| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Released from Activated Platelets Can Induce TNF-Stimulated Human Brain Endothelium Apoptosis: A New Mechanism for Microvascular Lesion during Cerebral Malaria1
* Centre National de la Recherche Scientifique, Unité Mixte de Recherche (UMR) 6020, Faculty of Medicine, Institut Fédératif de Recherches (IFR) 48, Université de la Méditerranée, Marseille, France;
Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom;
Laboratoire d’Hématologie, Hémostase, Fibrinolyse et Pathologie Vasculaire, Institut National de la Santé et de la Recherche Médicale, UMR 626, IFR 125, Faculty of Medicine, Université de la Méditerranée, Marseille, France;
Centers for Disease Control and Prevention, National Center for Infectious Diseases, Atlanta, GA 30333; and
¶ Department of Pathology, K25, Faculty of Medicine, University of Sydney, Sydney, Australia.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
1 was the main molecule involved in endothelial cell death, because its inhibition completely abrogated the activated-platelet supernatant effect. Our data illustrate another aspect of the duality of TGF-1 in malaria and may provide new insights into the pathogenesis of cerebral malaria. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Microvascular clogging with platelets is well documented in experimental murine CM (1, 2), and similar findings have been reported in human postmortem brain tissue from fatal pediatric CM cases. Importantly, such lesions are seen only in patients who died from CM but not in those dying from severe anemia (3). Supporting evidence of the importance of platelets comes from recent in vitro coculture studies showing that platelets can modulate and reorient the specificity of PRBC cytoadherence to brain endothelium (4). Platelets may also have proapoptotic effects on TNF-activated human brain microvascular endothelial cells (HBEC-5i)4 in vitro (25).
Inflammatory mediators such as TNF are key to the pathogenesis of infectious diseases (5), and fatal CM is associated with high circulating levels of this cytokine (6, 7), although other inflammatory mediators are also involved (8). Besides its well-documented effects on up-regulation of endothelial adhesion molecules, it is also evident that TNF in vitro acts as a potent inducer of endothelial cell apoptosis (9). This event may also be influenced by TGF-1 (10). TGF-1 is a pleiotropic cytokine expressed in a variety of tissues and stored in high amounts within 
granules of platelets (11). Regulation of TGF- production is crucial in determining the outcome of infection in mice (12); reduced levels promote inflammation and parasite clearance during early infection, whereas increased levels during the later phase of infection are associated with reduced pathology and disease control (12, 13). In humans the absolute ratios of proinflammatory and anti-inflammatory cytokines are important determinants of susceptibility to severe disease (14), and decreased levels of TGF- are associated with severity (15).
Recent studies have suggested that TNF and TGF- may act synergistically to enhance apoptosis of HUVEC (16). Therefore, in view of the prevailing conditions of high levels of TNF and elevated platelet numbers in the brain, we hypothesized that activated platelet-derived TGF- causes apoptosis of TNF-stimulated endothelial cells. Our data suggest that this is a plausible mechanism of microvascular damage during human CM.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Primary HBEC (HBEC-5i) were derived by Dorovini-Zis et al. (17) from small fragments of human cerebral cortex obtained from patients who had died of various causes. These brains were devoid of any pathologic abnormalities, as described, and the isolation and purification procedures were conducted according to the method developed by Bowman et al. (18), with minor modifications. These cells were then immortalized (17) and characterized in our laboratory, as described elsewhere (25). Briefly, they express stable patterns of endothelial cell markers such as VE-cadherin, von Willebrand factor VIII, and peripheral occludin, in addition to CD54, CD40, and CSA, and they show an up-regulation of CD54 and CD106 upon TNF activation. Moreover, HBEC-5i exhibit major features of cerebral endothelial cells, especially efficient tight-junction structures, as assessed by high trans endothelial electric resistance and very low permeability to 70-kDa dextran (25). HBEC-5i were cultured on 0.1% gelatin-coated 24-well culture plates (BD Falcon, BD Biosciences) and grown to confluence in DMEM:Ham’s F-12 (pH 7.4), supplemented with 10% FCS, 30 µg/ml endothelial cell growth supplement (E 0760; Sigma-Aldrich), and 10 µg/ml gentamicin.
