+
Journul of Invesrrgutrw Surgeri: Volume 4. pp. 487-494 Pnnted i n the UK. All nghts reserved.
0894-1939/91 $3.00 .MI Copyright (01991 Taylor & Francis
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Aortic Endothelial and Smooth Muscle Cell Co-Culture: An In Vitro Model of the Arterial Wall DEBRA J. GRAHAM, MD J. JEFFREY ALEXANDER, MD REMEDIOS MIGUEL, MS Cleveland Metropolitan General Hospital Case Western Reserve University Cleveland, OH 4412 1 Abstract Interactions between vascular endothelial (EC) and smooth muscle cells (SMC) contribute both to the norma1,functionof the vascular wall and to the pathogenesis of lesions such as atherosclerosis andjibrointimal hyperplasia. However, study of these interactions has been hampered by the dijiculty in growing these two cell types in simultaneous culture. Methods using conditioned media, shared media, and bilayer culture have been described, but none is well suiled to the study of vascular cell interactions. We report a method for EC-SMC co-culture that preserves bilayer morphology, allows independent study of Ihe cells and their matrices after intervention, remains stable over long periods in culture, and permits study of changes in cell-cell interaction with growth of the cells to confluence. This simple bilayer co-culture system simulates the in vivo situation and may enhance our understanding of EC-SMC interactions. Keywords Vascular endothelial cells, vascular smooth muscle cells, co-culture.
Introduction Interactions between vascular endothelial (EC) and smooth muscle cells (SMC) contribute both to the normal function of the vascular wall and to the pathogenesis of lesions such as atherosclerosis and fibrointimal hyperplasia. Each cell type produces substances that can affect cell growth, replication, and the formation of matrix elements. Intercellular interactions can alter these functions, as well as affect cellular lipid metabolism, prostanoid production, and peptide secretion.'-' Endothelial cells have also been shown to influence SMC contractility.' Advances in cell culture techniques have resulted in a greater understanding of the structure and function of both EC and SMC. However, most studies of these cells have been performed in single cell culture systems in which the potential influence of other cell type is absent. The study of EC-SMC interactions has been hampered by difficulties in growing these cells concurrently. When simultaneous cultures are established, smooth muscle cells rapidly overgrow and eliminate endothelial cells. Although a variety of interactive culture methods have been developed, most are not suited to the study of vascular cells. In this paper, we report a new method for co-cultivation of vascular endothelial and smooth muscle cells. This system is easy to Address correspondence to Debra J. Graham, MD, 1508 Parkhill, Cleveland, OH 44 12 1.
48 7
D. .I. Graham ct al.
488
establish and maintain. It preserves bilayer morphology and most physiologic relationships between these two cell types, and it allows easy separation of the cells for study purposes.
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Materials and Methods Co-culture of bovine aortic endothelial (BAEC) and smooth muscle cells (BASMC) was established using Transwells (Costar, Cambridge, MA). Transwells are inserts consisting of a plastic cylinder with a porous polycarbonate filter base. These inserts fit inside standard tissue culture wells (Fig l ) , creating two compartments within the well which are separated by the porous membrane. This allows communication between the cell types via pores of defined size. The distance between the porous membrane and the well bottom is 1 mm.
Cell Culture Endothelial and smooth muscle cells were harvested from fresh bovine aortas by means of collagenase digestion (collagenase type IV; Sigma, St Louis, MO) as previously described." BAECs were identified by their characteristic growth in culture and by fluorescent staining for factor VIII antigen. Smooth muscle cells were then harvested from the arterial media by the method of tissue explantation." Both cell types were maintained in Dulbecco's modified Eagles medium (DMEM; Whittaker Bioproducts, Walkersville, MD) supplemented with 10%fetal bovine serum (FBS; Hyclone Laboratories, Hyclone, UT), penicillin (50 units/mL), streptomycin (50 mg/ mL; Whittaker), L-glutamine (0.292 mg/mL), and fungizone (0.5 pg/mL) in a humidified 5% CO, incubator at 37 "C. Cells were passaged using trypsin-EDTA once cultures achieved approximately 90% confluence. Cells in passages 6- 12 were used for co-culture experiments.
Lower Compartment
Microporous Membrane
Figure 1. The Transwell.
