Development of a homogenous time-resolved fluorescence-based immunoassay for the evaluation of platelet thromboxane release

Nicholas Kirkby1,2, Sandrine Cabirol3, Marielle Mazille3, Jane Mitchell2, Tim Warner1. 1Barts & the London School of Medicine & Dentistry, London, UK, 2Imperial College, London, UK, 3Cisbio, Codolet, France.

Activated platelets synthesise thromboxane (Tx)A2 to amplify and perpetuate platelet aggregation, and inhibition of this process underlies the anti-thrombotic effect of aspirin. Measurement of platelet TxA2 release is therefore important for assessing drug effectiveness as well as for mechanistic investigations into platelet function. This is limited, however, by the expense and low throughput of traditional immunoassay techniques. Homogenous time-resolved fluorescence (HTRF®) technology is a liquid phase fluorescence-resonance energy transfer (FRET)-based assay system that scales readily to high density microtitre plate systems, thereby lowering reagent costs and increasing throughput. We sought to develop an HTRF®-based immunoassay suitable for the measurement of platelet TxA2 release.

A monoclonal antibody against the stable TxA2 degradation product, TxB2, was raised in mice and labelled with the HTRF® donor fluorophore Europium cryptate. This was paired with TxB2 conjugated to the HTRF® acceptor fluorophore d2. When mixed, the anti-TxB2 cryptate binds to TxB2-d2 in competition with any unlabelled TxB2 present. The excitation of the donor at 620nm triggers FRET towards the acceptor resulting in emission at 665nm. Measurements were taken 50μs after excitation to eliminate detection of auto-fluorescent emission using a Berthold Mithras LB 940 mutlimode plate reader. Assays were performed in a low volume 384 well microtitre plate format, requiring 5μl of each reagent and 10μl of sample. Samples for measurement of platelet TxA2 release were generated by stimulating human platelet-rich plasma with collagen (0.1-30μg/ml) in the presence or absence of aspirin (30μM) and monitoring aggregation by 96 well plate aggregometry (n = 4). After 16 mins, TxA2 formation was halted by addition of diclofenac (1mM). All data are quoted as mean ± SEM.

The incubation of anti-TxB2-Cryptate with TxB2-d2 produced a readily detectable FRET signal, which was attenuated in a concentration-dependent manner by unlabeled TxB2 and was stable from 30mins to 24hrs. The concentration of each reagent was optimized to maximize assay sensitivity, whilst maintaining an acceptable signal-to-background ratio. This yielded an assay with a detection range of 5-500ng/ml. When unlabeled TxB2 was spiked into human plasma, the most common matrix for the measurement of platelet TxA2 release, a 41% suppression of the FRET signal was observed, which could be overcome by dilution to <20% plasma. The assay exhibited low cross-reactivity to arachidonic acid (0.24±0.11%), U46619 (0.22±0.06%), PGE1 (0.10±0.01%), PGE2 (0.53±0.06%) and PGJ2 (0.19±0.07%; all n = 4). The suitability of the assay for measurement of platelet-derived TxA2 release was confirmed by assaying the supernatant of collagen-stimulated platelets. Collagen produced concentration-dependent increases in TxB2concentration and this response was almost abolished by pre-treatment with aspirin (e.g. at 30microg/ml collagen: Vehicle: 154.5±43.5ng/ml TxB2; Aspirin: 20.9±4.1ng/ml TxB2; n = 4; p<0.0001 by 2-way ANOVA).

In summary, here we describe an HTRF®-based competitive immunoassay for TxB2, which can be performed in total reaction volumes of 20ul minimising use of reagents and samples. This assay provides the means for the cost effective, high throughput screening of TxA2 release from platelets and other biological sources.