a significant reduction in both peak amplitude of maximum IBa (36.0 * 3.0 %, p < 0.05) and the maximum tail current amplitude (31.0 f 3.0 %, p < 0.05) in antibody-injected cells versus controls. No significant differences in the voltage-dependence of either activation or inactivation of maximum peak amplitude of IB, were observed between treatments. Thus, acute blockade of a pathway involving endogenous p21 -ras reduces calcium currents. Intracellular application of phosphopeptides (10-1 5 amino acids) with the sequences of tyrosine autophosphorylation sites on protein tyrosine kinases involved in growth factor activation of p21-ras (NGFR, TRK490; EGFR1068; and CRK adaptor protein, CRK221; at 0.1 mg/ml) also reduced both the peak amplitude of IBa, and maximum tail current amplitude. Effects of these phosphopeptides are thought to be due to the blockade of productive SH2-SH3 interactions which are initiated during the activation of p21-ras by growth factors. Inclusion of TRK490 in the patch pipette for up to 10 min caused a significant reduction of both peak IB, (1 5.8 f 2 %, p < 0.05) and tail current (34.0 f 6.4 %, p < 0.05). Application of EGFR1068 for 10 min resulted in a 19.8 f 4.0 % reduction of IB, (significant at p < 0.05) , and reduced the tail current by 12.8 f 3.0 %, which was not significant. CRK221 caused a significant reduction of peak I B ~ by 18.3 f 2.0 % (p < 0.05), and maximum tail current by 29.2 f 2.0 % (p < 0.01), with time. In control cells, neither the peak amplitude of Is,, nor the tail current amplitude changed significantly with time (up to 10 min). Similarly, internal application of a control ("scrambled") phosphopeptide, also had no ! . significant effects on either the maximum peak amplitude of IB,, or the maximum tail current amplitude. No significant changes in voltage-dependence of either current activation or inactivation were observed either within, or, between any of the above treatments. Thus, blocking a pathway possibly involving growth factor activation of p21-ras also reduced calcium current amplitude in these cells. In contrast, microinjection of oncogenic p21-K-ras (2 mg/ml, 24 h prior to experiments) caused a significant increase in the peak amplitude of maximum IBa (61 f 3.0 %, p < 0.01) when compared with control cells. Similarly, the maximum tail current amplitude also increased significantly by 59.7 * 3.0 % (p < 0.01) The voltagedependence of both activation and inactivation of the currents exhibited no significant differences in p21-K-ras-injected, versus control cells. Thus, we have provided evidence that endogenous p2 1 -ras, possibly activated via a protein tyrosine kinase pathway, is involved in the regulation of Ca2+ channels in rat DRGs, although unlike in cardiac myocytes, it is not involved in muscarinic inhibition of these Ca" currents. These results tie in with other studies in which the injection of oncogenic p21-ras has been found to increase Ca2+ current in neuronal cells [4,5,6].

1. The role of the voltage-dependent calcium channel (VDCC) beta-subunit has been examined in cultured rat dorsal root ganglion neurones (DRGs). An antipeptide antibody was raised and this recognized proteins corresponding to beta-subunits in a number of preparations. Immunoreactivity for the VDCC beta-subunit in DRGs was concentrated on the internal side of the plasma membrane but was also present in the cytoplasm. 2. A twenty-six-mer antisense oligonucleotide with homology to all published VDCC beta-subunit sequences was microinjected into individual cells, and maximal depletion of VDCC beta-subunit immunoreactivity was observed after 108 h suggesting a half-life for the turnover of the beta-subunit greater than 50 h. No depletion was obtained with nonsense oligonucleotide. 3. The effect of depletion of VDCC beta-subunit immunoreactivity on calcium channel currents in these cells was a reduction in amplitude of the maximum current of about 47%, and a shift in the voltage dependence of current activation of about +7 mV. These effects are the converse of those observed following co-expression of cloned beta- with alpha 1-subunits in oocytes and other expression systems. 4. The ability of the 1,4-dihydropyridine (DHP) agonist Bay K 8644 to enhance calcium channel currents was greatly reduced following depletion of beta-subunit immunoreactivity. This result is in agreement with the finding in several systems that co-expression of the beta-subunit with alpha 1-subunits results in an increased number of DHP binding sites. 5. These results show that calcium channel beta-subunits form part of native neuronal calcium channels and modify their biophysical and pharmacological properties.

Regulation of G protein mediated signal transduction is thought to occur primarily as the result of the occupancy of a receptor by the appropriate agonist. There is much regulation of this process, however, which occurs independently of straightforward agonist occupancy. It is demonstrated here that receptors themselves might have intrinsic activity and the expression of such intrinsically activated receptors might alter the activated state of a G protein and its effector within a given cell. The regulation of receptor display on the cell surface is also discussed and it is seen that the ability to sequester, internalize and restore receptors to the plasma membrane is quite different for different types of G protein coupled receptors. The suggestion that a given receptor uses specific G protein α, β, and γ subunits in order to couple receptor to effector is also explored. Such coding would allow a cell to shape its response to a given agonist. It further proposed that elements of the cytoskeleton (tubulin) form complexes with Gs, Gil or Gq and activate those proteins via the direct transfer of GTP. In this way, changes in cell shape could be conveyed into the intracellular mileu. Further, these cytoskeletal elements might act as the contact point for interplay among various signal transduction systems within the cell. Finally, it is observed that voltage-dependent calcium channels undergo a complex regulation in which G protein β subunits compete with one of the polypeptides making up the channel. Such a process allows for the neurotransmitter modulation of neurotransmitter release. Such modulation returns full circle back to the receptor-mediated activation of G proteins and the possibility that this process is modulated in the cellular interior.

