Complement-Induced Release of Monocyte Chemotactic Protein-1 From Human Smooth Muscle Cells
A Possible Initiating Event in Atherosclerotic Lesion Formation
Abstract Increasing evidence suggests that complement activation might represent an important mechanism in early atherogenesis. Thus, complement components, in particular the membrane attack complex (MAC) C5b-9(m), have been isolated from human atherosclerotic lesions. Furthermore, complement activation is known to occur in atherosclerotic lesions induced in experimental animals, and the severity of cholesterol-induced plaques is markedly reduced in complement-deficient animals. During atherogenesis monocytes are recruited into the arterial wall, and a potent chemoattractant for monocytes, monocyte chemotactic protein-1 (MCP-1), is expressed by vascular smooth muscle cells (SMCs). We hypothesized that generation of MACs on SMCs during the activation of complement might lead to the release of MCP-1 and hence to monocyte recruitment. In this study, MACs were generated on human SMCs in vitro by sequential addition of the purified complement components C5b6, C7, C8, and C9. The supernatant of the culture was chemotactic for freshly isolated peripheral blood monocytes in a modified Boyden chamber. The chemotactic activity of the supernatant was abolished by anti–MCP-1 blocking antibodies but not by an isotype-matched antibody against an irrelevant antigen. The release of chemotactic activity was dependent on the dose of MAC formed on SMCs and was demonstrated within 10 minutes of exposure of the cells. The data support the hypothesis that complement-mediated release of MCP-1 from SMCs might be important in the recruitment of monocytes into the developing atherosclerotic lesion and could be an important initiating event in atherogenesis.
- Received August 22, 1995.
- Accepted February 6, 1996.
Migration of monocytes into the arterial wall is an outstanding cellular event in atherosclerotic lesion formation. Monocytes are a major constituent of the growing lesion, whereas few if any neutrophils are found.1 The monocytes that migrate into the subendothelium via cell junctions respond to chemoattractants in the intima or media of the vessel wall.
MCP-1 seems to be the major chemotactic molecule generated within the vessel wall.2 It is chemotactic for monocytes but not neutrophils3 and is found in macrophage-rich areas of human and rabbit atherosclerotic lesions.4 Cellular sources of MCP-1 include monocytes and mesenchymal cells including fibroblasts, vascular endothelial cells, and SMCs. Its synthesis can be stimulated by platelet-derived growth factor, interleukin-1, tumor necrosis factor, lipopolysaccharide, and FCS, all of which induce MCP-1 gene transcription and mRNA synthesis.5
In addition to its cytolytic properties, the MAC of human complement [C5b-9(m), ie, the membrane-bound form of C5b-9] is now recognized as a mediator of a range of cellular processes in the absence of cell death.6 Rapid (within minutes) as well as slow (within hours) responses by attacked nucleated cells have been reported. Depending on the target cells, cellular responses include mitogen and eicosanoid release7 8 or increased collagen synthesis.9
Terminal complement complexes have been isolated from human arteriosclerotic lesions.10 The characterization of these complexes as C5b-9(m) provides conclusive evidence for the in situ activation of complement, because MAC/C5b-9(m) is not water soluble and cannot diffuse into the lesion. The component of lesions that activates complement is still a matter for discussion. A complement-activating lipid has been isolated from human atherosclerotic lesions,11 and enzymatic treatment of LDL generates a complement-activating lipid in vitro.12 Thus, a derivative of LDL might activate the alternative pathway of complement in the early lesion and cause a so-called “innocent bystander attack” on cells in the lesion.13 SMCs would be susceptible to such attack because, unlike other cells of the lesion, they do not express the MAC-protecting surface molecule CD59 constitutively, although under some circumstances it may be induced.14
In the light of the evidence for the presence of MACs in atherosclerotic lesions, we postulated that MACs might act on SMCs to cause the release of MCP-1, thus leading to chemoattraction of monocytes. Such an indirect pathway could explain, at least in part, the presence of monocyte/macrophages but the absence of neutrophils within the lesion.
