Upregulation of Proinflammatory Proteins Through NF-κB Pathway by Shed Membrane Microparticles Results in Vascular Hyporeactivity
Objective— Microparticles are membrane vesicles with procoagulant and proinflammatory properties released during cell activation, including apoptosis. The present study was designed in dissecting the effects evoked by microparticles on vascular reactivity.
Methods and Results— Microparticles from either apoptotic T lymphocytic cells or from plasma of diabetic patients with vascular complications induced vascular hyporeactivity in response to vasoconstrictor agents in mouse aorta. Hyporeactivity was reversed by nitric oxide (NO) synthase plus cyclooxygenase-2 inhibitors, and associated with an increased production of vasodilatory products such as NO and prostacyclin. Microparticles induced an upregulation of proinflammatory protein expressions, inducible NO-synthase and cyclooxygenase-2, mainly in the medial layer of the vessels as evidenced by immunochemical staining. In addition, microparticles evoke NF-κB activation probably through the interaction with the Fas/Fas Ligand pathway. Finally, in vivo treatment of mice with lymphocyte-derived MPs induces vascular hyporeactivity, which was reversed by the combination of NO and cyclooxygenase-2 inhibitors.
Conclusion— These data provide a rationale to explain the paracrine role of microparticles as vectors of transcellular exchange of message in promoting vascular dysfunction during inflammatory diseases.
Recent data suggest that inflammation play a central role in the origin and complications of cardiovascular disease. For instance, inflammation has a pivotal role in the development of atherosclerosis and the acute activation of the vascular wall with consecutive local thrombosis and altered vasoactivity. This process is orchestrated by the interactions between inflammatory cells, such as leukocytes and vascular cells, endothelial cells, and smooth muscle cells, which under activation or apoptosis, for example, lead to the release of circulating microparticles (MPs). MPs are vesicles shed from the blebbing plasma membrane of various cell types, such as platelets, T and B cells, monocytes, and endothelial cells during activation by agonists, shear stress,1,2 or apoptosis.3,4 MPs harbor cell surface proteins and contain cytoplasmic components of the original cell. They exhibit negatively charged phospholipids, chiefly phosphatidylserine, at their surface accounting for their procoagulant character1 and proinflammatory properties.5
Elevated levels of circulating MPs have been detected under pathological states, such as atherosclerosis, acute coronary syndrome, diabetes, preeclampsia, and sepsis.4,6–9 Furthermore, these pathologies are associated with vascular dysfunction including attenuation of endothelium-dependent vasodilatation, alteration of responsiveness of vascular smooth muscle to vasoconstrictor stimuli, or both. However, the mechanisms triggering the modifications of the vessel contraction/relaxation balance in inflammatory diseases are not fully elucidated. Also, the role for MPs in the regulation of vasoactivity is not known even though MPs from leukocytes were shown to act on endothelium as competent inflammatory agonists.10 These MPs are able to stimulate expression of inflammatory genes responsible for the release of cytokines and to promote upregulation of leukocyte-endothelial cell adhesion molecules.10
Recently, we have shown that MPs from apoptotic T cells affect endothelial function by impairment of nitric oxide (NO) pathway reflected by the decrease of endothelial NO synthase (eNOS) and increase of caveolin-1 expressions.11 Independently of the effect of MPs on endothelial function, now, we propose to determine the effects of shed MPs from T cells on vascular smooth muscle reactivity. It is interesting to note that under different pathological conditions, circulating levels of MPs from lymphocytes are enhanced. In diabetic patients the amount of MPs from leukocyte origin is 3-fold higher than in healthy donors.7 In addition, HIV-infected patients show elevated levels of MPs bearing CD4 antigens.3 In preeclampsia, elevated levels of MPs from granulocytes and lymphocytes have been reported.8 We have previously observed that MPs captured from blood samples from HIV-infected individuals have properties comparable to those of their counterparts from cultured apoptotic T CEM cells and from circulating plasma of diabetic patients.11 Hence, the use of MPs from T cells appears relevant in exploring their effects on vasoactivity.