Platelet purification
Blood was collected from healthy volunteers into vacutainers containing 0.129 M buffered sodium citrate as anticoagulant (ratio 1:9). The volunteers had not taken any drugs for at least 14 days. Platelets were isolated by centrifugation at 150 x g for 10 min. Platelet-rich plasma was collected and the platelets were pelleted by centrifugation for 10 min at 300 x g. The supernatant was discarded and the platelets were washed three times in Tyrode’s buffer (0.137 M NaCl, 3 mM KCl, 0.4 M NaH2PO4, 12 mM NaHCO3, 1 mM MgCl2, 14.7 mM HEPES, and 20 mM glucose (pH 6.9)) supplemented with 0.2% BSA. To avoid platelet activation, all centrifugations were performed at 37°C. Platelets were then resuspended in Tyrode’s buffer, and the density of the suspension was adjusted to 2 x 109 platelets/ml.
Platelet stimulation
Washed platelet suspensions of 1 ml containing 2 x 109 platelets were stimulated for 10 min with thrombin (0.1 U/ml) at 22°C without stirring, or with RBC parasitized with mature stages of P. falciparum FCR3 (Palo-Alto) strain for 10 min under gentle agitation. In the second case, platelet suspensions were mixed with an erythrocyte pellet containing 2.5 x 105, 5 x 105, or 1 x 106 PRBC/ml. For both conditions, after centrifugation (13,000 x g, 2 min at 20°C), the supernatant was filtered (0.22-µm filters; Millipore) and immediately added to endothelial cells with or without anti-human TGF-1 Ab (at a final concentration of 10 µg/ml) according to the experiment design, or stored at –80°C before use. Supernatants obtained from resting platelets and from platelets incubated with normal RBC (1 x 106 PRBC/ml) were also processed as described and used as controls.
Monoclonal Abs and other reagents
Surface CD40 and CD54 (ICAM-1) receptors blocked with specific mouse anti-human mAbs were purchased from Diaclone (clone B-B20) and Beckman Coulter Immunotech (clone 84H10), respectively. The platelet CD61/CD41 (GPIIbIIIa) was inhibited by a chimeric F(ab')2 receptor antagonist, abciximab (ReoPro; Centocor). A nonspecific mouse IgG1 (Beckman Coulter Immunotech) was used as control with the same concentration as blocking Abs. Mouse anti-human TGF-1 Ab (R&D Systems Europe) was used to inhibit the action of TGF-1 released from platelets upon thrombin (Diagnostica Stago) activation. Calcein-acetoxymethyl ester was purchased from Molecular Probes. Human rTNF- (Tebu-Bio) was used to stimulate endothelial cells to mimic inflammatory conditions. Paclitaxel (Taxol; Sigma-Aldrich) and human rTGF-1 (R&D Systems Europe) were used for HBEC-5i apoptosis positive control. Cells were fixed with paraformaldehyde (Sigma-Aldrich).
Platelet-endothelial cells coincubation
HBEC-5i were seeded (40,000–50,000 cells/well) in 24-well tissue culture plates, and medium was changed every day until confluence was reached. They were then either left unstimulated or incubated with TNF (10 ng/ml). Cells were washed in PBS before we added increasing platelet numbers with 0.5 ml of several suspensions ranging from 2.5 x 108 to 2 x 109 platelets/ml (ratio 300:1 to 2400:1, respectively). We incubated the cell mixture for 90 min at 37°C, and HBEC-5i were then washed vigorously to remove bound and unbound platelets. New medium was added and HBEC-5i were then incubated for 48 h before apoptosis assays.
Platelet adhesion assays
HBEC-5i were grown to confluence on 24-well culture plates and stimulated as described above. Purified platelets were labeled with the fluorescent dye calcein-AM (1 µM) at 37°C for 60 min, washed twice with Tyrode’s buffer, and incubated for another 30 min at 37°C to discard non-de-esterified dye. The labeled platelets were washed twice again and added simultaneously with or without several blocking molecules (0.5 ml of a 2.5 x 108 platelets/ml suspension per well). mAbs against CD40 and CD54 (ICAM-1) were used at the final concentration of 50 µg/ml, and CD61/CD41 (GPIIbIIIa) receptor antagonist was used in a final concentration that maximally inhibits platelet aggregation, i.e., 6.5 µg/ml. Cells were cocultured 90 min at 37°C before the plate was washed three times in PBS to remove unbound platelets, and then residual fluorescence intensity was determined with a fluorometer (FL600; Bio-Tek Instruments), using 480 and 530 nm as the excitation and emission wavelengths, respectively. To quantify the platelet adhesion, a calibration curve was performed with labeled platelets. Results are expressed as percentage of bound platelets.