Release of glutamate from cerebellar granule neurones was stimulated either by adding 50 mM K+ to normal Krebs medium, or by adding 5 mM Ca2+ to neurones continuously depolarised with 50 mM K+ in the absence of Ca2+. Pre-incubation of neurones for 16 h with pertussis toxin (PTX) increased the stimulated glutamate release in both K+-stimulated and continuously depolarised neurones. Under both conditions, the PTX-induced increase in release was abolished by cycloheximide. In contrast, in the presence of cycloheximide, PTX still prevented the GABAB agonist (−)-baclofen from inhibiting glutamate release. These results suggest that G-protein ADP-ribosylation by PTX in cerebellar granule neurones may increase synthesis of a protein associated with the L-type calcium channel.

In many types of excitable cell, there are several classes of voltage-dependent calcium channel (VDCC) as determined by the characteristic properties of the single channel activity. There is a clear distinction between low conductance channels, activated by moderate depolarizations (low voltage-activated, LVA) and high conductance channels activated by large depolarizations (high voltage- activated, HVA).2,36 In cardiac tissue these were termed T and L channels, respectively, because they were found to be pharmacologically, as well as biophysically, distinct.2 Because of this they can also be clearly differentiated in whole-cell current recordings.2,12 There is also evidence for a third class of single channel conductance (N type), whose biophysical properties were originally described as being intermediate between T and L, and which appears to be expressed only in cells of neuronal origin.12,38,51 The sensitivity of N-type channels to irreversible block by ω-conotoxin GVIA (ωCgTx),38 and the large number of ω-CgTx binding sites in neuronal tissue, indicates that they are likely to be important for neuronal function.31 High threshold Ca2+ currents insensitive to both ω-CgTx and dihydropyridines have been reported,27,38 and a selective blocker for at least part of this current is the peptide toxin from Agelenopsis Aperta, ω-agatoxin IVA (ω-aga IVA).

The effects of changing the intracellular concentrations of either free Mg2+ ions ([Mg2+]i) or Mg(2+)-bound adenosine triphosphate ([Mg.ATP]i) on Ca2+ channel currents were assessed in cultured rat cerebellar granule neurones using the whole-cell patch-clamp technique. Raising [Mg2+]i from 0.06 mM to 1.0 mM inhibited Ca2+ channel currents by approximately 50%. The action of omega-conotoxin GVIA (omega-CgTX), a selective inhibitor of "N"-type Ca2+ channels was also investigated. With increasing [Mg2+]i, the proportion of current irreversibly blocked by omega-CgTX was reduced, and was negligible (approximately 5 pA of current) in the presence of [Mg2+]i values of 0.5 mM or greater. Block of the omega-CgTX-sensitive current accounted for the reduction in total current by concentrations of [Mg2+]i to 0.5 mM. Raising [Mg2+]i had no effect on the rate of decay of Ca2+ currents, but did produce a negative shift in current activation, possibly due to a non-specific interaction with negative surface charge. Altering [Mg.ATP]i from 0.3 to 5.0 mM caused a twofold increase in the size of currents without affecting the proportion of current sensitive to omega-CgTX. [Mg2+]i was also effective in inhibiting the Ca2+ channel current following potentiation by increasing [Mg.ATP]i. These data suggest that omega-CgTX-sensitive current in these cells is selectively inhibited by internal Mg2+ whereas both omega-CgTX-sensitive and -resistant components of current are potentiated by internal Mg.ATP. The mechanism by which Mg2+ inhibits "N"-type channels is unclear, but may involve an open channel block.

High-voltage-activated (HVA) calcium channel currents (IBa) were recorded from acutely replated cultured dorsal root ganglion (DRG) neurons. IBa was irreversibly inhibited by 56.9±2.7% by 1 μM ω-conotoxin-GVIA (ω-CTx-GVIA), whereas the 1,4-dihydropyridine antagonist nicardipine was ineffective. The selective γ-aminobutyric acidB (GABAB) agonist, (−)-baclofen (50 μM), inhibited the HVA IBa by 30.7±5.4%. Prior application of ω-CTx-GVIA completely occluded inhibition of the HVA IBa by (−)-baclofen, indicating that in this preparation (−)-baclofen inhibits N-type current. To investigate which G protein subtype was involved, cells were replated in the presence of anti-G protein antisera. Under these conditions the antibodies were shown to enter the cells through transient pores created during the replating procedure. Replating DRGs in the presence of anti-Go antiserum, raised against the C-terminal decapeptide of the Gαo subunit, reduced (−)-baclofen inhibition of the HVA IBa, whereas replating DRGs in the presence of the anti-Gi antiseram did not. Using anti-Gαo antisera (1∶2000) and confocal laser microscopy, Gαo localisation was investigated in both unreplated and replated neurons. Gαo immunoreactivity was observed at the plasma membrane, neurites, attachment plaques and perinuclear region, and was particularly pronounced at points of cell-to-cell contact. The plasma membrane Gαo immunoreactivity was completely blocked by preincubation with the immunising Gαo undecapeptide (1 μg · ml−1) for 1 h at 37° C. A similar treatment also blocked recognition of Gαo in brain membranes on immunoblots. These results provide evidence that GABAB inhibition of N-type calcium channels in acutely replated DRGs occurs via Gαo.

