Flow cytometric analysis of platelets type 2 diabetes mellitus reveals ‘angry’ platelets

There is a strong indication that platelets also have relevant functions in inflammation [11]. In fact, it was shown that thrombosis and inflammation share many key molecular mechanisms and that they are fundamentally linked processes [7]. It is now recognised that vascular inflammation is the key underlying mechanism in atherogenesis and atheroprogression. The evidence of platelets being fundamental mediators in the initiation and maintenance of a chronic proinflammatory milieu is provided by the direct interactions with inflammatory cells and secretion of autocrine and paracrine effector molecules [12]. Another emerging concept is the significant role of platelet-mediated recruitment of leucocytes in the propagation, progression and pathogenesis of atherosclerotic disease. Platelets can interact with leucocytes: (a) during haemostasis, when there is vascular damage and recruit leucocytes to the growing thrombus, (b) when endothelial cells are stimulated thereby adhering and activating platelets and then bridge blood-borne leucocytes to the vessel wall and (c) in the formation of heterotypic aggregates prior to contact with endothelial cells when adhesion between platelets and leucocytes occur in the blood [13].

It is well known that in subjects with type 2 diabetes mellitus, function of platelets is impaired. In fact, a sub-threshold stimuli is needed to activate platelets which are constantly in activation despite the lack of a major plaque event and have thus been defined as ‘angry platelets’ [14]. This is significant as it has been postulated that circulating platelets in subjects with untreated type 2 diabetes mellitus are in a hyperactive state and are implicated as etiologic factors in thrombotic complications [15] which are accelerated in diabetics [16]. Of note is the finding of hyperactive platelets in metabolically controlled diabetics without cardiovascular complications [17]. In addition, expression of P-selectin is increased on the surface of platelets in patients presenting with symptomatic coronary artery disease, making it a marker of ‘angry platelets’ [18].

Diabetes with cardiovascular complication may also lead to acute conditions like thrombo-embolic ischemic stroke. Multiple studies regarding the activity of platelets in acute stroke have been performed. Results obtained from these studies (acute ischaemic stroke) showed increased mean platelet volume, platelet aggregation enhancement in post-ischaemic stroke, increased α-and dense granule release and statistically significant increase in expression of P-selectin (CD62P), CD63, and thrombospondin [19]. The study by Marquardt and co-workers investigated the time course of platelet activation after ischaemic stroke. They found a significant increase in CD62P and CD63 expression within 24 h post cerebral ischaemia. In addition, it was also shown that CD62P expression declines during the first weeks after stroke, whereas CD63 expression remains increased for at least 3 months after stroke [20].

Another confounding factor together with diabetes is cigarette smoking. Multiple studies provides evidence on the many adverse effects of smoking on the cardiovascular system. This includes: (a) it causes endothelial dysfunction [21, 22]; (b) increases inflammation [23]; (c) it alters the lipid profile and creates an atherogenic setting; (d) promotes atherosclerotic progression by enhancing oxidative stress, lipid peroxidation and mitochondrial damage [23, 24]; (e) it destabilizes atherosclerotic plaque by increasing matrix metalloproteinases [25]; (f) increases platelet activation and activates coagulation cascade with subsequent atherothrombosis [26, 27]. Flow cytometric findings in the research by Al-Dahr, showed a decrease in CD41b with an increase in CD40 and CD62 [28]. This paper, therefore investigates the functional role of platelets in diabetes, with and without cardiovascular involvement using flow cytometry and scanning electron microscopy to look at platelet ultrastructure.

Thirty healthy individuals were used as controls. These individuals were non-smokers, who did not use any chronic medication and did not have a history of thrombotic disease. Sixty diabetic subjects (type 2) were recruited from the Steve Biko Academic Hospital, diabetic clinic in South Africa. Inclusion criteria included: (a) subjects older than 18 years and willing to provide informed consent, (b) subjects with known diagnosis of diabetes, (c) for the cardiovascular group, history of previous myocardial infarction, peripheral arterial disease, stroke or coronary arterial bypass grafting. Exclusion criteria included: (a) subjects hemodynamically unstable and (b) subjected with documented life threatening disease (malignancy, HIV/AIDS). Two groups of thirty each were distinguished, with and without cardiovascular complications. Five millileters of blood was drawn into a citrate tube, from each participant. Ethical clearance was obtained for this study from the University of Pretoria Human Ethics Committee. Informed consent was obtained from all participants.

Scanning electron microscopy was used to prepare platelets from platelet rich plasma (PRP) according to previously described methods [29]. Platelets from individuals with diabetes, cerebrovascular disease and smoking were compared to platelets from healthy individuals.

