“Micro-Pharmacokinetics”: The Quantification Of Ligand Concentration At The Membrane Of Single Living Cells Using Fluorescence Correlation Spectroscopy

The vast majority of methods for determining pharmacological parameters such as binding affinity of a ligand for its target receptor assume a homogeneous distribution of ligand, with concentrations in the immediate vicinity of the receptor being the same as that in the bulk aqueous phase. It is, however, known that drugs are able to interact directly with the plasma membrane, potentially increasing the local concentration of ligand around the receptor. We have recently reported an influence of ligand-phospholipid interactions on ligand binding kinetics at the β2-adrenoceptor, proposing it is critical to consider the “micro-pharmacokinetic” properties of ligands when assessing their receptor pharmacology(1). Here, we assessed the “micro-pharmacokinetic” profile of BODIPY630/650-S-PEG8-propranolol (BY630-propranolol), afluorescent derivative of the classical β-blocker propranolol (2),by directly quantifyingits local concentration at various distances above single cell membranes using fluorescence correlation spectroscopy (FCS).

In this study, amixed population of CHO cells expressing C-terminally GFP-tagged β2-adrenoceptors(CHO-β2GFP) was used to select receptor-expressing and non-receptor expressing cells using GFP fluorescence. FCS experiments were carried out as previously described (3),and CHO-β2GFP cells were seeded two days prior to experimentation. On the day of the experiment, cells were exposed to 1.8 nMBY630-propranolol (2 h, 22°C). In antagonist experiments, cells were pre-incubated with 550 nMβ2-adrenoceptor antagonist ICI 118,551 (10 min, 22°C) prior to BY630-propranololaddition. FCS measurements (1x30s, 633nm excitation, LP650 emission) were then taken 2-200 µm above the membrane of an individual cell, selected based on its receptor expression level (GFP-fluorescence). Data were analysed using GraphPad Prism 6 and data are mean ± s.e.m. of n single cell experiments. Statistical analysis was performed using one-way ANOVA, followed by Dunnett’s post hoc test.

The concentration of BY630-propranololdetermined 2 µm above membranes ofβ2GFP expressing CHO cells was 37.9±11.5 nM (n=8), which was circa 20-fold higher than the nominal concentration (1.8 nM), and was significantly higher than the concentration measured at 200 µm (1.0±0.1 nM, n=8; P<0.05). In the same cells, the BY630-propranolol concentrationmeasured2 µm above the membrane in the presence of ICI 118,551 was 23.1±5.6 nM(n=6). In contrast, the BY630-propranolol concentrations measured 2 µm above the membrane of non-receptor expressing CHO cells in the absence and presence of ICI 188,551were 19.0±1.5 nM (n=8) and 25.3±7.5nM (n=7), respectively.

Here, we have quantified ligand concentrations immediately adjacent to the cell membrane for the first time, showing a significantly increased concentration compared to the bulk aqueous phase. This suggests a clear role of the cell membrane in determining local ligand concentrations. More importantly, the presence of the target receptor further enhanced ligand concentrations at the cell surface, likely as a direct consequence of drug rebinding (4). These data suggest that the affinity of propranolol for the β2-adrenoceptor is likely far lower than has previously been reported and highlights the critical importance of understanding the “micro-pharmacokinetic” profiles of ligands for membrane-associated proteins.

This work was funded by Novartis Institutes of Biomedical Research.

(1) Sykes DA et al. (2014). MolPharmacol 85: 608-617.

(2) Baker JG et al. (2011). J Med Chem 54: 6874-6887.

(3) Briddon SJ et al. (2004). PNAS 101: 4673-4678.

(4) Charlton SJ and Vauquelin G (2010). Br J Pharmacol 161: 488-508.