Mitochondrial Coupling Factor 6 Is Present on the Surface of Human Vascular Endothelial Cells and Is Released by Shear Stress
Background— We showed that mitochondrial coupling factor 6 (CF6), an endogenous inhibitor of prostacyclin synthesis, is present in the systemic circulation as a pressor substance in rats. We investigated the possibility of vascular endothelial cells as a source of circulating CF6.
Methods and Results— We used 2 cultured endothelial cell lines, human umbilical vein endothelial cells (HUVECs) and ECV 304 cells (transformed HUVECs), for this study. Immunofluorescence microscopy of both ECV 304 and HUVECs confirmed the surface-associated immunoreactivity of anti-CF6 antibody on the plasma membrane. The concentration of CF6 in the medium increased gradually with time in both ECV 304 and HUVECs in static conditions. Exposure of ECV 304 and HUVECs to a fluid shear stress enhanced the release of CF6: In ECV 304, the concentration of CF6 in the medium (ng · well−1 · 6 hours−1) was 2.1±0.8 at baseline, 4.3±0.8 after shear at 15 dynes/cm2, and 57.7±8.4 after shear at 25 dynes/cm2. CF6 contents in the cell homogenate and mitochondria were both significantly increased after exposure of ECV 304 to 6-hour shear at 15 dynes/cm2, whereas they were unchanged after shear stress at 25 dynes/cm2. The ratio of CF6 to GAPDH mRNA was enhanced significantly, by 1.8±0.2-fold, after 6-hour shear stress at 25 dynes/cm2. Flow cytometry analysis revealed that the surface-associated CF6 was significantly increased in a 3-hour static condition after the previous exposure of the cells to shear stress for 3 hours.
Conclusions— Vascular endothelial cells are a source of CF6, and shear stress regulates the release of the surface-associated CF6.
Received August 24, 2001; revision received October 15, 2001; accepted October 16, 2001.
Mitochondrial ATP synthase consists of 3 domains, namely the extrinsic and intrinsic membrane domains, F1 and F0, respectively, joined by a stalk.1,2 Four subunits of the stalk were identified and designated as coupling factor 6 (CF6), oligomycin sensitivity conferral protein, and subunits b and d.3–5 Of these 4 subunits, CF6 was reported to be essentially required for energy transduction.6 We recently demonstrated that CF6 is an endogenous inhibitory peptide of prostacyclin synthesis in vascular endothelial cells.7 This novel function of CF6 was discovered from the investigation of the mechanism of the discrepancy of prostacyclin synthesis seen in spontaneously hypertensive rats: The level of circulating prostacyclin in spontaneously hypertensive rats was decreased compared with that in normotensive control Wistar Kyoto rats, although the generation activity of prostacyclin in aortic strips was higher in spontaneously hypertensive rats than in Wistar Kyoto rats.8,9 We further demonstrated that mitochondrial CF6 is present in the systemic circulation as a pressor substance in rats.10
The passage of blood in the vessels generates hemodynamic forces, such as shear stress, and regulates the function of endothelial cells lining the intimal surface of the vasculature.11 Shear stress stimulates the gene expression of CF612 as well as the synthesis and secretion of bioactive molecules such as prostacyclin, nitric oxide (NO), tissue plasminogen activator, and platelet-derived growth factor.13–15 Recently, it was shown that the α and β subunits of ATP synthase are present on the surface in the vascular endothelial cells and that angiostatin binds to them to exert its effect,16–18 suggesting that ATP synthase on the plasma membrane may consist of full components. Taken together, CF6 may be present on the surface of the vascular endothelial cells and be released by shear stress into the systemic circulation in vivo. This study was designed to investigate the possibility of vascular endothelial cells as a source of circulating CF6 and the mechanism for regulating the level of circulating CF6.
Medium 199, FBS, trypsin (0.05%)/EDTA (0.02%), and PBS were purchased from Gibco. HuMedia was purchased from Kurabo Co, Ltd. An iodinated CF6 tracer was prepared by Suntory Biomedical Research Ltd. Restriction enzymes and Taq polymerase were purchased from Takara-shuzo. Anti-rabbit IgG goat serum was from Japan Antibody Research Institute. Anti-rabbit IgG was from Jackson Immunoresearch Laboratory, Inc. All other reagents were of the finest grade available from Sigma Chemical Co.