Previous studies have shown that pre-treatment of cerebellar granule neurones for 16h with pertussis toxin (PTX, 230ng/ml) increases K+-stimulated release of [ 3H] -glutamate newly synthesised from L3H1 glutamine (1). This may be due to an increased synthesls of Ca2' channels as suggested by the observed parallel increase in dihydropyridine binding sites (2). To further investigate this theory, cycloheximide (0.25pg/ml), a protein synthesis inhibitor, was included in the 16h incubation with PTX; here stimulated release was not increased. When neurones were continuously depolarised in 50mM K+ and release stimulated with pulses of 5mM Ca2+, 16h PTX pre-treatment increased basal glutamate release by 68 .5% from 0.107t0.024% to 0.208*0.028% of total glutamate (n=12,p

This study was designed to examine the ability of pertussis toxin to block various responses due to (-)-baclofen in cultured cerebellar granule neurons of the rat. Treatment with pertussis toxin for 3 h markedly reduced the ability of (-)-baclofen to stimulate GTPase in membranes, and its ability to inhibit forskolin-stimulated adenylyl cyclase in intact cells, whereas the ability of (-)-baclofen to inhibit glutamate release was not affected at 3 h, but was abolished after 16 and 48 h treatment with pertussis toxin. The amount of ADP-ribosylation of Gi/Go due to pertussis toxin in intact cells correlated well with the former two effects, but not with the prevention of the ability of baclofen to inhibit glutamate release. Pertussis toxin treatment for up to 48 h did not significantly affect the levels of Gs, Gi and Go in membranes from granule neurons determined by immunoblotting. Pertussis toxin treatment for 16 or 48 h but not 3 h increased the total amount of stimulated release of glutamate by about 40% under normal conditions, and by 84% under depolarizing conditions. In parallel experiments it was observed that pertussis toxin treatment for 16 h increased the number of dihydropyridine binding sites by about 90% on intact granule neurons. Whole-cell calcium channel currents, recorded under several conditions in the cells, were not increased in amplitude by pertussis toxin treatment for up to 48 h, although the ability of baclofen to inhibit calcium channel currents was blocked by pertussis toxin. These results indicate that the pertussis toxin-induced increase in glutamate release may be due to an increase in dihydropyridine binding sites, possibly localized to the presynaptic terminals.

The α 2 δ proteins are auxiliary subunits of voltage-gated calcium channels, but may also have other functions. Four mammalian α 2 δ genes have been cloned. There are several links between α 2 δ proteins and disease, including mice with mutated α 2 δ-2, which exhibit spike-wave epilepsy. Furthermore, the gabapentinoid drugs bind to α 2 δ-1 and α 2 δ-2. For an expanded treatment of this topic see Jasper's basic mechanisms of the epilepsies. 4th ed. (Noebels JL, Avoli M, Rogawski MA, Olsen RW, Delgado-Escueta AV, eds) published by Oxford University Press (available on the National Library of Medicine Bookshelf [NCBI] at www.ncbi.nlm. nih.gov/books).

The High Voltage-Activated (HVA) subgroup of voltage-gated calcium channels contain an α1 subunit, which forms the selective pore and determines the main functional properties of the channel. The α1 subunit is associated with auxiliary subunits including intracellular β and α2δ, which modulate trafficking and functional properties of the channels.α2δ subunits consists of two peptides: α2 which is entirely extracellular is disulfide-bonded to a δ subunit that links the protein into the plasma membrane. There are four genes encoding α2δ subunits, which are believed to have similar structure. We have shown previously that α2δ-2 subunits associate with lipid rafts, that are sub-domains of the cell membrane enriched in cholesterol and glycosphingolipids.We have addressed the ability of α2δ-3 to associate with lipid rafts in both native tissues (it is highly expressed in brain) and in overexpression systems. We have generated mutations which reduced expression of the subunit in lipid rafts as well as the surface expression of the protein. These mutations reduced the enhancing effect of α2δ-3 on calcium channel currents.The α2 and δ peptides are product of a single gene, and they are encoded as an uninterrupted α2δ pre-protein, which is further processed post-translationally. In native tissues we observed exclusively the mature form of the protein, which was strongly associated with the lipid rafts. However, in several overexpression systems we could also detect unprocessed α2δ-3 pre-protein coexisting with the mature α2δ3. The unprocessed form was localized both in the rafts and non-raft protein fractions, suggesting that maturation of the protein might occur in localized membrane domains.These results further demonstrate the role of lipid rafts in the regulation of Ca channel currents by α2δ and their involvement in the maturation of the α2δ protein.

Several lines of evidence suggest that L-type Ca2+ channels (1,4-dihydropyridine receptors) are modulated by GTP-binding proteins. We have further examined this interaction by measuring the effect of 1,4-dihydropyridines on GTPase activity in brain membranes. Dihydropyridine agonists significantly increased GTPase, reflected by an increase in the maximal rate of GTP hydrolysis, without affecting the affinity for GTP or the binding of a non-hydrolysable analogue of GTP. The stimulating effect on GTPase was abolished by antisera raised against Go alpha but not Gi alpha. L-type Ca2+ channels may act as endogenous GTPase activating proteins (GAPs) to stimulate GTP hydrolysis by Go.

To investigate the role of intracellular Ca 2+ in the survival of developing neurons before they become neurotrophic factor dependent, we have studied chick embryo nodose neurons, which have a particularly protracted period of neurorophic factor independence. Pharmacological reduction of intracellular free Ca 2+ or depletion of either Ca 2+ -regulated or inositol trisphosphate-regulated intracellular Ca 2+ stores kills early neurotrophic factor-independent neurons, but has a negligible effect on older neurons growing in the presence of brain-derived neutrotrophic factor. Shortly before they become dependent on brain-derived neurotrophic factor, nodose neurons express L-type Ca 2+ channels and their survival can be enhanced by depolarization-induced Ca 2+ influx. We conclude that intracellular Ca 2+ plays a role in regulating neuronal survival both prior to and after the onset of neurotrophic factor dependence, but does not mediate the survival-promoting effects of neurotrophic factors.

