Articles

Complement-Mediated Loss of Endothelium-Dependent Relaxation of Porcine Coronary Arteries

Role of the Terminal Membrane Attack Complex

Gregory L. Stahl, Wende R. Reenstra, Gyorgy Frendl
https://doi.org/10.1161/01.RES.76.4.575
Circulation Research. 1995;76:575-583
Originally published April 1, 1995

Jump to

Abstract

Abstract Reperfusion of the ischemic myocardium results in the loss of endothelium-dependent relaxation. We have shown recently that the alternate complement pathway is activated immediately on reperfusion of the ischemic porcine myocardium. We hypothesized that complement activation directly attenuates endothelium-dependent relaxation of porcine coronary arteries. Bradykinin (BK) or substance P concentration-dependently relaxed precontracted (U46619, 50 nmol/L) left anterior descending coronary artery (LAD) rings in vitro. Addition of zymosan to human (10%) or porcine (10%) serum for 30 minutes significantly (P<0.05) increased the EC50 of BK-induced LAD relaxation from 4±1 to 418±159 nmol/L (n=8) and from 9±3 to 281±132 nmol/L (n=7), respectively. Similarly, addition of zymosan to 10% human serum (HS) for 30 minutes increased the EC50 of substance P–induced LAD relaxation from 0.4±0.1 to 30±14 nmol/L (n=9, P<.05). Basal release of nitric oxide was reduced significantly in LAD rings exposed to zymosan-activated HS compared with HS alone. Addition of soluble CR1 (sCR1, 10 nmol/L) to zymosan-activated HS preserved BK-induced relaxation (EC50) of the LAD rings (control, 4±1 nmol/L; sCR1+zymosan+serum, 2±1 nmol/L; n=6). Zymosan-activated C8-depleted HS (10%) did not attenuate the EC50 of BK-induced coronary artery relaxation (3±1 to 3±1 nmol/L, n=7, P=NS). Zymosan-activated C8-depleted HS plus C8 (6 μg/mL) increased the EC50 of BK-induced coronary artery relaxation from 4±1 to 423±141 nmol/L (n=12, P<.05). We have further demonstrated that C5b-9 complexes can be found on the luminal surface of LAD endothelial cells after 5 minutes of exposure to zymosan-activated HS by using C5b-9 reactive monoclonal antibody fluorescent immunohistochemistry and confocal microscopy. We conclude that complement activation directly attenuates endothelium-dependent relaxation through the formation of the terminal membrane attack complex (C5b-9).

The obligatory role of endothelium in the regulation of vascular tone was first described by Furchgott and Zawadzki.1 This endothelium-dependent relaxing factor (EDRF) has been characterized to be at least partly composed of nitric oxide or a closely related compound.2 Since this discovery, many substances and physiological and pathophysiological conditions have been shown to augment, attenuate, or modulate the synthesis, production, or effects of EDRFs.

Studies of myocardial ischemia and reperfusion have shown marked alterations in endothelium-dependent relaxation of the coronary vasculature.3 4 5 Even brief periods of myocardial ischemia and reperfusion in swine alter endothelium-dependent, but not endothelium-independent, relaxation of coronary arteries.6 Although the mechanisms by which ischemia and reperfusion induce endothelial dysfunction remain unclear, there is reason to suspect that neutrophils (PMNs) and oxygen-derived free radicals play a role. Activated PMNs and oxygen-derived free radicals are toxic to endothelial cells.7 Further, superoxide ions inactivate EDRF and initiate lipid peroxidation, thus altering membrane permeability and leading to endothelial cell dysfunction.8 Additional experimental data demonstrate that monoclonal antibodies against neutrophil or endothelial cell adherence proteins provide not only myocardial protection but also preserve endothelium-dependent relaxation.9 10 11 12

A significant amount of evidence suggests that the complement cascade plays an important role in myocardial ischemia and reperfusion injury. First, the alternate complement pathway is activated immediately on reperfusion of the ischemic porcine myocardium.13 Second, subcellular membrane fragments released after myocardial ischemia have been shown to activate the complement cascade.14 15 Third, depletion of complement with cobra venom factor or inhibition of complement activation with soluble complement receptor type 1 (sCR1) provides protection against experimental models of myocardial ischemia and reperfusion injury.16 17 Fourth, C5b-9 directly induces cardiac dysfunction in the isolated human plasma–perfused rabbit heart.18 19 Fifth, intracoronary infusion of porcine C5a induces myocardial ischemia.20 21 Sixth, recent evidence from our laboratory demonstrates that a monoclonal antibody against porcine C5a protects the ischemic porcine myocardium against reperfusion injury.13 Thus, the complement system plays an important role in myocardial ischemia and reperfusion injury. However, the mechanisms of complement-induced myocardial injury and the importance of the various complement components have not been established fully.

There is evidence to suggest that complement may attenuate endothelial cell function. The terminal membrane attack complex (MAC, C5b-9) is observed along endothelial cells soon after reperfusion and may elicit endothelial cell injury by forming transmembrane channels.16 22 Additionally, complement activation releases heparan sulfate from cultured porcine endothelial cells.23 24 Heparan sulfate tethers superoxide dismutase to endothelial cells and attaches the endothelium to the extracellular matrix.23 Further, complement activation and C5b-9 formation have been shown to upregulate the PMN adherence protein, P-selectin, on endothelial cells.25 26 However, the direct role of the complement system on endothelial cell dysfunction and the loss of endothelium-dependent relaxation is unknown.

Recently, we have observed that the alternate complement cascade is activated immediately on reperfusion of the ischemic porcine myocardium.13 This time course of complement activation parallels the maximal loss of endothelium-dependent relaxation observed in another model of myocardial ischemia and reperfusion.3 In the present study, we investigated the potential direct effects of complement activation on endothelium-dependent relaxation of the porcine coronary vasculature. We hypothesized that complement activation would directly attenuate endothelium-dependent relaxation of porcine coronary arterial rings in vitro in the absence of blood-borne inflammatory cells (ie, PMNs). Our data demonstrate that complement activation directly attenuates endothelium-dependent relaxation of porcine coronary vessels. The presence of C5b-9 on the vascular endothelium, normal endothelium-dependent relaxation of left anterior descending coronary artery (LAD) rings after activation of C8-depleted serum, and abnormal endothelium-dependent relaxation of LAD rings after activation of C8-supplemented C8-depleted serum suggest that the loss of endothelium-dependent relaxation is mediated by the formation of C5b-9.

