A family of 4 rat muscarinic receptors (m1, m2, m3, and m4) have recently been cloned and sequenced (Bonner et al., 1987). Since pharmacological probes that are presently available do not appear to distinguish among 3 of these muscarinic receptors, we constructed oligonucleotide probes corresponding to the N-terminal sequences of the muscarinic receptors and used them to specifically localize m1, m2, m3, and m4 mRNA in sections of rat brain using in situ hybridization histochemistry. Northern analysis demonstrated a 3.1 kilobase (kb) m 1 mRNA, a 4.5 kb m3 mRNA, and a 3.3 kb m4 mRNA in cerebral cortex, striatum, hippocampus, and cerebellum. In situ hybridization histochemistry indicated a prevalence of m1 mRNA in the pyramidal cell layer of the hippocampus, the granule cell layer of the dentate gyrus, the olfactory bulb, amygdala, olfactory tubercule, and piriform cortex. Caudate putamen and cerebral cortex showed moderate levels of labeling. m2 mRNA was detectable in the medial septum, diagonal band, olfactory bulb, and pontine nuclei. m3 and m4 mRNA were also prevalent in the olfactory bulb and pyramidal cell layer of the hippocampus but were present only in low levels in the dentate gyrus. m3 mRNA was present in superficial and deep layers of the cerebral cortex, whereas m4 mRNA was more evenly distributed with a slightly more intense labeling evident in the midcortical layer. In addition, m3 mRNA was present in a number of thalamic nuclei and brain-stem nuclei, while m4 mRNA predominated in the caudate putamen. These data offer a new basis on which to interpret the heterogeneity of muscarinic actions in the CNS.

Treatment of CHO cells stably expressing the human M1 muscarinic acetylcholine (HM1) receptor with the cholinergic agonist carbachol results in a reduction in cellular levels of Gq alpha/G11 alpha. Half-maximal effects are produced by 3 h, and a new steady state of some 50% of the resting levels of Gq alpha/G11 alpha is subsequently established [Mullaney, Dodd, Buckley and Milligan, (1993) Biochem. J. 289, 125-131]. To analyse the mechanism of this effect, we examined the rate of turnover of Gq alpha/G11 alpha in these HM1-expressing cells in the presence and absence of carbachol (1 mM). In untreated cells the measured removal of 35S-labelled Gq alpha/G11 alpha was adequately described by a monoexponential curve with a half-time (t0.5) of 18.0 +/- 2.2 h. When the cells were treated with carbachol a more complex pattern of Gq alpha/G11 alpha degradation was observed. Upon addition of the agonist, the rate of degradation initially increased markedly (t0.5 = 2.9 +/- 0.2 h). The maintained presence of the agonist was unable, however, to sustain the enhanced rate of degradation. Beyond 8 h of treatment with carbachol, degradation of Gq alpha/G11 alpha returned to a rate close to that observed in untreated cells (t0.5 = 18.5 +/- 1.3 h). Parallel experiments indicated that the effect of carbachol was specific for Gq alpha/G11 alpha, as the t0.5 of Gi2 alpha (approx. 30 h) was not affected by the agonist. Analysis of Gq alpha/G11 alpha mRNA levels by reverse transcriptase/PCR indicated that there was no difference in cells maintained in the absence and presence of carbachol. Such data demonstrate that agonist-induced establishment of a new steady-state level of Gq alpha/G11 alpha results from an initial receptor-mediated enhancement of protein turnover followed by a desensitization of the receptor response.

CHO cells stably transfected with cDNA encoding the human M1 muscarinic acetylcholine (HM1) receptor were treated with the cholinergic agonist carbachol at various concentrations for differing times. Levels of the HM1 receptor and of a range of G-proteins were subsequently measured. Carbachol treatment of the transfected cells caused a substantial down-regulation of cellular levels of the alpha subunit of Gq (Gq alpha), but did not significantly alter cellular levels of the alpha subunits of Gs or Gi2. A small decrease in levels of G-protein beta-subunit was also produced. Parallel assessment of agonist-induced down-regulation of the HM1 receptor demonstrated that it was lost in concert with the G-protein. Similar concentrations of carbachol (5 microM) were required to produce half-maximal stimulation of inositol phosphate generation and loss of each of the HM1 receptor and Gq alpha, and half-maximal losses of both receptor and Gq alpha were produced by 3 h of treatment with 1 mM-carbachol. By contrast, treatment of the non-transfected parental CHO cells, which do not express detectable levels of the receptor, with carbachol had no effect on cellular Gq alpha levels. Concurrent treatment of the HM1-expressing CHO cells with carbachol and cycloheximide indicated that suppression of protein synthesis de novo did not mimic the effect of carbachol, and hence even complete inhibition of transcription of the Gq alpha gene and/or translation of pre-existing Gq alpha mRNA could not account for the agonist-induced effect. We have previously noted that cellular levels of both Gs alpha [McKenzie and Milligan (1990) J. Biol. Chem. 265, 17084-17093] and the alpha subunits of the pertussis-toxin-sensitive G-proteins Gi1, Gi2 and Gi3 [Green, Johnson and Milligan (1990) J. Biol. Chem. 265, 5206-5210] can be regulated in certain cell systems by agonist activation of receptors expected to interact with these G-proteins. These results demonstrate that the same is true of Gq alpha and suggest that agonist-induced co-ordinate loss of receptors and associated G-proteins may be a more common feature than has been appreciated to date.

Rat cultured ventromedial hypothalamic (VMH) neurones obtained from embryonichypothalamus were used to study the muscarinic (carbachol) modulation of voltage-gated K+ currents with the whole-cell patch-clamp technique. 2. Carbachol produced a potent and concentration-dependent (100 fM to 100 /M) decrease of the outward delayed rectifier K+ current (IK) with an IC50 of 44 pM and a Hill coefficient of 0 4. The carbachol-induced depression of IK was reduced by pirenzepine (1-10 /UM) and atropine (1 ,uM). Carbachol had no effect on the transient outward K+ current (IA).3. Intracellular dialysis with guanosine 5'-O-(2-thiodiophosphate) (GDP-,-S, 500 ,UM) significantly diminished the carbachol-induced depression of IK, suggesting GTP-binding protein (G-protein) involvement. Pre-incubation of VMH neurones with pertussis toxin (200-400 ng ml') or cholera toxin (1 jug ml') for 24-48 h had no effect on the carbacholinduced depression of IK. This suggested that the Ga., Gai, and Ga, G-protein a-subunits were not involved in mediating the carbachol-induced depression of IK in VMH neurones. 4. Treatment (24-48 h) of VMH neurones with antisense phosphothio-oligodeoxynucleotides to the Gall G-protein subunit (10 ,UM) significantly diminished the carbachol-induced depression of IK. Treatment with 10 /UM of either Ga,, sense or antisense to Gaq had no effect.5. These results demonstrate a novel and potent muscarinic depression of IK in VMN neurones, and that this depression is specifically mediated by the Ga,, G-protein subunit. This action of the Ga1, subunit also provides the first demonstration of an association between the Ga,, G-protein subclass (a member of the Gq class) and a neuronal membrane current.

No abstract