By using an in vitro method for generating C5b-9 complexes from purified complement components (reactive lysis),15 16 we were able to study the release of chemotactic activity by human SMCs without the confounding influences (eg, generation of chemotactic anaphylatoxins) that complement activation by either the classic or alternative pathway would have caused. We report here that formation of MACs on human SMCs in culture leads to release of chemotactic activity for freshly isolated HBMs. Experiments with blocking antibodies to MCP-1 provide evidence that this chemotactic activity is due to release of MCP-1. Thus, the data support the hypothesis that complement activation within the vessel wall may lead to the recruitment of monocytes by the release of MCP-1 from SMCs.
The culture medium used for most cell types was DMEM (GIBCO) that was buffered with 3.7 g/L NaHCO3, gassed with 5% CO2, and supplemented with 100 IU/mL penicillin (Sigma PEN-NA), 100 mg/mL streptomycin (Sigma S-6501), and 4 mmol/L l-glutamine (Sigma G-3126). Cells were maintained in a humidified incubator at 37°C. HBMs were purified from buffy coats (concentrated leukocytes) as described by Denholm and Wolber.17 The buffy coats were produced by the Blood Transfusion Service, Addenbrooke’s Hospital, Cambridge, and prevented from clotting by citrate. They were used on the day of production. The purity of monocytes was determined by flow cytometry using the size/granularity index and found to be about 75% to 80% of the total number of cells. SMCs were kindly provided by Prof Peter Weissberg, Department of Clinical Pharmacology, Addenbrooke’s Hospital. The SMCs were isolated from human aortas by the explant method of Ross18 and were demonstrated to be SMCs rather than fibroblasts by staining with an antibody to SMC α-actin. The secondary cultures of human SMCs were maintained in DMEM containing 20% FCS, fed once every 3 days, and split to a ratio of 1:2 upon reaching confluence. Cells were dissociated by using a trypsin/EDTA solution (GIBCO) in PBS. Subcultures 10 through 15 were used as target cells for complement attack.
Chemotaxis of monocytes and monocyte-derived macrophages was measured in a 48-well modified Boyden microchemotaxis chamber (lower- and upper-well volume, 25 and 50 μL, respectively; Neuroprobe). Incubation time was 75 minutes. To differentiate chemokinetic from chemotactic responses, a checkerboard analysis was performed.19 The migrated cells were quantified under a light microscope by using a specific grid. The migrated cells were counted in five random high-power fields for each well. Each sample was tested in triplicate.
MAC Formation on SMCs
C5b6 and C9 were isolated from acute-phase serum that had been activated with yeast cell walls.16 The euglobulin fraction of this activated serum was further purified by anion exchange on DE-Sephacel and size fractionation on Sephadex G-200. C7 and C8 were prepared from a 20% sodium sulfate precipitate of normal human serum. The precipitate was redissolved and further purified by anion exchange with an increasing salt gradient to elute the bound C7. C8 was further purified by a second precipitation by using Rivanol, by fractionation on CM-Cellulose with an increasing salt gradient, and by size fractionation on Sephadex G-200.
One MHD of C5b6 was defined as the amount of C5b6 needed to produce 50% lysis of 50 μL 1% guinea pig erythrocytes (1.25×106 cells). The equivalent dose of C7 and the doses of C8 and C9 required for lysis were found by titration. To form the MAC on SMCs, purified human terminal complement components diluted in DMEM were added in a 24-well plate to confluent quiescent human SMCs as follows: first, 300 μL DMEM, 25 μL PBS/C5b6 containing varying MHDs, and 2.5 μL PBS/C7, and 3 minutes later, 100 μL DMEM/C8+C9. After 45 minutes the culture medium was removed, spun down (5 minutes, 1500 rpm, 45°C), and then used in the chemotaxis assay for monocytes to test for chemotactic activity.