Bradykinin is one of several pro-inflammatory, pain-inducing substances produced during inflammation--the body's response to injury. In previous work we have shown that bradykinin and guanosine-5'-O-3-thiotriphosphate increase excitability in a subpopulation of cultured neonatal rat dorsal root ganglion neurons. We now describe experiments in which the mechanism underlying the stimulatory action of these two substances has been examined in more detail. Using the whole-cell voltage-clamp technique, bradykinin-sensitive cells were distinguished by their response to a 1-s depolarizing voltage-pulse which evoked more than one inward current during the step command. The secondary inward currents are likely to represent action potentials generated at the poorly clamped neurites of these cells. Bradykinin- and guanosine-5'-O-3-thiotriphosphate-induced changes in excitability were measured indirectly by a change in the number of inward currents recorded during the 1-s depolarizing voltage-step. The effect of activators and inhibitors of protein kinase C, arachidonic acid metabolism, G-protein activation and release of intracellular Ca2+ were examined on this response. In the presence of extracellular staurosporine (1.0 microM) or nordihydroguaiaretic acid (10 microM), these excitatory effects were reduced but not abolished, whilst indomethacin (20 microM) had no effect. Intracellular application of guanosine-5'-O-2-thiodiphosphate (10 mM) or ryanodine (100 microM) substantially reduced the effect of bradykinin. The excitatory effect of internal guanosine-5'-O-3-thiotriphosphate (500 microM) occurred gradually over time, and this was mimicked by internal application of myo-inositol 1,4,5-trisphosphorothioate (1.0 microM). From the results, it is proposed that G-protein activation is an essential component of the bradykinin response, which may also require a Ca(2+)-activated conductance modulated by protein kinase C and lipoxygenase metabolites of arachidonic acid.

Ca2+ channel currents have been recorded in cultured rat dorsal root ganglion neurones. The amplitude of IBa(GTP gamma S), recorded in the presence of GTP[ gamma S] (200 microM) in the patch pipette solution, is enhanced by external application of forskolin (10 microM), and there is an increase in the proportion of the rapidly activating component of the current. When forskolin (1 microM) is present in the bathing solution at the start of recording, or when 8-bromocyclic AMP (100 microM) is present in the patch pipette solution, the amplitude and rate of activation of IBa(GTP gamma S) are also increased compared to control IBa(GTP gamma S). The effect is mimicked by internal application of a 5 microM solution of a phosphopeptide fragment of inhibitor 1 (I1 PP), which inhibits phosphatase 1. The enhancement of IBa(GTP gamma S) caused by I1PP is not additive with that due to forskolin. Furthermore, the enhancement due to I1PP is reversibly lost when the holding potential is shifted from -80 mV to -30 mV, as was the enhancement due to forskolin and 8-bromocyclic AMP. I1PP also produced a less marked stimulation of the control Ca2+ channel current in the absence of G protein activation. The results suggest that phosphorylation regulates the interaction between calcium channels and G proteins in these neurones, and that phosphatase 1 is tonically active to dephosphorylate the relevant protein(s).

1. Toxins from invertebrates have proved useful tools for investigation of the properties of ion channels. In this study we describe the actions of arginine polyamine which is believed to be a close analogue of FTX, a polyamine isolated from the American funnel web spider, Agelenopsis aperta. 2. Voltage-activated Ca2+ currents and Ca(2+)-dependent Cl- currents recorded from rat cultured dorsal root ganglion neurones were reversibly inhibited by arginine polyamine (AP; 0.001 to 100 microM). Low voltage-activated T-type Ca2+ currents were significantly more sensitive to AP than high voltage-activated Ca2+ currents. The IC50 values for the actions of AP on low and high voltage-activated Ca2+ currents were 10 nM and 3 microM respectively. AP was equally effective in inhibiting high voltage-activated currents carried by Ba2+, Sr2+ or Ca2+. However, AP-induced inhibition of Ca2+ currents was attenuated by increasing the extracellular Ca2+ concentration from 2 mM to 10 mM. 3. The actions of AP on a Ca(2+)-independent K+ current were more complex, 1 microM AP enhanced this current but 10 microM AP had a dual action, initially enhancing but then inhibiting the K+ current. 4. gamma-Aminobutyric acid-activated Cl- currents were also reversibly inhibited by 1 to 10 microM AP. In contrast N-methyl-D-aspartate currents recorded from rat cultured cerebellar neurones were greatly enhanced by 10 microM AP. 5. We conclude that at a concentration of 10 nM, AP is a selective inhibitor of low threshold T-type voltage-activated Ca2+ currents. However, at higher concentrations 1-10 microM AP interacts with ion channels or other membrane constituents to produce a variety of actions on both voltage and ligand gated ion channels.

