Nerve growth factor mRNA stability is controlled by a cis-acting instability determinant in the 3'-untranslated region.

Nerve growth factor (NGF) mRNA is rapidly degraded in many non-neuronal cell types with a half-life of between 30 and 60 min. Similar to other short-lived mRNAs the 3'-untranslated region (3'-UTR) of the NGF mRNA contains a short AU nucleotide-rich sequence. To implicate this region as a cis-acting determinant of NGF mRNA instability, expression vectors containing NGF cDNA with and without the 3'-UTR, and vectors containing only the 3'-UTR were constructed and used in cell transfection experiments. Transfection of HEK293 or NIH3T3 cells with these expression vectors followed by measurement of NGF mRNA half-life indicated that NGF mRNA without the AU-rich 3'-UTR was approximately 3-fold more stable than NGF mRNA containing the 3'-UTR. Similar results were seen in a polysome-based cell-free RNA decay assay using NGF mRNA with and without the 3'-UTR prepared from transfected cells. Addition of a short RNA containing the AU-rich 3'-UTR to the cell-free RNA decay system prolonged the half-life of the full-length NGF mRNA, suggesting competition between these two RNA species for polysome-associated factors which degrade the NGF mRNA. Moreover, transfection of HEK293 or astroglial cells with vectors designed to express only the AU-rich region of the 3'-UTR resulted in enhanced expression of NGF mRNA. The results indicate that the 3'-UTR of the NGF mRNA contains a cis-acting instability determinant which, perhaps by interacting with trans-acting RNA-binding proteins, controls the rate of NGF mRNA turnover.

Okadaic acid stimulates nerve growth factor production via an induction of interleukin-1 in primary cultures of cortical astroglial cells.

Neonatal rat cortical astroglial cells in primary culture synthesize and secrete interleukin-1 beta (IL-1) and nerve growth factor (NGF). Treatment of astrocytes with okadaic acid (OA), an inhibitor of phosphoprotein phosphatases, dramatically increased both IL-1 and NGF mRNA content (about 50-fold) with maximal induction seen at 20-30 nM OA. The induction of IL-1 mRNA preceded that of NGF mRNA and was maximal after 9 h of treatment. OA increased IL-1 mRNA half-life by about 10-fold similar to the reported stabilization of the NGF mRNA. Addition of an IL-1 receptor antagonist dose-dependently inhibited the secretion of NGF stimulated by OA and IL-1. The results indicate that OA profoundly stimulates IL-1 expression in glial cells by enhancing IL-1 mRNA stability and that glial cell-derived IL-1 acts in a paracrine/autocrine manner to stimulate NGF production.

Okadaic acid increases nerve growth factor secretion, mRNA stability, and gene transcription in primary cultures of cortical astrocytes.

Neonatal rat cortical astrocytes in primary culture synthesize and secrete nerve growth factor (NGF) in response to cytokines, growth factors, and activators of protein kinases. To further implicate a protein phosphorylation mechanism in the regulation of NGF expression, astrocytes were treated with okadaic acid and calyculin A, inhibitors of phosphoprotein phosphatases 1 and 2A. Okadaic acid dramatically increased both NGF mRNA content (50-fold) and NGF secretion (100-fold) in astrocytes, while calyculin A, which has a spectrum of phosphatase inhibitory activity different from okadaic acid, failed to augment NGF expression. The increased mRNA accumulation was due mainly to an increase (4-fold) in the half-life of the NGF mRNA following 9 or 24 h of treatment. Nuclear run-on assays indicated that okadaic acid also activated NGF gene transcription, which was preceded by an induction of c-fos and c-jun gene transcription. The induction of NGF expression by okadaic acid appeared independent from protein kinase C activity because down-regulating protein kinase C activity failed to decrease the okadaic acid stimulation. In contrast, interleukin-1 beta acted synergistically with okadaic acid to stimulate NGF secretion. The results indicate that okadaic acid profoundly stimulates NGF expression in astrocytes mainly by enhancing NGF mRNA stability and suggest important roles for phosphoprotein phosphatases in regulating NGF production.

Transcriptional and posttranscriptional mechanisms involved in the interleukin-1, steroid, and protein kinase C regulation of nerve growth factor in cortical astrocytes.

