Ras-GRF, the activator of Ras, is expressed preferentially in mature neurons of the central nervous system.

In rodents, the Ras-specific guanine-nucleotide exchange factor (Ras-GRF) is expressed in different areas of the brain and, at a reduced level, also in the spinal cord. No expression of the 140 kDa Ras-GRF was detected in dorsal root ganglia and all other tissues tested. Analysis of primary cultures derived from brain reveals that this exchange factor is only present in neurons of the central nervous system. In primary hippocampal cultures, the expression of Ras-GRF increases in parallel with the onset of a neuronal network and in the whole brain it increases sharply after birth.

Differential distribution of synapsin IIa and IIb mRNAs in various brain structures and the effect of chronic morphine administration on the regional expression of these isoforms.

Quantitative reverse transcriptase-polymerase chain reaction and in situ hybridization techniques were used to determine the regional distribution of synapsin IIa and IIb mRNAs in rat central nervous system and to assess the effect of chronic morphine administration on the gene expression of these two isoforms of synapsin II. These isoforms are members of a family of neuron-specific phosphoproteins thought to be involved in the regulation of neurotransmitter release. Our data demonstrate the widespread distribution, yet regionally variable expression, of synapsin IIa and IIb mRNAs throughout the adult rat brain and spinal cord. The ratios of the relative abundance of synapsins IIa and IIb differed by up to 4.5-fold among the various regions studied. Synapsin IIa and IIb mRNAs were shown to be highly concentrated in the thalamus and in the hippocampus, whereas lower concentrations were found in most other central nervous system structures. In this study, we show differential regulation by morphine of synapsins IIa and IIb in various regions of the brain. In the striatum, a 2.4-fold increase was observed in the levels of synapsin IIa mRNA following chronic morphine regime, whereas no change was found for synapsin IIb. On the other hand, mRNA levels of synapsin IIb in spinal cord of chronically treated rats were markedly decreased (by 62%), while no alterations were observed in synapsin IIa. Selective regulation by morphine has also been demonstrated in several other central nervous system structures. The opiate-induced regulation of the gene expression of synapsin II isoforms could be viewed as one of the cellular adaptations to the persistent opiate effects and may be involved in the molecular mechanism underlying opiate tolerance and/or dependence.

Chronic morphine administration enhances the expression of Kv1.5 and Kv1.6 voltage-gated K+ channels in rat spinal cord.

Prolonged opiate administration leads to the development of tolerance and dependence. These phenomena are accompanied by selective regulation of distant cellular proteins and mRNAs, including ionic channels. Acute opiate administration differentially affects voltage-dependent K+ currents. Whereas, opiate activation of K+ channels is well established opioid-induced inhibition of K+ conductance has also been studied. In this study, we focused on the effect of chronic morphine exposure on voltage-dependent Shaker-related Kv1.5 and Kv1.6 K+ channel gene expression and on Kv1.5 protein levels in the rat spinal cord. Several experimental approaches including in-situ hybridization, RNAse protection, reverse transcriptase-polymerase chain reaction (RT-PCR), Western blotting and immunohistochemistry were employed. We found that motor neurons are highly enriched in Kv1.5 and Kv1.6 mRNA and in Kv1.5 channel protein. Moreover, we found significant increases in the amount of mRNA encoding for these two K+ channels and in Kv1.5 channel protein in the spinal cord of morphine-treated rats, compared with controls. For example, quantitative in-situ hybridization, revealed a 2.1 +/- 0.15- and 2.3 +/- 0.5-fold increase in Kv1.5 and Kv1.6 channel mRNA levels, respectively. Similar results were obtained by semiquantitative RT-PCR analyses. Kv1.5 protein level was increased by 1.9-fold in the spinal cord or morphine-treated rats. Our results suggest that Kv1.5 and Kv1.6 Shaker K+ channels play an important role in regulating motor activity that increases in mRNA and protein levels of the spinal cord K+ channels after chronic morphine exposure could be viewed as a cellular adaptation which compensates for a persistent opioid-induced inhibition of K+ channel activity. These alterations may account, in part, for the cellular events leading to opiate tolerance and dependence.

Adenylylcyclase supersensitization in mu-opioid receptor-transfected Chinese hamster ovary cells following chronic opioid treatment.

