Caroline Metais

Microvascular endothelial dysfunction and its mechanism in a rat model of subarachnoid hemorrhage

After subarachnoid hemorrhage (SAH), large cerebral arteries are prone to vasospasm. Using a rat model of SAH, we examined whether cortical microvessels demonstrate vasomotor changes that may make them prone to spasm and whether endothelial dysfunction may account for any observed changes. Two days after percutaneous catheterization into the cisterna magna, 0.3 mL of autologous blood was injected into the subarachnoid space. The brain tissue was harvested 20 min later, and microvessels were dissected from the parietal cortex. Vasomotor responses to the thromboxane analog U46619, the protein kinase C agonist phorbol acetate, endothelin-1, adenosine diphosphate, nitroprusside, and isoproterenol were examined in vitro in cerebral arterioles from the control, sham-operated, and SAH animals. Endothelial nitric oxide synthase (NOS3) messenger RNA and protein concentration was measured by northern and western blotting, respectively. Arterioles from the SAH animals demonstrated attenuated dilation to the endothelium-dependent dilator adenosine diphosphate and accentuated constriction to endothelin-1, while responses to the other agents tested were unchanged. NOS3 protein concentration was decreased, but NOS3 messenger RNA was increased after SAH. After SAH, cortical arterioles demonstrate endothelial dysfunction, which may be the basis for microvascular spasm. This is in part related to decreased NOS3, which occurs despite an increase in its transcription.

ANTI-C5A MONOCLONAL ANTIBODY REDUCES CARDIOPULMONARY BYPASS AND CARDIOPLEGIA-INDUCED CORONARY ENDOTHELIAL DYSFUNCTION

OBJECTIVE Because C5a induces tissue injury by activating polymorphonuclear leukocytes, the hypothesis was that inhibition of C5a activity would reduce cardioplegia-related injury. METHODS Pigs were placed on cardiopulmonary bypass. The hearts were arrested for 1 hour with hyperkalemic cardioplegia. Pigs were then separated from bypass, and the hearts were reperfused for 2 hours. Anti-porcine C5a monoclonal antibody (1.6 mg/kg, intravenously; n = 6) was administered 20 minutes before the onset of cardiopulmonary bypass. Six pigs received saline solution vehicle. Reactivity of coronary arterioles was studied in vitro with videomicroscopy. Microvessels from uninstrumented pigs served as controls for vascular studies. RESULTS Endothelium-dependent relaxation to adenosine diphosphate (percent relaxation of precontraction) was reduced after cardioplegic reperfusion (63% +/- 14% vs 77% +/- 10% in control at 10 micromol/L; P =.01). This impairment in endothelium-dependent relaxation was improved with anti-porcine C5a monoclonal antibody (80% +/- 22%; P =.01 vs saline solution), as was the impaired endothelium-dependent relaxation to clonidine (64% +/- 12% control; 26% +/- 17% saline solution; 55% +/- 24% anti-porcine C5a monoclonal antibody at 10 micromol/L; P =.01 saline solution vs control or anti-porcine C5a monoclonal antibody). Myeloperoxidase activity was significantly decreased (0.2 +/- 0.2 units/g protein; P =.04) in the anti-porcine C5a monoclonal antibody group compared with 5.2 +/- 2.7 in the saline solution group. CH50 2 hours after bypass was not statistically different (0.57 +/- 0.41 unit and 0.65 +/- 0.41 unit, respectively) between the anti-porcine C5a monoclonal antibody and saline solution groups. Despite less myocardial polymorphonuclear leukocyte infiltration after C5a inhibition, maximum rate of rise of left ventricular pressure, percent segmental shortening, and blood flow through the left anterior descending coronary artery were similar in the anti-porcine C5a monoclonal antibody and saline solution groups. CONCLUSIONS Inhibition of C5a limits neutrophil-mediated impairment of endothelium-dependent relaxation after cardiopulmonary bypass and cardioplegic reperfusion, but it has no effect on short-term myocardial functional preservation.

Mesenteric Dysfunction After Cardiopulmonary Bypass: Role of Complement C5a

BACKGROUND We investigated the effects of cardiopulmonary bypass (CPB) on ileal homeostasis, and the influence of functional inhibition of complement C5a on CPB-induced mesenteric injury. METHODS Pigs were perfused on CPB for 1 hour and then perfused off CPB for an additional 2 hours. Antiporcine C5a monoclonal antibody (C5a MAb) was administered 20 minutes before onset of CPB to 6 pigs; 6 controls received saline vehicle. Total complement activity, ileal myeloperoxidase, and indices of ileal integrity were examined. RESULTS Treatment with C5a MAb ameliorated CPB-induced abnormalities in endothelium-dependent relaxation to ADP and substance P, and the hypercontractile response to phenylephrine of ileal microvessels (88 to 168 microm). Ileal myeloperoxidase activity [units/g protein] was 41 +/- 11 in the C5a MAb group, compared to 83 +/- 13 in the saline group (19 +/- 10 base line). Total hemolytic complement activity was similar in the C5a MAb and saline groups (0.6 +/- 0.2 and 0.7 +/- 0.2 CH50 units). During CPB, ileal mucosal blood flow and mucosal pH, edema formation, and epithelial permeability deteriorated similarly in saline and C5a MAb groups. Inducible nitric oxide synthase (iNOS) mRNA expression was similar before and after CPB. CONCLUSIONS CPB is associated with significant physiologic alterations in mucosal perfusion, epithelial permeability, edema formation, and blood flow regulation. Inhibition of C5a limits neutrophil-mediated impairment of ileal microvascular regulation after bypass, but does not improve extravascular mesenteric dysfunction after CPB.

