Chimeric G proteins allow a high-throughput signaling assay of Gi-coupled receptors.

G-protein-coupled receptors are a major target for potential therapeutics; yet, a large number of these receptors couple to the Gi pathway, generating signals that are difficult to detect. We have combined chimeric G proteins, automated sample handling, and simultaneous 96-well fluorometric imaging to develop a high-throughput assay system for Gi signaling. The chimeric G proteins alter receptor coupling so that signaling can occur through Gq and result in mobilization of intracellular calcium stores. An automated signaling assay device, the fluorometric imaging plate reader (FLIPR), can simultaneously measure this response in real time in 96-well microplates, allowing two people to process more than 10,000 points per day. We used the chimeric G protein/FLIPR system to characterize signaling by the Gi-coupled human opioid receptors. We show that the mu, delta, and kappa opioid receptors and the related nociceptin receptor, ORL1, each couple to Galphaqi5, Galphaqo5, and Galpha16 (Galphaqi5 and Galphaqo5 refer to Galphaq proteins containing the five carboxyl-terminal amino acids from Galphai and Galphao, respectively) and that different receptor/G protein combinations show different levels of maximal activation. We tested 31 opioid ligands for agonist activity at the opioid receptors (124 ligand-receptor combinations); all 31 activated at least one receptor type, and several activated multiple receptors with differing potencies. This high-throughput assay could be useful for dissecting the complex ligand-receptor relationships that are common in nature.

Use of N-sigma-dansyl-L-lysine and flow cytometry to identify heat-killed mammalian cells.

We have employed the nontoxic fluorescent membrane probe, N-sigma-dansyl-L-lysine (DL) to study the effect of mild (45.5 degrees C) heat shock on a variety of mammalian cell lines. It has been previously proposed by Humphries and Lovejoy (1983) that DL selectively partitions into (and diffuses through) membranes whose component molecules have undergone lateral phase separation resulting in the formation of phospholipid domains. Excellent flow cytometric resolution of the DL staining cells from several cell lines was obtained by using bivariate (forward angle light scatter versus DL-fluorescence) analysis. Dye uptake and release data as well as measurement of the octanol: water partition coefficient (7.2) all indicated that the stain was likely associated with the plasma membrane. After heating, all cell lines exhibited a time-dependent increase in the fraction of cells stained by DL. Nearly all of the DL-staining cells were propidium iodide and trypan blue excluding. Exclusion of erythrosin B or inclusion of fluorescein showed a better correlation with colony formation, although neither was found to be as effective as DL in estimating cell killing. A comparison of cell survival curves as measured either by colony formation or by the fraction of cells not stained by DL 24 h after heating indicated a good, though not absolute correlation. These results indicate first that DL may have general usefulness as a stain indicating cell death following heat shock, and second, that DL may have utility as a probe of specific membrane damage induced by heat. Our results are consistent with the hypothesis that membrane lateral phospholipid domain partitioning is associated with hyperthermia-induced cell death in mammalian cells.

Membrane structure and the tenuously maintained resistance to staining with N epsilon-dansyl-L-lysine shown by many cells.

The ability to resist staining by N epsilon-dansyl-L-lysine is tenuously maintained in the majority of live nucleated cells taken from tissues concerned with immune function. Resistance is lost under a variety of nonphysiological conditions known to, or likely to, cause protein denaturation or aggregation. In contrast to that of dansyl-gamma-aminobutyrate, the fluorescence intensity of N epsilon-dansyl-L-lysine is only weakly enhanced by native proteins. This is further reduced on denaturation or aggregation of the proteins. It is unlikely, therefore, that cellular uptake of, and staining by, N epsilon-dansyl-L-lysine is a direct consequence of membrane protein denaturation/aggregation but may result from a decrease in protein-phospholipid interactions leading to formation of phospholipid domains. Previous work has indicated that such features are stained by N epsilon-dansyl-L-lysine (Humphries, G.M.K., Lovejoy, J.P., 1983, Biophys. J. 42:307-310; Humphries, G.M.K., Lovejoy, J.R., 1983, Biochem. Biophys. Res. Commun. 111:768-774). Although it appears likely that passage through a dansyl-lysine-staining state is a common, if not universal, prelude to cell death (as monitored by uptake of trypan blue), not all cells that lose resistance to dansyl-lysine staining are moribund. Resistance to staining is also lost by macrophages on binding to solid substrates and multivalent ligands. The possible physiological significance of this is discussed.