Neonatal rat cortical astrocytes in primary culture synthesize and secrete nerve growth factor (NGF). Treatment of astrocytes with interleukin-1 beta (IL-1) or the protein kinase C (PKC) activator 12-O-tetradecanoylphorbol 13-acetate (TPA) increased NGF mRNA content by six- to 10-fold, followed in time by increases in cell content and cell secretion of NGF. Dexamethasone potently inhibited the effects of IL-1 and TPA on astroglial cell NGF expression. The action of IL-1 was not mediated by PKC because treatment of cells with maximal concentrations of both IL-1 and TPA gave an additive increase in NGF mRNA content and NGF secretion, and because down-regulating PKC activity failed to inhibit the stimulatory effects of IL-1. Moreover, both agents increased NGF gene transcription in nuclear run-on assays, but only IL-1 significantly stabilized the NGF mRNA. An analysis of the effects of IL-1 and TPA on immediate early gene expression indicated that IL-1 preferentially induced c-jun gene expression, whereas TPA greatly increased c-fos and zif/268 gene expression. These results suggest that IL-1 activates c-jun and NGF gene expression, and NGF mRNA stabilization in astrocytes by a distinct PKC-independent signaling pathway.

Arachidonic acid lipoxygenation may mediate interleukin-1 stimulation of nerve growth factor secretion in astroglial cultures.

Interleukin-1 beta (IL-1) stimulates by about fivefold NGF secretion from rat neonatal cortical astrocytes in primary culture. We investigated the possible intracellular second messenger mechanisms involved in the IL-1 induced NGF secretion. Basal NGF secretion did not require extracellular Ca2+, whereas Ca2+ was necessary for the maximal NGF secretion stimulated by IL-1 (10 units/ml). The protein kinase C activator TPA stimulated by sixfold NGF secretion, but in this case, TPA acted synergistically with IL-1 to increase NGF secretion. Treatment of cells with the phospholipase A2 inhibitor mepacrine (30 microM) inhibited basal (by 50%) and IL-1 stimulated (by 80%) NGF secretion. Indomethacin, a cyclooxygenase inhibitor, produced a slight increase in basal NGF secretion at low concentrations, while PGE2 (10 microM) inhibited basal and IL-1 stimulated NGF secretion. In contrast, treatment of cells with nordihydroguaiaretic acid (NDGA), a lipoxygenase inhibitor, blocked in a concentration-dependent manner (IC50 = 10 microM) IL-1 stimulation of NGF secretion. The leukotriene LTB4 increased basal NGF secretion and this effect was not additive with IL-1 when both agents were added at saturating concentrations, indicating a common mechanism of action for these two agents. Thus, one possible mechanism by which IL-1 stimulates NGF secretion from astrocytes is by activation of the phospholipase A2-lipoxygenase pathway.

Cortical neurons inhibit basal and interleukin-1-stimulated astroglial cell secretion of nerve growth factor.

Primary cultures of neonatal rat cortical neurons and astrocytes synthesize and secrete nerve growth factor (NGF). Co-culturing neurons with astrocytes decreased NGF secretion in the co-cultures. The inhibition of co-culture NGF secretion was partially reversible upon selectively decreasing the number of neurons by glutamate treatment. Interleukin-1 beta (IL-1) stimulated NGF secretion from astrocytes, and the magnitude of this secretion was decreased in the co-cultures. Thus, co-culture with neurons decreases astroglial cell secretion of NGF and down-regulates astroglial responsiveness to IL-1.

Mechanism of nerve growth factor mRNA regulation by interleukin-1 and basic fibroblast growth factor in primary cultures of rat astrocytes.

Neonatal rat cortical astrocytes in primary culture synthesize and secrete nerve growth factor (NGF). Interleukin-1 beta(IL-1) and basic fibroblast growth factor (bFGF) treatment of astrocytes increased NGF mRNA content by about 2-fold. The effect of these two factors was specific, because other growth factors, such as tumor necrosis factor-alpha, insulin-like growth factor-1, and epidermal growth factor, failed to change NGF mRNA content. The concentrations of IL-1 and bFGF causing half-maximal stimulation were 1 unit/ml and 1 ng/ml, respectively. The increase in NGF mRNA elicited by IL-1 and bFGF was maximal at 3 hr of incubation. In the presence of IL-1 this increase persisted for 36 hr, whereas in the presence of bFGF the initial increase in NGF mRNA was followed by a decrease to 50% of control levels after 24 hr of incubation. Readdition of bFGF after 24 hr of treatment gave a similar increase in NGF mRNA content, suggesting that the decrease at 24 hr was not due to receptor desensitization. The effect of IL-1 was reversible, because removal of IL-1 after 3 hr of incubation resulted in a decrease of NGF mRNA content to control levels by 6 hr, whereas a readdition of IL-1 at this time led to a 2-3-fold increase in NGF mRNA content after an additional 3 hr of treatment. This second increase in NGF mRNA was also maintained for several hours. The combined treatment of astrocytes with maximally effective doses of IL-1 and bFGF produced an additive increase in NGF mRNA content, suggesting that different mechanisms are operative. Treatment of astrocytes with cycloheximide increased (about 6-fold) NGF mRNA content, and this content failed to increase further with IL-1 or bFGF treatment. Experiments using actinomycin D indicated that IL-1 increased the stability of the NGF mRNA. bFGF treatment failed to change this parameter. Thus, IL-1 increases NGF mRNA content in astrocytes, at least in part, by stabilizing mRNA, whereas bFGF does not affect mRNA stability but may act at the level of NGF gene transcription.