Aortic EC and SMC Co-Culture
48 9
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Co-Culture SMC were plated in 24-well tissue culture plates (Costar) in 750 p L DMEM at a density of 25,000 cells/cm2. These plates were then incubated for 4 h at 37 "C to allow even attachment of the SMC over the culture dish. The polycarbonate membranes of the inserts (5.0 pM pore size) were coated with a 1% gelatin solution (gelatin type B from bovine skin; Sigma). The coated inserts were then placed inside the wells containing the SMC, and EC were added to the upper chamber in 200 pL DMEM at a density of 25,000 cells/cm2. Cultures were maintained in the incubator, and media was replaced every 3 days in both the upper and lower chambers. Cell growth and the attainment of confluence were monitored closely. SMC were easily examined using standard phase-contrast microscopy. However, because of the properties of the polycarbonate filter, only faint outlines of EC could be seen with this technique. Thus, the degree of confluence could not be accurately determined. For this reason, scanning electron microscopy was used to document EC confluence on the filters.
Cell Proliferation To document EC-SMC interaction in this system, the effects of co-culture on cell proliferation were studied. The cell types were studied alone in their respective chambers and then in co-culture. Cell proliferation was determined using a standard ['Hlthymidine incorporation assay. After 48 h in culture, the subconfluent cell layers were labeled with 2 pCi [3H]thymidine in each chamber (New England Nuclear, Wilmington, DE). Cultures were incubated for 4 h, after which the media was removed and discarded. The cell layers were fixed with 5% cold trichloroacetic acid (Sigma), rinsed, and solubilized in 0.25N NaOH. Radioactivity was then determined in a liquid scintillation counter (TriCarb 2000 CA, Packard, Downer's Grove, IL).
Results Co-culture of bovine aortic EC and SMC was easily established using the Transwell system. Using the cell densities described, EC confluence was achieved in approximately 5 days. Electron micrography demonstrates a bare polycarbonate filter with 5.0-pM pores (Fig 2a), and a filter covered by an EC monolayer (Fig 2b). At confluence, the EC completely covered the filter, including the pores. The system remained stable for several weeks in culture, allowing the study of cells from very sparse densities to confluence, and of confluent cells over extended time periods.
Cell Proliferation When grown in subconfluent monolayers, the proliferation of both EC ( P < .004) and SMC ( P < .OOl), as determined by ['Hlthymidine incorporation, were significantly stimulated by the presence of the opposite cell type in the adjacent well (Fig 3).
Discussion Both EC and SMC have been extensively studied in tissue culture. Because cell-cell interactions are known to significantly influence the growth and metabolism of these
D.J. Graham et al.
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
490
Figure 2. ( a ) Bare polycarbonate filter with scattered 5.0-pM pores ( X 1200).( b )Confluent EC monolayer on a polycarbonate filter (X2000) showing complete coverage of the filter and pores.
Aortic EC and SMC Co-Culture
491
cpm/WELL (thousands)
l 80 o o r
60
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
40 20
0
EC 0 ALONE
SMC CO-CULTURED
Figure 3. Effects of co-culture on cellular proliferation.
two cell types, many investigators have tried to establish methods for the co-culture of vascular EC and SMC. However, co-cultivation of these cells is not easily achieved. When EC and SMC are plated together, SMC rapidly overgrow EC. This problem has led to the development of various systems for the study of EC-SMC interactions. One method available for study of cell-cell interactions is the use of “conditioned” media.12Media is first conditioned by using it to sustain a culture of a given cell type. After a period of incubation, the media is retrieved, processed, and then used to supplement the medium of a second cell type. The effect of the conditioned media on this second cell type can then be determined. This method is very easy to use, and measurement of an effect in a single cell system is easily accomplished. However, only stable, soluble mediators of cell-cell interactions can be detected and studied. In addition, study of changes in cellular interactions as they are influenced by changes in cell population density are extremely difficult with this method. The ability to study cells throughout their growth to confluence is particularly important when studying vascular cells, where physiologic differences are known to exist between confluent and subconfluent cells. Other culture systems have been described in which EC and SMC share media. Cells can be grown on microcamer beads in a single flask, with the two cell types separated by a porous filter. EC and SMC have also been grown on separate coverslips in a single culture dish.2 Davies and K e d 3 have described a method where EC are grown on microcamer beads in a chamber with a porous filter attached to the lid of a culture dish, while SMC are grown in the standard fashion on the plate surface. As with conditioned media, these systems are fairly simple to use, and the effects are easily measured. While only soluble mediators of cellular interactions can be studied, these latter techniques may allow the detection of less stable substances than could be preserved with the transfer of preconditioned media. Changes in EC-SMC interaction with cell growth to confluence can also be studied. However, this system is not