ECV 304 cells [spontaneously transformed human umbilical vein endothelial cells (HUVECs)] were cultured in medium M199 containing 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin (complete medium) at 37°C under 5% CO2. Primary HUVECs were cultured in HuMedia supplemented with 2% FBS, 10 ng/mL recombinant epidermal growth factor, 1 μg/mL hydrocortisone, 5 ng/mL recombinant fibroblast growth factor, and 10 μg/mL heparin (complete medium).
ECV 304 cells and HUVECs were incubated for 1 hour in PBS with either anti-human CF6 antibody or preimmune rabbit serum. The cells were then washed and incubated for 1 hour with goat anti-rabbit IgG conjugated to FITC. Control experiments were performed with preimmune serum or secondary antibody alone.
Confluent ECV 304 cells and HUVECs were exposed to a fluid shear stress with the use of a cone-plate viscometer.19 In the steady laminar mode, a cone of 1° angle spinning at 4 and 6.7 revolutions per second was used to achieve an average shear stress magnitude of 15 and 25 dynes/cm2. Viability of the cells after shear stress was assessed by protein measurement and trypan blue staining.
Confluent ECV 304 monolayers were exposed to shear stress at 15 and 25 dynes/cm2 for 3 or 6 hours, except that one monolayer was kept under static conditions (control). The medium was taken for the measurements of CF6, and the cells were used for either flow cytometry analysis against the cell surface–associated CF6 or the measurement of CF6 in the cell homogenate and the mitochondrial fraction. Cell fractionation was performed by centrifugation strategy: The cell homogenate was centrifuged at 900g for 10 minutes to remove unbroken cells and the nucleus. The undeposited material was then subjected to centrifugation at 5000g for 10 minutes, and the mitochondrial fraction was obtained. In HUVECs, the effect of shear stress at 25 dynes/cm2 for 3 hours on the release of CF6 was examined.
Preparation of Samples and Radioimmunoassay
Samples were loaded onto a Sep-Pak C18 cartridge, and the absorbed materials were eluted with 2 mL of 60% acetonitrile containing 0.1% TFA and lyophilized. The standard recombinant CF6 or the unknown sample was incubated with anti-CF6 antiserum diluent for 12 hours, and then the tracer solution (18 000 to 20 000 cpm) was added. After incubation for 24 hours, anti-rabbit IgG goat serum diluent containing 10% polyethylene glycol 6000 and rabbit IgG were added. After centrifugation at 2000g for 30 minutes, radioactivity of the precipitate was measured in gamma counter.
Synthesis of Recombinant CF6
Mature human CF6 was obtained from Escherichia coli by use of a cleavable fusion protein strategy.20 A cDNA fragment encoding mature human CF6 was inserted into the E coli expression vector. E coli JM109 cells were transformed and disrupted by sonication. Cleaved mature human CF6 was purified by high-performance liquid chromatography. Finally, the amino acid sequence and molecular mass of this peptide were checked by automated gas-phase peptide sequencer and mass spectrometry.
Synthesis of Antibody for CF6 and Its Cross-Reactivity
Synthetic CF6 fragment (human Cys 10 to 27 amino acid) solution was emulsified with an equal volume of Freund’s complete adjuvant and used for immunizing New Zealand White rabbits. The cross-reactivity of anti-CF6 antibody was examined by Western blot analysis. Briefly, the cell homogenate was subjected to SDS-PAGE using 16.5% separating gel. Proteins were transferred electrophoretically to a nitrocellulose membrane. The membranes were then treated with anti-CF6 antibody (1:1000 dilution) and stained by amplified alkaline phosphatase immunoblot kits.
Characterization of Immunoreactive Substances in the Medium
The immunoreactive species present in the medium were characterized by Western blot analysis. The culture medium obtained after exposure to shear stress at 25 dynes/cm2 was loaded onto a Sep-Pak C18 cartridge, and the absorbed materials were eluted with 2 mL of 60% acetonitrile containing 0.1% TFA and lyophilized. The residual materials were subjected to SDS-PAGE and stained with the immunoblot kit. The membrane fraction, which was obtained by ultracentrifugation (40 000g for 15 minutes) after the deposition of the mitochondria, and the mitochondrial fraction were used as controls for CF6.
Flow Cytometry Analysis
Confluent ECV-304 cells were trypsinized, washed, and reacted with saturating concentrations of anti-CF6 antibody (dilution of 1:1000) in PBS containing 1% BSA for 30 minutes on ice. After 3 washings with PBS, they were stained with FITC-conjugated goat anti-rabbit IgG in PBS for another 30 minutes and then analyzed in a FACScan.