Lateral phase separation of phospholipids as a basis for increased permeability of membranes towards fluorescein and other chemical species.

Using mouse spleen cells, before and after treatment with glutaraldehyde or mild hyperthermia, we observe a strong correlation between permeability to fluorescein and susceptibility to staining with N epsilon-dansyl-L-lysine (irrespective of the cells' ability to exclude trypan blue). We observe the same correlation using liposomes prepared from phosphatidylcholine and varying amounts of cholesterol. We have recently introduced N epsilon-dansyl-L-lysine as a fluorescent membrane stain, or "probe," whose uptake, we propose, is selective for phospholipid domains in membranes (G.M.K. Humphries & J.P. Lovejoy Biophys. J. 42:307-310, 1983; G.M.K. Humphries & J.P. Lovejoy J. Membrane Biol. 77:115-122, 1984). The results presented here are consistent with the hypothesis that the presence or absence of phospholipid domains in membranes also modifies their permeability toward fluorescein, and suggests that permeability towards other chemical species may be similarly affected. On the basis of work using liposomes, we believe this to be the case for carboxyfluorescein and for glucose.

Dansyl lysine: a structure-selective fluorescent membrane stain?

Dansyl lysine (DL) is a fluorescent compound that has significantly higher solubility in synthetic phosphatidylcholine (PC) membranes with a low cholesterol content than it does in water or in membranes having a high cholesterol content. Its fluorescence intensity is enhanced at least 50-fold when dissolved in PC membranes. Therefore, membranes with mole fractions of cholesterol (Xch) less than or equal to 0.5-0.3 are stained by aqueous solutions of DL: those with a higher cholesterol content, 0.3-0.4 less than or equal to Xch less than or equal to 0.5, are not. It is proposed that DL selects for a structural feature of membranes: cholesterol-free domains. The phenomenon has provided evidence for long-lived compositional heterogeneity in large multilamellar PC-cholesterol liposomes having Xch less than or equal to 0.2. This is not consistent with a model in which the homogeneous state is thermodynamically favored and both intermembrane transfer and transmembrane transfer (flip-flop) of cholesterol are fast. These studies are of potential importance for understanding cell membrane structure, in particular lipid-phase equilibria and the maintenance of compositional heterogeneity between the different membranes of cells.

Cholesterol-free phospholipid domains may be the membrane feature selected by N epsilon-dansyl-L-lysine and merocyanine 540.

We have used N epsilon-dansyl-L-lysine as a fluorescent membrane probe, to study cells taken from tissues concerned with immune function. There is a striking similarity between the staining selectivity of this compound and that reported by others for merocyanine 540. Both compounds stain leukemic, human, peripheral leukocytes, an erythroleukemia line, and some mouse bone marrow cells, suggesting common selectivity for a membrane feature of hemopoietic cells. Both compounds fail to stain red blood cells, normal human leukocytes, mouse spleen and thymus cells. We have recently reported that dansyl-lysine apparently selects for cholesterol-free phospholipid domains in liposomes and now report similar selectivity for merocyanine 540 staining of liposomes.

Difference in the ability of compactin and oxidized cholesterol, both known inhibitors of cholesterol biosynthesis, to suppress in vitro immune responses.

Compactin, a fungal metabolite, is a well-characterized competitive inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase, the enzyme which generally controls the rate of cholesterol biosynthesis in mammalian cells. Various products of cholesterol oxidation, e.g., 25-hydroperoxycholesterol and 25-hydroxycholesterol, also inhibit HMG CoA reductase but by an unknown mechanism which operates only in intact cells. In addition, oxidized cholesterol increases cholesterol esterification and suppresses the increase in low-density lipoprotein receptors which is the normal consequence of decreasing low-density lipoprotein supply. There are therefore, three ways in which the availability of nonesterified cholesterol to mammalian cells is decreased by oxidized cholesterol. It has been suggested that potent in vitro immunosuppression by compounds such as 25-hydroxycholesterol in the presence of cholesterol-containing fetal calf serum is a consequence of their ability to suppress cholesterol biosynthesis. This has been disputed in a previous paper (Humphries, G. M. K., and McConnell, H. M., J. Immunol., 122: 121-126, 1979) because mevalonate, the product of HMG CoA reductase activity, fails to abrogate the suppression. The present paper reports that compactin fails to suppress in vitro immune responses at concentrations known to inhibit HMG CoA reductase (either in the isolated or cellular form). This finding supports the previous conclusion that suppression of HMG CoA reductase activity is not sufficient to inhibit in vitro immune responses in the presence of an exogenous source of cholesterol.