Using CHO cells stably transfected with rat mu-opioid receptor cDNA, we show that the mu-agonists morphine and [D-Ala2,N-methyl-Phe4,Gly-ol5]enkephalin are negatively coupled to adenylylcyclase and inhibit forskolin-stimulated cAMP accumulation. Chronic exposure of cells to morphine leads to the rapid development of tolerance. Withdrawal of morphine or [D-Ala2,N-methyl-Phe4,Gly-ol5]enkephalin following chronic treatment (by wash or addition of the antagonist naloxone) leads to an immediate increase in cyclase activity (supersensitization or overshoot), which is gradually reversed upon further incubation with naloxone. Phosphodiesterase inhibitors do not affect the overshoot, indicating that it results from cyclase stimulation rather than phosphodiesterase regulation. Morphine's potency to inhibit cAMP accumulation is the same before and after chronic treatment, suggesting that the apparent tolerance results from cyclase activation, rather than from receptor desensitization. The similar kinetics of induction of tolerance and overshoot support this idea. Both the overshoot and acute opioid-induced cyclase inhibition are blocked by naloxone and are pertussis toxin-sensitive, indicating that both phenomena are mediated by the mu-receptor and Gi/G(o) proteins. The supersensitization is cycloheximide-insensitive, indicating that it does not require newly synthesized proteins. This is supported by the rapid development of supersensitization. Taken together, these results show that mu-transfected cells can serve as a model for investigating molecular and cellular mechanisms underlying opiate drug addiction.

Cannabinomimetic behavioral effects of and adenylate cyclase inhibition by two new endogenous anandamides.

We have previously shown that the endogenous putative cannabinoid ligand arachidonylethanolamide (anandamide, 20:4, n - 6) induces in vivo and in vitro effects typical of a cannabinoid agonist. We now report that two other endogenous anandamides, docosatetraenylethanolamide (anandamide, 22:4, n - 6) and homo-gamma-linolenylethanolamide (anandamide, 20:3, n - 6), have similar activities. The new anandamides bind to SV40-transformed African green monkey kidney cells transfected with the rat brain cannabinoid receptor cDNA and display K1 values of 253.4 +/- 41.1 and 244.8 +/- 38.7, respectively. The value found for arachidonylethanolamide was 155.1 +/- 13.8 nM. In addition, the new anandamides inhibit prostaglandin E1-stimulated adenylate cyclase activity in Chinese hamster ovary-K1 cells transfected with the cannabinoid receptor, as well as in N18TG2 mouse neuroblastoma cells that express the cannabinoid receptor naturally. The IC50 values for the inhibition of adenylate cyclase in transfected Chinese hamster ovary-K1 cells were 116.8 +/- 8.7 and 109.3 +/- 8.6 nM for docosatetraenylethanolamide and homo-gamma-linolenylethanolamide, respectively. These values were similar to that obtained with arachidonylethanolamide (100.5 +/- 7.7 nM), but were significantly higher than the IC50 value observed with the plant cannabinoid delta9-tetrahydrocannabinol (9.2 +/- 8.6 nM). The inhibitory effects of the anandamides on adenylate cyclase activity were blocked by pertussis toxin, indicating the involvement of pertussis toxin-sensitive GTP-binding protein(s). In a tetrad of behavioral assays for cannabinoid-like effects, the two new anandamides exerted similar behavioral effects to those observed with delta9-tetrahydrocannabinol and arachidonylethanolamide: inhibition of motor activity in an open field, hypothermia, catalepsy on a ring, and analgesia on a hot plate.

Increased expression of synapsin I mRNA in defined areas of the rat central nervous system following chronic morphine treatment.

Chronic opiate administration leads to a selective regulation of several cellular proteins and mRNAs. This phenomenon has been viewed as a compensatory mechanism to the opiate signaling leading to the development of opiate addiction. In this study, in situ hybridization histochemistry experiments were employed to investigate the effect of chronic morphine treatment on synapsin I gene expression. We show here for the first time that prolonged morphine exposure causes a selective increase in the mRNA levels of synapsin I in several brain regions which are considered to be important for opiate action. Quantitative analysis of the signals, obtained by hybridization of digoxigenin-labeled antisense RNA probe, revealed a 5.8- and 7-fold increase of synapsin I mRNA levels in the locus coeruleus and the amygdala of morphine-treated rats, respectively, as compared with control untreated rats. Increased expression of synapsin I mRNA was also observed in the spinal cord of morphine-treated rats (by 3.8-fold). Since opiates were shown to attenuate neurotransmitter release and reduce synapsin I phosphorylation, it is suggested that the increase in synapsin I levels would lead to the requirement of higher amounts of opiate agonists to obtain the opiate physiological effects. These results suggest that the increases in mRNA levels of synapsin I in these specific areas can be part of the molecular mechanism(s) underlying opiate tolerance and withdrawal.

kappa-Opioid receptor-transfected cell lines: modulation of adenylyl cyclase activity following acute and chronic opioid treatments.