Materials and Methods

Animal and Human Research Guidelines

All procedures in the present study were approved by the Human Subjects Review Committee and the Animal Use and Care Administrative Advisory Committee of the University of California, Davis, and the Harvard Medical Area Standing Committee on Animals.

Materials

Bradykinin, substance P, Nω-nitro-l-arginine methyl ester (L-NAME), and zymosan were purchased from Sigma Chemical Co. U46619 was a gift from Upjohn. Complement compounds, including C8 and C8-depleted human serum, were purchased from Quidel. sCR1 was a gift from SmithKline and Beecham. Mouse anti-human C5b-9 antibodies were purchased from Dako and Quidel.

Coronary Vessel Preparation

Domestic swine (35 to 45 kg) were premedicated and anesthetized as described previously.21 The hearts were removed, 2 to 3 cm of the proximal portion of the LAD was dissected free, and connective tissue was removed. The vessels were cut into 2- to 3-mm rings and mounted between a pair of stainless steel hooks. One hook was attached to a force transducer (Grass model FT03). The other hook was attached to the bottom of a warmed (37°C) organ bath (5 mL). The organ bath was filled and oxygenated (95% O2/5% CO2) with Krebs-Henseleit (KH) buffer as we have described previously.27 28 Preliminary length-tension studies in our laboratory demonstrated that a preload of 4 to 6 g was optimal for contraction and relaxation studies with this size porcine coronary artery. The KH buffer was replaced every 20 to 30 minutes. The analog signals were digitized, analyzed, and stored by a commercially available software system (po-ne-mah) on a 486-based computer.

After an equilibration period of 1 hour, the rings were contracted with 100 mmol/L K+ KH buffer and washed. This procedure was repeated three times. If necessary, the preload was adjusted back to 4 to 6 g after each 100 mmol/L K+ KH buffer–induced contraction. The LAD rings then were contracted with U46619 (50 nmol/L), which resulted in a stable contraction for 15 to 20 minutes. Use of other vasoconstricting agents (ie, norepinephrine, phenylephrine, prostaglandin F2α, and acetylcholine) in preliminary studies resulted in poor or unstable contractions. At the peak of contraction, a dose-response curve to bradykinin (BK, 10−9 to 10−6 mol/L) or substance P (10−10 to 10−8 mol/L) was performed by adding cumulative amounts of these endothelium-dependent relaxing agents.29 30 31 The LAD rings then were washed with KH buffer and allowed to return to baseline.

After 1 hour, each of the rings was then bathed in one of the following supplemented KH buffers: (1) 10% porcine serum, (2) 10% human serum, (3) zymosan (1 mg/mL), (4) 1%, 3%, or 10% porcine serum plus zymosan (1 mg/mL), (5) 1%, 3%, or 10% human serum plus zymosan (1 mg/mL), (6) 10% porcine serum plus zymosan (1 mg/mL) plus sCR1 (10 nmol/L), (7) 10% human serum plus zymosan (1 mg/mL) plus sCR1 (1 nmol/L), (8) C8-depleted human serum (10%) plus zymosan (1 mg/mL), (9) C8-depleted human serum (10%) plus zymosan (1 mg/mL) plus C8 (6 μg/mL), (10) zymosan-activated human serum (10%), or (11) human recombinant C5a (0.1 μmol/L). After 30 minutes, each bath was then washed with normal KH buffer and contracted with U46619 (50 nmol/L), and a second dose-response curve to BK or substance P was generated.

Inhibition of Basal Release of Nitric Oxide

Additional LAD rings were exposed to human serum (10%) or human serum (10%) plus zymosan (1 mg/mL) for 30 minutes. These vessels then were washed and exposed to L-NAME (100 μmol/L) to inhibit the basal release of nitric oxide in these vessels.26

Immunohistochemistry

Purified mouse anti-human C5b-9 (MAC) IgG2 monoclonal antibody (anti-human MAC, aE11, Dakopats A/S) was used for immunohistochemical detection of human C5b-9 in the tissues. Fluorescein isothiocyanate (FITC)–labeled goat anti-mouse IgG F(ab′)2 (Jackson Immuno Research Laboratories) secondary antibody was used for visualization.

Tissue samples were fixed in 2% paraformaldehyde for 2 to 3 hours, then placed in 30% sucrose for 14 to 16 hours at 4°C, and frozen in 2-methyl-butane at −70°C. Frozen tissue sections (10 to 12 μm thin) were prepared after imbedding into Tissue-Tek OCT compound (Miles Diagnostics) on a Zeiss Microm HM 505E cryostat. Multiple sections were mounted on precleaned and coated Superfrost/Plus (Fisher Scientific) microscopic slides, fixed with acetone, air-dried, and processed for immunohistochemical staining.

Tissue samples were blocked with 1% goat serum for 20 minutes at room temperature, washed, and then incubated with anti-human C5b-9 monoclonal antibody in phosphate-buffered saline (PBS) containing 1% bovine serum albumin for 1 hour. After the slides were washed, they were incubated with the FITC-labeled goat anti-mouse secondary antibody for 30 minutes, washed with PBS several times, and covered with antifade solution (SlowFade, Molecular Probes Inc) and coverslips for confocal microscopic analysis.

Confocal Microscopic Analysis

The specimens were analyzed with the Leica confocal laser scanning microscope equipped with an argon ion laser with an output power of 2 to 50 mW, two photomultiplier tubes, and narrow band filters. The argon laser has two excitation wavelengths, 488 and 514 nm. The apertures were set at the minimum size for optimal signal. Smaller apertures (pinhole) allow a narrower optical cross section and less background.32 33 The images are en face optical sections through the vertical axis of the tissue. Each image of the series was taken at 1.0-μm intervals in the 512×512 pixel format. The tissue sections for immunohistochemistry were viewed with a ×100 (numerical aperture, 1.0) water immersion lens. Optical sections were taken through 3 to 10 layers, depending on the intensity of the stain. The images were analyzed, enhanced, and stored on an optical disc. Pseudocolor images were computer-generated with a twofold computer enhancement. All samples were collected, analyzed, and enhanced under the same conditions.

Statistical Analysis

The data are presented as mean±SEM. Relaxation of the preconstricted LAD rings is presented as a percentage of the U46619-induced contraction. The median effective concentration (EC50) of BK or substance P inducing a 50% relaxation of the preconstricted LAD ring was calculated for each vessel. Distribution of data was assessed with a Shapiro-Wilk test.34 A paired t test or the Wilcoxon signed rank test (analogous nonparametric t test) was used to establish significance of the EC50 data.35 Differences in relaxation between the control and experimental group at the various BK or substance P concentrations were analyzed with Friedman’s repeated-measures ANOVA followed by the Student-Newman-Keuls post hoc test. Statistical significance was taken at P≤.05. Analyses were performed using crunch 4 (Crunch Software Corp) and sigmastat (Jandel Scientific) for the IBM computer.

