1 Neuromuscular blocking drugs (NMBD's) are known to produce cardiovascular side e ects manifesting as brady/tachycardias. In this study we have examined the interaction of a range of steroidal NMBD's with recombinant human m1 ± m5 muscarinic receptors expressed in Chinese hamster ovary cells. Our main hypothesis is that NMBD's may interact with m2 (cardiac) muscarinic receptors. 2 All binding studies were performed with cell membranes prepared from CHO m1 ± m5 cells in 1 ml volumes of 20 mM HEPES, 1 mM MgCl 2 at pH 7.4 for 1 h. Muscarinic receptors were labelled with [ 3 H]-NMS and displacement studies were performed with pancuronium, vecuronium, pipecuronium, rocuronium and gallamine. In addition a range of muscarinic receptor subtype selective reference compounds were included. In order to determine the nature of any interaction the e ects of pancuronium, rocuronium and vecuronium on methacholine inhibition of forskolin stimulated cyclic AMP formation in CHO m2 cells was examined. Cyclic AMP formation was assessed in whole cells using a radioreceptor assay. All data are mean+s.e.mean (n55). 50 ) by pirenzepine in CHO m1 membranes (7.97+0.04), methoctramine in CHO m2 membranes (8.55+0.1), 4-diphenylacetoxy-Nmethyl piperidine methiodide (4-DAMP) in CHO m3 membranes (9.38+0.03), tropicamide in CHO m4 membranes (6.98+0.01). 4-DAMP, pirenzepine, tropicamide and methoctramine displaced [ 3 H]NMS in CHO m5 membranes with pK 50 values of 9.20+0.14, 6.59+0.04, 6.89+0.05 and 7.22+0.01 respectively. These data con®rm homogenous subtype expression in CHO m1 ± m5 cells. 5 [ 3 H]NMS binding was displaced dose-dependently (pK 50 ) by pancuronium (m1, 6.43+0.12; m2, 7.68+0.02; m3, 6.53+0.06; m4, 6.56+0.03; m5, 5.79+0.10), vecuronium (m1, 6.14+0.04; m2, 6.90+0.05; m3, 6.17+0.04; m4, 7.31+0.02; m5, 6.20+0.07), pipecuronium (m1, 6.34+0.11; m2, 6.58+0.03; m3, 5.94+0.01; m4, 6.60+0.06; m5, 4.80+0.03), rocuronium (m1, 5.42+0.01; m2, 5.40+0.02; m3, 4.34+0.02; m4, 5.02+0.04; m5, 5.10+0.03) and gallamine (m1, 6.83+0.05; m2, 7.67+0.04; m3, 6.06+0.06; m4, 6.20+0.03; m5, 5.34+0.03). 6 Cyclic AMP formation was inhibited dose dependently by methacholine in CHO m2 cells pEC 50 for control and pancuronium (300 nM) treated cells were 6.18+0.34 and 3.57+0.36 respectively. Methacholine dose-response curves in the absence and presence of rocuronium (1 mM) and vecuronium (1 mM) did not di er signi®cantly. Pancuronium, vecuronium and rocuronium did not inhibit cyclic AMP formation alone indicating no agonist activity. 7 With the exception of rocuronium there was a signi®cant interaction with m2 muscarinic receptors with all NMBD's at clinically achievable concentrations suggesting that the brady/tachycardias associated with these agents may result from an interaction with cardiac muscarinic receptors. Furthermore pancuronium at clinically achievable concentrations antagonised methacholine inhibition of cyclic AMP formation in CHO m2 cells further suggesting that the tachycardia produced by this agent results from muscarinic antagonism. The mech...