Effects of coronary artery disease on expression and microvascular response to VEGF

The effects of coronary artery disease (CAD) on human coronary microvascular responses to vascular endothelial growth factor (VEGF) and the alterations of the myocardial expressions of VEGF and its flk-1 and flt-1 receptors were examined in 48 patients. Microvascular studies were performed in vitro with video microscopy. The expressions of VEGF and its receptors were examined using Northern analysis of total mRNA, and the expressions of constitutive nitric oxide synthase (cNOS) and inducible nitric oxide synthase (iNOS) were examined by RT-PCR. VEGF and hepatocyte growth factor (HGF) caused potent relaxations of microvessels. These responses were reduced in the presence of N G-nitro-l-arginine and the tyrosine kinase inhibitor genistein or in microvessels from patients with CAD. Relaxations to substance P and sodium nitroprusside were similar in both groups. The substance P response was abolished in the presence of N G-nitro-l-arginine. The expression of VEGF and its receptors and the expression of cNOS and iNOS were not altered in patients with CAD. In conclusion, VEGF and HGF elicit the release of nitric oxide through activation of tyrosine kinase receptors. CAD is associated with reduced vascular responses to both VEGF and HGF; this is not likely due to a reduced expression of VEGF or flt-1 or flk-1 receptors and not due to a generalized endothelium dysfunction despite the presence of mild hypercholesterolemia in these patients with CAD. These findings may have important implications regarding the efficacy of endogenous and exogenous VEGF in patients with risk factor for CAD.

Attenuation of Endothelium-Dependent Dilation of Pig Pulmonary Arterioles After Cardiopulmonary Bypass Is Prevented by Monoclonal Antibody to Complement C5a

UNLABELLED We examined whether pulmonary endothelial dysfunction associated with cardiopulmonary bypass (CPB) may be mediated by complement C5a in pigs. Pigs were placed on normothermic CPB for 1 h with or without a previous administration of 1.6 mg/kg anti-C5a monoclonal antibody (MAb), then reperfused for 2 h. Pulmonary tissue myeloperoxidase activity was measured. Expression of nitric oxide synthase (NOS) was measured by reverse transcriptase polymerase chain reaction and Western blotting. Pulmonary arterioles approximately 100 microm in diameter were preconstricted with the thromboxane analog U46619 1 microM, and relaxation responses to adenosine diphosphate 10(-9)-10(-4) M, substance P 10(-12)-10(-6) M, and sodium nitroprusside 10(-9)-10(-4) M were examined in vitro by videomicroscopy. Relaxation to the endothelium-dependent dilators adenosine diphosphate and substance P was attenuated after CPB; this attenuation was prevented by the previous administration of MAb. Relaxation to sodium nitroprusside was not affected by CPB. Neutrophil sequestration, as measured by MPO activity, increased after CPB, either with or without MAb. Transcription of NOS was unchanged by CPB, but translation of constitutive NOS was decreased after CPB, and this decrease was prevented by a previous administration of MAb. We conclude that pig pulmonary endothelial dysfunction associated with CPB may be mediated by C5a. The mechanism may involve changes in NOS translation. IMPLICATIONS In pigs, pulmonary endothelial dysfunction may occur after cardiopulmonary bypass due to product(s) of complement activation.

Serotonin-induced human coronary microvascular contraction during acute myocardial ischemia is blocked by COX-2 inhibition.

Abstract Since serotonin (5-HAT) is implicated in exacerbating acute coronary syndromes, we studied the reactivity of atrial coronary arterioles, (70–140 μm) of atherosclerotic patients undergoing cardiac surgery to 5-HAT, substance P (Sub P), and sodium nitroprusside by video-microscopy. Before ischemia 5-HAT-induced relaxation was not affected by NS398 (cyclooxygenase inhibitor), H2O2 or U63557A (thromboxane A2 synthase inhibitor), but was reduced by L-NNA. 5-HAT elicited a potent contractile response after ischemia that was inhibited by NS398, Indo, and U63557A. While Sub P relaxation was decreased after ischemia, SNP relaxation was unchanged. The nRNA steady-state levels of NOS-3, NOS-2, prostacyclin synthase, and COX-1 were not altered by ischemia. COX-2 mRNA and protein levels (Westernblotting), however, were increased (mean ± SEM) 2.4 ± 0.4 and 3.2 ± 0.7 fold, respectively, in ischemic atrium corroborating with the immunohistochemistry of atrial tissue. It is concluded that myocardial ischemia enhanced contractile response of coronary arterioles to 5-HAT maybe due to the stimulated prostaglandin release (likely thromboxane A2) secondary to induction of COX-2 expression. These findings may have implications regarding the cause of coronary spasm during acute myocardial ischemia.