Chronic treatment with scopolamine and physostigmine changes nerve growth factor (NGF) receptor density and NGF content in rat brain.

Nerve growth factor (NGF) and NGF receptors were measured in cortex and hippocampus of rats treated with drugs affecting cholinergic neurotransmission. High (Kd = 0.045 nM) and low (Kd = 21 nM) affinity 125I-NGF binding sites were present in both cortical and hippocampal membranes with hippocampus containing higher numbers of both sites than cortex. Chronic treatment of rats with the muscarinic receptor antagonist scopolamine (5 mg/kg, twice daily) decreased the density of high- and low-affinity sites by 50-90% in cortical and hippocampal membranes. These changes were seen after 7 days, but not 3 days, of scopolamine treatment. Chronic infusion of physostigmine (1 mg/kg/day) using minipumps increased the number of high- and low-affinity sites in cortex 3- and 6-fold, respectively. The changes in receptor-binding parameters induced by physostigmine were transient as they were evident after 3 days of treatment, but returned to control levels after 7 days. NGF content in cortex and hippocampus was reduced by about 50% following 7, but not 3, days of chronic physostigmine infusion. In contrast, scopolamine treatment failed to change NGF levels in the cholinergic neuronal target regions but it decreased NGF content in the septal area. The content of NGF mRNA in the cortex measured by Northern blot analysis failed to change following either scopolamine or physostigmine treatment. The results suggest that the levels of NGF and NGF receptors in the target regions of cholinergic neurons are regulated by the extent of cholinergic neurotransmitter activity.

Regulation by interleukin-1 of nerve growth factor secretion and nerve growth factor mRNA expression in rat primary astroglial cultures.

Primary cultures of neonatal rat cortical astrocytes contain low cellular levels (about 2 pg/mg of protein) of nerve growth factor (NGF), but secrete significant amounts of NGF into the culture medium (about 540 pg of NGF/mg of cell protein/38-h incubation). Incubation of astrocytes with interleukin-1 (IL-1) increased the cellular content of NGF and the amount secreted by about threefold. In comparison, cerebellar astrocytes secreted significant amounts of NGF, and the secretion was also stimulated by IL-1. The stimulatory action of IL-1 on astrocytes prepared from cortex was dose- and time-dependent. Concentrations of IL-1 causing half-maximal and maximal stimulation of NGF secretion were 1 and 10 U/ml, respectively). Maximal NGF secretion induced by IL-1 (10 U/ml) was seen following 38 h of incubation. The basal secretion of NGF was reduced by about 50% under Ca2(+)-free conditions; however, the percent stimulation of NGF secretion by IL-1 was the same in the absence or presence of Ca2+. The stimulatory action of IL-1 was specific, because other glial growth factors and cytokines were almost ineffective in stimulating NGF secretion from cortical astroglial cells. IL-1 treatment also increased cellular NGF mRNA content twofold. The results indicate that IL-1 specifically triggers a cascade of events, independent of cell growth, which regulate NGF mRNA content and NGF secretion by astrocytes.

Serotonin-stimulated protein phosphorylation in aortic smooth muscle cells.

The effects of serotonin on the formation of inositol phosphates and protein phosphorylation were examined in cultured smooth muscle cells. Serotonin stimulated the formation of [3H]inositol monophosphate, [3H]inositol bisphosphate and [3H]inositol trisphosphate. This effect was prevented by 5-HT2 specific antagonist, 6-methyl-1-(1-methylethyl)ergoline-8-carboxylic acid, 2-hydroxy-1-methylpropyl ester [Z]-2-butenedioate (LY53857). Serotonin stimulated the phosphorylation of many polypeptides, among which a 20 kDa polypeptide was the most prominent. The phosphorylation was also inhibited by LY53857. LY53857 alone produced no effects on protein phosphorylation. The 20 kDa polypeptides were also phosphorylated by the addition of 12-O-tetradecanoylphorbol-13-acetate. These results suggest that serotonin stimulates protein phosphorylation through 5-HT2 receptors and possibly activates protein kinase C in intact vascular smooth muscle cells.