Assay for Toxicity by Measurement of [3H]Adenine Release
After the SMCs were incubated in 0.2% FCS, they were loaded with 1 μCi [3H]adenine (Amersham) in 0.2% FCS/DMEM, and the incubation period was continued for 2 hours. This medium was then removed, cells were washed with PBS, and the test medium was added. Streptolysin S (50 U/mL; Sigma) served as a positive control. At the end of the incubation with test medium (45 minutes; see above) the medium from each well was removed and placed in scintillation vials. The wells were then washed in PBS, and the washings were added to separate scintillation vials. Finally, the cells remaining in each well were treated with 0.1% Triton X-100 in PBS for 5 minutes to lyse the cells. The lysate was also counted for 3H on a scintillation counter, and the percentage of [3H]adenine release from cells was calculated.
Neutralization of Chemotactic Activity With Specific Antibodies
Supernatant (100 μL) from MAC-treated SMCs was preincubated with 1 μL anti–human monocyte chemotactic and activating factor antibody (MCAF/MCP-1, Anogen Ltd, clone S-101, 1 mg/mL) at 37°C for 30 minutes and subsequently used in the chemotaxis assay.
Results are expressed as mean±SEM. Sample means were compared by using a Student’s t test; n=number of assays. A value of P<.05 was considered significant.
HBMs Lose Their Chemotactic Responsiveness to MCP-1 During In Vitro Maturation
To study the dynamics of monocyte responsiveness to MCP-1, commercially available purified MCP-1 (Pepro Tech) in its maximum chemotactic concentration (20 ng/mL, as specified by the manufacturer) was offered to freshly isolated HBMs and 1- and 7-day–matured monocytes. The results showed a marked decrease in the chemotactic responses of monocytes to purified MCP-1 during the process of in vitro maturation (Fig 1⇓). After only 1 day of incubation in 20% AB serum, monocytes had lost all their responsiveness to MCP-1.
The chemotactic response of freshly isolated monocytes to the commercial MCP-1 could be blocked completely when the MCP-1 was preincubated with neutralizing anti–MCP-1 antibodies (Fig 2⇓). To exclude nonspecific effects, the antibody was tested in an assay with C5a as a chemoattractant. No blockade of chemotactic response to C5a in concentrations from 2 to 50 ng/mL could be observed. Furthermore, a murine monoclonal antibody to C5b-9 had no effect on the response of HBMs to MCP-1.
Human Confluent SMCs Attacked by Sublethal Concentrations of Terminal Complement Proteins Release MCP-1
Terminal complement proteins were generated on human SMCs (see “Methods”). The supernatant of nontreated SMCs, the individual complement components, the complement components in combination, and DMEM served as controls. The results showed a significantly increased migration to the supernatant of MAC-treated SMCs (1000 MHD) compared with controls (Fig 3⇓). The chemotactic index (number of migrated cells in the sample/number of migrated cells in the control) was ≈2. Checkerboard analysis indicated a true chemotactic rather than chemokinetic response. Experiments with the blocking antibody provided evidence that this chemotactic response was due to MCP-1. Thus, the results suggested a release of MCP-1 by human SMCs in response to MAC formation.
A dose-response curve in the presence of equivalent amounts of C7, C8, and C9 was provided with increasing concentrations of C5b6. Fig 4⇓ shows an increase of chemotactic response that reached a plateau at 1000 MHD. Measurement of [3H]adenine release showed no lytic effect of the MAC up to a concentration of 1000 MHD/well, providing evidence for a release of chemotactic activity in response to a sublytic dose of the MAC.
The time curve (Fig 5⇓) showed a chemotactic response to the supernatant only 10 minutes after exposure of the cells to the MAC, suggesting a very quick release of chemoattractants.
We have demonstrated that treatment of human confluent, 48-hour quiescent SMCs with human purified C5b-9(m) proteins, the MAC of human complement, results in the release of chemotactic activity for HBMs. Experiments with blocking antibodies provided evidence that this chemotactic activity is due to MCP-1. The release was MAC dose–dependent and could be observed after only 10 minutes following treatment of the cells with the MAC.