Background Descending facilitation, from the brainstem, promotes spinal neuronal hyperexcitability and behavioural hypersensitivity in many chronic pain states. We have previously demonstrated enhanced descending facilitation onto dorsal horn neurones in a neuropathic pain model, and shown this to enable the analgesic effectiveness of gabapentin. Here we have tested if this hypothesis applies to other pain states by using a combination of approaches in a rat model of osteoarthritis (OA) to ascertain if 1) a role for descending 5HT mediated facilitation exists, and 2) if pregabalin (a newer analogue of gabapentin) is an effective antinociceptive agent in this model. Further, quantitative-PCR experiments were undertaken to analyse the α2δ-1 and 5-HT3A subunit mRNA levels in L3–6 DRG in order to assess whether changes in these molecular substrates have a bearing on the pharmacological effects of ondansetron and pregabalin in OA. Results Osteoarthritis was induced via intra-articular injection of monosodium iodoacetate (MIA) into the knee joint. Control animals were injected with 0.9% saline. Two weeks later in vivo electrophysiology was performed, comparing the effects of spinal ondansetron (10–100 μg/50 μl) or systemic pregabalin (0.3 – 10 mg/kg) on evoked responses of dorsal horn neurones to electrical, mechanical and thermal stimuli in MIA or control rats. In MIA rats, ondansetron significantly inhibited the evoked responses to both innocuous and noxious natural evoked neuronal responses, whereas only inhibition of noxious evoked responses was seen in controls. Pregabalin significantly inhibited neuronal responses in the MIA rats only; this effect was blocked by a pre-administration of spinal ondansetron. Analysis of α2δ-1 and 5-HT3A subunit mRNA levels in L3–6 DRG revealed a significant increase in α2δ-1 levels in ipsilateral L3&4 DRG in MIA rats. 5-HT3A subunit mRNA levels were unchanged. Conclusion These data suggest descending serotonergic facilitation plays a role in mediating the brush and innocuous mechanical punctate evoked neuronal responses in MIA rats, suggesting an adaptive change in the excitatory serotonergic drive modulating low threshold evoked neuronal responses in MIA-induced OA pain. This alteration in excitatory serotonergic drive, alongside an increase in α2δ-1 mRNA levels, may underlie pregabalin's state dependent effects in this model of chronic pain.

1. The characteristics have been examined of the high threshold calcium channel current in cultured rat dorsal root ganglion (DRG) neurones recorded in the presence of guanosine-5'-O-(3-thiotriphosphate) (GTP gamma S; 200 microM in the patch pipette). This current, termed IBa, GTP gamma S, was slowly activating and showed little inactivation over 100 ms. 2. External application of forskolin (10 microM) to elevate internal cyclic AMP levels increased the amplitude of IBa, GTP gamma S whereas it had no effect on the control IBa. This cyclic AMP-dependent protein kinase (PKI; 25 microM). 3. The cyclic AMP-dependent phosphorylation induced enhancement of IBa, GTP gamma S was voltage dependent and either did not occur or was observed only transiently at a holding potential (VH) of -30 mV. The forskolin-stimulated enhancement seen at VH -80 mV was lost with a t1/2 of about 1 min when VH was depolarized to -30 mV. Cholera toxin pre-treatment also increased the amplitude of IBa, GTP gamma S at VH -80 mV but not at VH -30 mV. 4. The calcium channel antagonist (-)-202-791 (5 microM) increased the amplitude of IBa, GTP gamma S when applied at VH -80 mV, but either not, or only transiently, at VH -30 mV, as previously observed. This 'agonist' effect of (-)-202-791 was prevented by PKI and was occluded by prior enhancement of IBa, GTP gamma S with forskolin. (-)-202-791 did not increase cyclic AMP levels in DRG neurones. 5. The 'agonist' response of IBa, GTP gamma S to D600 (10 microM) was also occluded by application of forskolin (10 microM) in the patch pipette. Forskolin alone, applied in this manner, increased IBa, GTP gamma S to a similar extent to D600 applied alone. 6. The agonist effect of (+)-202-791 (5 microM) on IBa, GTP gamma S was not prevented by prior enhancement with forskolin, nor was it prevented by PKI. 7. In conclusion, internal GTP gamma S activates G proteins which may interact directly with calcium channels to influence the kinetics of activation and to reduce steady-state inactivation of the channels. There is also an indirect effect on the generation of second messengers such as cyclic AMP. It is likely that forskolin enhances IBa, GTP gamma S by increasing activated Gs coupling to adenylyl cyclase and increasing cyclic AMP generation. The mechanism of action of (-)-202-791 to enhance IBa, GTP gamma S also involves cyclic AMP-dependent phosphorylation.(ABSTRACT TRUNCATED AT 400 WORDS)

In a previous communication, we reported an increase in protein efflux into perfusates obtained from push-pull cannulation of the dentate gyrus, following induction of long-term potentiation (LTP) in the perforant path3,14. LTP was accompanied by a delayed but general increase in protein efflux. Protein B5 (MW 14 kDa), because of its relatively high concentration in the perfusates and absence from serum, has been chosen for further characterization. Spectrophotometric analysis, in situ proteolysis, two-dimensional electrophoresis and immunoblotting, revealed that protein B5 is indistinguishable from haemoglobin. A persistent increase in an 80 kDa protease, detected by SDS-PAGE zymography, was also seen after the induction of LTP. We consider the possibility that the increased haemoglobin content of perfusates after the induction of LTP may reflect an LTP-associated increase in the release (or activation) of proteases into extracellular space, leading to proteolytic breakdown of blood clots in the vicinity of the cannula. Evidence in favour of this hypothesis is provided.

The ability of a depolarizing prepulse to increase the rate of activation of IBa has been examined in cultured sensory neurones of the rat. Both in control neurones and in the presence internally of the guanine nucleotide analogue, guanosine 5'-O-3-thiotriphosphate (GTP gamma S) which markedly slows the rate of activation of IBa and reduces its amplitude, a depolarizing prepulse increased the rate of activation of IBa, but did not increase its amplitude measured at the end of the 100 ms voltage step. The calcium channel antagonist (-)-202-791, which we have previously shown to increase the amplitude of IBa in the presence of GTP gamma S, did not occlude the response to a depolarizing prepulse, suggesting that the mechanism of action of (-)-202-791 is not to disrupt the interaction of the channels with activated G proteins.

The present chapter summarizes the results of experiments aimed at identifying macromolecules whose extracellular concentration is increased during long-term potentiation (LTP). The main approach has been sampling of extracellular fluid, using a push-pull cannula, before and after the induction of LTP.