The opioid receptors mu, delta and kappa have recently been cloned. Here we show that kappa-agonists inhibit adenylyl cyclase activity in Chinese hamster ovary cells stably transfected with rat kappa-opioid receptor cDNA. Chronic exposure of the cells to kappa-agonists did not lead to significant desensitization of the capacity of the agonists to inhibit adenylyl cyclase. On the other hand, withdrawal of the agonist following the chronic treatment led to the phenomenon of supersensitivity ('overshoot') of adenylyl cyclase activity. Both the inhibition of adenylyl cyclase activity by the acute opioid treatment and the chronic agonist-induced supersensitivity are pertussis toxin sensitive, demonstrating involvement of Gi/Go proteins in both processes.

Truncation of the thyrotropin-releasing hormone receptor carboxyl tail causes constitutive activity and leads to impaired responsiveness in Xenopus oocytes and AtT20 cells.

We studied the activity of a truncated thyrotropin-releasing hormone receptor (TRH-R), which lacks the last 59 amino acids of the carboxyl tail, where Cys-335 was mutated to a stop codon (C335Stop) (Nussenzveig, D. R., Heinflink, M., and Gershengorn, M. C. (1993) J. Biol. Chem. 268, 2389-2392). In Xenopus laevis oocytes expressing C335Stop TRH-Rs, TRH binding was higher, whereas chloride current, 45Ca2+ efflux, and [Ca2+]i responses evoked by TRH were 23, 39, and 21%, respectively, of those in oocytes expressing wild type mouse pituitary TRH-Rs (WT TRH-Rs). In oocytes expressing C335Stop TRH-Rs, basal 45Ca2+ efflux and [Ca2+]i were twice those in oocytes expressing WT TRH-Rs; chelation of Ca2+ caused a rapid increase in holding current, which is consistent with basal activation; and coexpression with other receptors caused inhibition of the responses to the other cognate agonists. In AtT20 pituitary cells stably expressing C335Stop TRH-Rs, thyrotropin-releasing hormone (TRH)-independent inositol phosphate formation was 1.32 +/- 0.11-fold higher, basal [Ca2+]i was 1.8 +/- 0.2-fold higher, and the [Ca2+]i response to TRH was much lower than in cells expressing WT TRH-Rs. We conclude that a TRH-R mutant truncated at Cys-335 exhibits constitutive activity that results in desensitization of the response to TRH.

The hemispheric functional expression of the thyrotropin-releasing-hormone receptor is not determined by the receptors' physical distribution.

The thyrotropin-releasing-hormone receptor (TRH-R) is a member of a family of the G-protein-coupled receptors that share structural similarities and exert their physiological action via the inositol lipid signal-transduction pathway. The TRH-R when expressed in Xenopus oocytes exhibits marked preference of the response (increased chloride conductance) for the animal hemisphere. Whereas the rat TRH-R functional distribution was strongly asymmetric (animal/vegetal ratio = 9.5), the mouse TRH-R exhibited a significantly lower ratio (3.9). Truncation of the last 59 amino acids of the C-terminal region of the mouse TRH-R did not lead to any changes in the functional hemispheric distribution. Despite the polarization of response, receptor number was similar on both hemispheres. Moreover, the apparent half-life of the functional expression of the TRH-R was approx. 4 h on both hemispheres when the expression was inhibited by a specific antisense oligonucleotide. Inhibition of total protein synthesis with cycloheximide affected hemispheric responses mediated by each of the three TRH-Rs tested in a qualitatively different way. These results suggest that an additional, rapidly degraded, protein modulates the functional hemispheric expression of the TRH-Rs.

Differential effects of cytoskeletal agents on hemispheric functional expression of cell membrane receptors in Xenopus oocytes.

