Platelets stored at 4°C contribute to superior clot properties compared to current standard-of-care through fibrin-crosslinking

Abstract

Currently, platelets for transfusion are stored at room temperature (RT) for 5-7 days with gentle agitation, but this is less than optimal because of loss of function and risk of bacterial contamination. We have previously demonstrated that cold (4°C) storage is an attractive alternative because it preserves platelet metabolic reserves, in vitro responses to agonists of activation, aggregation and physiological inhibitors, as well as adhesion to thrombogenic surfaces better than RT storage. Recently, the US Food and Drug Administration clarified that apheresis platelets stored at 4°C for up to 72 h may be used for treating active haemorrhage. In this work, we tested the hypothesis that cold-stored platelets contribute to generating clots with superior mechanical properties compared to RT-stored platelets. Rheological studies demonstrate that the clots formed from platelets stored at 4°C for 5 days are significantly stiffer (higher elastic modulus) and stronger (higher critical stress) than those formed from RT-stored platelets. Morphological analysis shows that clot fibres from cold-stored platelets were denser, thinner, straighter and with more branch points or crosslinks than those from RT-stored platelets. Our results also show that the enhanced clot strength and packed structure is due to cold-induced plasma factor XIII binding to platelet surfaces, and the consequent increase in crosslinking.

Keywords: factor XIII; clot strength; refrigeration; rheology; ultrastructure.

Conflict of interest statement

Conflict of Interest Disclosures

The authors declare no conflict of interest.

© 2017 John Wiley & Sons Ltd.

Figures

Figure 1. Effect of platelets on clot stiffness

Rheological properties of evolving clots were investigated using dynamic mechanical analysis at 37°C for 30 min. (A) Representative traces of storage modulus, G′ (clot stiffness) for clots formed from plasma, plasma with platelets (300 ×109 platelets /l) and plasma with lyophilized platelets (300 ×109 platelets /l); (B) Clot stiffness increases by 6-fold with physiological concentration of live platelets in plasma (n=3; *, P<0.05). (C) Platelets failed to form stable clots on blocking glycoprotein IIb/IIIa receptor (Eptifibatide) or actin polymerization (Cytochalasin B). [n=3; ns, not significant (P >0.05); *, P < 0.05; **, P<0.01; ***, P<0.001]. Lyo, lyophilized platelets; PPP, platelet poor plasma; PRP, platelet rich plasma.

Figure 2. Effect of storage temperature on clot mechanical properties

Stiffness of clots formed from fresh platelets or platelets stored for 5 days at 4°C (4C) or room temperature (RT) were analysed. (A) Clots from 4C-stored platelet rich plasma (PRP) showed stiffness comparable to that from fresh platelets, and was higher than that from RT-stored PRP; (B) Clots from platelet poor plasma (PPP) stored at either 4°C or RT were much weaker than from fresh PPP (n = 5); (C) Representative rheological traces of fresh/ stored platelets with fresh frozen plasma (FFP); (D) RT-stored platelets form clots with less stiffness, whereas 4C-stored platelets have similar stiffness as fresh platelets (n = 5); (E) Cold storage maintains clot stiffness over longer periods of storage, unlike RT storage. Data normalized to stiffness of fresh platelets (n = 4, * compared to Fresh, δ compared to 5 day RT-stored, P < 0.05); (F) Representative traces from amplitude sweep analysis of pre-formed clots; Clot strength and cross-linking density were quantified as the stress at which the clot deformed irreversibly and the slope of the stress-strain graph in the linear regime, respectively; (G) Cold, but not RT storage maintains clot strength; and (H) Relative crosslinking density (RCD) compared to fresh platelets (= 1.0). Clots from cold-stored platelets have higher cross-linking density (n=5). ns, not significant (P >0.05); *, P<0.05; **, P<0.01.

Figure 3. Clot architecture

Clots were formed from fresh/stored platelets with 20 mM calcium, fixed, stained, dehydrated and analysed by scanning electron microscopy (5000x at 20 kV, Jeol 6610LV; Jeol, Tokyo, Japan). (A) Representative images of clots formed from fresh and stored platelets; (B–E) Quantification of morphology of clots from fresh and stored platelets: (B) Density; (C) Curvature; (D, E) Frequency distribution of fibre lengths and mode (D) and fibre diameters and mode (E). Based on the diameter, the fibres were classified either as thin (300 nm). (n=3). ns, not significant (P >0.05); *, P<0.05; **, P<0.01. 4C, platelets stored at 4°C; RT, platelets stored at room temperature.

Figure 4. Thrombin generation from fresh and stored platelets

Thrombin generation in apheresis platelet (AP) pellets resuspended in fresh frozen plasma (150,000/ml) was measured using the calibrated automated thrombogram (CAT) for 50 min. (A) Representative thrombogram from the CAT assay; (B,C) Less lag time and increased peak height for thrombin generation in stored platelets compared to fresh platelets (D) Total thrombin generated over time (ETP) is the same in stored as well as in fresh platelets (n=4).; ns, not significant (P >0.05); *, P<0.05. 4C, platelets stored at 4°C; RT, platelets stored at room temperature.

Figure 5. Cold induced plasma factor XIII (FXIII) binding

Platelets were stained with anti-FXIIIB antibodies to confirm the source of FXIII binding in cold-stored platelets.(A) Representative histogram of FXIIIB expression in stored vs. fresh platelets (B) Increased FXIIIB positive platelets compared to fresh platelets at baseline (C) Geometric mean fluorescence (GMFI) indicates more FXIIIB binding on individual cold-stored platelet surfaces (D) FXIIIB expression in aggregates and microparticles in stored platelets implies fibrinogen-mediated binding; FXIIIA was probed in immunoblots. (E) Representative immunoblot showing increased relative density of FXIIIA content in cold-stored platelets (F) Decreased FXIII binding in PAS storage compared to AP (n=4). ns, not significant (P >0.05); *, P<0.05; **, P<0.01. 4C, platelets stored at 4°C; aggs, aggregates; AP, apheresis platelets; MP, microparticles; PAS, platelet additive solution; RT, platelets stored at room temperature.

Figure 6. Effect of FXIII on clot mechanics

Rheological analysis was carried out on stored platelets stored in PAS. (A, B) Cold-stored platelets in PAS have similar clot stiffness and crosslinking density to fresh platelets (fresh = 1.0, n=4). (C) Cold-stored platelets in autologous plasma form clots in FXIII-deficient plasma with higher stiffness than fresh platelets (n=4). ns, not significant (P >0.05); *, P<0.05. 4C, platelets stored at 4°C; AP, apheresis platelets; PAS, platelet additive solution; RCD, relative crosslinking density; RT, platelets stored at room temperature.

Figure 7. Mechanism of FXIII-mediated cross linking in cold-stored platelets

(A) Cold storage leads to Intracellular calcium leakage in platelets, resulting in shape change and activation of GPIIb/IIIa receptor through calcium inside-out signalling. Fibrinogen (complexed with FXIII on γ chain) in stored autologous plasma binds to GPIIb/IIIa (B) Upon activation, thrombin cleaves fibrinopeptides to convert fibrinogen monomer to fibrin, as well as FXIIIB to activate FXIIIA. Platelets act as nucleation site for fibrin polymerization, and FXIIIA mediates γ–dimer (longitudinal and transverse), γ-tetramer, α-polymer, αγ-hybrid crosslinks (covalent bonds) in laterally aggregated fibrin polymer (soft clot) to form a strong clot (hard clot) with thinner fibres. D, fibrinopeptide D; E, fibrinopeptide E; GP, glycoprotein.