In the present study, we provide evidence that MPs from either apoptotic T lymphocytes or from diabetic patients with vascular complications promote vascular hyporeactivity by inducing an increased production of vasodilatory products such as NO and prostacyclin. These MPs also induced upregulation of proinflammatory protein expression, the inducible NO-synthase (iNOS) and COX-2, mainly in the medial layer of vessels through NF-κB–dependent transcription via Fas/FasL pathway. Interestingly, in vivo treatment of mice with lymphocyte-derived MPs induces vascular hyporeactivity, suggesting their pathophysiological relevance during inflammatory diseases.
For a detailed Methods section, please see http://atvb.ahajournals.org.
MPs From T Cell Line Induce Vascular Hyporeactivity to Vasoconstrictor Agonists
5-HT produced a concentration-dependent increase in tension in aortic rings with or without functional endothelium. Incubation of mice aortic rings with 30 nmol/L PS eq MPs for 24-hour decreased vascular reactivity to the agonist both in endothelium-intact and endothelium-denuded vessels (Figure 1A and 1B). MP treatment-induced vascular hyporeactivity may result from the release of vasodilatory products from different cellular origin (ie, endothelial, smooth muscle, or fibroblast). However, the fact that 5-HT–induced contractions were also reduced in vessels without functional endothelium, strongly suggests that vasodilatory products from non-endothelial origin are implicated in the hyporeactivity produced by MP treatment. Therefore, the following experiments were performed in vessels without functional endothelium.
To investigate the role of NO and COX metabolites, the effect of the NO-synthase inhibitor, L-NA, the selective inhibitor of COX-2, NS-398, alone or in combination, were studied on the response to 5-HT. L-NA, NS-398, nor L-NA plus NS-398 had a significant effect on contractile responses to 5-HT in control arteries (Figure 1C and 1E). In aorta incubated with MPs, L-NA alone or NS-398 alone increased the contractile response to the agonist (Figure 1D). In vessels treated with MPs, the combination of the 2 inhibitors produced further potentiation of contraction when compared with that observed in the presence of either L-NA or NS-398 alone (Figure 1F) and restored the contractile response to 5-HT toward that of aortic rings not exposed to MPs. Similar data were obtained with the nonselective COX inhibitor, indomethacin (not shown). Also, MPs induced hyporeactivity in response to another agonist phenylephrine that was reversed in the combination of L-NA and NS-398 (not shown).
MPs Generated From T Cell Line Stimulate NO and 6-keto-PGF1α Production
Direct in situ measurements of NO production were performed by EPR spectroscopy using Fe(DETC)2 as a spin trap. Both control and MP-treated aortae with functional endothelium, preincubated with Fe(DETC)2, exhibited an EPR feature of signals derived from NO-Fe(DETC)2. The quantitative measurement of the NO-Fe(DETC)2 signal amplitude was reported to the relative units for weight in mg of the dried sample (Wds). The NO-Fe(DETC)2 EPR signal was greater in aortae treated with MPs for 24 hours (429±80 A/Wds) compared with nontreated vessels (179±18 A/Wds), (*P<0.05, n=8).
Assay of 6-keto-PGF1α, the stable product of PGI2, showed increased production in aortae treated with MPs for 24 hours (3907±625 pmol/mg Wds) compared with nontreated vessels (1875±390 pmol/mg Wds), (**P<0.01, n=6).
MPs From T Cell Line Induce iNOS and COX-2 Expression
Immunohistochemical detection of iNOS and COX-2 was conducted by confocal microscopy. Whereas weak or no staining of either iNOS or COX-2 was found in the vessel wall of control aorta (Figure 2A and 2H), marked iNOS and COX-2 labeling were observed in the medial layer of aorta treated with MPs (Figure 2B and 2I). The negative control obtained by incubation with the secondary murine fluorescence-labeled antibody did not display any staining (Figure 2C and 2J). Western blot analysis showed an increase of both iNOS and COX-2 expression in aorta treated with MPs compared with control vessels (Figure 2G and 2N). The ability of MPs from T CEM cells to interact with human smooth muscle cells (HSMCs) has been studied by their capacity in promoting upregulation of iNOS and COX-2 expression. As shown in Figure 2O to 2R, HSMCs treated with T cell MPs exhibit a marked labeling of iNOS and COX-2 when compared with controls (Figure 2P and 2R versus 2O and 2Q, respectively).