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
4 92
D. J. Graham et al.
physiologic. In the normal vessel wall, the confluent endothelium serves as a barrier between luminally secreted substances and the underlying SMC. In the shared media systems, EC confluence does not have this barrier effect. An ideal physiologic model would involve the culture of EC directly on SMC. Several such bilayer methods have been described. EC can be plated directly on top of established, growth-arrested cultures of SMC,4or they can be plated with an interven~*~~~~*~ ing layer of some matrix element, generally either fibronectin or c ~ l l a g e n . ~Van Buul-Wortelboer et all6have developed a method in which SMC are allowed to grow into a collagen gel, after which EC are plated on the gel. All of these methods produce bilayers that more closely resemble the vessel wall. These bilayer systems are the most physiologic of any available, preserving both direct cell-cell contact and polarity of EC-SMC interaction. Unfortunately, they are difficult to establish and maintain. Prior to the addition of EC, the SMC culture must be quiescent or overgrowth of SMC will occur. When EC are plated, heparin can be added to inhibit further SMC proliferation. While this is generally effective in preventing SMC overgrowth, the use of heparin is a confounding factor that may significantly affect the outcome of an experiment. If EC confluence is not achieved shortly after the cells are plated, SMC overgrowth may occur despite the use of heparin. In addition, cell associated collagenases can degrade collagen gels, resulting in a shortened lifespan for these types of bilayers. Because of these limitations, such systems cannot be used to study the growth of cells to confluence, and, generally, are not sufficiently stable to allow time course experiments. Separating the cells for study is also very difficult. The use of collagenase digestion to selectively remove the EC layer from the underlying SMC layer is complicated by variations in enzyme activity, which can result in an unreliable separation of the two cell types. Therefore, while these methods are more physiologic, they are often impractical. The Transwell co-culture method described in this paper appears to offer some distinct advantages over the methods currently available for the study of cell-cell interactions. Use of the Transwell system allows co-culture to be easily established and maintained. The SMC are plated in standard tissue culture wells. An initial incubation period before the co-culture is established assures an even distribution of SMC over the surface of the culture dish. If the inserts are placed before the SMC have attached to the dish, the population density of the SMC will be less uniform. Gelatin coating ofthe filter is not essential, but EC monolayers grown on gelatincoated filters appear to be more stable than those grown on bare filters. Cells are easily dislodged from bare filters and major cell loss can occur with media changes. Younger cells adhere better to the bare polycarbonate filters than older cells. Gelatin coating increases the adherence of older cells, allowing their use in this system. The plating of cells and their subsequent management is no different than with standard culture systems. Cells are easily separated for study using standard tissue culture methods, such as enzymatic release with either trypsin or collagenase, or by various dissolution methods. In addition, extracellular matrices can be studied separately. Because the cells are in separate compartments, growth studies from sparse population densities to confluence are possible without SMC overgrowth. These cocultures can be maintained for several weeks. Transwell co-culture preserves most of the basic physiologic relationships between EC and SMC. The system has a bilayer morphology. When the EC in the upper chamber are subconfluent, there is free communication between the two chambers. However, once EC confluence is achieved, the endothelial monolayer regulates trans-
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Aortic EC and SMC Co-Culture
493
port between the two compartments. Polarity ofthe cell interactions is also preserved, with communication between the albuminal surface of the EC and the SMC, despite EC confluence. Using the Transwell method, we have documented interaction between EC and SMC. In subconfluent layers of both cell types, cocultivation with the opposite cell type produced a significant increase in cell proliferation. This effect is most likely the result of free diffusion of substances secreted by these cells which pass through the porous polycarbonate filter. Although the distance between the EC and SMC is less than 1 mm, the fact that the cells are not in direct contact is a major flaw of this system. While most of the in vivo interactions of these two cell types should be preserved, the activity of substances such as FGF and heparan sulfates, which are secreted into the extracellular matrix, may be attenuated or lost altogether in the absence of direct cell-cell or cellmatrix contact. Despite this lack of direct contact between the cell types, it is felt that the Transwell method offers significant advantages over currently available methodology. In summary, co-culture of EC and SMC using the Transwell system provides a useful method to study the interaction of EC and SMC. The system is simple to establish and remains stable over time. Cell growth can be studied from very sparse populations to confluence and beyond. In addition, cells and their matrices can be easily and reliably separated for study. Despite the lack of direct cell-cell contact, the Transwell system provides a co-culture system that allows study of the dynamic interaction between EC and SMC. We believe that this method provides an opportunity to greatly expand our understanding of how cells of the vascular wall interact in both normal and disease states, and may provide a system with which to assess interventions directed toward the alteration of disease processes.
