LETTERS TO THE EDITOR
Thrombin induces de novo protein synthesis of stromal cell-derived factor-1α but not angiostatin in human platelets
Article first published online: 1 OCT 2012
© 2012 International Society on Thrombosis and Haemostasis
Journal of Thrombosis and Haemostasis
Volume 10, Issue 10, pages 2202–2205, October 2012
How to Cite
HUANG, Z., RAHMAN, M. F.-U., JIANG, L., XIE, H., HU, H., LUI, W.-O. and LI, N. (2012), Thrombin induces de novo protein synthesis of stromal cell-derived factor-1α but not angiostatin in human platelets. Journal of Thrombosis and Haemostasis, 10: 2202–2205. doi: 10.1111/j.1538-7836.2012.04888.x
- Issue published online: 1 OCT 2012
- Article first published online: 1 OCT 2012
- Accepted manuscript online: 20 AUG 2012 10:20AM EST
- Received 25 June 2012, accepted 24 July 2012
Platelets are versatile cells that function in multiple physiologic/pathophysiologic processes other than thrombosis and hemostasis . It has become clear that platelets contribute importantly to both angiogenesis and the maintenance of vessel integrity [1,2]. Platelets contain both proangiogenic (e.g. stromal cell-derived factor-1α [SDF-1α] and vascular endothelial growth factor) and antiangiogenic (angiostatin and endostatin) regulators. Interestingly, platelets package proangiogenic and antiangiogenic regulators in distinct α-granules, and release them separately upon different stimuli [3–5]. The selective release of platelet proangiogenic regulators and of antiangiogenic regulators have been demonstrated to promote and inhibit angiogenesis, respectively . On the other hand, platelets retain a certain capacity for protein synthesis upon activation [7–9], albeit they are anucleate cells. We therefore investigated whether platelet activation could lead to de novo synthesis of angiogenic regulatory proteins, namely SDF-1α and angiostatin.
Twelve healthy subjects (seven males and five females, aged 25–46 years) gave informed consent to participate in the study, which was approved by the Ethics Committee of the Karolinska Institute. Platelets were isolated from venous blood by combining gradient centrifugation and leukocyte depletion with CD45 magnetic beads, as detailed previously . Washed platelets (leukocyte contamination of < 0.001%) were resuspended in serum-free Dulbecco’s modified Eagle’s medium at a concentration of 109 mL−1, and treated (37 °C, 30 min, or 16 h) with vehicle or thrombin (0.1 U mL−1), which increased platelet P-selectin expression up to 95%. Platelets were pelleted (2000 × g, 5 min) after 30 min or 16 h of treatment. For immunoblotting, platelet pellets were lysed with an NP-40 lysis buffer (Invitrogen, Carlsbad, CA, USA) containing a protease inhibitor cocktail (Sigma, St Louis, MO, USA) and 1 mm phenylmethanesulfonyl fluoride (Sigma). Platelet total RNA was extracted from the samples containing at least 109 washed platelets with an Ambion mirVana microRNA isolation kit (Ambion/Applied Biosystems, Austin, TX, USA).
Figure 1A shows that thrombin stimulation for 30 min decreased the intensity of SDF-1α-immunoreactive bands, indicating reduced levels of SDF-1α in platelets, i.e. SDF-1α release upon stimulation. The SDF-1α immunoblotting intensity was, however, partially restored after 16 h of culture, suggesting de novo protein synthesis of SDF-1α. As expected, the immunoblotting intensity of the loading control glyceraldehyde-3-phosphate dehydrogenase remained largely unchanged throughout the experiments. When SDF-1α immunoblotting intensity was quantified, SDF-1α levels were significantly decreased by 30 min of thrombin stimulation, and showed a rebound after 16 h, albeit to a level lower than that before stimulation (Fig. 1C). Similar to that of SDF-1α, thrombin stimulation markedly reduced the intensity of angiostatin-immunoreactive bands (Fig. 1B,D), indicating that thrombin also induced platelet secretion of angiostatin. In contrast to the rebound seen with SDF-1α, the reduction in platelet angiostatin content remained after 16 h of cell culture.