Hippocampal membranes contain a neurotrophic activity that stimulates cholinergic properties of fetal rat septal neurons cultured under serum-free conditions.

Primary cultures of fetal rat septal neurons were used to identify a membrane-associated cholinergic neurotrophic activity. Under serum-free culture conditions, approximately 98% of the septal cells are neurons, and approximately 6% of the neurons are cholinergic as determined immunocytochemically. Crude membranes prepared from rat hippocampal homogenates stimulate choline acetyltransferase (ChAT) activity in treated septal neurons. The membrane-associated trophic activity is apparent at lower protein concentrations than activity present in the soluble fraction and is unevenly distributed in various brain regions; it is highest in hippocampus and striatum and negligible in cerebellum. Membrane trophic activity is developmentally regulated, is heat and trypsin sensitive, and increases the rate of expression of ChAT in septal neurons. Upon gel filtration chromatography of a high-salt membrane extract, trophic activity elutes as a broad peak in the 500 kilodalton (kD) molecular mass range. Stimulation of septal neuronal ChAT activity by either crude membranes or partially purified preparations is not inhibited by antibodies against nerve growth factor (NGF), and its maximal activity is additive to maximally active doses of NGF. The results indicate that hippocampal membranes contain cholinergic neurotrophic activity which may be important for the development of septal cholinergic neurons.

Negative feedback regulation of the content of proenkephalin mRNA in chromaffin cell cultures.

The content of proenkephalin messenger RNA (PEmRNA) in cultured bovine adrenal chromaffin cells was reduced in the presence of reserpine (1 nM to 0.1 microM) with a return to basal levels 3 days after removal of the drug. In these cells, the basal release of Met5-enkephalin-Arg6-Phe7 immunoreactivity (MERF-IR) into the medium was significantly decreased when the cultures were pretreated with 0.2 microM reserpine for 3 days. The addition of 0.1 microM etorphine for 3 days also decreased the basal release of MERF-IR without depleting stores of catecholamines. Neither drug modified the total (cells + medium) amount of MERF-IR. In contrast, reserpine was without effect on levels of PEmRNA or release of Met5-enkephalin immunoreactivity (ME-IR) in primary cultures of the striatum of the fetal rat. The present data establish a correlation between inhibition of the secretion of enkephalin and reduced accumulation of its specific mRNA, suggesting a negative feedback inhibition by low molecular weight enkephalins.

Substrates for protein kinase C in a cell free preparation of rat aorta smooth muscles.

Protein phosphorylation has been studied in a cell free system of rat aorta smooth muscles. Addition of Ca2+ caused phosphorylation of several proteins. The addition of phosphatidylserine or calmodulin together with Ca2+ further increased the phosphorylation of proteins with apparent molecular weights of 20 and 92.5 kilodaltons. The activators of protein kinase C, 12-0-tetradecanoylphorbol-13-acetate and 1,2-diolein, increased phosphorylation of the protein bands of similar molecular weight to those increased by phosphatidylserine in the presence of Ca2+, whereas the biologically inactive phorbol ester, 4 alpha-phorbol-12,13 didecanoate (4 alpha PDD) failed to change the pattern of protein phosphorylation. These results show that proteins present in smooth muscle of rat aorta with molecular weights of 20 and 92.5 kilodaltons are substrates for protein kinase C.

gamma-Aminobutyric acid- and benzodiazepine-induced modulation of [35S]-t-butylbicyclophosphorothionate binding to cerebellar granule cells.