Determination of CF6 Gene Expression
Total RNA was prepared from ECV 304 cells with the Trizol RNA purification system. cDNA was prepared from mRNA with oligo (dT) primers. Oligonucleotide primers designed against CF621 were forward primer 5′-GAATAGAAATCTAAGCGACAG-3′ and reverse primer 5′-TACAACTAATCCGTGACAAAT-3′. The relative quantities of cDNA were assessed by the second polymerase chain reaction amplification of GAPDH as previously described.22 All polymerase chain reaction procedures were performed as follows: 25 cycles for CF6 and GAPDH (45 seconds at 94°C, 45 seconds at 52°C for CF6 and 62°C for GAPDH, and 1 minute at 72°C) and final elongation (5 minutes at 72°C).
Immunofluorescence microscopy of ECV 304 cells and HUVECs confirmed the surface-associated immunoreactivity of anti-CF6 antibody on cell membranes (Figure 1A and 1B). The same fields of ECV 304 cells and HUVECs under visible light are shown in Figure 1C and 1D, respectively. Control experiments were performed with preimmune serum alone (Figure 1E) and secondary antibody alone (Figure 1F), and there was no staining in these conditions.
Radioimmunoassay for CF6
The antiserum to CF6 detected the peptide with high affinity at a final dilution of 1:1000. Half-maximum inhibition of radioiodinated ligand binding by human recombinant CF6 was observed at 300 pg/tube. An appropriate amount of cold recombinant CF6 added to the radioimmunoassay sample was precisely determined by the present radioimmunoassay. The recovery rate was >90% when the culture sample was treated with a Sep-Pak C-18 cartridge. The intra-assay and interassay coefficients of variance were 8.0% and 10.2%, respectively. As shown in Figure 2A, the antibody reacted to only one substance (9 kDa), which is compatible with authentic CF6.
Effect of Shear Stress on CF6 Synthesis
In static conditions (Figure 2B), the concentration of CF6 in medium M199 of ECV 304 cells was increased gradually with time to the levels of 16.1±3.1 ng/well at 24 hours, 46.8±5.7 ng/well at 48 hours, and 62.3±8.9 ng/well at 72 hours (n=8). In HUVECs, the concentration of its peptide was 12.5±5.0 ng/well at 24 hours, 52.5±14.4 ng/well at 48 hours, and 66.5±0.2 ng/well at 72 hours (n=6).
As shown in Figure 3, shear stress for 6 hours enhanced the release of CF6 from ECV 304 cells in a magnitude-dependent manner (2.1±0.8 ng/well at baseline versus 4.3±0.8 ng/well at 15 dynes/cm2 and 57.7±8.4 ng/well at 25 dynes/cm2, n=7, both P<0.05). Exposure of HUVECs to a fluid shear stress at 25 dynes/cm2 for 3 hours enhanced the release of CF6 (1.6±0.6 ng/well at baseline versus 15.8±6.2 ng/well after shear, n=4, P<0.05). Under these conditions, trypan blue–positive cells were undetectable and protein content was unchanged.
CF6 in the cell homogenate of ECV 304 cells (ng/mg protein) was 28.7±1.4 at baseline (n=8), 41.6±2.1 after 3-hour exposure to 15 dynes/cm2 (n=4, P<0.05 versus baseline), and 37.8±1.4 after 6-hour exposure to the same shear (n=4, P<0.05 versus baseline). In contrast, it was unchanged after 3- and 6-hour exposure of the cells to shear stress at 25 dynes/cm2 (27.3±2.4 at baseline, 34.5±1.7 after 3 hours, and 25.5±1.2 after 6 hours, n=4, respectively). CF6 in the mitochondrial fraction of ECV 304 cells (ng/mg protein) was significantly increased after 6-hour exposure of the cells to shear stress at 15 dynes/cm2 (4.5±0.4 at baseline versus 8.0±0.5 after shear, n=6, P<0.05) and unchanged after 6-hour exposure of the cells to shear stress at 25 dynes/cm2 (4.3±1.0 before shear versus 4.7±0.7 after shear, n=6).
