Platelet granules: surprise packages

In this issue of Blood, Kamykowski et al use high-resolution immunofluorescence microscopy to examine the distribution of α granule cargo proteins to determine whether platelets contain cargo-specific granule subsets.¹ 

Platelet secretion is essential in the normal response to vascular damage but may also contribute to pathogenic sequllae. To date, more than 300 distinct molecules have been detected in platelet releasates.²  Dense granules predominantly contain small molecules (eg, serotonin, ADP, polyphosphates). Alpha granules contain a plethora of proteins that comprise the bulk of the platelet secretome, including hemostatic factors (eg, Factor V, VWF, fibrinogen), angiogenic factors (eg, angiogenin, VEGF), anti-angiogenic factors (eg, angiostatin, PF4), growth factors (eg, PDGF, bFGF, SDF1α), proteases (eg, MMP2, MMP9), necrotic factors (eg, TNFα, TNFβ), and other cytokines. Some of these are produced by megakaryocytes and packaged into granules during biosynthesis. Other cargo (eg, fibrinogen and Factor V) are thought to be endocytosed by circulating platelets and then transported to α granules.³  The releasate catalog suggests that platelet secretion is pivotal to establishing and controlling the microenvironment at a wound site. However, it is not clear whether platelet secretion is a controlled, contextually responsive process or a random, stochastic event.

Italiano et al proposed that platelets differentially store and release their granular cargo in a thematic way; for example, promoting angiogenesis by releasing VEGF from a granule subset under one circumstance and dampening angiogenesis in other circumstances by releasing angiostatin from a different granule subset.⁴  This hypothesis grew from 2 observations: First, some cargo proteins (ie, pro- and anti-angiogenic factors) were detected in distinct, non-overlapping compartments.^4,5  Second, Ma et al suggested that specific cargo proteins were differentially released in response to distinct agonists (ie, PAR1 and PAR4 activating peptides).⁶  This hypothesis has drawn great attention because it implies that platelets are “smart” delivery devises, which can release contextually appropriate cargo.⁷  Despite this excitement, it is not clear whether and to what extent α granules differ and whether the secretory machinery present in platelets is sufficient for differential control of granule exocyosis.

Kamykowski et al examined the first aspect of this hypothesis, differential cargo packaging, using high-resolution immunofluorescence microscopy techniques. One method, called 3-dimensional structured illumination microscopy, allows resolution beyond the diffraction limits of a standard light microscope.⁸  Kamykowski et al examined the colocalization of 15 different α granule cargo proteins in 28 different pairwise combinations. Granule proteins were selected based on their functional diversity and their potential source (ie, endocytosed or synthesized). Detailed colocalization analysis showed considerable variation in codistributions and little if any thematic patterns in cargo clustering. In fact, the data suggested a Gaussian cargo distribution that implies random sorting of granule cargo. In short, there did not appear to be α granule subsets with distinct cargo as was predicted. One feature that was apparent from the analysis was the distinct spatial distributions of cargo within a granule. Cargo was not uniformly spread throughout the granule, but instead appeared to be distributed into clusters. Similar segregation of cargo within an α granule was reported in an elegant ultrastructural analysis done by van Nispen tot Pannerden et al.⁹ 

The second aspect of the hypothesis, differential release, has yet to be fully addressed. Much of the extant data has been generated from single time point or single agonist dose experiments. It is not clear if the differential release measured is because of differences in the degree of platelet activation or in the rate of the release process. It should be noted that there are clear differences comparing dense and α granule cargo release kinetics.¹⁰  Regardless of agonist used for stimulation, small molecules such as serotonin and ADP are released at faster rates than either α granule or lysosomal cargo proteins. A full analysis of the kinetics of platelet secretion will be required to more completely determine whether release of specific α granule cargo proteins is differential.

Does platelet secretion need to be differential? It certainly is when one compares dense granule to α granule cargo release. It is also attractive to think of platelets as “smart,” releasing specifically needed α granule cargo at the site of vascular injury; but is that required? By simultaneously releasing factors with opposing functions, the platelet could exert a finely balanced effect at the damage site. Kamykowski et al discuss a process in Caenorhabditis elegans where coincident release of opposing factors is important for development.¹  Alternatively, kinetically controlled release could exert phasic effects. Rapidly released cargo could mediate immediate/early events (ie, platelet activation), whereas slow or prolonged release could have effects on processes that persist during and after thrombosis. While the distinction between dense and α granule cargo is clear because they come from granules with different properties, it is not clear how α granule cargo could be released at different rates if they origiate from the same class of granules. One possible mechanism for differential release of α granule cargo may lie in the apparent distribution of cargo proteins. Cargo is heterogeneously distributed in a granule.^1,5,9  Could heterogeneity in cargo release be a simple factor of differential solubilization of these cargo clusters once the granule membrane has fused to the plasma membrane? More detailed analysis of platelet secretion is needed before these questions can be answered.

While the data presented by Kamykowski et al fail to support one aspect of the differential release hypothesis, it is clear that there are questions yet to be resolved. Are there conditions where platelet secretion is thematic? Can platelet secretion be explained by differences in release kinetics or are the apparent difference because of differential protein solubilities? How “smart” a transfer device are these small cellular fragments? Given the recent revelations of the intricacies of platelet cell biology, it is clear that platelets have many more surprises to deliver.