Results

A total of 38 pigs were used in the present study. High K+ (100 mmol/L) KH buffer contracted the LAD rings by 10.5±0.6 g. The first U46619-induced contraction was 73±4% of the 100 mmol/L K+ KH buffer response.

Time Control Experiments

Zymosan, porcine (10%), or human (10%) serum did not significantly (P>.05) alter U46619-induced contraction of the LAD rings compared with the first U46619 contraction (7.7±2.4 versus 9.0±3.1 g, 3.9±0.4 versus 5.7±0.7 g, and 4.5±1.1 versus 5.4±1.2 g, respectively.

Fig 1⇓ summarizes the effect of zymosan (Fig 1A⇓), porcine (Fig 1B⇓), or human (Fig 1C⇓) serum on BK-induced relaxation. BK concentration-dependently relaxed the LAD rings during control states in the presence of KH buffer (Fig 1A⇓ through 1C). Repeating the BK dose-response curve after incubation of the LAD rings with zymosan (Fig 1A⇓), 10% porcine serum (Fig 1B⇓), or 10% human serum (Fig 1C⇓) for 30 minutes did not attenuate BK-induced relaxation. The EC50 for BK-induced relaxation of the LAD rings was not significantly different from the control value when compared with zymosan, porcine, or human serum (3±1 versus 3±1 nmol/L, 3±1 versus 2±1 nmol/L, and 3±1 versus 4±1 nmol/L, respectively).

Figure 1.
Figure 1.

Graphs showing the effect of zymosan (A), porcine serum (B), and human serum (C) on bradykinin (BK)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. BK dose-dependently relaxed the LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in zymosan (1 mg/mL), porcine serum (10%), or human serum (10%), BK-induced relaxation was not significantly attenuated. Open and closed circles represent the means of n experiments. Brackets represent ±SEM.

Effect of Zymosan-Activated Human Serum on LAD Ring Contraction and Relaxation

Zymosan-activated human serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=8) compared with the first U46619-induced contraction (7.1±1.3 versus 7.8±1.2 g, P=NS).

Zymosan-activated human serum concentration-dependently attenuated BK-induced relaxation of the LAD rings (Fig 2⇓). No significant loss of BK-induced relaxation was observed after a 30-minute incubation with zymosan-activated 1% human serum in six LAD rings (Fig 2A⇓). The EC50 for BK-induced relaxation of the LAD rings in the KH buffer (ie, control) was 3±1 nmol/L and increased to 11±3 nmol/L (P<.05) after incubation of the LAD rings in 3% human serum activated with zymosan for 30 minutes. A greater loss of endothelium-dependent relaxation was observed after incubation (ie, 30 minutes) of eight LAD rings in 10% human serum activated with zymosan, with the EC50 increasing from 4±1 to 418±159 nmol/L (P<.05). However, nitroglycerin (10−5 mol/L) completely relaxed all the LAD rings to the baseline value.

Figure 2.
Figure 2.

Graphs showing concentration-dependent effects of 1 mg/mL zymosan–activated human serum (HS) on bradykinin (BK)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. LAD rings were bathed in 1% (A), 3% (B), and 10% (C) HS and activated with zymosan for 30 minutes. BK dose-dependently relaxed all LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in zymosan-activated HS (10%, C), BK-induced relaxation was attenuated significantly. Open and closed circles represent the means of n experiments. Brackets represent ±SEM. *P<.05 vs control.

Effect of Zymosan-Activated Porcine Serum on LAD Ring Contraction and Relaxation

Zymosan-activated porcine serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=7) compared with the first U46619-induced contraction (6.1±1.4 versus 7.2±1.7 g, P=NS).

Zymosan-activated porcine serum attenuated BK-induced relaxation of the LAD rings (Fig 3⇓). No significant loss of BK-induced relaxation was observed after a 30-minute incubation with zymosan-activated 1% or 3% porcine serum in three and seven LAD rings, respectively (Fig 3A⇓ and Fig 3B⇓, respectively). The EC50 for BK-induced relaxation of the LAD rings was 4±1 nmol/L before and after incubation of the LAD rings in 3% zymosan-induced activation of porcine serum. The EC50 for BK-induced relaxation of the LAD rings in the KH buffer (ie, control) was 2±1 nmol/L and increased to 10±1 nmol/L (P<.05) after incubation of seven additional LAD rings in 7% zymosan-activated porcine serum for 30 minutes. A greater loss of endothelium-dependent relaxation was observed after incubation of seven LAD rings in 10% zymosan-activated porcine serum, with the EC50 increasing from 9±3 to 281±132 nmol/L (Fig 3C⇓, P<.05). However, nitroglycerin (10−5 mol/L) completely relaxed all the LAD rings to the baseline value.

Figure 3.
Figure 3.

Graphs showing concentration-dependent effects of 1 mg/mL zymosan–activated porcine serum (PS) on bradykinin (BK)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. LAD rings were bathed in 1% (A), 3% (B), and 10% (C) PS and activated with zymosan for 30 minutes. BK dose-dependently relaxed all LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in zymosan-activated PS (10%, C), BK-induced relaxation was attenuated significantly. Open and closed circles represent the means of n experiments. Brackets represent ±SEM. *P<.05 vs control.

Effect of Zymosan-Activated Human Serum on Substance P–Induced Relaxation

LAD rings incubated with 10% human serum did not attenuate the EC50 for substance P–induced relaxation (Fig 4⇓, top). The EC50 for substance P–induced relaxation of the LAD rings was 0.3±0.1 and 0.5±0.1 nmol/L (n=7, P=NS) after incubation of the LAD rings in KH buffer (ie, control) and human serum (10%), respectively. Substance P–induced relaxation was significantly attenuated when the LAD rings were incubated with zymosan-activated human serum (Fig 4⇓, bottom). The EC50 for substance P–induced relaxation for the LAD rings in the KH buffer was 0.4±0.1 nmol/L and increased to 30±14 nmol/L (n=9, P<.05) after incubation of nine LAD rings in zymosan-activated human serum for 30 minutes.

Figure 4.
Figure 4.

Graphs showing the effect of human serum (HS, top) and 1 mg/mL zymosan–activated HS (bottom) on substance P (SP)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. Substance P dose-dependently relaxed the LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in HS (10%, n=7), substance P–induced relaxation was not attenuated significantly (top). After incubation (30 minutes) of the LAD rings (n=9) in zymosan-activated HS (10%), substance P–induced relaxation was attenuated significantly (bottom). Open and closed circles represent the means of n experiments. Brackets represent ±SEM. *P<.05 vs control.