HBEC-5i-platelet supernatants coincubation
HBEC-5i were cultured and stimulated with TNF as described. Paclitaxel (Taxol; 100 nM) was used as a positive control for apoptosis induction and was added overnight before analyses. HBEC-5i were then incubated for 90 min at 37°C with recombinant human TGF-1. The concentration (40 pM) was selected on the basis of a dose-response experiment (data not shown). Other conditions included resting or thrombin-activated platelet supernatants in the presence or absence of anti-TGF-1 (10 µg/ml). Thrombin (0.1 U/ml) alone was used as a control. Cell monolayers were finally washed with PBS, new culture medium was added, and cells were incubated for 48 h before the HBEC-5i apoptosis quantitation.
HBEC-5i apoptosis measurement
Cells were harvested, fixed, and stained according to the procedure of the APO-Direct kit (BD Pharmingen). Briefly, HBEC were fixed first in PBS-1% (w/v) paraformaldehyde, then in 75% (v/v) ethanol before a staining step performed with FITC-coupled dUTP, to label DNA breaks of apoptotic cells. Cells were then analyzed by flow cytometry.
Statistical analysis
Statistical analysis was performed with Prism 4.0 from GraphPad. Data were analyzed by Mann-Whitney U test to compare pairs of groups. Results were expressed as means ± SD of individual experimental groups. A value of p < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The effect of increasing numbers of resting platelets on the viability of resting or TNF-stimulated HBEC-5i was analyzed by quantifying DNA fragmentation by flow cytometry. Cell viability was measured 6, 24, and 48 h after cocultures; 48 h was selected as this period led to substantial and reproducible degrees of apoptosis. In these conditions, when HBEC-5i were left unstimulated, the rate of apoptosis was not significantly modified in the presence of platelets, irrespective of the numbers of platelets added (Fig. 1). In contrast, in TNF-stimulated HBEC-5i, increasing numbers of platelets significantly potentiated death by apoptosis, reaching up to 34.2 ± 2.3%. Thus, even resting platelets can dramatically enhance HBEC-5i killing, provided they have been prestimulated by TNF.
|
We then analyzed the importance of platelet-endothelium binding in the induction of apoptosis of HBEC-5i cells. After cocultures of resting or TNF-stimulated HBEC-5i with platelets or platelets with receptor antagonists, we measured platelet binding and its effects on endothelial cell viability. The numbers of bound platelets and the associated endothelial cell death by apoptosis were significantly higher when HBEC-5i were previously stimulated by TNF than in resting conditions (Fig. 2). For ensuing experiments a platelet:endothelial cell ratio of 300:1 was selected. Addition of the CD41/61 antagonist dramatically inhibited platelet binding to HBEC-5i and significantly reduced the endothelial apoptosis. Platelet binding and endothelial killing were also significantly but partially reduced in the presence of blocking mAbs against ICAM-1 or CD40, suggesting a physical effect of platelets on apoptosis enhancement. The concomitant addition of these inhibitors also led to the inhibition of platelet binding and the associated elevated death rate, but these effects were not significantly more pronounced than with anti-CD41/61 alone.