t-Butylbicyclophosphorothionate (TBPS) is a bicyclophosphate derivative with potent picrotoxin-like convulsant activity that binds with high affinity and specificity to a Cl- channel-modulatory site of the gamma-aminobutyric acid (GABA)/benzodiazepine receptor complex. Using intact cerebellar granule cells maintained in primary culture, we have studied the modifications induced by GABA and diazepam on the ion channel-modulatory binding site labeled by [35S]TBPS. At 25 degrees C, and in a modified Locke solution, the [35S]TBPS specific binding, determined by displacing the radioligand with an excess (10(-4) M) of picrotoxin, was approximately 70% of the total radioactivity bound to the cells. [35S]TBPS specific binding was saturable with a Kd of approximately 100 nM, a Bmax of approximately 440 fmol/mg of protein, and a Hill coefficient of 1.18. Neither cerebellar astrocytes maintained in culture for 2 weeks nor a neuroblastoma cell line (NB-2A) exhibited any specific [35S]TBPS binding. Muscimol (0.3 to 5 microM) enhanced and bicuculline (0.1 to 5 microM) inhibited [35S]TBPS specific binding to intact cerebellar granule cells. The effect of muscimol and bicuculline on [35S]TBPS binding was noncompetitive. Muscimol (0.1 to 5 microM) reversed bicuculline inhibition in a dose-dependent fashion but failed to reverse picrotoxin-induced inhibition. [35S]TBPS binding was also modulated by benzodiazepine receptor ligands. The binding was increased by diazepam and decreased by 6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylic acid methylester. Muscimol (0.05 microM) failed to reverse bicuculline inhibition in the absence of diazepam, but it became effective in the presence of 0.1 to 1 microM diazepam.(ABSTRACT TRUNCATED AT 250 WORDS)

Ca2+ and phospholipid-dependent protein kinase activity and phosphorylation of endogenous proteins in bovine adrenal medulla.

Soluble and membrane fractions of bovine adrenal medulla contain several substrates for the Ca2+/phospholipid-dependent and cyclic AMP-dependent protein kinases. The phosphorylation of soluble proteins (36 and 17.7 kilodaltons) and a membrane protein (22.5 kilodaltons) showed an absolute requirement for the presence of both Ca2+ and phosphatidylserine; other substrates showed less stringent phosphorylation requirements and many of these proteins were specific for each of the protein kinases. The Ca2+/phospholipid-dependent phosphorylation was rapid, with effects seen as early as at 30 s of incubation. Measurement of enzyme activities with histone H1 as an exogenous substrate demonstrated that the Ca2+/phospholipid-dependent protein kinase was equally distributed between the soluble and membrane fractions whereas the cyclic AMP-dependent enzyme was predominantly membrane-bound in adrenal medulla and chromaffin cells. The activity of the soluble Ca2+/phospholipid-dependent protein kinase of adrenal medulla was found to be about 50% of the enzyme level present in rat brain, a tissue previously shown to contain a very high enzyme activity. These results suggest a prominent role for the Ca2+/phospholipid-dependent protein kinase in chromaffin cell function.

Regulation of the GABA receptor complex by a phosphorylation mechanism.

GABA- modulin , a regulatory component of the GABA/benzodiazepine receptor complex, is phosphorylated by cyclic-AMP-, Ca/calmodulin-, and Ca/phospholipid-dependent protein kinases at distinct sites in the molecule. Phosphorylation of GM by the cyclic-AMP-dependent process results in a complete loss of GM inhibitory activity on specific 3H-GABA binding to synaptic membrane recognition sites. The effect of the Ca2+-dependent phosphorylation, dependent on CaM or PS, of GM is presently unknown but may involve a synergistic or antagonistic action on the cyclic-AMP-dependent phosphorylation. Alternatively, the Ca2+-dependent protein kinases may regulate another function of GM, perhaps its postulated role as a coupler of the GABA/benzodiazepine recognition sites. The results of these experiments strongly implicate a role for protein phosphorylation in the regulation or modulation of GABA receptor function, and such a mechanism may be extrapolated to other neurotransmitter receptor complexes.

GABAergic synapses. Supramolecular organization and biochemical regulation.

Extraneurally released gamma-aminobutyric acid (GABA) interacts with specific recognition sites associated with proteins located in postsynaptic neuronal membranes that function as chloride (Cl-)ionophores. As a result of the interaction between GABA and the recognition sites, Cl- ionophores are opened causing an influx or an efflux of Cl-, depending on the values of the Cl- equilibrium potential and of the membrane potential. Hyperpolarization or depolarization will result from inward or outward Cl- fluxes, respectively. Independently of the change in conductivity elicited by GABA, this amino acid transmitter will reduce the effectiveness of the sodium ion (Na+) excitatory potential. In attempts to elucidate the molecular mechanism, whereby benzodiazepines facilitate the action of GABA on membrane conductance without changing the activity of Cl- or other ionophore, a basic protein (GABA-modulin, GM) has been isolated from rat brain which is similar in structure to the small molecular weight myelin basic protein, found in rodent brain. While GABA-modulin is located in synaptosomes, the small molecular weight myelin basic protein is located in the myelin fraction: more important, GABA-modulin inhibited the high affinity binding of GABA to crude synaptic membranes while the basic myelin protein did not. Also, amino acid composition and molecular weight differentiate the two proteins. The GABA-modulin can be phosphorylated with different stoichiometry by cyclic AMP-dependent protein kinase (4 mol PO4(-3)) or Ca2+-dependent protein kinase (1 mol PO4(-3)). Only cyclic AMP-dependent phosphorylation inhibited the action of GABA-modulin on GABA binding.