Effect of Zymosan-Induced Activation of Human Serum on Basal Release of Nitric Oxide

Fig 5A⇓ shows a representative tracing of a porcine LAD ring receiving L-NAME (100 μmol/L) after exposure to 10% human serum for 30 minutes. Exposure of six LAD rings to L-NAME after incubation in 10% human serum increased baseline tension by 8±2%.

Figure 5.
Figure 5.

A and B, Tracings showing the effect of NG-nitro-l-arginine methyl ester (L-NAME, 100 μmol/L) on basal tension development in porcine left anterior descending coronary artery (LAD) rings after 30 minutes of exposure to 10% human serum (HS, A) or 10% HS and zymosan (B). C, Bar graph summarizing the data from six LAD rings exposed to HS and from nine rings exposed to HS and zymosan (1 mg/mL). *P<.05 vs HS.

Fig 5B⇑ shows a representative tracing of a porcine LAD ring receiving L-NAME (100 μmol/L) after exposure to 10% human serum and zymosan for 30 minutes. Exposure of nine LAD rings to L-NAME after incubation in 10% human serum and zymosan increased baseline tension by 2±1%. We observed a significantly greater increase in tension above baseline after application of L-NAME to LAD rings exposed to 10% human serum compared with 10% human serum plus zymosan (Fig 5C⇑).

Role of Complement in Loss of Endothelium-Dependent Relaxation

Incubation of four LAD rings with 10% heat-inactivated (56°C for 30 minutes) human serum plus zymosan (1 mg/mL) for 30 minutes did not attenuate the EC50 for BK-induced relaxation compared with the control value (33±7 versus 39±19 nmol/L, respectively; P=NS).

Administration of sCR1 (1 nmol/L) to zymosan-activated human serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=6) compared with the first U46619-induced contraction (7.0±1.9 versus 10.0±2.6 g, respectively; P=NS). Similarly, sCR1 (10 nmol/L) in the presence of zymosan-activated porcine serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=6) compared with the first U46619-induced contraction (4.8±0.5 versus 5.9±0.4 g, respectively; P=NS).

sCR1 preserved BK-induced relaxation of the LAD rings incubated with zymosan-activated human serum (Fig 6A⇓) or zymosan-activated porcine serum (Fig 6B⇓). The EC50 of BK-induced relaxation of LAD rings in the KH buffer (ie, control) was 4±1 and 2±1 nmol/L before and after sCR1 treatment of zymosan-activated human serum (Fig 6A⇓, P=NS). Similarly, 10 nmol/L sCR1 preserved BK-induced relaxation of LAD rings (n=6) incubated with zymosan-activated porcine serum. The EC50 for BK-induced relaxation was 9±4 and 10±3 nmol/L (P=NS) before and after sCR1 treatment of zymosan-activated porcine serum, respectively.

Figure 6.
Figure 6.

Graphs showing the effect of soluble CR1 (sCR1) and 1 mg/mL zymosan–activated human serum (HS, A) or zymosan-activated porcine serum (PS, B) on bradykinin (BK)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. BK dose-dependently relaxed the LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in the presence of sCR1 and zymosan-activated HS or zymosan-activated PS (10%), BK-induced relaxation was not attenuated significantly. Open and closed circles represent the means of n experiments. Brackets represent ±SEM. *P<.05 vs control.

Effect of C5a on the Loss of Endothelium-Dependent Relaxation

Recombinant human C5a (0.1 μmol/L, Sigma) failed to attenuate BK-induced relaxation in six LAD rings. The EC50 (3±1 nmol/L) for BK-induced relaxation remained unchanged after incubation of the rings with C5a for 30 minutes.

Zymosan-activated human serum was made in a test tube, as we described previously,21 and then added to the LAD rings. In this protocol, addition of zymosan-activated human serum (10%) failed to attenuate BK-induced relaxation in four additional LAD rings (EC50, 4±1 versus 2±1 nmol/L).

Role of C8 in the Loss of Endothelium- Dependent Relaxation

Zymosan-activated C8-depleted human serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=7) compared with the first U46619-induced contraction (5.8±0.7 versus 4.1±0.9 g, respectively; P=NS). Similarly, addition of C8 (6 μg/mL) to zymosan-activated C8-depleted human serum (10%) did not significantly alter U46619-induced contraction of the LAD rings (n=12) compared with the first U46619-induced contraction (3.1±0.7 versus 5.1±0.8 g, respectively; P=NS).

LAD rings incubated with zymosan-activated C8-depleted human serum demonstrated no significant loss of BK-induced relaxation (Fig 7A⇓). The EC50 for BK-induced relaxation of the LAD rings was 3±1 nmol/L before and after incubation of the LAD rings with zymosan-activated C8-depleted human serum. In contrast, addition of C8 (6 μg/mL) to zymosan-activated C8-depleted human serum significantly attenuated endothelium-dependent relaxation (Fig 7B⇓). The EC50 for BK-induced relaxation of the LAD rings in KH buffer was 4±1 nmol/L and increased significantly (P<.05) to 423±141 nmol/L after the addition of C8 (6 μg/mL) to zymosan-activated C8-depleted human serum. The EC50 for BK-induced relaxation of LAD rings after incubation with 10% C8-depleted serum plus C8 plus zymosan was not significantly different from that found with 10% human serum plus zymosan (423±141 versus 418±159 nmol/L, respectively).

Figure 7.
Figure 7.

Graphs showing the effect of C8-depleted human serum (HS) plus 1 mg/mL zymosan (A) and C8-depleted HS plus C8 (6 μg/mL) (B) on bradykinin (BK)–induced relaxation of precontracted (U46619, 50 nmol/L) porcine left anterior descending coronary artery (LAD) rings. BK dose-dependently relaxed the LAD rings in the presence of Krebs-Henseleit (control) buffer. After incubation (30 minutes) of the LAD rings in the presence of soluble CR1 and zymosan-activated HS or PS (10%), BK-induced relaxation was not attenuated significantly. Open and closed circles represent the means of n experiments. Brackets represent ±SEM. *P<.05 vs control.