Additionally, we report here that freshly isolated HBMs rapidly lose their responsiveness to purified MCP-1 during the process of in vitro maturation in 20% AB serum. These observations are in line with findings by Yoshimura and Leonard,20 who report that monocytes lose most of their MCP-1 receptors and responsiveness after overnight culture in RPMI 1640 medium containing 10% human AB serum at 37°C.
MCP-1, a low-molecular-weight protein of 8.7 kD, seems to be the major chemotactic molecule generated within the vessel wall and has received much attention recently. It is chemotactic for monocytes but not neutrophils.3 Because of the high content of monocyte-derived cells and the lack of neutrophils in arteriosclerotic lesions, MCP-1 is of considerable interest in lesion development. SMCs are known to produce MCP-1, and its synthesis and secretion in these cells is stimulated by FCS, platelet-derived growth factor, lipopolysaccharide, and interleukin-1, all of which induce increased MCP-1 gene transcription and mRNA synthesis.5
Increasing evidence suggests that complement activation might be an important initiating event in atherogenesis. In animal experiments complement activation reaches completion at a very early stage, paralleling lipid accumulation and preceding monocyte infiltration.21 The severity of cholesterol-induced atheroma is markedly reduced in rabbits depleted of complement or C6.22 An aberrant lipid in the arterial lesion might be the complement-activating structure in the arteriosclerotic lesion.11 Terminal complement complexes have been isolated from human aortic lesions.10 Interestingly, the highest concentrations have been eluted from an early stage of lesion development, the intimal thickening. In situ formation of MACs in human lesions has been demonstrated.11 The fact that human SMCs are not as well protected against complement attack as other cell types in the lesion suggests SMCs as target cells for MAC formation.14
In other studies, sublethal concentrations of MACs have resulted in the release of basic fibroblast growth factor and platelet-derived growth factor within minutes after MAC treatment.7 The mechanisms that may cause this release include MAC-induced ionic changes (especially an increase in intracellular calcium), which activate exocytic pathways, and mitogen release through the MAC pores themselves. Other cellular responses to MACs are slower and may involve cellular enzyme systems such as the phosphatidylinositol-directed phospholipase C23 and increased mRNA synthesis. Examples of this delayed response are increased eicosanoid release and interleukin-1 release from human glomerular epithelial cells8 as well as the enhanced synthesis of type IV collagen by glomerular epithelial cells.9
The short time period of the response observed in our experiments makes requirement of increased mRNA synthesis or protein synthesis very unlikely, and thus it indicates a release of intracellular MCP-1 stores. Nothing is known about intracellular storage of MCP-1 so far, but an intracellular storage is conceivable and has been described for many other secreted proteins. This observation raises the question of the state of the SMC in vivo. Previous stimulation of SMCs by MCP-1 synthesis (see above) might be the precondition for the response of the cells to terminal complement components.
As yet little is known about the interaction of complement and SMCs. Apart from the observation made nearly 30 years ago that C5a causes contraction of SMCs,24 this issue has received little emphasis. Under normal circumstances in the arterial wall SMCs are hardly exposed to complement because the endothelial monolayer and the internal elastic lamina largely inhibit the penetration of the plasma molecules.25 In case of endothelial damage the transudation of macromolecules, including complement components, into the arterial wall might increase. The complement-activating structure in the lesion is still a matter for discussion. In any event, whatever the complement-activating structure is, MAC formation has been demonstrated in the early lesion, and the target cell most likely is the SMC. In this context our data suggest that MCP-1 release from human SMCs in response to MAC formation might be an important mechanism in specific monocyte recruitment into the arterial wall.