References 1 . Clowes AW. Arterial Endothelial-Smooth Muscle Cell Interactions: Evaluation and Treatment of’ Upper and Lower Extremity Diseases. New York: Grune 8~ Stratton;
1984:25-38. 2. Staiano-Coico L, Hajjar DP, Hefton JM, Hajjar KA, Kimmel M. Interactions of arterial cells, 111: stathmokinetic analysis of smooth muscle cells cocultured with endothelial cells. J Cell Phys. 1988;134:485-490. 3. Campbell JH, Campbell GR. Endothelial cell influence on vascular smooth muscle cell phenotype. Annu Rev Physiol. 1986;48:295-306. 4. Memlees MJ, Scott L. Interaction of aortic endothelial and smooth muscle cells in culture: effect on glycosaminoglycan levels. Atherosclerosis. 198 1 ;39:147- 16 1 . 5. Castellot JJ. Addonizio ML, Rosenberg R, Karovsky MJ. Cultured endothelial cells produce a heparin like inhibitor of smooth muscle cell growth. J Cell Biol. 198 1;90:372-379. 6. Hajjar DP, Falcone DJ, Amberson JB, Hefton JM. Interaction of arterial cells, I: endothelial cells alter cholesterol metabolism in co-cultured smooth muscle cells. J Lipid Res. 1985;26:1212-1223. 7. Davies PF, Truskey GA, Warren HB, OConnor SE, Eisenhaure BH. Metabolic cooperation between vascular endothelial cells and smooth muscle cells in co-culture: changes in low-density lipoprotein metabolism. J Cell Biol. 1985;101:871-879. 8. Hajjar DP, Marcus AJ, Hajjar KA. Interactions ofarterial cells: studies in the mechanisms of endothelial cell modulation of cholesterol metabolism in co-cultured smooth muscle cells. J Biol Chem. 1987;262:6976-698 1 .
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
4 94
D.J. Graham et al.
9. Vanhoutle PM, Rubanyi GM, Miller VM, Houston DS. Modulation of vascular smooth muscle cell contraction by the endothelium. Annu Rev Physiol. 1986;48:307-320. 10. Booyse FM, Selak BJ, Rafelson ME. Culture of arterial endothelial cells: characterization and growth of bovine aortic cells. Thromb Haemorrh. 1973;34:825-839. 1 I . Chamley-Campbell J, Campbell GR, Ross R. The smooth muscle cell in culture. Phys Rev. 1979;59:1-6 1. 12. Gadjusek CM, DiCorleto P, Ross R, Schwartz S. An endothelial cell-derived growth factor. J Cell B i d . 1980;85:467-472. 13. Davies PF, Kerr C. Co-cultivation of vascular endothelial and smooth muscle cells using microcarrier techniques. Exp Cell Res. 1982;141:455-459. 14. Jones PA. Construction of an artificial blood vessel wall from cultured endothelial and smooth muscle cells. Proc Natl Acad Sci USA. 1979;76:1882-1886. 15. Navab M, Hough GP, Stevenson LW, Drinkwater DC,Laks H, Fogelman AM. Monocyte migration into the subendothelial space of a coculture of adult human aortic endothelial and smooth muscle cells. J Clin Invest. 1988;82:1853-1863. 16. Van Bud-Wortelboer MF, Brinkman HJM, Dingeman KP, DeGroot PG, van Aken WG, van Mourik JA. Reconstruction of the vascular wall in vitro: a novel model to study interaction between endothelial and smooth muscle cells. Exp Cell Rex 1986;162:I5 1-1 58.