Immunoreactive C5b-9 Is Present on Coronary Vascular Endothelium After Complement Activation

Representative confocal microscopic images of porcine LAD coronary arteries are presented in Fig 8⇓. The luminal side of the LAD ring was exposed to zymosan (1 mg/mL) and 10% HS for 5 (Fig 8A⇓ and 8D⇓) or 30 (Fig 8B⇓ and 8E⇓) minutes and then processed for frozen sectioning and immunofluorescent histochemical evaluation. A mouse anti-human C5b-9 monoclonal antibody (aE11, Dako) combined with an FITC-labeled goat anti-mouse secondary antibody (Jackson Immunochemicals) was used for the detection of C5b-9 protein complexes. Confocal microscopic analysis was used for visualization. C5b-9 immunoreactivity (intense fluorescence) can be detected after a 5-minute exposure of the samples (Fig 8A⇓) to zymosan-activated 10% human serum on the luminal surface (arrows denote surface) of endothelial cells, with a further increase after 30 minutes of incubation (Fig 8B⇓). Identical tissue samples stained with the secondary antibody only (Fig 8D⇓ through 8F; control staining for Fig 8A⇓ through 8C, respectively) showed no staining in this region. Furthermore, LAD rings incubated with 10% human serum for 30 minutes did not demonstrate the presence of C5b-9 (Fig 8C⇓). However, we noted a high level of autofluorescence in the region of the elastic lamina in all tissue samples regardless of treatment. The autofluorescence made it impossible to use conventional fluorescence microscopy for the evaluation of the samples and necessitated the use of confocal microscopy. It should be noted that conventional immunohistochemical techniques (ie, ABC kit with color reaction for visualization) were only able to demonstrate the presence of C5b-9 on endothelial cells, when extremely high concentrations of complement (zymosan-induced activation of 100% human serum) were used. However, those studies have also demonstrated that the microscopic structure of the blood vessels and endothelial cells was intact (data not shown here) when zymosan activation of 10% human serum was used (ie, same concentration of activated serum that demonstrated functional loss of endothelium-dependent relaxation).

Figure 8.
Figure 8.

Confocal microscopic images demonstrating the presence of immunoreactive C5b-9 on coronary vascular endothelial cells after complement activation. A, Presence of C5b-9 on the vascular endothelium after incubation of the coronary rings with 10% human serum (HS) plus zymosan for 5 minutes. B, Increased presence of C5b-9 staining of the vascular endothelium after incubation of the coronary rings with 10% HS plus zymosan for 30 minutes. C, Lack of C5b-9 staining of the vascular endothelium after incubation of the coronary rings with 10% HS for 30 minutes. D through F, Antibody controls to HS plus zymosan for 5 minutes, zymosan for 30 minutes, and HS alone, respectively. Scale bar represents computer-generated scale of fluorescent staining intensity. The scale ranges from black/rust (low-intensity staining) through orange/yellow (moderate staining) to white/blue (high-intensity staining). The high autofluorescence of the elastic lamina is apparent in all figures. Original magnification ×250 for all panels.

Discussion

Several important findings were observed in the present study. Activation of 10% human or porcine serum with zymosan attenuated endothelium-dependent relaxation of porcine LAD rings. Basal release of nitric oxide was significantly inhibited by exposure of LAD rings to zymosan-activated human serum. sCR1 prevented zymosan-activated porcine serum– or human serum–induced loss of endothelium-dependent relaxation. Addition of recombinant human C5a or previously generated complement products (ie, zymosan-activated human serum) failed to attenuate endothelium-dependent relaxation. Removal of the complement component C8 in human serum also prevented the loss of endothelium-dependent relaxation induced by zymosan-activated human serum. The addition of C8 to C8-depleted human serum followed by zymosan activation significantly attenuated endothelium-dependent relaxation. Immunohistochemical staining of the coronary vessels demonstrated the presence of immunoreactive C5b-9 on the luminal surface of the vascular endothelium. Thus, we have demonstrated that complement activation directly attenuates endothelium-dependent relaxation of the porcine coronary vasculature, with the concomitant presence of C5b-9 on the endothelial cell membrane. These results suggest that the terminal MAC (ie, C5b-9) initiates the loss of endothelium-dependent relaxation in this model.

Role of Complement in the Loss of Endothelium-Dependent Relaxation of Coronary Vessels

The role of complement in the loss of endothelium-dependent relaxation of the coronary vasculature has not been studied. Complement activation may attenuate endothelium-dependent relaxation directly or indirectly. Complement activation results in the production of C5a, a potent chemotaxin and PMN-aggregating and -activating anaphylatoxin.36 37 Thus, C5a production following reperfusion could attenuate endothelium-dependent relaxation of coronary vessels indirectly by activating PMNs. Additionally, C5a has been shown to release heparan sulfate from porcine endothelial cells. Heparan sulfate tethers superoxide dismutase to the endothelial cell and secures the endothelial cell to the extracellular matrix.23 24 In the present study, the addition of a high concentration of human C5a (0.1 μmol/L) did not attenuate BK-induced relaxation of the porcine LAD rings. Additionally, the addition of zymosan-activated human serum, which contains C3a/C3a des-Arg, C5a/C5a des-Arg, and soluble C5b-9, did not attenuate endothelium-dependent relaxation. These data are similar to a preliminary study by Rendig et al,38 which demonstrated that C5a does not attenuate endothelium-dependent relaxation of the porcine microvasculature. Further, zymosan-activated C8-depleted human serum did not attenuate BK-induced relaxation in the present study. One would expect C5a and C3a to be produced after zymosan activation of C8-depleted human serum, although C5a and C3a were not measured in these experiments. Thus, it appears that the anaphylatoxins, C5a and C3a, do not directly attenuate endothelium-dependent relaxation of porcine coronary arteries in this model.

The formation of membrane-bound C5b-9 could directly attenuate endothelium-dependent relaxation of the porcine coronary artery rings. It is well known that C5b-9 forms transmembrane pores; in nonnucleated cells, this activity results in cellular lysis. It is less appreciated that the formation of C5b-9 may directly cause cellular activation in the absence of cellular lysis. Lysis of nucleated cells is less likely because of the presence of various membrane-bound complement regulatory proteins or homologous restriction factors, including CD59 (protectin), CD46 (membrane cofactor protein, MCP), CD55 (decay-accelerating factor), and C8 binding protein.39 40 41 42 Endothelial cells have two well-characterized membrane-bound proteins, CD59 and CD55, that restrict complement activation at the plasma membrane.43 The regulation of C5b-9 formation by CD59 and CD55 is generally considered to be species restricted. However, recent observations by Van den Berg and Morgan44 suggest that porcine CD59 may not be homologously restricted. We observed concentration-dependent inhibition of endothelium-dependent relaxation after activation of 3% to 10% human serum. However, we observed a significant loss of endothelium-dependent relaxation of porcine coronary arteries only after activation of 7% to 10% porcine serum. These data suggest that porcine endothelial cell complement-regulatory mechanisms do not efficiently recognize human complement but effectively attenuate low levels of porcine complement activation.

