Régis Bobe

High-dose irradiation can trigger numerous endothelial dysfunctions, including apoptosis, the overexpression of adhesion molecules, and alteration of adherens junctions. Altogether, these endothelial dysfunctions contribute to the development of tissue inflammation and organ damage. The development of endothelial dysfunctions may depend on protein phosphorylation by various protein kinases, but the possible role of protein kinase A (PKA) has not been investigated so far, and efficient compounds able to protect the endothelium from irradiation effects are needed. Here we report the beneficial effects of the PKA inhibitor KT5720 on a panel of irradiation-induced endothelial dysfunctions in human pulmonary microvascular endothelial cells (HPMECs). High-dose X-irradiation (15 Gy) triggered the late apoptosis of HPMECs independent of the ceramide/P38 MAP kinase pathway or p53. In contrast, the treatment of HPMECs with KT5720 completely prevented irradiation-induced apoptosis, whether applied before or after cell irradiation. Immunostainings of irradiated monolayers revealed that KT5720 treatment preserved the overall integrity of endothelial monolayers and adherens junctions linking endothelial cells. Real-time impedance measurements performed in HPMEC monolayers confirmed the overall protective role of KT5720 against irradiation. Treatment with KT5720 before or after irradiation also reduced irradiation-induced ICAM-1 overexpression. Finally, the possible role for PKA in the development of endothelial dysfunctions is discussed, but the potency of KT5720 to inhibit the development of a panel of irradiation-induced endothelial dysfunctions, whether applied before or after irradiation, suggests that this compound could be of great interest for both the prevention and treatment of vascular damages in the event of exposure to a high dose of radiation.

Endoglin (Eng) is an endothelial cell (EC) transmembrane glycoprotein involved in adhesion and angiogenesis. Eng mutations result in vessel abnormalities as observed in hereditary hemorrhagic telangiectasia of type 1. The role of Eng was investigated in endothelial functions and permeability under inflammatory conditions, focusing on the actin dynamic signaling pathway. Endothelial Colony-Forming Cells (ECFC) from human cord blood and mouse lung/aortic EC (MLEC, MAEC) from Eng^+/+ and Eng^+/− mice were used. ECFC silenced for Eng with Eng-siRNA and ctr-siRNA were used to test tubulogenesis and permeability +/− TNFα and +/− LIM kinase inhibitors (LIMKi). In silico modeling of TNFα–Eng interactions was carried out from PDB IDs 5HZW and 5HZV. Calcium ions (Ca²⁺) flux was studied by Oregon Green 488 in epifluorescence microscopy. Levels of cofilin phosphorylation and tubulin post-translational modifications were evaluated by Western blot. F-actin and actin–tubulin distribution/co-localization were evaluated in cells by confocal microscopy. Eng silencing in ECFCs resulted in a decrease of cell sprouting by 50 ± 15% (p < 0.05) and an increase in pseudo-tube width (41 ± 4.5%; p < 0.001) compared to control. Upon TNFα stimulation, ECFC Eng–siRNA displayed a significant higher permeability compared to ctr-siRNA (p < 0.01), which is associated to a higher Ca²⁺ mobilization (p < 0.01). Computational analysis suggested that Eng mitigated TNFα activity. F-actin polymerization was significantly increased in ECFC Eng-siRNA, MAEC^+/−, and MLEC^+/− compared to controls (p < 0.001, p < 0.01, and p < 0.01, respectively) as well as actin/tubulin distribution (p < 0.01). Furthermore, the inactive form of cofilin (P-cofilin at Ser3) was significantly decreased by 36.7 ± 4.8% in ECFC Eng-siRNA compared to ctr-siRNA (p < 0.001). Interestingly, LIMKi reproduced the absence of Eng on TNFα-induced ECFC-increased permeability. Our data suggest that Eng plays a critical role in the homeostasis regulation of endothelial cells under inflammatory conditions (TNFα), and loss of Eng influences ECFC-related permeability through the LIMK/cofilin/actin rearrangement-signaling pathway.