Journul of Invesrrgutrw Surgeri: Volume 4. pp. 487-494 Pnnted i n the UK. All nghts reserved.
0894-1939/91 $3.00 .MI Copyright (01991 Taylor & Francis
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Aortic Endothelial and Smooth Muscle Cell Co-Culture: An In Vitro Model of the Arterial Wall DEBRA J. GRAHAM, MD J. JEFFREY ALEXANDER, MD REMEDIOS MIGUEL, MS Cleveland Metropolitan General Hospital Case Western Reserve University Cleveland, OH 4412 1 Abstract Interactions between vascular endothelial (EC) and smooth muscle cells (SMC) contribute both to the norma1,functionof the vascular wall and to the pathogenesis of lesions such as atherosclerosis andjibrointimal hyperplasia. However, study of these interactions has been hampered by the dijiculty in growing these two cell types in simultaneous culture. Methods using conditioned media, shared media, and bilayer culture have been described, but none is well suiled to the study of vascular cell interactions. We report a method for EC-SMC co-culture that preserves bilayer morphology, allows independent study of Ihe cells and their matrices after intervention, remains stable over long periods in culture, and permits study of changes in cell-cell interaction with growth of the cells to confluence. This simple bilayer co-culture system simulates the in vivo situation and may enhance our understanding of EC-SMC interactions. Keywords Vascular endothelial cells, vascular smooth muscle cells, co-culture.
Introduction Interactions between vascular endothelial (EC) and smooth muscle cells (SMC) contribute both to the normal function of the vascular wall and to the pathogenesis of lesions such as atherosclerosis and fibrointimal hyperplasia. Each cell type produces substances that can affect cell growth, replication, and the formation of matrix elements. Intercellular interactions can alter these functions, as well as affect cellular lipid metabolism, prostanoid production, and peptide secretion.'-' Endothelial cells have also been shown to influence SMC contractility.' Advances in cell culture techniques have resulted in a greater understanding of the structure and function of both EC and SMC. However, most studies of these cells have been performed in single cell culture systems in which the potential influence of other cell type is absent. The study of EC-SMC interactions has been hampered by difficulties in growing these cells concurrently. When simultaneous cultures are established, smooth muscle cells rapidly overgrow and eliminate endothelial cells. Although a variety of interactive culture methods have been developed, most are not suited to the study of vascular cells. In this paper, we report a new method for co-cultivation of vascular endothelial and smooth muscle cells. This system is easy to Address correspondence to Debra J. Graham, MD, 1508 Parkhill, Cleveland, OH 44 12 1.
48 7
D. .I. Graham ct al.
488
establish and maintain. It preserves bilayer morphology and most physiologic relationships between these two cell types, and it allows easy separation of the cells for study purposes.
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Materials and Methods Co-culture of bovine aortic endothelial (BAEC) and smooth muscle cells (BASMC) was established using Transwells (Costar, Cambridge, MA). Transwells are inserts consisting of a plastic cylinder with a porous polycarbonate filter base. These inserts fit inside standard tissue culture wells (Fig l ) , creating two compartments within the well which are separated by the porous membrane. This allows communication between the cell types via pores of defined size. The distance between the porous membrane and the well bottom is 1 mm.
Cell Culture Endothelial and smooth muscle cells were harvested from fresh bovine aortas by means of collagenase digestion (collagenase type IV; Sigma, St Louis, MO) as previously described." BAECs were identified by their characteristic growth in culture and by fluorescent staining for factor VIII antigen. Smooth muscle cells were then harvested from the arterial media by the method of tissue explantation." Both cell types were maintained in Dulbecco's modified Eagles medium (DMEM; Whittaker Bioproducts, Walkersville, MD) supplemented with 10%fetal bovine serum (FBS; Hyclone Laboratories, Hyclone, UT), penicillin (50 units/mL), streptomycin (50 mg/ mL; Whittaker), L-glutamine (0.292 mg/mL), and fungizone (0.5 pg/mL) in a humidified 5% CO, incubator at 37 "C. Cells were passaged using trypsin-EDTA once cultures achieved approximately 90% confluence. Cells in passages 6- 12 were used for co-culture experiments.
Lower Compartment
Microporous Membrane
Figure 1. The Transwell.