It is unlikely that zymosan-activated serum resulted in the direct lysis of porcine endothelial cells in the present study for several reasons. First, in a series of histochemical studies attempting to demonstrate the presence of C5b-9 on endothelial cells, we observed no damage to the microscopic appearance of the endothelial cells, except when extremely high concentrations of activated complement (100% activated serum) were used. Second, the membrane regulators of complement activation (ie, CD59 and CD55) would have been able to effectively inhibit lytic concentrations of C5b-9 in only 10% serum. Third, endothelial cells exposed to other species of complement are not easily lysed even after 48 hours of exposure to complement.45 Fourth, one would expect to observe a significant attenuation or augmentation of the contractile effect of U46619 in the present study, a finding that was not observed. Therefore, we postulate that it is more likely that C5b-9 formed sublytic concentrations of transmembrane pores in the present study. We speculate that C5b-9 leads to the activation of the endothelium and the loss of endothelium-dependent relaxation. However, this assumption is speculative and warrants further investigation.

Endothelium-Dependent Relaxation and Nitric Oxide Formation

It is well known that porcine coronary arteries and endothelial cells release multiple EDRFs when exposed to BK.30 31 Thus, the loss of BK-induced porcine coronary artery relaxation resulted from a loss of endothelium-“dependent” relaxation in the present study. However, substance P releases only nitric oxide from porcine endothelial cells.29 Thus, we conclude that the attenuated substance P–induced relaxation of LAD rings, exposed to zymosan-activated human serum, was a result of altered nitric oxide–induced relaxation.

Vascular ring preparations, like those performed in the present study, release basal amounts of nitric oxide from normal functioning endothelial cells.26 We observed that the addition of L-NAME to LAD rings exposed to human serum resulted in contraction of the vascular smooth muscle. Further, LAD rings exposed to 10% human serum relaxed in the presence of BK or substance P. These data demonstrate that both basal nitric oxide and mediator-induced release of nitric oxide are not inhibited by 10% human serum in this model. In contrast, LAD rings exposed to human serum and zymosan contracted significantly less to L-NAME and relaxed significantly less to substance P or BK. Thus, complement attenuates not only the pharmacologically induced release of nitric oxide but the basal release as well. Future studies will investigate the mechanism(s) by which C5b-9 attenuates nitric oxide–induced relaxation.

Possible Role of C5b-9 in Myocardial Ischemia and Reperfusion

There is reason to suspect that the terminal MAC (C5b-9) is an important mediator of ischemia and reperfusion injury. First, C5b-9 directly induces myocardial dysfunction in the isolated perfused rabbit heart.18 19 46 Second, C5b-9 directly modifies myocardial contractility and intracellular calcium in isolated myocytes.47 Third, deposition of C5b-9 has been demonstrated on myocardial cells in infarcted areas obtained from autopsies, in plasma, and along endothelial cells soon after reperfusion of ischemic human myocardium.16 22 48 49 However, the importance of the presence of C5b-9 on endothelial cells is unknown.

In the present study, we demonstrate that either zymosan-activated porcine or human serum results in a loss of endothelium-dependent relaxation of porcine coronary arteries and is associated with deposition of C5b-9 on the endothelial cells. Further, we have shown that the basal release of nitric oxide was also attenuated by zymosan-activated serum. The loss of endothelium-dependent relaxation appears to be mediated by the formation of C5b-9, since depletion and subsequent addition of C8 resulted in normal and abnormal endothelium-dependent relaxation of the LAD rings, respectively. The loss of a functional endothelial cell lining in the myocardium could be potentially harmful to an already ischemic vascular bed. Loss of endothelium-dependent relaxation may predispose the coronary vasculature to vasoconstrictive metabolites formed during ischemia and reperfusion and lead to vasospasm. Along these lines, complement activation has been shown to play a significant role in cerebral vasospasm following subarachnoid hemorrhage in humans.50 These data suggest that complement activation during myocardial ischemia and reperfusion may result in the loss or attenuation of endothelium-dependent relaxation.

Complement activation and C5b-9 formation have been shown to upregulate P-selectin on endothelial cells.25 26 Recent data suggest that inhibition of nitric oxide synthase with L-NAME induces the upregulation of P-selectin.51 P-selectin is necessary for PMN–endothelial cell adherence and the resulting transendothelial migration.9 Furthermore, inhibition of P-selectin function protects the feline myocardium from ischemia and reperfusion injury.9 52 Thus, complement activation may be an early and possibly a necessary factor in the upregulation of P-selectin and transcellular migration of PMNs after ischemia and reperfusion.

Limitations of the Present Model

We have shown recently that complement activation is observed immediately on reperfusion of the ischemic porcine myocardium.13 However, the present LAD ring preparation is not a model of ischemia and reperfusion. Whether complement-mediated loss of endothelium-dependent relaxation takes place in vivo remains unknown. Further, regulation of coronary blood flow is at the arteriolar level. It is unknown at present whether complement attenuates endothelium-dependent relaxation of the microvasculature. Future studies in our laboratory will examine the effects of complement activation on the microvasculature in vitro and after ischemia and reperfusion in vivo.

Summary and Conclusion

We have shown that activation of the complement cascade directly attenuates the endothelium-dependent relaxation of porcine coronary arteries. The formation of C5b-9 is the contributing factor in the loss of endothelium-dependent relaxation after complement activation. These studies demonstrate that in addition to oxygen-derived free radicals and activated PMNs, complement can directly attenuate the endothelium-dependent relaxation of coronary arteries and possibly lead to coronary vasospasm.

Acknowledgments

This study was supported by the American Heart Association, California Affiliate, Inc (91-32A), the Golden Empire Chapter, and the Brigham and Women’s Hospital Anesthesia Foundation. Dr Stahl was a research fellow of the American Heart Association, California Affiliate, Inc, during a portion of these studies. We acknowledge the technical assistance of Chris Longhurst and Margaret Morrissey.

Footnotes

  • Previously presented in part at the Experimental Biology meetings, Anaheim, Calif, April 23-28, 1994, and the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994.

  • This manuscript was sent to Harold C. Strauss, MD, Consulting Editor, for review by expert referees, editorial decision, and final disposition.

  • Received May 11, 1994.
  • Accepted December 16, 1994.