Evidence for direct control of an in vitro plaque-forming cell response by quantitative properties of intact, fluid, haptenated liposomes: a potential model system for antigen presentation by macrophages.

Stimulation of the primary in vitro plaque-forming cell (PFC)2 response to fluid, haptenated liposomes by spleen cells depleted of a Sephadex-adherent population has been made possible by the addition of an uncharacterized cell-derived soluble factor(s). In its absence, PFC responses by such cells are greatly reduced or absent. The factor(s) is present in the supernatant from Concanavalin A-stimulated spleen cells. This phenomenon has permitted a comparison of the behavior of cultures with and without adherent cells, with a view to determining the relative likelihood of liposomes or adherent cells functioning as antigen presenters in this system. Two modes of control by quantitative properties of liposomes have been studied. PFC stimulation is controlled, in a biphasic manner, by varying either liposomal epitope density at constant liposome concentration or liposome concentration at fixed epitope density. Overall hapten concentration in the cultures is only of significance as a consequence of epitope density and liposome concentration; it does not, itself, control the response. Whole spleen cell cultures and cultures depleted of Sephadex-adherent cells but supplemented with soluble factor(s) exhibit the same biphasic response profiles when these quantitative properties of liposomes are varied. The results argue against the role of Sephadex-adherent cells as antigen presenters in whole spleen cell cultures. A model in which antigen presentation is accomplished by intact liposomes, in both whole spleen and adherent cell-depleted cultures, is consistent with the data.

Specific stimulation and suppression of a primary in vitro plaque-forming cell response by monovalent lipid haptens in fluid liposomal membranes.

Multilamellar, "fluid" liposomes containing an appropriate concentration of a synthetic dinitrophenylated lipid hapten behave as T-independent (TI), adherent cell-dependent antigens for primary, in vitro, stimulation of hapten-specific IgM plaque-forming cells (PFC). The appropriate concentration of hapten is approximately 1 mol% with respect to the other liposomal lipids, cholesterol, and dimyristoylphosphatidylcholine (DMPC). At a higher hapten concentration, 4 mol%, liposomes will suppress the anti-dinitrophenyl PFC response to co-cultured immunogenic liposomes, but not the anti-sheep cell response to co-cultured sheep cells. This suppression is also TI. The ability of "fluid," haptenated liposomes specifically to stimulate or to suppress the TI PFC response is determined by the ratio of hapten to other lipids in a manner that is independent of the total hapten concentration in culture.

Membrane-controlled depletion of complement activity by spin-label-specific IgM.

Complement depletion mediated by high molecular weight (IgM) rabbit antibodies specifically bound to spin-label lipid haptens dispersed in model membranes is controlled by various physical attributes of those membranes other than the total number of exposed determinants that they provide. Carrier lipids used at 32 degrees were (i) a "fluid" phosphatidylcholine (PC), (ii) a "solid" PC, and (iii) a cholesterol/PC mixture. The concentration of hapten in the plane of the membranes (two-dimensional concentration) was varied while the overall hapten molarity (three-dimensional concentration) was kept constant. Both specific binding and the efficiency of depletion by IgM are markedly enhanced by systematically decreasing the average distance between haptens (infinity --> 26 A). Heterogeneous distribution was found to be more favorable than a random homogeneous distribution of the same number of haptens in the same total quantity of lipids. IgM efficiency is also markedly increased by the inclusion of cholesterol in PC membranes, an effect thought to result from enhanced projection of the determinant from the surface of the membrane and hence increased accessibility to the antibody-binding site. Furthermore, the efficiency of IgM was increased by using haptens dispersed in fluid rather than in solid PC membranes. The results are consistent with the hypothesis that IgM molecules must be bound to a critical multiple of antigenic determinants at a membrane surface in order to induce complement-mediated attack and that subtle variation of the physical state of membrane antigens can be the crucial factor in determining the outcome of this type of efferent immune response.