Aortic EC and SMC Co-Culture
48 9
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Co-Culture SMC were plated in 24-well tissue culture plates (Costar) in 750 p L DMEM at a density of 25,000 cells/cm2. These plates were then incubated for 4 h at 37 "C to allow even attachment of the SMC over the culture dish. The polycarbonate membranes of the inserts (5.0 pM pore size) were coated with a 1% gelatin solution (gelatin type B from bovine skin; Sigma). The coated inserts were then placed inside the wells containing the SMC, and EC were added to the upper chamber in 200 pL DMEM at a density of 25,000 cells/cm2. Cultures were maintained in the incubator, and media was replaced every 3 days in both the upper and lower chambers. Cell growth and the attainment of confluence were monitored closely. SMC were easily examined using standard phase-contrast microscopy. However, because of the properties of the polycarbonate filter, only faint outlines of EC could be seen with this technique. Thus, the degree of confluence could not be accurately determined. For this reason, scanning electron microscopy was used to document EC confluence on the filters.
Cell Proliferation To document EC-SMC interaction in this system, the effects of co-culture on cell proliferation were studied. The cell types were studied alone in their respective chambers and then in co-culture. Cell proliferation was determined using a standard ['Hlthymidine incorporation assay. After 48 h in culture, the subconfluent cell layers were labeled with 2 pCi [3H]thymidine in each chamber (New England Nuclear, Wilmington, DE). Cultures were incubated for 4 h, after which the media was removed and discarded. The cell layers were fixed with 5% cold trichloroacetic acid (Sigma), rinsed, and solubilized in 0.25N NaOH. Radioactivity was then determined in a liquid scintillation counter (TriCarb 2000 CA, Packard, Downer's Grove, IL).
Results Co-culture of bovine aortic EC and SMC was easily established using the Transwell system. Using the cell densities described, EC confluence was achieved in approximately 5 days. Electron micrography demonstrates a bare polycarbonate filter with 5.0-pM pores (Fig 2a), and a filter covered by an EC monolayer (Fig 2b). At confluence, the EC completely covered the filter, including the pores. The system remained stable for several weeks in culture, allowing the study of cells from very sparse densities to confluence, and of confluent cells over extended time periods.
Cell Proliferation When grown in subconfluent monolayers, the proliferation of both EC ( P < .004) and SMC ( P < .OOl), as determined by ['Hlthymidine incorporation, were significantly stimulated by the presence of the opposite cell type in the adjacent well (Fig 3).
Discussion Both EC and SMC have been extensively studied in tissue culture. Because cell-cell interactions are known to significantly influence the growth and metabolism of these
D.J. Graham et al.
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
490
Figure 2. ( a ) Bare polycarbonate filter with scattered 5.0-pM pores ( X 1200).( b )Confluent EC monolayer on a polycarbonate filter (X2000) showing complete coverage of the filter and pores.
Aortic EC and SMC Co-Culture
491
cpm/WELL (thousands)
l 80 o o r
60
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
40 20
0
EC 0 ALONE
SMC CO-CULTURED
Figure 3. Effects of co-culture on cellular proliferation.
two cell types, many investigators have tried to establish methods for the co-culture of vascular EC and SMC. However, co-cultivation of these cells is not easily achieved. When EC and SMC are plated together, SMC rapidly overgrow EC. This problem has led to the development of various systems for the study of EC-SMC interactions. One method available for study of cell-cell interactions is the use of “conditioned” media.12Media is first conditioned by using it to sustain a culture of a given cell type. After a period of incubation, the media is retrieved, processed, and then used to supplement the medium of a second cell type. The effect of the conditioned media on this second cell type can then be determined. This method is very easy to use, and measurement of an effect in a single cell system is easily accomplished. However, only stable, soluble mediators of cell-cell interactions can be detected and studied. In addition, study of changes in cellular interactions as they are influenced by changes in cell population density are extremely difficult with this method. The ability to study cells throughout their growth to confluence is particularly important when studying vascular cells, where physiologic differences are known to exist between confluent and subconfluent cells. Other culture systems have been described in which EC and SMC share media. Cells can be grown on microcamer beads in a single flask, with the two cell types separated by a porous filter. EC and SMC have also been grown on separate coverslips in a single culture dish.2 Davies and K e d 3 have described a method where EC are grown on microcamer beads in a chamber with a porous filter attached to the lid of a culture dish, while SMC are grown in the standard fashion on the plate surface. As with conditioned media, these systems are fairly simple to use, and the effects are easily measured. While only soluble mediators of cellular interactions can be studied, these latter techniques may allow the detection of less stable substances than could be preserved with the transfer of preconditioned media. Changes in EC-SMC interaction with cell growth to confluence can also be studied. However, this system is not