References

  1. ↵
    Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells on the relaxation of arterial smooth muscle by acetylcholine. Nature. 1980;288:373-376.
    OpenUrlCrossRefPubMed
  2. ↵
    Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev. 1991;43:109-142.
    OpenUrlPubMed
  3. ↵
    Tsao PS, Aoki N, Lefer DJ, Johnson G III, Lefer AM. Time course of endothelial dysfunction and myocardial injury during myocardial ischemia and reperfusion in the cat. Circulation. 1990;82:1402-1412.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    van Benthuysen KM, McMurtry IF, Horowitz LD. Reperfusion after acute coronary occlusion in dogs impairs endothelium-dependent relaxation to acetylcholine and augments contractile reactivity in vivo. J Clin Invest. 1987;79:265-274.
  5. ↵
    Mehta JL, Nichols WW, Donnelly WH, Lawson DL, Saldeen TGP. Impaired canine coronary vasodilator response to acetylcholine and bradykinin after occlusion-reperfusion. Circ Res. 1989;64:43-54.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Headrick JP, Angello DA, Berne RM. Effects of brief coronary occlusion and reperfusion on porcine coronary artery reactivity. Circulation. 1990;82:2163-2169.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Tsao PS, Ma X, Lefer AM. Activated neutrophils aggravate endothelial dysfunction after reperfusion of the ischemic feline myocardium. Am Heart J. 1992;123:1464-1471.
    OpenUrlCrossRefPubMed
  8. ↵
    Halliwell B, Gutteridge JMC, Cross CE. Free radicals, antioxidants, and human disease: where are we now? J Lab Clin Med. 1992;119: 598-620.
  9. ↵
    Weyrich AS, Ma X, Lefer DJ, Albertine KH, Lefer AM. In vivo neutralization of P-selectin protects feline heart and endothelium in myocardial ischemia and reperfusion injury. J Clin Invest. 1993;91:2620-2629.
  10. ↵
    Ma X-L, Tsao PS, Lefer AM. Antibody to CD-18 exerts endothelial and cardiac protective effects in myocardial ischemia and reperfusion. J Clin Invest. 1991;88:1237-1243.
  11. ↵
    Ma X, Weyrich AS, Lefer DJ, Buerke M, Albertine KH, Kishimoto TK, Lefer AM. Monoclonal antibody to L-selectin attenuates neutrophil accumulation and protects ischemic reperfused cat myocardium. Circulation. 1993;88:649-658.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Ma X, Lefer DJ, Lefer AM, Rothlein R. Coronary endothelial and cardiac protective effects of a monoclonal antibody to intercellular adhesion molecule-1 in myocardial ischemia and reperfusion. Circulation. 1992;86:937-946.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Amsterdam EA, Stahl GL, Pan H-L, Rendig SV, Fletcher MP, Longhurst JC. Limitation of reperfusion injury by a monoclonal antibody to C5a during myocardial infarction in pigs. Am J Physiol. 1995;268(Heart Circ Physiol 37):H448-H457.
  14. ↵
    Kagiyama A, Savage HE, Michael LH, Hanson G, Entman ML, Rossen RD. Molecular basis of complement activation in ischemic myocardium: identification of specific molecules of mitochondrial origin that bind human C1q and fix complement. Circ Res. 1989;64:607-615.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Rossen RD, Michael LH, Kagiyama A, Savage HE, Hanson G, Reisberg MA, Moake JN, Kim SH, Self D, Weakley S, Giannini E, Entman ML. Mechanism of complement activation after coronary artery occlusion: evidence that myocardial ischemia in dogs causes release of constituents of myocardial subcellular origin that complex with human C1q in vivo. Circ Res. 1988;62:572-584.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Weisman HF, Bartow T, Leppo MK, Marsh HCJ, Carson GR, Concino MF, Boyle MP, Roux KH, Weisfeldt ML, Fearon DT. Soluble human complement receptor type 1: in vivo inhibitor of complement suppressing post-ischemic myocardial inflammation and necrosis. Science. 1990;249:146-151.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Crawford MH, Grover FL, Kolb WP, McMahan CA, O’Rourke RA, McManus LM, Pinckard RN. Complement and neutrophil activation in the pathogenesis of ischemic myocardial injury. Circulation. 1988;78:1449-1458.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Homeister JW, Satoh P, Lucchesi BR. Effects of complement activation in the isolated heart: role of the terminal complement components. Circ Res. 1992;71:303-319.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Homeister JW, Satoh PS, Kilgore KS, Lucchesi BR. Soluble complement receptor type 1 prevents human complement-mediated damage of the rabbit isolated heart. J Immunol. 1993;150:1055-1064.
    OpenUrlAbstract
  20. ↵
    Stahl GL, Amsterdam EA, Symons JD, Longhurst JC. Role of thromboxane A2 in the cardiovascular response to intracoronary C5a. Circ Res. 1990;66:1103-1111.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Stahl GL, Fletcher MP, Amsterdam EA, Longhurst JC. Role of granulocytes and C5a in myocardial response to zymosan-activated serum. Am J Physiol. 1991;261(Heart Circ Physiol 30):H29-H37.
  22. ↵
    Rus HG, Niculescu F, Vlaicu R. Presence of C5b-9 complement complex and S-protein in human myocardial areas with necrosis and sclerosis. Immunol Lett. 1987;16:15-20.
    OpenUrlCrossRefPubMed
  23. ↵
    Platt JL, Dalmasso AP, Lindman BJ, Ihrcke NS, Bach FH. The role of C5a and antibody in the release of heparan sulfate from endothelial cells. Eur J Immunol. 1991;21:2887-2890.
    OpenUrlPubMed
  24. ↵
    Lindman BJ, Noreen HJ, Geller RL, Dalmasso AP, Bach FH, Platt JL. The role of cell-surface glycoproteins in the activation of endothelial cells by antibody and complement. Transplant Proc. 1992;24:586-587.
    OpenUrlPubMed
  25. ↵
    Mulligan MS, Polley MJ, Bayer RJ, Nunn MF, Paulson JC, Ward PA. Neutrophil-dependent acute lung injury: requirement for P-selectin (GMP-140). J Clin Invest. 1992;90:1600-1607.
  26. ↵
    Ma X, Weyrich AS, Lefer DJ, Lefer AM. Diminished basal nitric oxide release after myocardial ischemia and reperfusion promotes neutrophil adherence to coronary endothelium. Circ Res. 1993;72:403-412.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Stahl GL, Lefer AM. Heterogeneity of vascular smooth muscle responsiveness to lipid vasoactive mediators. Blood Vessels. 1987;24:24-30.
