The mu opioid receptor, the opioid receptor most closely associated with both analgesia and behavioral reinforcement, is a member of the extensive family of G protein-coupled receptors. To characterize the receptors, most binding studies from the earliest to the present, use cell homogenates, which disperse receptor-membrane fragments as a fine suspension. An advantage of this method is that the receptors behave more like a homogeneous solution and obey the corresponding laws pertaining to the kinetics and equilibrium of a reversible binary association. However, the cellular function of the receptors is to bind opioids, both the endogenous ones as well as opioid drugs, and to generate cellular responses by means of a cascade of reactions that occur within the cell. As these reactions themselves can alter the affinity of the receptor, it is more relevant to study drug binding to whole, living cells in which all of the coupling machinery remains dynamically intact. The simplest such system consists of a monolayer of cells grown in culture, but the question remains as to whether such a system, as simple as it is, still obeys the simple chemical models for equilibrium and kinetics, or whether the attachment of receptors to whole cells confined to a thin layer in space presents new constraints. Testing this hypothesis has occupied much of this year using Chinese hamster ovary (CHO) cells permanently transfected with the mu opioid receptor as well as non-transfected cells for controls. We employed mostly radioligand binding methods, but we have also brought quantitative confocal microscopy to bear on the subject. We found that (1) the presence of receptors slows the dissociation of [3H]-naloxone from the cells by about 4-fold in accordance to the """"""""retention effect"""""""", (2) that even nonspecific binding of [3H]-naloxone is retarded in its diffusion into the cell monolayer, (3) that this nonspecific binding is not saturable but is blocked completely by azure a, a dye believed to have specificity for cerebroside sulfate, (4) that confocal microscopy, by means of its exquisite optical sectioning capability, can detect differences in binding times at the upper versus the lower surfaces of the monolayer of cells, even though the spacial difference chosen was only 4 micrometers, (5) that drug hydrophobicity and stirring rate have large influences over deviations from ideality of the cell monolayer. The second facet of our work is a pursuit of the puzzling observation seen by us and two other labs, namely that the mu opioid receptor population estimated by [3H]- naloxone under some conditions is greater than that estimated by the antagonist [3H]- beta-FNA. We find the ratio is 2:1 in our CHO cells, that it occurs only when the cells are first osmotically stressed by hypotonic medium, and that the ratio returns to 1:1 by 30 minutes of recovery time after the stress. We showed, by confocal microscopy, that the stress causes the release of calcium from intracellular stores, though we have not proven yet that this release is causative. We hypothesized that the stress shifts the receptor monomer-dimer ratio. To test this hypothesis, we have run western blots of cell lysates under both unstressed and stressed conditions using crosslinkers to preserve dimers. While the monomer clearly appears as a diffuse band at about 65 kDa, as expected, we have not yet found any trace of the dimer. For comparison, we have tested SH-SY5Y cells under various conditions and have seen mu and delta receptor monomers but never dimers. As a tangential investigation, the CHO cells were checked for possible use in electrophysiological experiments under voltage clamp conditions. However, no sign of reported ion channels or of any reactivity to the agonist normorphine was found.