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
4 92
D. J. Graham et al.
physiologic. In the normal vessel wall, the confluent endothelium serves as a barrier between luminally secreted substances and the underlying SMC. In the shared media systems, EC confluence does not have this barrier effect. An ideal physiologic model would involve the culture of EC directly on SMC. Several such bilayer methods have been described. EC can be plated directly on top of established, growth-arrested cultures of SMC,4or they can be plated with an interven~*~~~~*~ ing layer of some matrix element, generally either fibronectin or c ~ l l a g e n . ~Van Buul-Wortelboer et all6have developed a method in which SMC are allowed to grow into a collagen gel, after which EC are plated on the gel. All of these methods produce bilayers that more closely resemble the vessel wall. These bilayer systems are the most physiologic of any available, preserving both direct cell-cell contact and polarity of EC-SMC interaction. Unfortunately, they are difficult to establish and maintain. Prior to the addition of EC, the SMC culture must be quiescent or overgrowth of SMC will occur. When EC are plated, heparin can be added to inhibit further SMC proliferation. While this is generally effective in preventing SMC overgrowth, the use of heparin is a confounding factor that may significantly affect the outcome of an experiment. If EC confluence is not achieved shortly after the cells are plated, SMC overgrowth may occur despite the use of heparin. In addition, cell associated collagenases can degrade collagen gels, resulting in a shortened lifespan for these types of bilayers. Because of these limitations, such systems cannot be used to study the growth of cells to confluence, and, generally, are not sufficiently stable to allow time course experiments. Separating the cells for study is also very difficult. The use of collagenase digestion to selectively remove the EC layer from the underlying SMC layer is complicated by variations in enzyme activity, which can result in an unreliable separation of the two cell types. Therefore, while these methods are more physiologic, they are often impractical. The Transwell co-culture method described in this paper appears to offer some distinct advantages over the methods currently available for the study of cell-cell interactions. Use of the Transwell system allows co-culture to be easily established and maintained. The SMC are plated in standard tissue culture wells. An initial incubation period before the co-culture is established assures an even distribution of SMC over the surface of the culture dish. If the inserts are placed before the SMC have attached to the dish, the population density of the SMC will be less uniform. Gelatin coating ofthe filter is not essential, but EC monolayers grown on gelatincoated filters appear to be more stable than those grown on bare filters. Cells are easily dislodged from bare filters and major cell loss can occur with media changes. Younger cells adhere better to the bare polycarbonate filters than older cells. Gelatin coating increases the adherence of older cells, allowing their use in this system. The plating of cells and their subsequent management is no different than with standard culture systems. Cells are easily separated for study using standard tissue culture methods, such as enzymatic release with either trypsin or collagenase, or by various dissolution methods. In addition, extracellular matrices can be studied separately. Because the cells are in separate compartments, growth studies from sparse population densities to confluence are possible without SMC overgrowth. These cocultures can be maintained for several weeks. Transwell co-culture preserves most of the basic physiologic relationships between EC and SMC. The system has a bilayer morphology. When the EC in the upper chamber are subconfluent, there is free communication between the two chambers. However, once EC confluence is achieved, the endothelial monolayer regulates trans-
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
Aortic EC and SMC Co-Culture
493
port between the two compartments. Polarity ofthe cell interactions is also preserved, with communication between the albuminal surface of the EC and the SMC, despite EC confluence. Using the Transwell method, we have documented interaction between EC and SMC. In subconfluent layers of both cell types, cocultivation with the opposite cell type produced a significant increase in cell proliferation. This effect is most likely the result of free diffusion of substances secreted by these cells which pass through the porous polycarbonate filter. Although the distance between the EC and SMC is less than 1 mm, the fact that the cells are not in direct contact is a major flaw of this system. While most of the in vivo interactions of these two cell types should be preserved, the activity of substances such as FGF and heparan sulfates, which are secreted into the extracellular matrix, may be attenuated or lost altogether in the absence of direct cell-cell or cellmatrix contact. Despite this lack of direct contact between the cell types, it is felt that the Transwell method offers significant advantages over currently available methodology. In summary, co-culture of EC and SMC using the Transwell system provides a useful method to study the interaction of EC and SMC. The system is simple to establish and remains stable over time. Cell growth can be studied from very sparse populations to confluence and beyond. In addition, cells and their matrices can be easily and reliably separated for study. Despite the lack of direct cell-cell contact, the Transwell system provides a co-culture system that allows study of the dynamic interaction between EC and SMC. We believe that this method provides an opportunity to greatly expand our understanding of how cells of the vascular wall interact in both normal and disease states, and may provide a system with which to assess interventions directed toward the alteration of disease processes.