Shed MPs From T Cell Line Activate NF-κB Detected by p65/Rel A and Phosphorylated IκB-Alpha Staining
Because enhanced expression of proinflammatory enzymes such as iNOS and COX-2 is under the control of the NF-κB/Rel family of transcription factors, the activation of the latter was assessed. NF-κB family members are heterodimers consisting of p65/RelA and p50/NF-κB1,14 but only the p65 subunit has transactivation domains capable of initiating transcription. Immunohistochemical studies showed marked aortic staining of p65/RelA subunit of NF-κB in the medial layer and weaker in the adventitial layer of aorta incubated with MPs (Figure 3B). No specific staining was found in control aorta (Figure 3A). Negative controls obtained by incubation with the secondary rabbit fluorescence-labeled antibody did not display any staining (Figure 3C).
As illustrated in Figure 3G to 3H, T cells MPs were able to activate NF-κB p65/Rel family in HSMC. Indeed, p65 translocates to the nucleus of cultured smooth muscle cells as shown by the merged picture after MPs stimulation (Figure 3H). These data were confirmed by the phosphorylation of IκB-alpha at Ser32, which is essential for release of active NF-κB p65/Rel A. Whereas weak or no staining was found in control cells, a marked labeling of phosphorylated IκB-alpha was found in HSMC treated with MPs (Figure 3I and 3J).
Fas/FasL Signaling Accounts for Vascular Hyporeactivity Induced by MPs From T Cell Line
The possibility that a direct interaction of membrane MPs with vascular cell (ie, smooth muscle) surface may account for activation of intracellular pathway was assessed. For this, the interaction between FasL, harbored by lymphocytic MPs, and Fas receptor expressed by vascular cells was considered using Fas receptor labeling by immunohistochemical studies. The vessel wall of control aorta that has not been in contact with MPs displayed a marked Fas receptor labeling in accordance with the expression level (Figure I, available online at http://atvb.ahajournals.org). In contrast, when the aorta was incubated for 24 hours with MPs, weak or no staining of Fas receptor was observed suggesting an interaction with MP-borne FasL, significantly blunting the labeling (Figure I). Negative controls obtained by incubation with streptavidin–phycoerythrin did not display any staining demonstrating the specificity of the labeling.
In addition and as shown in Figure I, MPs induced iNOS and COX-2 expression in cultured smooth muscle cells. Moreover, in the presence of the antibody anti-FasL, the increase in iNOS and COX-2 expression was completely abolished (Figure I). These results indicate that interaction of Fas/FasL is required for induction of inflammatory enzymes (iNOS and COX-2) in MP-treated smooth muscle cells. In agreement with these observations, MP-induced hyporeactivity to the vasconstrictor agents such as phenylephrine was lost after pre-incubation in the presence of anti-FasL antibody (Figure I). The response to phenylephrine was not significantly different between control vessels and those corresponding to incubation with MPs in which the FasL/Fas receptor interaction has been previously neutralized by anti-FasL antibody.
In Vivo Treatment of Mice With Lymphocyte-Derived MPs Induce Vascular Hyporeactivity
To study the in vivo relevance of our study, we injected 30 nmol/L PS eq MPs generated from T cells or supernatant corresponding to the last MP washing medium in the tail vein of mice. Twenty-four hours after the injection, vessels were isolated from mice to study vascular reactivity. In endothelium intact preparations, the contractile response to 5-HT was lower in aorta taken from mice treated with MPs compared with the response obtained in vessels from mice treated with vehicle (Figure 4A). Furthermore MP-induced vascular hyporeactivity to 5-HT was reversed by the combination of L-NA and NS-398 (Figure 4A). Moreover, increased CD4 labeling was found in the media layer of aortas from MP-treated mice, whereas weak CD4 staining was detected in control vessels (Figure 4B through 4G).