Calcium channel currents have been recorded from cultured rat sensory neurons at clamp potentials of between -30 and +120 mV. At large depolarizing potentials between +50 and +120 mV, the current was outward. This outward current was shown to be largely due to ions passing through calcium channels, because it was substantially although generally incompletely blocked by Cd2+ (1 mM) and omega-conotoxin (1 microM). Internal GTP-gamma-S (100 microM) and to a lesser extent GTP (1 mM) reduced the amplitude and slowed the activation of the outward, as well as the inward calcium channel current. Baclofen (100 microM) reversibly inhibited both the inward and outward currents. These results suggest that the effect of baclofen and G protein activation on calcium channel currents is not due to a shift in the voltage-dependence of channel availability.

Glutamate release has been examined from cultured cerebellar granule neurons in the rat using the technique of prelabelling the releasable pool of glutamate with [3H]glutamine. Glutamate release was stimulated in control neurons by 2-min incubation with 50 mM K+, or in neurons continuously depolarized in Ca2(+)-free 50 mM K+ medium, by 2-min incubation with medium containing 5 mM Ca2+. The ability of the Ca2(+)-channel agonist (+)-202-791 to increase the stimulated release of [3H]glutamate was approximately doubled in the depolarized condition. The antagonist enantiomer (-)-202-791 produced a small inhibition of K(+)-stimulated release, whereas (-)-202-791 completely inhibited Ca2(+)-stimulated release from depolarized neurons at concentrations greater than 10 nM. (-)-Baclofen (100 microM) inhibited transmitter release similarly (25-30%) under the two conditions. Calcium-channel currents were recorded from cultured dorsal root ganglion neurons under control conditions at a holding potential of -80 mV, or in neurons depolarized to -30 mV. (-)-202-791 produced a greater effect at -30 than at -80 mV although even at -30 mV the inhibition was slow in onset and incomplete. (-)-Baclofen (100 microM) inhibited the amplitude of the calcium-channel current at both holding potentials by 30-50%, although it did not clearly slow activation of the current at the depolarized holding potential. The GABAB receptors associated with inhibition of glutamate release and of calcium-channel currents were both markedly blocked by phaclofen but not by 2-OH-saclofen. These findings suggest that the GABAB receptor associated with inhibitory modulation of transmitter release, and that associated with inhibition of calcium-channel currents show pharmacological similarities, and are able to exert their action even at levels of steady depolarization at which most N-type channels should be inactivated.

Some facts and figures to illuminate the debate on women in science: only 1 of 44 recipients of the Colworth medal, given annually to a promising British biochemist under 35, is a woman. And just 1 of 27 plenary lectures at recent Biochemical Society annual meetings was given by a woman. Across the Atlantic, however, there have been some advances. The National Institutes of Health has made significant changes to its pioneer awards to give women a better chance to compete. This issue is still a major topic on the Nature news blog; go to http://tinyurl.com/kacrg to join in.

Stargazin (γ2) and the closely related γ3, and γ4 transmembrane proteins are part of a family of proteins that may act as both neuronal voltage-dependent calcium channel (VDCC) γ subunits and transmembrane α-amino-3-hydroxy-5-methyl-4-isoxazoleproponinc (AMPA) receptor regulatory proteins (TARPs). In this investigation, we examined the distribution patterns of the stargazin-like proteins γ2, γ3, and γ4 in the human central nervous system (CNS). In addition, we investigated whether human γ2 or γ4 could modulate the electrophysiological properties of a neuronal VDCC complex transiently expressed in Xenopus oocytes. The mRNA encoding human γ2 is highly expressed in cerebellum, cerebral cortex, hippocampus and thalamus, whereas γ3 is abundant in cerebral cortex and amygdala and γ4 in the basal ganglia. Immunohistochemical analysis of the cerebellum determined that both γ2 and γ4 are present in the molecular layer, particularly in Purkinje cell bodies and dendrites, but have an inverse expression pattern to one another in the dentate cerebellar nucleus. They are also detected in the interneurons of the granule cell layer though only γ2 is clearly detected in granule cells. The hippocampus stains for γ2 and γ4 throughout the layers of the every CA region and the dentate gyrus, whilst γ3 appears to be localized particularly to the pyramidal and granule cell bodies. When co-expressed in Xenopus oocytes with a CaV2.1/β4 VDCC complex, either in the absence or presence of an α2δ2 subunit, neither γ2 nor γ4 significantly modulated the VDCC peak current amplitude, voltage-dependence of activation or voltage-dependence of steady-state inactivation. The human γ2, γ3 and γ4 stargazin-like proteins are detected only in the CNS and display differential distributions among brain regions and several cell types in found in the cerebellum and hippocampus. These distribution patterns closely resemble those reported by other laboratories for the rodent orthologues of each protein. Whilst the fact that neither γ2 nor γ4 modulated the properties of a VDCC complex with which they could associate in vivo in Purkinje cells adds weight to the hypothesis that the principal role of these proteins is not as auxiliary subunits of VDCCs, it does not exclude the possibility that they play another role in VDCC function.

In contrast to excitable tissues where calcium channels are well characterized, the nature of the B lymphocyte calcium channel is unresolved. Here, we demonstrate by single cell analysis of freshly isolated rat B cells that the anti-immunoglobulin (Ig)-induced calcium influx takes place through a channel which shares pharmacologic and serologic properties with the L-type calcium channel found in excitable tissues. It is sensitive to the dihydropyridines nicardipine and Bay K 8644, to calciseptine, and to an anti-peptide antibody raised against the alpha1 subunit of the L-type calcium channel, but is voltage-insensitive. Anti-alpha1 and anti-alpha2 antibodies stain B but not T lymphocytes. Application of a cGMP agonist, measurement of cGMP levels in anti-Ig-stimulated B cells, and examining the effect of a guanylyl cyclase inhibitor on the anti-Ig response show that cGMP mediates the influx. This possibly involves a cGMP-dependent protein kinase. The anti-Ig-induced response is not abolished by prior treatment of B cells with a high dose of thapsigargin. These findings undermine the widely held belief of a categorical divide between excitable and non-excitable tissue calcium channels, demonstrate the limitations of the capacitative calcium influx theory, and point to a distinction between the calcium response mechanisms utilized by B and T lymphocytes.