Subsequent work was undertaken to investigate whether changes in platelet protein content were related to mRNA maturation. To demonstrate the expression of functional SDF-1α mRNA, we chose a quantitative real-time RT-PCR (qRT-PCR) with primers spanning an exon–exon junction of SDF-1α/CXCL12 (ID: Hs00171022_m1; Applied Biosystems), which allowed measurement of mature mRNA of SDF-1α but not of pre-mRNA; 18S rRNA was used as an endogenous control (ID: Hs99999901_s1). The cDNA synthesis was performed with 200 ng of total RNA and a High Capacity cDNA Reverse Transcription kit (Applied Biosystems). Figure 1E shows that qRT-PCR readily detected 18S rRNA in both unstimulated and thrombin-activated platelets. For SDF-1α, however, no mature mRNA was detected in unstimulated platelets, whereas significant amplification of mature SDF-1α mRNA was found in thrombin-activated platelets after approximately 30 PCR cycles. These results suggest that thrombin stimulation induced maturation of SDF-1α mRNA, and that the mRNA was expressed at a relatively low level as compared with 18S rRNA. When the angiostatin/plasminogen (PLG) qRT-PCR assay Hs00264877_m1 (Applied Biosystems) was used, no significant angiostatin/PLG mRNA was detected in either unstimulated or thrombin-stimulated platelets (Fig. 1F).
Platelets were previously considered to be incapable of protein synthesis. However, a number of transcripts have been found in polysomes of platelets, and accumulating evidence is showing that platelets can undertake significant protein synthesis . The best-known examples of proteins synthesized de novo in platelets are the regulatory protein B-cell lymphoma-3  and the proinflammatory cytokine interleukin-1β . It is of note that de novo synthesis of both proteins was seen in activated platelets but not in quiescent platelets [7,8]. The presence of functional SDF-1α mRNA in platelets has previously been investigated by us and others [9,11]; in these studies, mature mRNA was not detected in platelets without in vitro stimulation. In agreement with those findings, the present work demonstrated that mature SDF-1α mRNA was absent in unstimulated platelets. Mature SDF-1α mRNA was, however, detected in thrombin-stimulated platelets. Hence, thrombin stimulation triggers the maturation process of SDF-1α pre-mRNA, leading to de novo protein synthesis of SDF-1α. Functional angiostatin/PLG mRNA was not detectable in either unstimulated or thrombin-stimulated platelets, indicating that plasminogen is a presynthesized protein stored in platelets. The absence of functional angiostatin/PLG mRNA highlights the fact that platelet activation-initiated de novo protein synthesis is selective. Thus, apart from distinct packaging and secretion of platelet proangiogenic and antiangiogenic regulators [4,5], the present findings of platelet activation-induced de novo protein synthesis of selective angiogenic regulators represent a new aspect of finely tuned platelet angiogenic activities.
Platelet activation leads to SDF-1α release and expression on the cell surface [5,11]. It has been demonstrated that platelet-expressed SDF-1α, in collaboration with other platelet adhesion molecules (e.g. P-selectin and glycoprotein IIb–IIIa), supports bone marrow-derived progenitor cell adhesion and recruitment at sites of vessel injury , and that SDF-1α promotes differentiation of bone marrow-derived progenitor cells into endothelial progenitor cells . Therefore, selective release of proangiogenic and antiangiogenic regulators [4,5] is important for platelet angiogenic activities during the acute phase of vessel reparation, whereas de novo synthesis of platelet angiogenic regulators may represent a delayed phase of platelet angiogenic activity.
In conclusion, platelets contain SDF-1α mRNA transcripts. Thrombin stimulation induces SDF-1a mRNA maturation, which leads to de novo synthesis of SDF-1α after activation. The newly synthesized SDF-1α may reinforce platelet angiogenic activities in remodeling and repair of the injured vessels.
This work was supported by grants from the Swedish Research Council, the Swedish Heart-Lung Foundation, the Karolinska Institute, and the Stockholm County Council.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.