In Vivo Circulating MPs and Lymphocyte-Derived MPs from Diabetic Patients Induce Vascular Hyporeactivity
To assess the pathophysiologic relevance of the present study, the effect of circulating MPs from diabetic patients that had typical symptoms, such as microvascular complications, was examined. As shown in Figure 5A, treatment of mouse aorta with circulating MPs from 5 diabetic patients significantly reduced the contractile response to phenylephrine, and reactivity was restored in the presence of anti-FasL antibody. Furthermore, MPs shed by the actinomycin D-challenged lymphocytes of the same diabetic patients were also able to induce vascular hyporeactivity (Figure 5B). These data confirm that MPs play a significant role in the development of vascular dysfunction under several pathophysiological situations such as cardiovascular disorders (ie, diabetes).
In the present study, MPs stemming from apoptotic T lymphocytes or circulating in diabetic patients with vascular complications promote vascular hyporeactivity by inducing the production of vasodilatory mediators such as NO and prostacyclin. This effect resulted in an upregulation of iNOS and COX-2 through NF-κB–dependent transcription via the Fas/FasL pathway. These data provide valuable information to explain the paracrine role of MPs as vectors of transcellular exchange of message in promoting vascular dysfunction during inflammatory diseases.
In general, studies of the impact of MPs on cell activation or function have been performed using MPs from stimulated platelets, because they represent the main source of circulating procoagulant MPs and play a role in hematopoiesis and cell activation.15,16 Elevated levels of circulating MPs of endothelial origin have also been detected in peripheral blood from patients with acute coronary syndrome, diabetes or lupus anticoagulant.6,17,18 However, MPs are not only released from platelets or endothelial cells, but also from various cell types, including T cells.3,7 Here, we have used 30 nmol/L PS eq MPs because this concentration is frequently observed in several pathologies like paroxysmal nocturnal hemoglobinuria, diabetes, or unstable angina.6,7,12 Thus, this concentration of MPs may well be reached in vivo, in the plaque vicinity, for instance.4
Several pathologies are associated with vascular dysfunction including attenuation of endothelium-dependent vasodilatation, alteration of responsiveness of vascular smooth muscle to vasoconstrictor stimuli, or both. For instance, in inflammatory disorders such as cirrhosis, portal hypertension,19 or sepsis20 both vascular hypo-responsiveness to vasoconstrictors and reduced endothelium-dependent relaxation have been reported. We have previously shown that the same type of MPs used in the present study is able to induce endothelial dysfunction.11
Here, MPs from apoptotic T cells and from plasma of diabetic patients impaired the contraction induced by agonists, and this hyporeactivity was observed even in vessels without functional endothelium. These results and those obtained on cultured HSMC indicate that MPs are able to act on the smooth muscle cells and induce the release of vasodilatory factors, or alternatively, MPs can evoke an alteration of the balance between relaxant and constricting factors in smooth muscle cells.
With regard to the vasodilatory factors, the inhibition of proinflammatory enzymes, iNOS or COX-2, reduced the hyporeactivity in MP-treated vessels. In addition, the combined inhibition of both enzymes restored the contractile response. Moreover, MPs stimulated the production and release of NO and prostacyclin, account for the hyporeactivity induced by MPs. This is corroborated by the expression of iNOS and COX-2 in the vessel wall. Interestingly, the observations by confocal microscopy show that the increases of iNOS and COX-2 expressions were mainly localized in the medial layer of aorta treated with MPs. This is partially in accordance with the upregulation reported by Barry16,21 in HUVECs, platelets, or monocytic cells subjected to platelet MPs, but in their study arachidonic acid was the key mediator for these effects.16 The difference of the vasodilatory product released by COX-2 might be related to the cell origin of MPs, the stimuli responsible for MP release or target cells. Concerning the effect of MPs on iNOS, to the best of our knowledge the present study shows for the first time upregulation of iNOS induced by MP treatment. Our previous results on endothelial dysfunction through alteration of eNOS11 and those obtained in the present study show that MPs from T cells can affect the expression of different types of NO-synthases to alter vascular function and this includes iNOS for its activating effect on the release of vasodilatory products from smooth muscle origin.