Characterization of Immunoreactive Substances in the Medium
As shown in Figure 4, Western blot analysis showed that there was a single immunoreactive band in the lanes of the medium (A), the membrane (B), and the mitochondria (C). The molecular weight of these immunoreactive substances was ≈9 kDa and was identical to that of authentic CF6.
Figure 5 showed the representative data of fluorescence-activated cell sorter (FACS) for the cell surface–associated CF6 in ECV 304 cells. The cell surface–associated CF6 was unchanged 3 hours after exposure of the cells to shear stress at 25 dynes/cm2 (n=4), whereas it was increased significantly, by 22.2±2.8%, in a 3-hour static condition after previous exposure of the cells to shear stress for 3 hours (P<0.05, n=4).
CF6 Gene Expression
As illustrated in Figure 6A, the expression of CF6 mRNA was increased significantly, by ≈2-fold, after 6-hour exposure of the cells to shear stress at 25 dynes/cm2. The ratio of CF6 to GAPDH mRNA was increased significantly, by 1.8±0.2-fold, after shear stress compared with that after static conditions (n=5, P<0.05, Figure 6B).
The present study provided evidence that mitochondrial CF6 is present on the surface of cultured human vascular endothelial cells, that CF6 is released from the intact vascular endothelial cells in static conditions, and that shear stress increases the gene expression of CF6, the amount of cell surface–associated CF6, and the release of CF6 from the cell surface.
Presence of CF6 on the Cell Surface
α-Enolase23 and ATP synthase,24 which are normally found in the cytoplasm, can be present on the cell surface and function to bind plasminogen and aid in lymphocyte-mediated cytotoxicity, respectively. The β-subunit of mitochondrial ATP synthase is present on the surface of several tumor cell lines and functions to transport H+ across the plasma membrane, resulting in cytolysis. This finding is supported by studies demonstrating that addition of ATP synthase to cultures of tumor cell lines induces membrane depolarization, changes in permeability, and eventual lysis of a variety of transformed cells.17,18 In this study, we first showed that CF6, another component of ATP synthase, is present on the surface of HUVECs and ECV 304 cells. Vascular endothelial cells play a strategic role within the vasculature, serving as a barrier between the intravascular compartment and the underlying tissues, and often are exposed to hypoxic stress. Compared with other cell types, endothelial cells are relatively resistant to hypoxic challenge by their ability to maintain a high level of intracellular ATP.25 Thus, it is interesting to speculate that a plasma membrane–associated ATP synthase may produce extracellular ATP, providing an additional ATP source.26 The presence of CF6 on the cell surface would support the concept that ATP synthase on the plasma membrane participates in an important function of vascular endothelial cells.
Vascular Endothelial Cells as a Source of Circulating CF6
The presence of CF6 on the surface of vascular endothelial cells stimulated us to examine a possible role of these cells as the source of circulating CF6. Because mitochondria are a major storage site of CF6, cell damage would be expected to be a major mechanism for the increase in its concentration in the culture medium. No cell damage was verified, however, by the measurement of protein content on the dishes and staining with trypan blue. It is therefore proposed that mechanisms other than cell damage are responsible for the increase in CF6 concentration in the medium. The result clearly demonstrated that CF6 was released from HUVECs and ECV 304 in static conditions and that its release was enhanced by shear stress.
It is noted that CF6 contents in the cell homogenate and the mitochondrial fraction were both increased after shear stress at 15 dynes/cm2, whereas they were unchanged after shear stress at 25 dynes/cm2. The gene expression of CF6 was enhanced after shear stress at 25 dynes/cm2. Thus, it seems reasonable to presume that high shear stress (25 dynes/cm2) stimulates not only gene expression of CF6 but also the release of its peptide from the cells, resulting in no increase in the contents in the cell homogenate and the mitochondrial fraction. The fashion of release of the surface-associated CF6 still remains to be elucidated.
In Vivo Role of CF6
The in vivo role of cell surface–associated ATP synthase has been increasingly recognized as the receptor of angiostatin16 and the source of the circulating prostacyclin inhibitor CF6. By binding to the α- and β-subunits of plasma membrane–localized ATP synthase, angiostatin disrupts the production of ATP, rendering endothelial cells more vulnerable to hypoxic challenge and eventual irreversible cell damage. In the microenvironment of a growing tumor, angiostatin decreases endothelial cell survival by abolishing the ability to resist low oxygen tension.