References 1 . Clowes AW. Arterial Endothelial-Smooth Muscle Cell Interactions: Evaluation and Treatment of’ Upper and Lower Extremity Diseases. New York: Grune 8~ Stratton;
1984:25-38. 2. Staiano-Coico L, Hajjar DP, Hefton JM, Hajjar KA, Kimmel M. Interactions of arterial cells, 111: stathmokinetic analysis of smooth muscle cells cocultured with endothelial cells. J Cell Phys. 1988;134:485-490. 3. Campbell JH, Campbell GR. Endothelial cell influence on vascular smooth muscle cell phenotype. Annu Rev Physiol. 1986;48:295-306. 4. Memlees MJ, Scott L. Interaction of aortic endothelial and smooth muscle cells in culture: effect on glycosaminoglycan levels. Atherosclerosis. 198 1 ;39:147- 16 1 . 5. Castellot JJ. Addonizio ML, Rosenberg R, Karovsky MJ. Cultured endothelial cells produce a heparin like inhibitor of smooth muscle cell growth. J Cell Biol. 198 1;90:372-379. 6. Hajjar DP, Falcone DJ, Amberson JB, Hefton JM. Interaction of arterial cells, I: endothelial cells alter cholesterol metabolism in co-cultured smooth muscle cells. J Lipid Res. 1985;26:1212-1223. 7. Davies PF, Truskey GA, Warren HB, OConnor SE, Eisenhaure BH. Metabolic cooperation between vascular endothelial cells and smooth muscle cells in co-culture: changes in low-density lipoprotein metabolism. J Cell Biol. 1985;101:871-879. 8. Hajjar DP, Marcus AJ, Hajjar KA. Interactions ofarterial cells: studies in the mechanisms of endothelial cell modulation of cholesterol metabolism in co-cultured smooth muscle cells. J Biol Chem. 1987;262:6976-698 1 .
J Invest Surg Downloaded from informahealthcare.com by Flinders University of South Australia on 01/17/15 For personal use only.
4 94
D.J. Graham et al.
9. Vanhoutle PM, Rubanyi GM, Miller VM, Houston DS. Modulation of vascular smooth muscle cell contraction by the endothelium. Annu Rev Physiol. 1986;48:307-320. 10. Booyse FM, Selak BJ, Rafelson ME. Culture of arterial endothelial cells: characterization and growth of bovine aortic cells. Thromb Haemorrh. 1973;34:825-839. 1 I . Chamley-Campbell J, Campbell GR, Ross R. The smooth muscle cell in culture. Phys Rev. 1979;59:1-6 1. 12. Gadjusek CM, DiCorleto P, Ross R, Schwartz S. An endothelial cell-derived growth factor. J Cell B i d . 1980;85:467-472. 13. Davies PF, Kerr C. Co-cultivation of vascular endothelial and smooth muscle cells using microcarrier techniques. Exp Cell Res. 1982;141:455-459. 14. Jones PA. Construction of an artificial blood vessel wall from cultured endothelial and smooth muscle cells. Proc Natl Acad Sci USA. 1979;76:1882-1886. 15. Navab M, Hough GP, Stevenson LW, Drinkwater DC,Laks H, Fogelman AM. Monocyte migration into the subendothelial space of a coculture of adult human aortic endothelial and smooth muscle cells. J Clin Invest. 1988;82:1853-1863. 16. Van Bud-Wortelboer MF, Brinkman HJM, Dingeman KP, DeGroot PG, van Aken WG, van Mourik JA. Reconstruction of the vascular wall in vitro: a novel model to study interaction between endothelial and smooth muscle cells. Exp Cell Rex 1986;162:I5 1-1 58.