NF-κB activation is upstream of the synthesis of acute phase inflammatory mediators. Among the genes known to be positively regulated by NF-κB are iNOS and COX-2. One possible explanation is that both promoter regions of iNOS and COX-2 genes have NF-κB binding sequences, and, thus, the same pathophysiologic stimuli (MPs in this case) may turn on the expression of both genes simultaneously.17 Alternatively, NO produced through NF-κB–induced iNOS expression may affect COX-2 expression and/or activity as shown in other models.22 Nevertheless, here MPs from T cells were able to induce NF-κB activation in the vessel wall and in HSMC, as revealed staining of its p65 subunit and phosphorylation of IκB-alpha.
In vascular smooth muscle cells it has been shown that the activation of the Fas/FasL pathway results in the increased expression of a specific program of inflammatory genes.23,24 We have previously shown that MPs used in the present study expressed FasL.11 Here, MP-treated vessels displayed a lack of Fas labeling. It cannot be excluded that MPs induced downregulation of Fas expression at the surface of smooth muscle cells. However, the fact that after incubation of MPs with an anti-FasL antibody, the vessel reactivity was restored is in favor of an interaction of FasL from MPs with Fas from the vessel wall, leading to intracellular signaling. Concerning the possible pathway linked to MPs, it has been shown that MP-borne FasL, but not the soluble form, is cytotoxic through activation of NF-κB.25,26 These works are in accordance with the present study where MPs from apoptotic T cells activated NF-κB through the Fas/FasL pathway independently of its ability to induce smooth muscle cell apoptosis.
Finally, we show that in vivo treatment of mice with MPs generated from T cells induces similar results than those obtained in ex vivo incubation of mice vessels with MPs. In addition, CD4 staining was observed in vessels from in vivo-treated mice. Altogether, these results strongly suggest that circulating MPs are able to induce in vivo vascular hyporeactivity that is associated with an increase of CD4 expression in the media layer of the vessel wall. Although CD4 labeling could be related to an infiltration of in vivo monocytes-macrophages into the media layer of the vessel wall, one can advanced the hypothesis that MPs reach smooth muscle and mediate their effect through Fas/FasL interaction. The mechanism by which MPs cross the endothelial barrier and affect vascular smooth muscle function remains to be determined. Nevertheless, the present data demonstrate for the first time to our knowledge the in vivo effect of MPs and thus their pathophysiological relevance in terms of vascular hyporeactivity.
In summary, we propose that MPs interacting with smooth muscle cells through the Fas/FasL pathway evoke NF-κB activation, which in turn upregulates iNOS and COX-2 expression leading to the production and release of the vasodilatory factors NO and prostacyclin. Both NO and prostacyclin account for the MP-induced hyporeactivity. Hatano et al27 have shown that NO from endogenous iNOS, which is activated by NF-κB, provides partial protection from Fas-mediated inflammation. In this respect, vasodilatory factors released from smooth muscle cells may counteract the cytotoxic effects of Fas/FasL interaction. Our results indicate that MPs released from T cells can act as true mediators of an inflammatory signal. The hyporeactivity to vasoconstrictor agents observed on vessel treatment with MPs from diabetic patients emphasizes the role of MPs as vectors of transcellular exchange of message in promoting vascular dysfunction accompanying inflammatory diseases.
We thank Pr J.L. Pasquali for providing blood samples from diabetic patients and F. Zobairi for help in MP preparation.
- Received April 7, 2005.
- Accepted September 13, 2005.
Zwaal RF, Schroit AJ. Pathophysiologic implications of membrane phospholipid asymmetry in blood cells. Blood. 1997; 89: 1121–1132.