Rationale: Ca ²⁺ signaling is a key and ubiquitous actor of cell organization and its modulation controls many cellular responses. SERCAs (sarco-endoplasmic reticulum Ca ²⁺ -ATPases) pump Ca ²⁺ into internal stores that play a major role in the cytosolic Ca ²⁺ concentration rise upon cell activation. Platelets exhibit 2 types of SERCAs, SERCA2b and SERCA3 (SERCA3 deficient mice), which may exert specific roles, yet ill-defined. We have recently shown that Ca ²⁺ mobilization from SERCA3-dependent stores was required for full platelet activation in weak stimulation conditions. Objective: To uncover the signaling mechanisms associated with Ca ²⁺ mobilization from SERCA3-dependent stores leading to ADP secretion. Methods and Results: Using platelets from wild-type or Serca3 -deficient mice, we demonstrated that an early (within 5–10 s following stimulation) secretion of ADP specifically dependent on SERCA3 stored Ca ²⁺ is exclusively mobilized by nicotinic acid adenosine dinucleotide-phosphate (NAADP): both Ca ²⁺ mobilization from SERCA3-dependent stores and primary ADP secretion are blocked by the NAADP receptor antagonist Ned-19, and reciprocally both are stimulated by permeant NAADP. In contrast, Ca ²⁺ mobilization from SERCA3-dependent stores and primary ADP secretion were unaffected by inhibition of the production of IP3 (inositol-1,4,5-trisphosphate) by phospholipase-C and accordingly were not stimulated by permeant IP3. Conclusions: Upon activation, an NAADP/SERCA3 Ca ²⁺ mobilization pathway initiates an early ADP secretion, potentiating platelet activation, and a secondary wave of ADP secretion driven by both an IP3/SERCA2b-dependent Ca ²⁺ stores pathway and the NAADP/SERCA3 pathway. This does not exclude that Ca ²⁺ mobilized from SERCA3 stores may also enhance platelet global reactivity to agonists. Because of its modulating effect on platelet activation, this NAADP-SERCA3 pathway may be a relevant target for anti-thrombotic therapy. Graphic A graphic abstract is available for this article.

In obesity, platelets are described as hyperactive, mainly based on increased platelet size and presence of pro-thrombotic plasmatic molecules. We explored platelet functions, including calcium signalling in obesity, and the effect of weight loss. We included 40 obese patients (women with body mass index [BMI] of ≥ 35 kg/m²) who were to undergo gastric bypass surgery and 40 healthy lean subjects (women with BMI of < 25 kg/m²) as a control group. Approximately 1 year after surgery, the obese patients lost weight (75% had a BMI < 35 kg/m²). They were explored a second time with the same healthy control for the same platelet experiments. Compared with controls, obese patients' platelets displayed reduced sensitivity to thrombin (aggregation EC₅₀ increased by 1.9 ± 0.3-fold, p = 0.005) and a lower Ca²⁺ response (70 ± 7% decrease, p < 10⁻⁴). In 17 pairs of patients, we performed additional experiments: in obese patients' platelets, thrombin-induced αIIbβ3 activation was significantly lower (p = 0.003) and sarco-endoplasmic reticulum Ca²⁺ATPase (SERCA3) expression was decreased (48 ± 6% decrease, p < 10⁻⁴). These differences were abolished after weight loss. Interestingly, pharmacological inhibition of SERCA3 activity in control group's platelets mimicked similar alterations than in obese patients' platelets and was associated with defective adenosine diphosphate (ADP) secretion. Addition of ADP to agonist restored platelet functions in obese patients and in SERCA3-inhibited control platelets (five experiments) confirming the direct involvement of the SERCA3-dependent ADP secretion pathway. This is the first study demonstrating that platelets from obese patients are hypo-reactive, due to a deficiency of SERCA3-dependent ADP secretion. Weight loss restores SERCA3 activity and subsequent calcium signalling, αIIbβ3 activation, platelet aggregation and ADP secretion. ^* These authors contributed equally to this work. ^** These authors are joint senior authors on this work.

Objective—: Here, we provide evidence for the role of FLNA (filamin A) in the modulation of store-operated calcium entry (SOCE). Approach and Results—: SOCE is a major mechanism for calcium influx controlled by the intracellular Ca ²⁺ stores. On store depletion, the endoplasmic reticulum calcium sensor STIM1 (stromal interaction molecule 1) redistributes into puncta at endoplasmic reticulum/plasma membrane junctions, a process supported by the cytoskeleton, where it interacts with the calcium channels; however, the mechanism for fine-tuning SOCE is not completely understood. Our results demonstrate that STIM1 interacts with FLNA on calcium store depletion in human platelets. The interaction is dependent on the phosphorylation of FLNA at Ser ²¹⁵² by the cAMP-dependent protein kinase. Impairment of FLNA phosphorylation and knockdown of FLNA expression using siRNA increased SOCE in platelets. Similarly, SOCE was significantly greater in FLNA-deficient melanoma M2 cells than in the FLNA-expressing M2 subclone A7. Expression of FLNA in M2 cells attenuated SOCE, an effect prevented when the cells were transfected with the nonphosphorylatable FLNA S2152A mutant. Transfection of M2 cells with the STIM1(K684,685E) mutant reduced the STIM1–FLNA interaction. In platelets, attenuation of FLNA expression using siRNA resulted in enhanced association of STIM1 with the cytoskeleton, greater STIM1–Orai1 interaction, and SOCE. Introduction of an anti-FLNA (2597–2647) antibody attenuated the STIM1–FLNA interaction and enhanced thrombin-induced platelet aggregation. Conclusions—: Our results indicate that FLNA modulates SOCE and then the correct platelet function, by fine-tuning the distribution of STIM1 in the cytoskeleton and the interaction with Orai1 channels.

