Two populations of Chinese hamster ovary (CHO) cells expressing similar numbers of recombinant human alpha2A-adrenergic receptors (alpha(2A)-AR) showed different capacity to inhibit adenylyl cyclase (AC) activity. Cells transfected with an integrating vector exhibited agonist-dependent inhibition of forskolin-stirmilated AC, whereas cells transfected with a non-integrating episomal vector showed no inhibition. Fluorescent microscopy and flow cytometry revealed a very uneven receptor distribution in the episomally transfected cell population. Monoclonal cell populations were expanded from this parent population. Most clones lacked significant amounts of receptors, while a few expressed receptors at high density; these exhibited efficient agonist-dependent inhibition of forskolin-stimulated AC activity. Thus, dense receptor expression in only a few cells is not sufficient to evoke a significant inhibitory response in a functional assay where AC is stimulated in all cells. Consequently, a false negative result was produced. Furthermore, the cell population transfected with an integrating vector showed loss of homogeneity with increasing passage number. (c) 2004 Elsevier B.V. All rights reserved.