Inclusion of a phosphatase inhibitor prevents agonist metabolism in sphingosine-1-phosphate stimulated 35SGTP-gamma-S assays

Michelle E Bradley, Mark R Dowling, Neil McGuiness, John Atkinson and Steven J Charlton. Novartis Institutes for Biomedical Research, Horsham, West Sussex, U.K. RH12 5AB.

A previous study using [35S]GTPγS assays found that the potency of sphingosine-1-phosphate (S1P) at S1P2 and S1P3 receptors decreased over time (Dowling et al, this meeting). This was attributed to the metabolism of S1P by phosphatase enzymes present on the membranes of Chinese hamster ovary (CHO) cells, clearly precluding the use of this assay for quantitative pharmacology. The aim of this study was, therefore, to discover an inhibitor of S1P breakdown that was suitable for inclusion in [35S]GTPγS binding assays.

Membranes derived from CHO cells stably expressing either S1P2 or S1P3 receptors (1.25 and 2.5 µg/well, respectively), were pre-incubated with S1P, GDP (0.3 µM and 3 µM, respectively) and SPA beads, with or without phosphatase inhibitor for 0, 60, 120 and 180 minutes prior to the addition of 300 pM [35S]GTPγS. This was followed by a further 60 minute incubation, after which bound [35S]GTPγS was quantified using a Topcount (Packard). Statistical significance was determined using an unpaired Student’s t-test.

The inclusion of a cocktail of broad-spectrum phosphatase inhibitors (Cocktail II, Sigma) significantly (p < 0.05) increased the pEC50 of S1P at both receptors following a 60 minute incubation (7.90 ± 0.11 and 8.89 ± 0.07 at S1P2; 8.86 ± 0.13 and 9.75 ± 0.19 at S1P3; n = 3). The inclusion of this cocktail, however, significantly (p < 0.05) reduced total [35S]GTPγS binding by 59.9 ± 0.6 % at S1P2 and 51.8 ± 2.8 % at S1P3.

The components of this phosphatase inhibitor cocktail were then tested individually to determine which compounds were responsible for inhibition of the S1P metabolism. Only two components of the cocktail, sodium molybdate (Na2MoO4) and sodium orthovanadate (Na3VO4) were found to prevent the decrease in potency of S1P (see Table 1 for data with Na3VO4). Sodium molybdate, however, also significantly (p < 0.05) decreased total [35S]GTPγS binding at both receptors (53.5% ± 3.4 at S1P2 and 34.6% ± 6.5 at S1P3, all n=3 ± s.e.m). In contrast, sodium orthovanadate, had no significant effect on the magnitude of [35S]GTPγS binding.

Table 1: S1P pEC50 at the S1P2 receptor in the presence and absence of Na3VO4 at different incubation times. Data expressed as mean ± s.e.m (n = 3) * indicates statistically different to pEC50 in the absence of Na3VO4 at same pre-incubation time (p < 0.05; unpaired student’s t-test).

In conclusion, sodium orthovanadate prevents metabolism of S1P without reducing the magnitude of [35S]GTPγS binding, thus allowing the determination of accurate potency values at S1P2 and S1P3 receptors. Assays run in the absence of a suitable inhibitor should only be used to qualitatively study S1P pharmacology.