The aim of the present study was to validate the Cytosensor microphysiometer, a novel system that measures the extracellular acidification rate as a reliable index of the integrated functional response to receptor activation, as a method for studying NK3 receptor pharmacology, and then to use this system to assess the functional activity of novel compounds at this receptor.The selective NK3 agonist senktide caused reproducible, concentration‐related increases in acidification ratein CHO‐NK3 cells, with a pEC50 value of 8.72±0.11 (n=15). [β‐Ala8]NKA(4–10), the selective NK2 agonist, elicited a much weaker response (pEC50=6.68±0.08, n=4), while the NK1‐selective agonist substance P methylester only caused a very weak response at concentrations 3 μM (n=2). The rank order of potency for the endogenous tachykinins NKB>NKA>substance P (n=3) confirmed the response was mediated by the NK3 receptor. Moreover, the actual potencies obtained were consistent with affinities measured in radioligand binding studies.The novel compounds PD156319‐121 (0.3–1 μM), PD161182 (10–300 nM), PD168001 (10–100 nM) and PD168073 (10–100 nM) all acted as surmountable antagonists of the senktide‐induced acidification response, with pA2 values of 7.49, 8.67, 9.17 and 9.25 respectively (n=3–5). In comparison the known NK3 antagonist SR142801 (10–100 nM) had a pA2 value of 8.83 (n=8) for the interaction with senktide. Again, these values are consistent with the radioligand binding data.Amiloride (1 mM) inhibited the senktide‐induced acidification response by 68.3±3.3 (n=4), indicating that the Na+/H+ antiporter plays an important role in this response, and this is consistent with the importance of this antiporter in other acidification responses.Inhibition of protein kinase C with staurosporine (0.1 μM), or depletion of the intracellular Ca2+ stores with thapsigargin (1 μM), both resulted in a reduction in the maximum response to senktide (63.3±1.7 and 68.9±3.2% respectively, n=3–5), and co‐application of these inhibitors abolished the response (n=3). This strongly suggested that the NK3 receptor was coupling via phospholipase C (PLC), as would be expected, although this could not be confirmed by the use of the putative PLC/PLA2 inhibitor U73122.In conclusion, we have demonstrated the utility of the Cytosensor in the characterization of functional responses to agonists, and assessment of the affinities of antagonists in CHO cells expressing the human NK3, and have shown that our series of novel compounds are non‐peptide NK3 antagonists of high affinity, as exemplified by PD168073.British Journal of Pharmacology (1998) 125, 761–766; doi:10.1038/sj.bjp.0702156