Isoflurane does not further impair microvascular vasomotion in a rat model of subarachnoid hemorrhage

PurposeSince isoflurane is known to attenuate endotheliumdependent dilation (EDD) in normal cerebral arterioles, we examined whether the anesthetic has a similar effect and further impairs EDD in vessels exposed to SAH.MethodsAutologous blood was introduced in the subarachnoid space and the parietal lobe harvested. Control animals were sacrificed without introduction of blood. The response of microvessies to the endothelium-dependent dilator adenosine diphosphate (ADP) 10−9–10−4 M, the endothelium-independent dilator nitroprusside 10−9–10−4 M, and ET-1 10−13–10−8 M was measured by videomicroscopy in the presence of 0–2 minimum alveolar concentration (MAC) of isoflurane.ResultsIsoflurane attenuated EDD to ADP in control vessels [66 ± 5% (control) vs 27 ± 11% (2 MAC) dilation to ADP 10−4 M, P < 0.05], Although SAH was associated with reduced dilation to ADP exposure to isoflurane did not further impair dilation to ADP after SAH [26 ± 3% (SAH) vs 21 ± 5% (SAH/2 MAC) dilation to ADP 10−4 M,P = NS]. Dilation to nitroprusside was not affected by isoflurane or SAH. Constriction to ET-1 was reduced by 2 MAC ofisoflurane[21 ± 1% (control) vs 13 ± 5% (2 MAC) constriction to ET-1 10−8 M,P < 0.05], but not by I MAC of isoflurane in control vessels. Constriction to ET-1 was greatly attenuated by 1 or 2 MAC of isoflurane after SAH [32 ± 5% (SAH) vs 18 ± 4% (SAH/2 MAC) constriction to ET-1 10−8 M,P < 0.05],ConclusionIn rats, isoflurane does not further impair EDD after SAH and modulates the constrictive response to ET-1. Such an effect of isoflurane would not predispose the SAH-exposed vessels to vasospasm.RésuméObjectifL’isofurane est connu pour atténuer la dilatation dépendante de l’endothélium (DDE) dans les artérioles cérébrales normales. Nous avons vérifié si l’anesthésique a un effet similaire, et s’il affecte davantage la DDE, dans les vaisseaux sanguins exposés à l’hémorragie sous-arachnoïdienne (HSA).MéthodeDu sang autologue a été introduit dans l’espace sousarachnoïdien et le lobe pariétal a été prélevé. Les animaux témoins ont été sacrifiés sans introduction de sang. La réaction des microvaisseaux à 10−9–10−4 M d’adénosine diphosphate (ADP), à 10−9–10−4 M de nitroprussiate (deux dilatateurs dépendants de l’endothélium) et à 10−13–10−8 M de ET-1 a été mesurée par vidéomicroscopie en présence d’une concentration alvéolaire minimale (CAM) de 0–2 d’isoflurane.RésultatsL’isoflurane a diminué la DDE liée à l’ADP dans les vaisseaux témoins [66 ± 5 % (témoin) vs 27 ± 11 % (2 CAM) de dilatation à 10−4 M d’ADP, P < 0,05]. Bien que l’HSA soit associée à une dilatation réduite liée à l’ADP, l’exposition à l’isoflurane n’a pas accentué la dilatation liée à l’ADP à la suite d’une HSA [26 ± 3 % (HSA) vs 21 ± 5% (HSA/2 CAM) de dilatation à 10−4 M d’ADP, P = NS]. La dilatation liée au nitroprussiate n’a pas été modifiée par l’isofurane ou l’HSA. La constriction liée à l’ET-1 a été réduite par 2 CAM d’isofurane [21 ± 1 % (témoin)vs 13 ± 5%(2 CAM) de constriction à 10−8 M d’ET-1, P < 0,05], mais non par 1 CAM d’isofurane dans les vaisseaux témoins. La constriction liée à l’ET-1 a été grandement atténuée par 1 ou 2 CAM d’isofurane après une HSA [32 ±5% (HSA) vs 18 ± 4% (HSA/2 CAM) de constriction à 10−8 M d’ET-1, P < 0,05].ConclusionChez les rats, l’isofurane ne produit pas d’altération subséquente de la DDE à la suite d’une HSA et il module la réaction constrictive liée à l’ET-1. Cet effet de l’isofurane ne prédisposerait pas au vasospasme les vaisseaux exposés à l’HSA.