Selected Abbreviations and Acronyms
|DMEM||=||Dulbecco’s modified Eagle’s medium|
|FCS||=||fetal calf serum|
|HBM||=||human blood monocyte|
|MAC||=||membrane attack complex|
|MCP-1||=||monocyte chemotactic protein-1|
|MHD||=||minimum hemolytic dose|
|SMC||=||smooth muscle cell|
We gratefully acknowledge Prof Dr Sucharit Bhakdi for focusing our interest on the potential role of complement in atherogenesis and for helpful discussions and critical reading of the manuscript. We thank the Blood Transfusion Service, Addenbrooke’s Hospital, Cambridge, for providing buffy coats. We gratefully appreciate the expert technical assistance of Cheryl Godliman.
Faruqi MR, Di Corleto PE. Mechanisms of monocyte recruitment and accumulation. Br Heart J. 1993;69:19-29.
Leonard EJ, Yoshimura T. Human monocyte chemotactic protein-1 (MCP-1). Immunol Today. 1990;3:97-101.
Yla Hertualla SY, Lipton BA, Rosenfeld ME, Sarkioja T, Yoshimura T, Leonard EJ, Witztum JL, Steinberg D. Expression of monocyte chemotactic protein 1 in monocyte-rich areas of human and rabbit atherosclerotic lesions. Proc Natl Acad Sci U S A. 1993;88:5252-5256.
Wang JM, Sica A, Peri G, Walter S, Padura IM, Libby P, Ceska M, Lindley I, Colotta F, Monotovani A. Expression of monocyte chemotactic protein–1 and interleukin-8 by cytokine-activated human vascular smooth muscle cells. Arterioscler Thromb. 1991;11:1166-1174.
Bhakdi S, Hugo F, Tranum-Jensen J. Functions and relevance of terminal complement sequence. Blood. 1990;60:309-318.
Benzaquen LR, Nicholson-Weller A, Halperin JA. Terminal complement protein C5b-9 releases bFGF and PDGF from endothelial cells. J Exp Med. 1994;179:985-992.
Lovett DH, Haensch GM, Goppelt M, Resch K, Gemsa F. Activation of glomerular mesangial cells by the terminal membrane attack complex of complement. J Immunol. 1987;138:2473-2480.
Seifert PS, Hugo F, Tranum-Jensen J, Zahringer U, Muhly M, Bhakdi S. Isolation and characterization of a complement-activating lipid extracted from human atherosclerotic lesions. J Exp Med. 1990;172:547-557.
Bhakdi S, Dorweiler B, Kirchmann R, Torzewski J, Weise E, Tranum-Jensen J, Walev I, Wieland E. On the pathogenesis of atherosclerosis: enzymatic modification of low-density lipoprotein to an atherogenic moiety. J Exp Med. 1995;182:1959-1971.
Thompson MB, Lachmann PJ. Reactive lysis: the complement-mediated lysis of unsensitized cells I. J Exp Med. 1970;131:629-642.
Lachmann PJ, Thompson MB. Reactive lysis: the complement-mediated lysis of unsensitized cells II. J Exp Med. 1970;131:643-655.
Ross R. The smooth muscle cell, II: growth of smooth muscle cells in culture and formation of elastic fibers. J Cell Biol. 1971;50:172-177.
Yoshimura T, Leonard EJ. Identification of high affinity-receptors for human monocyte chemoattractant protein-1 on human monocytes. J Immunol. 1989;145:292-297.
Geertinger P, Sørensen H. On the reduced atherogenic effect of cholesterol feeding on rabbits with congenital (Cb) deficiency. Artery. 1974;1:177-184.
Cybulski A, Cyr M. Phosphatidyl-choline-directed phospholipase C: activation by complement C5b-9. Am J Physiol. 1993;265:551-556.
Cochrane CG, Mueller-Eberhard HJ. The derivative of two distinct anaphylatoxin activities from the third and fifth component of human complement. J Exp Med. 1968;127:371-380.
Penn MS, Saidel GM, Chisolm GM. Relative significance of endothelium and internal elastic lamina in regulating the entry of macromolecules into arteries in vivo. Circ Res. 1994;74:74-82.