|
1 released by activated platelets kills TNF-stimulated HBEC-5i
We investigated the possible role of TGF-1 in the platelet-induced HBEC-5i apoptosis. Cell death by apoptosis was measured 6, 24, and 48 h after cocultures of HBEC-5i with Tyrode’s buffer, human rTGF-1, resting platelets, or thrombin-activated platelet supernatant. Drastic and significant results were only obtained after a 48-h incubation (data not shown). First, there was no modification of viability in resting HBEC-5i, in all conditions tested, confirming that TNF is a prerequisite for the platelet-induced endothelium apoptosis (Fig. 3A). Second, the most drastic proapoptotic effect was obtained when stimulated HBEC-5i were cocultured with the supernatant obtained from thrombin-activated platelets. This culture condition led to a dramatic increase of the HBEC-5i apoptotic rate, from 9.81 ± 0.51% to 20.1 ± 1.99%. Human rTGF-1 caused a similar level of endothelial apoptosis (17.8 ± 1.2%). Interestingly, HBEC-5i incubated in the presence of thrombin for the same time did not show any viability alteration, ruling out an involvement of thrombin in this supernatant effect. Third, when incubated simultaneously with thrombin-activated supernatant and anti-human TGF-1 blocking mAb, the apoptosis in HBEC-5i was dramatically reduced and brought back to the level seen with resting platelet supernatants. Finally, the HBEC-5i apoptotic rate obtained with supernatant of thrombin-activated platelets was as high as that obtained when activated HBEC-5i were incubated with supernatants obtained from platelets activated by PRBC contact (18.9 ± 2.4%; Fig. 3B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
.
Our studies are consistent with a sequence of events leading up to endothelial cell damage. The initial steps involve activation of both endothelial cells by TNF and activation of platelets by parasite-derived products. Conceivably, both reactions may occur together during severe P. falciparum infection. Parasite-derived products released at schizogony would not only induce TNF production but, additionally, they would act as triggers for platelet activation. The binding of activated platelets to TNF-primed endothelial cells would lead to platelet adhesion with release of TGF- and subsequent induction of apoptosis of endothelial cells. Moreover, resting platelets can also bind to activated endothelium, and promote the adherence of PRBC on endothelial cells (4). This contact between PRBC and platelets might result in the activation of the latter cell type (19), with the same potential proapoptotic consequences on the endothelium. However, bound PRBC alone also have been reported to induce apoptosis of endothelial cells (20), and this cytotoxic effect, coupled to the phenomenon we describe, may result in a strong endothelial injury in vivo. Although anti-TGF- Ab prevented apoptosis, the effects of other platelet-derived endothelial cell-damaging molecules such as the eicosanoids (HETE, thromboxane A2, etc.) cannot be excluded (2, 21). Compared with resting cells, TNF-stimulated HBEC-5i exhibited a higher susceptibility to killing in response to TGF-1, confirming previous results of Emmanuel et al. (16).
The anti-inflammatory properties of TGF-1 in malaria are well documented (reviewed in Refs.8 and 22). In mice, TGF- in regulated quantities promotes protective immune responses, with slower parasite growth during early infection; increased quantities appear to down-regulate pathology during late infection. Indeed, spleen TGF- mRNA (12), circulating levels of TGF-1, and production of bioactive TGF-1 by splenocytes (13) were found to be low during lethal infections with Plasmodium berghei ANKA. In contrast, resolving infections with the nonlethal parasites Plasmodium chabaudi chabaudi and Plasmodium yoelii were correlated with a significant TGF-1 production. Similar results were obtained in studies of human malaria in Ghana, which showed an association between cytokine production and a significantly reduced risk of fever (13).
However, circulating levels of the cytokine reach up to 1 pg/ml during human malaria infection, whereas in in vitro assays >40 pg/ml are necessary to induce apoptosis. In vivo, apoptosis induction is likely to involve a massive local release of TGF-1 at the site of platelet sequestration, which is consistent with the high numbers of platelets accumulated within brain microvasculature in CM patients (3). Moreover, malaria parasites also are able to directly activate latent TGF- (13, 23), possibly aggravating the apoptosis of brain microvasculature in vivo.
Despite these anti-inflammatory properties in malaria infection, we showed in the present study that platelet-derived TGF-1 may synergize with TNF to induce the microvascular lesion of CM. This massive, local, and pathogenic release of the cytokine must be dissociated from the T cell/macrophage-derived TGF-, which can be induced by Plasmodium-derived products to create an anti-inflammatory environment (24). Indeed, it is conceivable that there are distinct effects between systemic and local TGF concentrations. Decreased tissue (12) and circulating (13) levels of TGF- are associated with CM severity both in mice and in humans, whereas increased local production of the cytokine by high numbers of sequestered platelets in contact with PRBC and brain endothelium (3) may lead to damage of the latter cells. Thus, local TGF-1 release in brain microvessels may be responsible for endothelial cell apoptosis, permeability changes, cerebral edema, and hemorrhages that characterize CM pathology. Our data illustrate the effects of cytokine interaction and underline an important aspect of the duality of TGF-1 in malaria, possibly providing new insights into CM pathogenesis.