Prostacyclin exerts a number of profound effects, including inflammation, thrombogenesis, cell growth, and peripheral circulation.27 In addition, its effect is widespread, because prostacyclin receptor is localized in various sites, such as heart, aorta, liver, kidney, and brain.28 Thus, the regulatory mechanism for the release of plasma membrane–localized CF6 influences the state of the vasculature. Shear stress stimulates the production of prostacyclin and is associated with the activation of phospholipase A2, which is mediated by the pertussis toxin–sensitive GTP-binding proteins Gi3.14,29 The shear-induced release of CF6 would counteract the prostacyclin production and regulate systemic circulation and thrombogenesis. We recently reported that CF6 circulates in the vascular system of the rat.10 Intravenous injection of recombinant CF6 increased blood pressure, apparently by suppressing prostacyclin synthesis, whereas a specific antibody to CF6 decreased systemic blood pressure concomitantly with an increase in plasma prostacyclin. The hypotensive effect of the antibody was abolished by treatment with the cyclooxygenase inhibitor indomethacin. These findings indicate that mitochondrial CF6 functions as a potent endogenous vasoconstrictor in the fashion of a circulating hormone and may suggest a new mechanism for hypertension.
In conclusion, this report first showed that mitochondrial CF6, a potent endogenous vasoconstrictor, is present on the surface of human vascular endothelial cells as a source of its circulating peptide. In addition, shear stress regulates the release of the surface-associated CF6 into the systemic circulation.
This study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture (C 13670686), Tokyo, Japan.
Kagawa Y, Racker E. Partial resolution of the enzymes catalyzing oxidative phosphorylation: correlation of morphology and function in submitochondrial particles. J Biol Chem. 1966; 241: 2475–2482.
Knowles AF, Guillory RJ, Racker E. Partial resolution of the enzymes catalyzing oxidative phosphorylation, XXIV: a factor required for the binding of mitochondrial adenosine triphosphatase to the inner mitochondrial membrane. J Biol Chem. 1971; 246: 2672–2679.
Osanai T, Kamada T, Fujiwara N, et al. A novel inhibitory effect on prostacyclin synthesis of coupling factor 6 extracted from the heart of spontaneously hypertensive rats. J Biol Chem. 1998; 273: 31778–31783.
Kroll MH, Hellums JD, McIntire LV, et al. Platelets and shear stress. Blood. 1996; 85: 1525–1541.
Diamond SL, Eskin SG, McIntire LV. Fluid flow stimulates tissue plasminogen activator secretion by cultured endothelial cells. Science. 1989; 243: 1483–1485.
Frangos JA, McIntire LV, Eskin SG, et al. Flow effects on prostacyclin production by cultured human endothelial cells. Science. 1985; 227: 1477–1479.
Mitsumata M, Fishel RS, Nerem RM, et al. Fluid shear stress stimulates platelet-derived growth factor expression in endothelial cells. Am J Physiol. 1993; 265: H3–H8.
Moser TL, Stack MS, Asplin I, et al. Angiostatin binds ATP synthase on the surface of human endothelial cells. Proc Natl Acad Sci U S A. 1999; 96: 2811–2816.
Zanovello P, Bronte V, Rosato A, et al. Responses of mouse lymphocytes to extracellular ATP, II: extracellular ATP causes cell type-dependent lysis and DNA fragmentation. Immunology. 1990; 145: 1545–1550.
Malek A, Izumo S. Physiological fluid shear stress causes downregulation of endothelin-1 mRNA in bovine aortic endothelium. Am J Physiol. 1992; 263: C389–C396.
Das B, Mondragon MO, Sadeghian M, et al. A novel ligand in lymphocyte-mediated cytotoxicity: expression of the beta subunit of H+ transporting ATP synthase on the surface of tumor cell lines. J Exp Med. 1994; 180: 273–281.
Unno N, Menconi MJ, Salzman AL, et al. Hyperpermeability and ATP depletion induced by chronic hypoxia or glycolytic inhibition in Caco-2BBe monolayers. Am J Physiol. 1996; 270: G1010–G1021.
Nakagawa O, Tanaka I, Usui T, et al. Molecular cloning of human prostacyclin receptor cDNA and its gene expression in the cardiovascular system. Circulation. 1995; 90: 1643–1647.
Gudi SRP, Clark CB, Frangos JA. Fluid flow rapidly activates G proteins in human endothelial cells: involvement of G proteins in mechanochemical signal transduction. Circ Res. 1996; 79: 834–839.