    OpenUrlPubMed
  28. ↵
    Stahl GL, Tsao P, Lefer AM, Ramphal JY, Nicolaou KC. Pharmacologic profile of lipoxins A5 and B5: new biologically active eicosanoids. Eur J Pharmacol. 1989;163:55-60.
    OpenUrlCrossRefPubMed
  29. ↵
    Sharma NR, Davis MJ. Mechanism of substance P-induced hyperpolarization of porcine coronary artery endothelial cells. Am J Physiol. 1994;266:H156-H164.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Myers PR, Guerra R Jr, Harrison DG. Release of multiple endothelium-derived relaxing factors from porcine coronary arteries. J Cardiovasc Pharmacol. 1992;20:392-400.
    OpenUrlPubMed
  31. ↵
    Cowan CL, Cohen RA. Two mechanisms mediate relaxation by bradykinin of pig coronary artery: NO-dependent and -independent responses. Am J Physiol. 1991;261(Heart Circ Physiol 30):H830-H835.
  32. ↵
    Pawley JB. Fundamental limits in confocal microscopy. In: Pawley JB, ed. Handbook of Biological Confocal Microscopy. New York, NY: Plenum Publishing Corp; 1990:15-26.
  33. ↵
    Wilson T. The role of pinhole in confocal imaging systems. In: Pawley JB, ed. Handbook of Biological Confocal Microscopy. New York, NY: Plenum Publishing Corp; 1990:113-126.
  34. ↵
    Shapiro SS, Wilk MB, Chen HJ. A comparative study of various tests for normality. J Am Stat Assoc. 1968;63:1343-1372.
    OpenUrlCrossRef
  35. ↵
    Glantz SA. Primer of Biostatistics. New York, NY: McGraw-Hill Publishing Co; 1992:1-440.
  36. ↵
    Chenoweth DE. The properties of human C5a anaphylatoxin: the significance of C5a formation during hemodialysis. Contrib Nephrol. 1987;59:51-71.
    OpenUrlPubMed
  37. ↵
    Fletcher MP, Stahl GL, Longhurst JC. C5a-induced myocardial ischemia: role for CD18-dependent PMN localization and PMN-platelet interactions. Am J Physiol. 1993;265:H1750-H1761.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Rendig S, Gray S, Longhurst J, Amsterdam E. Differential effects of complement C5a on porcine coronary conductance and resistance arteries. FASEB J. 1994;8:A857. Abstract.
    OpenUrl
  39. ↵
    Nicholson-Weller A, Burge J, Fearon DT, Weller PF, Austen KF. Isolation of a human erythrocyte membrane glycoprotein with decay-accelerating activity for C3 convertase of the complement system. J Immunol. 1982;129:184-189.
    OpenUrlPubMed
  40. ↵
    Cole J, Housley GA, Dykman TR, MacDermott RP, Atkinson JP. Identification of an additional class of C3-binding membrane proteins of human peripheral blood leukocytes and cell lines. Proc Natl Acad Sci U S A. 1986;82:859-864.
  41. ↵
    Davies A, Simmons DL, Hale G, Harrison RA, Tighe H, Lachmann PJ, Waldmann H. CD59, and Ly-6 like protein expressed in human lymphoid cells, regulates the action of the complement membrane attack complex on homologous cells. J Exp Med. 1989;170:637-654.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Zalman LS, Wood LM. Isolation of a human erythrocyte membrane protein capable of inhibiting expression of homologous complement transmembrane channels. Proc Natl Acad Sci U S A. 1986;83:6975-6982.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Brooimans RA, Van Wieringen PAM, Van Es LA, Daha MR. Relative roles of decay-accelerating factor, membrane cofactor protein, and CD59 in the protection of human endothelial cells against complement-mediated lysis. Eur J Immunol. 1992;22:3135-3140.
    OpenUrlPubMed
  44. ↵
    Van den Berg CW, Morgan BP. Complement-inhibiting activities of human CD59 and analogues from rat, sheep, and pig are not homologously restricted. J Immunol. 1994;152:4095-4101.
    OpenUrlAbstract
  45. ↵
    Benzaquen LR, Nicholson-Weller A, Halperin JA. Terminal complement proteins C5b-9 release basic fibroblast growth factor and platelet-derived growth factor from endothelial cells. J Exp Med. 1994;179:985-992.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Kilgore KS, Homeister JW, Satoh PS, Lucchesi BR. Sulfhydryl compounds, captopril, and MPG inhibit complement-mediated myocardial injury. Am J Physiol. 1994;266(Heart Circ Physiol 35):H28-H35.
  47. ↵
    Berger H-J, Taratuska A, Smith TW, Halperin JA. Activated complement directly modifies the performance of isolated heart muscle cells from guinea pig and rat. Am J Physiol. 1993;265(Heart Circ Physiol 34):H267-H272.
  48. ↵
    Vakeva A, Laurila P, Meri S. Regulation of complement membrane attack complex formation in myocardial infarction. Am J Pathol. 1993;143:65-75.
    OpenUrlPubMed
  49. ↵
    Langlois PF, Gawryl MS. Detection of the terminal complement complex in patient plasma following acute myocardial infarction. Atherosclerosis. 1988;70:95-105.
    OpenUrlCrossRefPubMed
  50. ↵
    Yanamoto H, Kikuchi H, Ishikawa J, Shimizu Y, Sato M, Tokuriki Y, Okamoto S, Matsumoto M, Matsumoto K, Nakamura M. Intravenous FUT-175 inhibits complement activation in the cerebrospinal fluid and vasospasm-related delayed ischemic neurological deficit following subarachnoid hemorrhage. In: Findlay JM, ed. Cerebral Vasospasm. New York, NY: Elsevier Science Publishing Co; 1993:431-434.
  51. ↵
    Davenpeck KL, Gauthier TW, Lefer AM. Inhibition of endothelial-derived nitric oxide promotes P-selectin expression and actions in the rat microcirculation. Gastroenterology. 1994;107:1050-1058.
    OpenUrlPubMed
  52. ↵
    Buerke M, Weyrich AS, Zheng Z, Gaeta FCA, Forrest MJ, Lefer AM. Sialyl Lewisx-containing oligosaccharide attenuates myocardial reperfusion injury in cats. J Clin Invest. 1994;93:1140-1148.
View Abstract

This Issue

Jump to

Article Tools

Share this Article

  • Email
  • Complement-Mediated Loss of Endothelium-Dependent Relaxation of Porcine Coronary Arteries
    Gregory L. Stahl, Wende R. Reenstra and Gyorgy Frendl
    Circulation Research. 1995;76:575-583, originally published April 1, 1995
    https://doi.org/10.1161/01.RES.76.4.575
    Permalink:
    del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Related Articles

Cited By...