Unilateral electrolytic lesions were made in the areas of the dorsal or the ventral noradrenergic bundles in rats to see if the motor and biochemical effects of unilateral electrolytic locus coeruleus lesions could be reproduced. Lesions of the locus coeruleus result in transient contraversive circling in response to systemically administered apomorphine and d -amphetamine, a transient rise in ipsilateral striatal dopamine and a permanent fall in ipsilateral cerebral cortical noradrenaline. Electrolytic lesions of the dorsal noradrenergic bundle area produced the opposite motor effect, with permanent ipsiversive circling in response to systemically administered apomorphine or d -amphetamine and a reduction in ipselateral cerebral cortical noradrenaline, but no increase in striatal dopamine. Control unilateral electrolytic lesions placed in the adjacent tectum sparing the dorsal bundle area produced neither rotational behaviour nor changes in brain noradrenaline or dopamine levels. Although the damaged area producing ipsiversive rotation included the dorsal noradrenergic bundle, it seems probable that damage to other structures caused this response. Electrolytic lesions of the dorsal bundle did not mimic the effects of electrolytic lesions of the locus coeruleus, and it is unlikely that this ascending pathway carries the fibres responsible for the circling behaviour and changes in striatal dopamine seen with locus coeruleus lesions. Unilateral electrolytic lesions restricted to the ventral noradrenergic bundle caused transient postoperative contraversive circling in response to apomorphine or d -amphetamine, and a transient elevation in the ipsilateral striatal dopamine content; these findings were present 5 days after operation and had disappeared by 1–3 months. Thus lesions of the ventral noradrenergic bundle alone reproduced the effects of lesions of the locus coeruleus. Extension of the lesion to involve the region of the dorsal noradrenergic bundle resulted in ipsiversive rotation in response to systemic apomorphine and d -amphetamine, even though the increase in ipsilateral striatal dopamine content still occurred. It is suggested that the similar findings produced by locus coeruleus and ventral bundle lesions are due to damage to noradrenergic fibres which pass from the locus coeruleus to the nigrostriatal system via the ventral bundle. This proposed pathway is envisaged as normally enhancing nerve impulse flow between the substantia nigra and the striatum. Alternatively, the circling produced by locus coeruleus lesions may be due to secondary alterations in raphe 5-hydroxytryptamine activity, which subsequently affects the nigrostriatal dopamine system.

Unilateral electrolytic lesions in the region of the locus coeruleus in rats result in contraversive circling behaviour following the administration of systemic apomorphine or dexamphetamine. In a large group of such animals there was a very good correlation between histological damage to the locus coeruleus and the presence of such rotation. There was a progressive increase in the percentage of rats showing this drug-induced behaviour as ipsilateral cerebral cortical noradrenaline content decreased. Unilateral injections of 6-OHDA produce more selective damage to the noradrenergic locus coeruleus and result in a similar, although less marked, apomorphine- and dexamphetamine-induced rotation. Unlike the rats with electrolytic lesions of the locus coeruleus, those with 6-OHDA injections did not show an increase in ipsilateral striatal dopamine content. It seems likely that the motor asymmetry seen with lesions which involve the locus coeruleus is due to damage to its noradrenergic cells. A possible explanation of the differences between electrolytic and 6-OHDA induced lesion is that the 6-OHDA produces damage to the ventral noradrenergic bundle as well as the locus coeruleus and results in an increase in ipsilateral cerebral cortical 5-hydroxytryptamine.

Voltage-activated Ca2+ channel currents were recorded from cultured rat dorsal root ganglion neurones using the whole cell clamp technique. Inclusion of 500 microM guanosine 5'-O(3-thio)triphosphate (GTP-gamma-S) in the patch pipette solution resulted in inhibition of the transient component of the Ca2+ channel current. The slowly activating, relatively non-activating Ca2+ channel current which remained in the presence of intracellular GTP-gamma-S was potentiated by the 1,4-dihydropyridine Ca2+ channel agonist Bay K 8644 (5 microM). The percentage increase induced by Bay K 8644 was greater in cells with internal GTP-gamma-S than in control cells. Pretreating cells with pertussis toxin resulted in Bay K 8644 inhibiting rather than increasing Ca2+ channel currents. These results support the hypothesis that L-type Ca2+ channel activity underlies the sustained Ca2+ channel currents which are not inhibited by GTP-gamma-S and that Ca2+ channel ligands require activation of a pertussis toxin-sensitive G protein to exert their agonist effects.

