Background: Light transmission aggregometry (LTA) with simultaneous assay of ATP released from platelet dense granules (lumi-aggregation) provides the ability to assess platelet function in a variety of clinical settings. This methodology, however, requires the ability to obtain adequate sample volume from the patient, if a number of platelet agonists are to be assessed. The platelet count within the patient's blood is also a limiting factor, and LTA interpretations can become difficult when platelet counts in the sample fall significantly below 150,000/µL. Analysis of platelet function by flow cytometry circumvents the limitations of both blood volume and low platelet counts, typically providing information as to glycoprotein IIb/IIIa activation state and release of platelet alpha granules in response to platelet agonists. Flow cytometry, however, is not presently capable of providing quantitative analysis of the release of platelet dense granule constituents. Given the clinical importance of platelet dense granule secretion, we sought to develop a non-radioactive method for quantitation of agonist-induced release of serotonin (5-HT) from blood sample volumes obtainable from small infants and from patients with at least moderate thrombocytopenia.

Methods: In the initial studies, total platelet content and released 5-HT following stimulation by a panel of agonists were assessed using 200 µL of freshly drawn, normal citrated blood. In subsequent studies, uptake and agonist-induced secretion of non-radioactive, deuterated serotonin (d-5HT) were additionally measured in parallel with the endogenous 5-HT in these 200 µL samples. In brief, the citrated blood was initially incubated for 15' at 37° with d-5HT, then diluted 10-fold with buffer, and platelet stimulus (collagen, TRAP, ADP, epinephrine, ionophore A23187, thromboxane analogue U46619, ristocetin, or buffer blank) added. Following stimulation, samples were further diluted with buffer, prostaglandin E1 added, and then centrifuged to yield supernatants for analysis. 5-HT and d-5HT total content in unstimulated platelets was additionally assayed. N-methyl-5HT was employed as an internal standard. 5-HT, d-5HT and N-methyl-5HT were subsequently detected and quantified simultaneously by liquid chromatography/tandem mass spectrometry (MS) (Eksigent MicroLC 200/QTrap 6500 Mass Spectrometer, Sciex, Framingham, MA).

Results and Conclusions: MS analysis of 5-HT was linear over the range of 50-2000 pg/mL. For whole blood platelet counts of at least 120,000/µL, this sensitivity was sufficient to permit analysis of 5-HT released in response not only to strong stimuli, but even to weaker stimuli such as low concentration (2 µg/mL) collagen. Since the goal of this project, however, was to be able to extend functional analysis to patients with more significant degrees of thrombocytopenia, further efforts were made to improve sensitivity. Assay of agonist-induced release, following platelet uptake, of d-5HT was assessed in parallel with the assay of released endogenous 5-HT. Release of 5-HT and d-5HT in response to increasing concentrations of collagen showed generally similar kinetics. However, not only was the absolute signal increased for the d-5HT, but the analytic assay itself proved significantly more sensitive as compared to that for 5-HT. Combined, these effects increased the overall sensitivity of the assay by at least an order of magnitude. For example, d-5HT released in response to 2 µg/mL collagen in a donor blood having a platelet count of 188,000/µL produced an MS signal approximately 30-fold higher than the lower limit of detection for the assay. Although it will require actual study of thrombocytopenic patients to verify, these results suggest that by use of this MS approach with d-5HT it may be possible to measure platelet dense granule secretion to a panel of agonists from 200 µL of blood in patients with thrombocytopenia potentially as severe as 10,000/µL.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.