1. We studied the effects of three cytoskeleton-disrupting agents, colchicine (COL), vinblastine (VIN), cytochalasins, on the functional hemispheric expression of native muscarinic and acquired thyrotropin-releasing hormone receptors TRH-Rs). Responses in oocytes of common donors, which express M3-like receptors (M3Rs), were not affected by either COL or VIN on the animal hemisphere. The functional expression of M3Rs on the vegetal hemisphere was inhibited by 50%. Cytochalasin B caused a uniform inhibition (by 31-33%) of receptor functional expression on either hemisphere. 2. Oocytes of variant donors express predominantly M1-like receptors (M1Rs) on the animal and M3Rs on the vegetal hemisphere. In these oocytes, both COL and VIN caused approximately 50% inhibition of functional expression on either hemisphere. Cytochalasin B caused more extensive, though variable inhibition on both hemispheres. Both antitubulin agents had no effect on the functional expression of the TRH-Rs on either hemisphere. Cytochalasin B, however, caused an extensive inhibition of the functional expression of this receptor (by 70-75%). 3. Induction of maturation of oocytes (7-hr incubation with progesterone) resulted in a 66% decrease in the response to TRH, reflecting mainly a decrease on the animal hemisphere. Maturation in the presence of colchicine had no further effect on the activity measured on the animal hemisphere but caused a major increase in the activity on the vegetal hemisphere. This resulted in a dramatic change in animal/vegetal activity ratio (4.8 +/- 1.5 to 0.8 +/- 0.2). 4. It appears that while antitubulin drugs affect the functional expression of the three receptors at the two hemispheres differently, disruption of the microfilaments interferes uniformly with receptor functional expression. We suggest that microfilaments may be involved in a common component of the signal transduction pathway in oocytes or in the anchoring of receptors coupled to the guaninine nucleotide-binding regulatory proteins. Moreover, progesterone-induced changes in the functional organization of the signal transduction pathway appear to be controlled to a large extent by the tubulin component of the cytoskeleton.

The hemispheric distribution of Torpedo nicotinic receptors expressed in Xenopus oocytes.

The physical and the functional distribution of Torpedo nicotinic-cholinergic receptors expressed in Xenopus oocytes was assayed. Physical hemispheric receptor distribution was tested by binding of 125[I]-bungarotoxin. The density of the expressed nicotinic receptors was equal on both hemispheres (ratio animal/vegetal = 1.1 +/- 0.2). Functional distribution was tested either by whole hemispheric response assay or by monitoring responses from small areas on the two hemispheres. While the first method yielded results that suggested uniform receptor density distribution, the second method indicated two-fold higher responsiveness on the animal hemisphere, when compared with the vegetal hemisphere. Direct comparison on oocytes of the same donors did not reveal significant differences between the two assays. We did see, however, a high variability among the different donors (animal/vegetal activity ratio range 0.5-4.7). Overall, in 35 experiments in 18 donors, the animal/vegetal ratio of hemispheric responsiveness was 1.4. The possible source of this high variability may have been the large excess of bungarotoxin-binding sites over the number of active channels. We have also tested hemispheric responsiveness ratio with different concentrations of acetylcholine. When acetylcholine concentration was below 10 microM, the animal/ vegetal ratio was significantly lower than 1.0. Similar results were obtained with nicotinic receptors expressed after injection of RNA transcribed in vitro from cloned mouse nicotinic receptor subunits. These results imply that hemispheric membrane heterogeneity may affect receptor and/or channel activities to yield polarized channel activity despite nearly homogeneous receptor distribution.

Kinetics of the functional loss of different muscarinic receptor isoforms in Xenopus oocytes.

Native Xenopus oocytes express two isoforms of muscarinic receptors that mediate qualitatively different physiological responses. Oocytes of the majority of donors (common) express M3-like receptors (M3Rs) at comparable densities at both the animal and vegetal hemispheres of the cell. Rare (variant) donors possess oocytes that express mainly M1-like receptors (M1Rs), localized predominantly at the animal hemisphere. We have investigated the apparent degradation of these two isoforms and its relationship to their hemispheric distribution. Cycloheximide (CHX) caused a time-dependent decrease in receptor-mediated responses and [3H]quinuclidinyl benzylate (QNB) binding in oocytes from both types of donors. The t1/2 values ranged between 3 and 7 h. Removal of CHX resulted in rapid recovery of the response. This implied rapid degradation and turnover of both types of receptors. The loss of M1Rs was more than that of M3Rs. Moreover, the decrease was more rapid and more extensive on the animal hemisphere in both types of donors. Injection of oocytes expressing either receptor isoform with specific antisense oligonucleotides complementary to either m1 or m3 muscarinic receptors (from mouse) showed receptor loss at approximately the same rate as that calculated from experiments with CHX. Furthermore, oocytes of variant donors express M1Rs exclusively on the animal hemisphere, while the residual activity found on the vegetal hemisphere of the cell was mediated by M3Rs. Inhibition of putative receptor glycosylation with tunicamycin caused a rapid decrease in receptor-mediated responses and radioligand binding on M1Rs, but had virtually no effect on M3Rs. The expression of cloned m1 muscarinic receptors, however, was not affected by tunicamycin, suggesting that glycosylation is not a general prerequisite for the functional expression of muscarinic receptors.