The effect of the glutamate antagonist γ- d -glutamylglycine on the induction of long-term potentiation in the dentate gyrus has been investigated in vivo. γ- d -glutamyglycine (10−3 M) perfused through a push-pull cannula into the dentate gyrus, rapidly reduced perforant path evoked potentials. Application to the perforant path of a high-frequency train (250 Hz, 500 ms), which in control animals reliably produced long term potentiation, had no effect on the evoked potentials when applied during blockade by γ- d -glutamylglycine. This result was obtained despite the inability of γ- d -glutamylglycine completely to inhibit the evoked potentials. However, when standard medium was reintroduced, potentiation was revealed in animals that had received the high-frequency train, whereas in animals that had received no high-frequency train during γ- d -glutamylglycine inhibition the potentials returned only to pre-drug levels. In additional experiments, in which the dentate gyrus was continuously perfused with [3H]glutamine, and the steady state outflow of [3H]glutamate was measured, it was observed that γ- d -glutamylglycine (10−3 M) increased the steady state release of [3H]glutamate into the perfusate. From this result it is likely that γ- d -glutamylglycine does not have any presynaptic inhibitory activity at the perforant path-granule cell synapse. The results indicate that a high frequency train applied to the perforant path during a period of inhibition by γ- d -glutamylglycine was able to induce long term potentiation, whose expression was, however, masked until the glutamate antagonist was removed. This leads to the conclusion that activation of the class of post-synaptic glutamate receptors blocked by γ- d -glutamylglycine is not essential for the induction of long term potentiation, although involvement of other postsynaptic receptors not blocked by this antagonist cannot be ruled out. However, these results are also consistent with the hypothesis that there is a presynaptic component, involving increased transmitter release from perforant path terminals, in the mechanism of long-term potentiation.

1. The effect of intracellular application of the hydrolysis-resistant GTP and GDP analogues, guanosine 5'-O-3-thiotriphosphate (GTP-gamma-S), and guanosine 5'-O-2-thiodiphosphate (GDP-beta-S) has been examined on voltage-activated calcium-channel currents and the ability of the gamma-aminobutyric acid B agonist baclofen to inhibit them, in cultured rat dorsal root ganglion (d.r.g.) neurones. 2. Under control conditions, the calcium-channel current, recorded using the whole-cell patch technique with Ba2+ rather than Ca2+ as the permeant divalent cation, consists of an inactivating and a sustained current. In the presence of 500 microM-GTP-gamma-S included in the patch pipette, the calcium-channel current was activated more slowly and was largely non-inactivating during the 100 ms depolarization voltage step. The effects of GTP-gamma-S were abolished by pre-treatment of cells with pertussis toxin. 3. The calcium-channel current recorded in the presence of 500 microM-GDP-beta-S had a more marked transient component than the control calcium-channel current. The proportion of transient calcium-channel current in the presence of GDP-beta-S was not reduced in Na+-free medium. 4. No statistically significant effects of GTP-gamma-S and GDP-beta-S were observed on the calcium-activated potassium current IK(Ca), the transient outward potassium current activated in Ca2+-free medium, or on the inwardly rectifying current (Ih) activated by hyperpolarization. 5. GTP-gamma-S increased the ability of baclofen to inhibit calcium-channel currents, whereas this was decreased by GDP-beta-S and by pre-treatment of cells with pertussis toxin. The half-maximal effective dose (EC50) for baclofen was 2 microM in the presence of GTP-gamma-S, 15 microM for control and 50 microM in the presence of GDP-beta-S. Comparable results were obtained using a single concentration of the adenosine agonist 2-chloroadenosine (2-CA, 0.05 microM) to inhibit calcium-channel currents; its effect was significantly increased by GTP-gamma-S and reduced by GDP-beta-S. 6. The ability of baclofen to inhibit calcium-channel currents was not affected by 1 microM-forskolin or 50 microM-intracellular cyclic AMP. 7. It is concluded that calcium channels in d.r.g.s are associated with a nucleotide binding protein, and that this mediates the effect of baclofen and 2-CA on calcium-channel currents. The ability of GTP-gamma-S to inhibit the transient component of calcium-channel currents in the absence of agonist may represent a means of differentially regulating calcium-channel activity.

To examine the role played by endogenous adenosine in the modulation of transmitter release in the CNS, the effect of adenosine antagonists has been studied. Two systems have been used: EPSPs recorded from pyramidal cells in organotypic hippocampal cultures; and release of newly synthesized [3H]glutamate from cerebellar granule cells in dissociated culture. Bath application of 0.1-1 microM 8-phenyltheophylline (8-PT) reversibly increased both the number and size of spontaneous EPSPs and caused bursting activity in some cells. This effect was blocked by the glutamate antagonist gamma-D-glutamylglycine (DGG) (1 mM) but not by atropine (10 microM) or bicuculline (100 microM). Another adenosine antagonist isobutylmethylxanthine (IBMX, 10 microM) had a similar effect to 8-PT. Spontaneous activity in pyramidal cells and that induced by adenosine antagonists was blocked by the adenosine agonist 2-chloroadenosine (2-CA) (0.2-20 microM). 8-PT (10 microM) markedly potentiated K+-stimulated release of newly synthesized glutamate, and also enhanced basal glutamate release. The agonist (-)-phenylisopropyladenosine ((-)-PIA, 2 microM) which is relatively selective for A1 receptors, reduced by 19 +/- 5% the 8-PT-induced enhancement, and reduced K+-stimulated glutamate release in the absence of 8-PT to a similar extent. In contrast 5'-N-ethylcarboxamido adenosine (NECA, 2 microM), which is a relatively selective A2 agonist, slightly enhanced glutamate release. From these results it is likely that 8-PT potentiates glutamate release in both systems by blocking the effect of endogenous adenosine on presynaptic A1 receptors.

The levels of cAMP-dependent protein kinase (type I), or cGMP-dependent protein kinase, or protein I, and of a 23,000 MW substrate for the cGMP-dependent protein kinase were measured in cerebella from normal rats and in the cerebella from rats in which a selective loss of interneurons in the cerebellar cortex had been produced by X-irradiation. A decrease was observed in the concentrations of cAMP-dependent protein kinase and of protein I, whereas an increase was observed in the concentrations of cGMP-dependent protein kinase and of the 23,000 MW substrate. The data, taken together with the results of other studies, support the interpretation that cAMP-dependent protein kinase and protein I are distributed throughout the cerebellum, but that cGMP-dependent protein kinase and the 23,000 MW substrate are highly concentrated in Purkinje cells.