The role of the sarco-endoplasmic reticulum calcium (Ca²⁺)ATPase (SERCA)3 in platelet physiology remains poorly understood. Here, we show that SERCA3^-/- mice exhibit prolonged tail bleeding time and re-bleeding. Thrombus formation was delayed both in arteries and venules in an in vivo ferric chloride-induced thrombosis model. Defective platelet adhesion and thrombus growth over collagen was confirmed in vitro. ADP removal by apyrase diminished adhesion and thrombus growth of control platelets to the level of SERCA3^-/- platelets. Aggregation, dense granule secretion and Ca²⁺ mobilization of SERCA3^-/- platelets induced by low collagen or low thrombin concentration were weaker than controls. Accordingly, SERCA3^-/- platelets exhibited a partial defect in total stored Ca²⁺, and in Ca²⁺ store re-uptake following thrombin stimulation. Importantly ADP but not serotonin, rescued aggregation, secretion and Ca²⁺ mobilization in SERCA3^-/- platelets, suggesting specificity. Dense granules appeared normal upon electron microscopy, mepacrine staining, and total serotonin content, ruling out a dense granule defect. ADP induced normal platelet aggregation, excluding a defect in ADP activation pathways. The SERCA3-specific inhibitor tBHQ diminished both Ca²⁺ mobilization and secretion of control platelets, as opposed to the SERCA2b inhibitor thapsigargin. This confirmed the specific role of catalytically active SERCA3 in ADP secretion. Accordingly, SERCA3-dependent Ca²⁺ stores appeared depleted in SERCA3^-/- platelets. Finally α_IIbβ₃ integrin blockade did not affect SERCA3-dependent secretion, therefore proving independent of α_IIbβ₃ engagement. Altogether these results show that SERCA3-dependent Ca²⁺ stores control a specific ADP secretion pathway required for full platelet secretion induced by agonists at low concentration, and independent of α_IIbβ₃.

Calcium ions (Ca(2+)) are versatile messengers that need to be tidily regulated in time and space in order to create a large number of signals. The coupling between Ca(2+) entry and Ca(2+) refilling is playing a central role in this Ca(2+) homeostasis. Since the capacitative Ca(2+) entry has been described, different mechanisms have been proposed in order to explain how the Ca(2+) entry could be under control of intracellular store Ca(2+) depletion. Today, in addition of STIM1 and Orai1, the two major elements of SOCe, increasing attention is put on the role of the transient receptor potential canonical (TRPC), that can form protein clusters with Orai1, and Sarco/endoplasmic reticulum Ca(2+)ATPases (SERCAs), that refill the stores and are also located in the same environment than SOC clusters. Altogether, these proteins elaborate either Ca(2+) microdomains in the vicinity of the membrane or larger Ca(2+) increases overtaking the whole cell. The coupling between Ca(2+) entry and Ca(2+) refilling can possibly act much further away from the plasma membrane. Ca(2+), uptaken by SERCAs, have been described to move faster and further in the ER than in the cytosol and to create specific signal that depends on Ca(2+) entry but at longer distance from it. The complexity of such created Ca(2+) currents resides in the heteromeric nature of channels as well as the presence of different intracellular stores controlled by SERCA2b and SERCA3, respectively. A role for mitochondria has also been explored. To date, mitochondria are other crucial compartments that play an important role in Ca(2+) homeostasis. Although mitochondria mostly interact with intracellular stores, coupling of Ca(2+) entry and mitochondria cannot be completely rule out.

Apelin peptide and its receptor APJ are directly implicated in various physiological processes ranging from cardiovascular homeostasis to immune signaling. Here, we show that apelin is a key player in hemostasis with an ability to inhibit thrombin- and collagen-mediated platelet activation. Mice lacking apelin displayed a shorter bleeding time and a prothrombotic profile. Their platelets exhibited increased adhesion and a reduced occlusion time in venules, and displayed a higher aggregation rate after their activation by thrombin compared with wild-type platelets. Consequently, human and mouse platelets express apelin and its receptor APJ. Apelin directly interferes with thrombin-mediated signaling pathways and platelet activation, secretion, and aggregation, but not with ADP and thromboxane A2-mediated pathways. IV apelin administration induced excessive bleeding and prevented thrombosis in mice. Taken together, these findings suggest that apelin and/or APJ agonists could potentially be useful adducts in antiplatelet therapies and may provide a promising perspective for patients who continue to display adverse thrombotic events with current antiplatelet therapies.