For each blood sample taken four tubes was prepared; each tube containing 1 ml sheath fluid from Beckmann and Coulter and 20 µl of blood. The various tubes were stained with 20 µl of CD41-FITC (fluorescein isothiocyanate) and 20 µl of one of the following probes: CD41-PE (phycoerythrin), CD42b-PE, CD62P-PE and CD63-PE (from Beckman Coulter). The samples stained with different probes, were incubated at room temperature in the dark for 20 min before being analyzed by a flow cytometer (FC 500, Beckman Coulter). The surface expression of platelet receptors was determined by flow cytometry using the different monoclonal antibodies as indicated in Table 1.

Forward scatter and 90º side scatter were displayed on logarithmic scales. Two platelet gates were set. The first gate was set according to the morphological characteristics of platelets while the second gate was set according to CD41-FITC fluorescence, a platelet specific marker. The fluorescence of the different antibodies was plotted on 256-channel log histograms. The results were expressed in arbitrary units as mean channel fluorescence intensity (MCFI).

For each participant the MCFI was calculated as the mean fluorescence of a large sample of platelets (10,000 platelets per individual), the well-known Central Limit Theorem assures us that the Normal distribution is a close approximation for the distribution of the MCFIs for the experimental groups. GraphPad Prism 5 was employed to perform one-way ANOVA for all statistical analysis, with a p value of ≤0.005 considered significant. Post-hoc Dunnett’s Multiple Comparison Test was performed to compare the two diabetic groups to the controls.

Increased expression of platelet activation markers CD31, CD36, CD49b, CD62P and CD63 was confirmed by Eibl and co-workers when type 2 diabetics were compared with normal individuals [30]. In fact, increased expression of CD63 and CD62, enhanced platelet activation, and aggregation are viewed as one of the major causes of atherosclerosis and thrombosis in diabetes [31]. CD62P is found in the α-granules of platelets and are used as markers for activated platelets, while their absence suggests a resting state [32]. A feature that appears strongly in diabetics is that of platelet hyperaggregation. This is prevalent in both type 1 and type 2 diabetics [33]. From a pathophysiological view, this is significant as hyper-aggregated platelets have a tendency to block blood vessels [5], contributing to atherothrombotic complications in diabetics. CD63 is a 53 kDa lysosomal membrane protein identified on surface of activated platelets after release reaction [34, 35].

Prevention of early platelet adhesion to the damaged vessel wall by blocking platelet surface receptors GPIbα or GPVI protects from stroke without provoking bleeding complications. In addition, downstream signalling of GPIbα and GPVI has a key role in platelet calcium homeostasis and activation [36]. The CD42b MoAb used in this research, specifically binds to the platelet GPIbα. GPIbα forms part of the GPIb-IX-V complex which is the receptor for von Willebrand’s factor and is known as von Willebrand’s factor-dependant adhesion receptor. According to De Meyer and co-workers in 2011, the importance of GPIbα far exceeds that of VWF in arterial thrombosis and GPIbα is a central receptor in different vascular processes of thrombosis and inflammation, all of which may contribute to the progression of ischemic stroke [37]. Furthermore, engagement of GPIb-IX-V by von Willebrand factor (VWF) mediates platelet adhesion to damaged vessels and triggers platelet activation and thrombus formation in heart attacks and stroke [38].

Flow cytometric analysis found a significant increase in platelet activation regarding CD62 (P-selectin) for the diabetics group while the CD62 MCFI values decreased compared to the controls as shown in Table 3. This unexpected finding may be attributed to the fact that P-selectin can be cleaved from the membrane surface after activation releasing P-selectin into the plasma known as soluble P-selectin (sP-selectin). The exact mechanism of this shedding is unknown but several mechanisms have been suggested including cleavage by serum proteases or non-specific enzymes or by simple shedding [39]. However, studies have shown an increase in both P-selectin on the surface of platelets and sP-selectin indicating that diabetes with or without cardiovascular complications are associated with chronic activation of platelets as P-selectin is being shed from activated platelets and as new P-selectin is being expressed on recently activated platelet [40]. This study finding echoes results found by Véricel et al. [17] whom also discovered hyperactive platelets in metabolically controlled diabetics without cardiovascular complications. The diabetic subjects with cardiovascular disease, recruited in this study were those with ischaemic events many months and years prior to recruitment into this study. Our finding is in keeping with persistently hyperactivated platelets.

The CD63 MoAb recognizes the activation-specific fusion of the lysosomal granule membrane with the plasma membrane, therefore it only binds on the surface of activated platelets and is a useful tool to use in the identification of activated platelets [41]. In our study CD63 percentage activated platelets were significantly increased compared to the healthy controls. The diabetic group with cardiovascular complication showed the greatest percentage of platelet activation (61.24 % activation) while the diabetic group without cardiovascular complications showed a slightly lower activation percentage (54.39 % activation), as shown in Fig. 2. It appears as if CVD may play a role in platelet hyperactivation with lysosomal involvement.