Modes of inhibition by acylcarnitines, adriamycin and trifluoperazine of cardiac phospholipid-sensitive calcium-dependent protein kinase.

Palmitoylcarnitine, adriamycin, and trifluoperazine competively inhibited, with respect to phosphatidylserine (a phospholipid cofactor), purified cardiac phospholipid-sensitive Ca2+-dependent protein kinase, with apparent Ki values of 3, 49 and 14 microM respectively. These compounds also inhibited the enzyme competitively with respect to Ca2+ (a metal activator), with corresponding apparent Ki values of 0.8, 140 and 9 microM. A synergistic inhibition was observed when palmitoylcarnitine and trifluoperazine were present in combination. A simple addition inhibition on the other hand, was observed for the combination of either palmitoylcarnitine and adriamycin, or trifluoperazine and adriamycin. 1,3-Diolein decreased the inhibitory effect of trifluoperazine by increasing the affinity of the enzyme for phosphatidylserine. The results indicate that the recently identified phospholipid-sensitive species of Ca2+-dependent protein kinase was inhibited by a variety of agents, probably via their abilities to interfere with a hydrophobic interaction between phospholipid and the enzyme, an interaction presumably required to confer upon the enzyme a Ca2+ sensitivity. Because other long-chain fatty acylcarnitines (stearoyl- and linoleoylcarnitine), short-chain fatty acylcarnitines (such as octanoylcarnitine) and palmitoyl CoA, compared to palmitoylcarnitine, were less active as inhibitors, it is further suggested that lipophilicity as well as other structural determinants are crucial for the ability of compounds to regulate the enzyme activity.

Phosphorylation induces a decrease in the biological activity of the protein inhibitor (GABA-modulin) of gamma-aminobutyric acid binding sites.

gamma-Aminobutyric acid (GABA)-modulin is a brain protein of Mr 16,500 that down-regulates the high-affinity binding site for GABA which is located in crude synaptic membranes. This protein can be phosphorylated in vitro by the catalytic subunit of cAMP-dependent protein kinase and by a partially purified preparation of calmodulin-sensitive Ca2+-dependent protein kinase. The GABA-modulin sites that are phosphorylated by the two enzymes are different, as revealed by HPLC analysis of tryptic digests. The capacity of GABA-modulin to decrease the number of sites that bind [3H]muscimol was completely abolished by phosphorylation of this protein with the cAMP-dependent protein kinase but not with the Ca2+-dependent enzyme. GABA-modulin present in crude synaptic membranes prepared from rat cortex also was shown to be phosphorylated by endogenous protein kinases activated by cAMP, Ca2+ and calmodulin, and Ca2+ and phosphatidylserine. These results suggest a potentially important role for protein kinase and GABA-modulin in the regulation of the number of GABA recognition sites.

Cotransmitters: pharmacological implications.

The discovery that two or more neuroactive substances coexist in the same nerve terminal suggests that two or more neuroactive compounds can be released by nerve impulses simultaneously and probably act cooperatively at postsynaptic sites. This interaction changes the models of synaptic transmission we have used in the past and imposes a reevaluation of current understanding of synaptic pharmacology. Neuroactive substances co-existing in the same axon terminal can function as "primary transmitter" if they activate the receptor-transducer system or as "cotransmitter" if they modulate the gain of the system. Two examples of synaptic mechanisms in which two neuroactive substances coexisting in the same axon terminal appear to function as primary transmitter and cotransmitter are discussed. These examples are: 1. the modulation of the function of nicotinic receptors of chromaffin cells by endogenous opiate peptides stored in the splanchnic nerve and 2. the modulation of GABA receptor function by benzodiazepines. The understanding of the mechanisms by which primary transmitter and cotransmitter interact at the postsynaptic site may be of obvious importance in elucidating the integrative and discriminative function of the nervous system, in interpreting the action of drugs and in developing new therapeutic agents devoid of untoward side effects.