Native Xenopus oocytes express two types of muscarinic receptors.

We have recently described two types of muscarinic responses in native Xenopus oocytes of different donors (common and variant) that display qualitative and quantitative differences (Lupu-Meiri et al., 1990). Here we characterized the muscarinic receptors mediating these two types. The anti-muscarinic toxins from Dentroaspis significantly inhibited responses in oocytes of common donors, but had little effect on responses in oocytes of variant donors, possibly indicating expression of different receptor subtypes. Using specific muscarinic antagonists, we found that oocytes of common donors exhibit a pattern compatible with the M3 subtype of muscarinic receptors, while oocytes of variant donors appear to possess receptors of the M1 subtype. To more directly determine the subtypes of muscarinic receptors in oocytes of both populations of donors, we have microinjected antisense oligonucleotides into native oocytes. Antisense oligonucleotides to unique sequences in the N-terminal and the third cytoplasmic loop of M3 muscarinic receptors caused a significant inhibition of the response of common oocytes, but had virtually no effect on responses in oocytes of variant donors. Conversely, oligonucleotides complementary to the unique sequences of the m1 muscarinic receptors inhibited the response in variant oocytes, but not in oocytes of common donors. We conclude that native Xenopus oocytes of different donors phenotypically express either M3-like (majority) or M1-like (minority) muscarinic receptor subtypes. The differences in receptor subtype expression may explain the different characteristics of responses in the two populations.

Two types of intrinsic muscarinic responses in Xenopus oocytes. I. Differences in latencies and 45Ca efflux kinetics.

Oocytes of 40% of Xenopus laevis frogs respond to acetylcholine (ACh). Oocytes of the majority of responders exhibit the common two-component depolarizing muscarinic response (mean amplitude of the rapid component, 54 nA). Oocytes of approximately 10% of the responders ("variant" donors) exhibit a muscarinic response characterized by a very large transient, rapid current (mean amplitude 1242 nA, reversal potential -33 mV). Responses in oocytes of variant donors exhibit further qualitative differences: pronounced desensitization (absent in oocytes of common donors), characteristic prolonged latency (5.4 vs 0.9 s in oocytes of common donors) and marked inhibition of the response by activators of protein kinase C. Rapid responses in oocytes of variant donors are usually increased by treatment with collagenase, which, in common oocytes, often results in a complete loss of the response that correlates with the loss of muscarinic ligand binding. The number of muscarinic receptors was similar in oocytes of both types of donors (2.2 vs 3.0 fmol/oocyte). Also, the responses of oocytes of variant donors to microinjections of CaCl2 or inositol 1,4,5-trisphosphate were similar to those found in cells of common donors. These findings imply that altered receptor number, calcium stores and/or chloride channel density are not responsible for the variant responses. However, ACh caused an sixteen-fold greater efflux of 45Ca in oocytes of variant donors (35 vs 2.2% of total label in oocytes of common donors). Hence, the characteristics of the variant response may be related to a more efficient coupling between receptor stimulation and the mobilization of cellular calcium.

Two types of intrinsic muscarinic responses in Xenopus oocytes. II. Hemispheric asymmetry of responses and receptor distribution.

Fully grown Xenopus laevis oocytes display marked morphological asymmetry. The giant cell is divided into animal (pigmented) and vegetal hemispheres. We have developed methodology aimed at easy determination of hemispheric responses to the application of acetylcholine (ACh) and determination of the distribution of muscarinic receptors. Oocytes of common donors exhibit muscarinic responses that are similar when either the animal or the vegetal hemisphere of the cell is exposed to ACh. Oocytes of variant donors, however, exhibit markedly larger muscarinic responses when the animal hemisphere is exposed to ACh (ratio animal/vegetal, 5.8). The differences in hemispheric responsiveness correlate well with the hemispheric distribution of muscarinic receptors. While oocytes of common donors exhibit a modest excess of receptor number at the animal hemisphere (ratio animal/vegetal hemispheres, 1.4), oocytes of variant donors exhibit a large excess of receptors on the animal hemisphere (ratio animal/vegetal, 5.6). Upon further examination, we have found that the distribution of muscarinic receptors is non-homogeneous in either hemisphere in oocytes of both common and variant donors. The asymmetric distribution of receptors may be related to increased efficiency of signal transduction coupling in oocytes of variant donors.