Ligands, their receptors and ... plasma membranes

Abstract

Ligand-receptor interactions are customarily described by equations that apply to solutes. Yet, most receptors are present in cell membranes so that sufficiently lipophilic ligands could reach the receptor by a two-dimensional approach within the membrane. As summarized in this review, this may affect the ligand-receptor interaction in many ways. Biophysicians calculated that, compared to a three-dimensional approach from the liquid phase, such approach could alter the time the ligands need to find a receptor. Biochemists found that ligand incorporation in lipid bilayers modifies their conformation. This, along with the depth at which the ligands reside in the bilayer, will affect the probability of successful receptor interaction. Novel mechanisms were also introduced, including "exosite" binding and ligand translocation between the receptor's alpha-helical transmembrane domains. Pharmacologists focused attention at ligand concentrations in membrane, their adsorption and release rates and the effects thereof on ligand potency and residence time at the receptor.

Fig. 1

Representative example of the path taken by a ligand to reach its receptor target at the surface of a membrane (black dot) via an exclusive 3D approach (left panel) or via a mixed 3D–2D approach (right panel) (Wang et al., 1992).

Fig. 2

Schematic representation of drug (L)–receptor (R) binding with the formation of an ephemeric encounter complex from where the drug can either bind with the reaction forward rate constant k₁ or diffuse away from the receptor.

Fig. 3

Dihydropyridine partition coefficients into biological (sarcoplasmic reticulum) membranes and octanol: adapted form Mason et al. (1991).

Fig. 4

Locations of a membrane-spanning protein (hydrophobic TM domains are shaded) and the ligands amiodarone, nimodipine and propranolol at different depths within the lipid bilayer (from Herbette et al., 1988).

Fig. 5

Panel A, proposed model in where hydrophilic ligands approach a GPCR from the aqueous phase while amphiphilic/hydrophobic ligands approach the receptor by lateral diffusion within the membrane and then translocate via the receptor's TM domains to their central binding pocket (TM domain in front of translocation is semi-transparent). Panel B, structure of amphiphilic ligands that allegedly reach their receptors via a combined 3D–2D approach. Panel C, binding of salmeterol via its hydrophobic phenylalkoxyalkyl tail to an alleged exosite in the vicinity of the β₂-adrenergic receptor with occupancy (right side) or not (left side) of the receptor's central binding pocket by the saligenin head (TM domains in front of the saligenin head are semi-transparent). Panel D, binding of a lipid-conjugated gastrin molecule to its receptor according to a molecular modeling study by Lutz et al. (1997) (TM domains in front of the peptide section are semi-transparent).

Fig. 6

Panel A, insertion of the hydrophobic Aβ1–40 peptide segment into membranes (Bokvist et al., 2004). Panel B, simplified scheme for the “membrane catalysis” model of Sargent and Schwyzer (1986) in where partitioning and conformational change of a peptide ligand is described as a single equilibrium step defined by K_p. ${K^{\prime }}_{\text{D}}$ is the local equilibrium dissociation constant and refers to the ligand concentration in the membrane at which half of the receptors are occupied. The overall equilibrium dissociation constant for the ligand–receptor interaction (K_D, expressed in moles l⁻¹ of aqueous solution) corresponds to $K_{\text{D}}={K^{\prime }}_{\text{D}}\cdot K_{\text{p}}^{-1}$.

Fig. 7

Panel A, influence of the initial receptor occupancy (abscissa, in percent of total receptor concentration) on the residence half-life (ordinate, the free ligand half-life being taken as unit) when the free ligand concentration decreases exponentially with time. Panel B, a high K_p and a slow ligand desorption rate act in concert to prolong the concentration of a membrane-associated ligand within the therapeutic window (adapted from Herbette, 1994).

Fig. 8

Panel A, translocation of [³H]-spiperone from “non-specific” sites in the plasma membrane from recombinant intact Chinese hamster ovary cells (CHO cells) to the therein-expressed D_2L-receptors. Experiments were done as in Packeu et al. (2008). In short, intact cells were pre-incubated for 30 min at 37 °C with 1 nM [³H]-spiperone after which its specific (i.e. 1 μM (+)-butaclamol-displaceable) binding is measured (here denoted as control binding). Co-addition of 10⁻⁵ M raclopride reduces the [³H]-spiperone binding to 15% of the control (Lane 1). When [³H]-spiperone + raclopride-pretreated cells are washed twice (to remove the free ligands present in the aqueous solution) and finally incubated for 10 min at 37 °C with buffer alone, specific binding almost doubles (Lane 2). This increase reflects genuine receptor occupancy as it can be blocked by raclopride in a concentration-dependent fashion (10⁻⁷ M in Lane 3 and 10⁻⁵ M in Lane 4) and with the same potency as in competition binding experiments. N = 3–5. Panel B, different rates of [³H]-spiperone dissociation from D_2L-receptors in recombinant CHO cells and membrane preparations thereof. Intact cells (♦) and membrane preparations thereof (●) were incubated for 30 min at 37 °C with 1 nM [³H]-spiperone. [³H]-Spiperone dissociation at 37 °C was initiated with 1 μM (+)-butaclamol and, at the times indicated, its remaining specific binding was measured. The same dissociation experiments were also done with leaky cells (■), i.e. in the throughout presence of 0.01 mg/ml of the pore-forming agent filipin. N = 3.