Holme PA, Orvim U, Hamers MJ, Solum NO, Brosstad FR, Barstad RM, Sakariassen KS. Shear-induced platelet activation and platelet microparticle formation at blood flow conditions as in arteries with a severe stenosis. Arterioscler Thromb Vasc Biol. 1997; 17: 646–653.
Mallat Z, Hugel B, Ohan J, Lesèche G, Freyssinet JM, Tedgui A. Shed membrane microparticles with procoagulant potential in human atherosclerotic plaques: a role for apoptosis in plaque thrombogenicity. Circulation. 1999; 99: 348–353.
Mallat Z, Benamer H, Hugel B, Benessiano J, Steg PG, Freyssinet JM, Tedgui A. Elevated levels of shed membrane microparticles with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes. Circulation. 2000; 101: 841–843.
Sabatier F, Darmon P, Hugel B, Combes V, Sanmarco M, Velut JG, Arnoux D, Charpiot P, Freyssinet JM, Oliver C, Sampol J. Dignat-George F. Type 1 and type 2 diabetic patients display different patterns of cellular microparticles. Diabetes. 2002; 51: 2840–2845.
Nieuwland R, Berckmans RJ, McGregor S, Boing AN, Romijn FP, Westerdorp RG, Hack CE, Sturk A. Cellular origin and procoagulant properties of microparticles in meningococcal sepsis. Blood. 2000; 95: 930–935.
Mesri M, Altieri DC. Endothelial cell activation by leukocyte microparticles. J Immunol. 1998; 161: 4382–4387.
Martin S, Tesse A, Hugel B, Martínez MC, Morel O, Freyssinet JM, Andriantsitohaina R. Shed membrane particles from T lymphocytes impair endothelial function and regulate endothelial protein expression. Circulation. 2004; 109: 1653–1659.
Hugel B, Socie G, Vu T, Toti F, Gluckman E, Freyssinet J. M, Scrobohaci ML. Elevated levels of circulating procoagulant microparticles in patients with paroxysmal nocturnal hemoglobinuria and aplastic anemia. Blood. 1999; 93: 3451–3456.
Mulsch A, Mordvintcev P, Bassenge E, Jung F, Clement B, Busse R. In vivo spin trapping of glyceryl trinitrate-derived nitric oxide in rabbit blood vessels and organs. Circulation. 1995; 92: 1876–1882.
De Martin R, Hoeth M, Hofer-Warbinek R, Schmid JA. The transcription factor NF-kappa B and the regulation of vascular cell function. Arterioscler Thromb Vasc Biol. 2000; 20: e83–e88.
Barry OP, Kazanietz MG, Pratico D, Fitzgerald GA. Arachidonic acid in platelet microparticles up-regulates cyclooxygenase-2-dependent prostaglandin formation via a protein kinase C/mitogen-activated protein kinase-dependent pathway. J Biol Chem. 1999; 274: 7545–7556.
Salvemini D, Misko TP, Masferrer JL, Seibert K, Curie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A. 1993; 90: 7240–7244.
Schaub FJ, Liles WC, Ferri N, Sayson K, Seifert RA, Bowen-Pope DF. Fas and Fas-associated death domain protein regulate monocyte chemoattractant protein-1 expression by human smooth muscle cells through caspase- and calpain-dependent release of interleukin-1alpha. Circ Res. 2003; 93: 515–522.
Jodo S, Xiao S, Hohlbaum AM, Strehlow D, Marshak-Rothstein A, Ju ST. CD95 (Fas) ligand-expressing vesicles display antibody-mediated, FcR-dependent enhancement of cytotoxicity. J Biol Chem. 2001; 276: 39938–39944.
Xiao S, Jodo S, Sung SS, Marshak-Rothstein A, Ju ST. A novel signaling mechanism for soluble CD95 ligand. Synergy with anti-CD95 monoclonal antibodies for apoptosis and NF-kappaB nuclear translocation. J Biol Chem. 2002; 277: 50907–50913.