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was funded by grants from the PAL+ program 2000 and program 2002 from the French Ministry of Research and Technology (to G.E.G.). S.C.W. was a fellow supported by a grant from the PAL+ program from the French Ministry of Research and Technology (to G.E.G.) and by the Fondation Recherche Médicale (Paris, France). 
2 Current address: Malawi-Liverpool Wellcome Trust Clinical Research Programme, Chichiri, Bantyr, Malawi.
3 Address correspondence and reprint requests to Dr. Georges E. Grau, Department of Pathology, K25, Faculty of Medicine, University of Sydney, Sydney, New South Wales 2402, Australia. E-mail address: g.grau{at}med.usyd.edu.au or georges.grau{at}medecine.univ-mrs.fr
4 Abbreviations used in this paper: PRBC, Plasmodium falciparum-parasitized RBC; HBEC, human brain microvascular endothelial cell; CM, cerebral malaria.
Received for publication August 5, 2005. Accepted for publication October 25, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
1 in endothelial cells versus smooth muscle cells. Circulation 99: 2019-2026. in human platelets: identification of a major storage site, purification, and characterization. J. Biol. Chem. 258: 7155-7160. production is inversely correlated with severity of murine malaria infection. J. Exp. Med. 188: 39-48. 1, interleukin-12 and increased migration inhibitory factor are associated with severe malaria. Acta Trop. 89: 319-327. [Medline] and transforming growth factor-. Cytokine 18: 237-241. [Medline]?. Parasitol. Today 16: 18-23. [Medline] by malaria parasite-derived metalloproteinases and a thrombospondin-like molecule. J. Exp. Med. 198: 1817-1827. regulates proinflammatory cytokine production and determines the outcome of lethal and nonlethal Plasmodium yoelii infections. J. Immunol. 171: 5430-5436. Related articles in The JI:
This article has been cited by other articles:
|
|
 |
|
  B. J. McMorran, V. M. Marshall, C. de Graaf, K. E. Drysdale, M. Shabbar, G. K. Smyth, J. E. Corbin, W. S. Alexander, and S. J. Foote Platelets Kill Intraerythrocytic Malarial Parasites and Mediate Survival to Infection Science, February 6, 2009; 323(5915): 797 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-F. Penet, M. Abou-Hamdan, N. Coltel, E. Cornille, G. E. Grau, M. de Reggi, and B. Gharib Protection against cerebral malaria by the low-molecular-weight thiol pantethine PNAS, January 29, 2008; 105(4): 1321 - 1326. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Gillrie, G. Krishnegowda, K. Lee, A. G. Buret, S. M. Robbins, S. Looareesuwan, D. C. Gowda, and M. Ho Src-family kinase dependent disruption of endothelial barrier function by Plasmodium falciparum merozoite proteins Blood, November 1, 2007; 110(9): 3426 - 3435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. F. Langer, K. Daub, G. Braun, T. Schonberger, A. E. May, M. Schaller, G. M. Stein, K. Stellos, A. Bueltmann, D. Siegel-Axel, et al. Platelets Recruit Human Dendritic Cells Via Mac-1/JAM-C Interaction and Modulate Dendritic Cell Function In Vitro Arterioscler Thromb Vasc Biol, June 1, 2007; 27(6): 1463 - 1470. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Basuroy, S. Bhattacharya, D. Tcheranova, Y. Qu, R. F. Regan, C. W. Leffler, and H. Parfenova HO-2 provides endogenous protection against oxidative stress and apoptosis caused by TNF-{alpha} in cerebral vascular endothelial cells Am J Physiol Cell Physiol, November 1, 2006; 291(5): C897 - C908. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bobik Transforming Growth Factor-{beta}s and Vascular Disorders Arterioscler Thromb Vasc Biol, August 1, 2006; 26(8): 1712 - 1720. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), normal RBC (NRBC), PRBC, or thrombin. Results are expressed in percentage of apoptotic endothelial cells, and bars represent SDs of four experiments (Mann-Whitney U test).