ИСТИНА |
Войти в систему Регистрация |
|
ФНКЦ РР |
||
Background: Human platelets have a unique feature: one relatively simple signal simultaneously controls different functional responses of the platelet depending on type and amount of an activator. Platelets exhibit series of calcium spikes upon activation, so we subjected platelet calcium profiles to a number of tests in order to understand the principles of signal encoding in them. Aims: Our goal was to understand how a platelet interprets extracellular stimuli by a series of calcium spikes. Methods: We obtained calcium profiles using TIRF microscopy and microfluidic systems that allowed us to control platelet activation. We analyzed calcium profiles for resting latelets, platelets activated with 10 μM ADP and non-fibrillar collagen type I. We used FFT to extract the possible fundamental frequency from a signal. Using thresholding to eliminate the noise, we performed feature extraction which resulted in distributions of peak amplitudes, widths, areas, numbers per signal and interspike intervals. Statistical differences between types of platelet activation were measured using Kolmogorov-Smirnov criterion. Also, we tested interspike intervals to be of Poissonian nature or not. Results: Fourier analysis did not reveal a fundamental frequency of platelet calcium scillations due to stochasticity of calcium spikes and Gibbs effect caused by peaks. We observed statistical differences between platelet calcium spikes’ features for all types of activation, with interspike intervals which displayed 3 s shifts between non-activated and activated cells. Also, the large (more than 10 s) interspike intervals were observed only for resting platelets. Resting platelets exhibited 0.02 - 0.05 peaks per second, while in response to collagen or ADP platelets tend to 0.25 - 0.35 peaks per second. We showed that means and standard deviations of interspike intervals were in linear relation for all types of activation, and there was a minimum interspike interval of 1.7 ± 0.3 s predicted. Conclusions: We showed that distributions of platelet calcium peak features were statistically different between each other upon activation, with interspike intervals and number of peaks per second giving the most dramatic results, indicating the importance of temporal properties of the signals. We showed that occurence of platelet calcium peaks can be interpreted as a Poissonian process with a deterministic part that is constant between types of platelet activation. Fast Fourier Transform is not applicable for platelet calcium dynamics. Analysis of fibrin formation and thrombin generation dynamics