Annals of Biomedical Engineering,Vol.19,pp.529-545,1991 0090-6964/91$3.00+.00 Printed in the USA.All rights reserved. Copyright1991 Pergamon Press plc Cellular Engineering Robert M.Nerem The 1991 ALZA Distinguished Lecture Biomedical Engineering Society Annual Meeting Atlanta,GA (Received 4/22/91) Cellular engineering applies the principles and methods of engineering to the prob- lems of cell and molecular biology of both a basic and applied nature.As biomedi- cal engineering has shifted from the organ and tissue level to the cellular and sub-cellular level,cellular engineering has emerged as a new area.A cornerstone of much of this activity is cell culture technology,i.e.,the ability to grow living cells in the artificial environment of a laboratory.Cellular engineering includes the role of engineering in both basic cell biology research and in the making of products which use living cells,e.g.,tissue engineering and bioprocess engineering.The former in- volves the use of living cells in the development of biological substitutes for the res- toration or replacement of function,and the latter the use of living cells to manufacture a biochemical product,e.g.,through the use of recombinant DNA tech- nology.In fact,as biomedical engineering has expanded to include the cellular level, and bioprocess engineering has shifted in interest from microbial organisms to include mammalian cells,there are intellectual issues in which an interest is shared by these two formerly separate areas of engineering activity.Cellular engineering thus tran- scends the field of biomedical engineering. Keywords-Cellular engineering,Tissue engineering,Bioprocess engineering,Cell and molecular biology. INTRODUCTION Over the past twenty-five years the field of biomedical engineering has undergone significant development,evolving in many different ways.For example,the contri- butions by engineers to the archival literature for biomedical research has been steadily increasing.The medical device and product industry,including instrumen- Acknowledgments-This review was written based on the author's own research activities which are currently supported by National Science Foundation Grant ECS-8815656 and National Institutes of Health Grants HL-26890 and HL-41175.The author thanks P.Girard,T.Sambanis,T.Wick,and C.Zhu,his col- leagues at Georgia Tech who work in the area of cellular engineering.The author also thanks his research collaborators,R.W.Alexander,B.C.Berk,P.Delafontaine,M.J.Levesque,M.Sato,C.J.Schwartz,and E.A.Sprague,for their contributions to the work and ideas reflected in this paper.Finally,the author thanks his students who keep coming up with new ideas and who will be the cellular engineers of tomorrow *The lecture was delivered on April 22,1991 by R.M.Nerem,Parker H.Petit Professor for Engireering in Medicine,School of Mechanical Engineering,Georgia Institute of Technology,Atlanta,GA 30332-0405. 529
Annals ofBiomedicalEngineering, Vol. 19, pp. 529-545, 1991 0090-6964/91 $3.00 + .00 Printed in the USA. All rights reserved. Copyright 9 1991 Pergamon Press plc Cellular Engineering Robert M. Nerem The 1991 ALZA Distinguished Lecture* Biomedical Engineering Society Annual Meeting Atlanta, GA (Received 4/22/91) Cellular engineering applies the principles and methods of engineering to the problems of cell and molecular biology of both a basic and applied nature. As biomedical engineering has shifted from the organ and tissue level to the cellular and sub-cellular level, cellular engineering has emerged as a new area. A cornerstone of much of this activity is cell culture technology, i.e., the ability to grow living cells in the artificial environment of a laboratory. Cellular engineering includes the role of engineering in both basic cell biology research and in the making of products which use living cells, e.g., tissue engineering and bioprocess engineering. The former involves the use of living cells in the development of biological substitutes for the restoration or replacement of function, and the latter the use of living cells to manufacture a biochemical product, e.g., through the use of recombinant DNA technology. In fact, as biomedical engineering has expanded to include the cellular level, and bioprocess engineering has shifted in interest from microbial organisms to include mammalian cells, there are intellectual issues in which an interest is shared by these two formerly separate areas of engineering activity. Cellular engineering thus transcends the field of biomedical engineering. Keywords- Cellular engineering, Tissue engineering, Bioprocess engineering, Cell and molecular biology. INTRODUCTION Over the past twenty-five years the field of biomedical engineering has undergone significant development, evolving in many different ways. For example, the contributions by engineers to the archival literature for biomedical research has been steadily increasing. The medical device and product industry, including instrumenAcknowledgments-This review was written based on the author's own research activities which are currently supported by National Science Foundation Grant ECS-8815656 and National Institutes of Health Grants HL-26890 and HL-41175. The author thanks P. Girard, T. Sambanis, T. Wick, and C. Zhu, his colleagues at Georgia Tech who work in the area of cellular engineering. The author also thanks his research collaborators, R.W. Alexander, B.C. Berk, P. Delafontaine, M.J. Levesque, M. Sato, C.J. Schwartz, and E.A. Sprague, for their contributions to the work and ideas reflected in this paper. Finally, the author thanks his students who keep coming up with new ideas and who will be the cellular engineers of tomorrow. *The lecture was delivered on April 22, 1991 by R.M. Nerem, Parker H. Petit Professor for Engiv.eering in Medicine, School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0405. 529
530 R.M.Nerem tation and imaging,has grown substantially,and although,in general,one would not characterize this as a high technology industry,there clearly are areas of commercial application where the limits of our knowledge are being pushed.In the last two de- cades the number of individuals who have identified themselves as biomedical engi- neers,or bioengineers,has dramatically increased,and the number of programs at academic institutions,and the number of students participating in these programs, continues to grow.Finally,both the engineer and the physical scientist are increas- ingly being recognized as important members of the interdisciplinary teams,and their approaches as necessary for research and development at the forefront of the biomed- ical field. With this evolution comes a different type of change-one in which the spectrum of activity has been extended from the organ and tissue level to include cellular phenomena.Just as medicine in general has moved to focus on cell and molecular biology,so have engineers participating in biomedical research and in health care tech- nology applications.This has given rise to what some might call a sub-specialty of bio medical engineering,i.e.,the field of cellular engineering.However,this emerging area of cellular engineering is not just a sub-specialty,because it transcends the nor- mal borders of biomedical engineering,reaching beyond to provide a bridge to the field of biochemical engineering or bioprocess engineering. Just what is cellular engineering?As defined here it is the application of the princi- ples and methods of engineering to problems in cell and molecular biology of both a basic and applied nature.To elaborate on this,let us examine the various elements of this definition.The application of the principles and methods of engineering means to include the ability to quantify information and to establish interrelationships,as well as to model biological,chemical,and physical phenomena.In applying this to problems in cell and molecular biology,it is both cellular and sub-cellular processes which are of interest.Finally,in including applications of both a basic and applied nature,the spectrum of activities ranges from very basic research,i.e.,investigations in cell biology conducted with an engineering perspective,to the development and manufacturing of products based on this science,i.e.,the commercialization of the technology arising out of the science of cell and molecular biology. The development of the technology necessary to grow cells in the laboratory has been critical in the advancement of cellular engineering.In fact,if there is a subtitle to the theme of this presentation,it might be "sex and the single cell."This is because the act of cell replication is an underlying topic,a thread,interwoven throughout the remainder of this text.In the next section we will briefly examine cell culture tech- nology,i.e.,the technology of growing cells in the laboratory. CELL CULTURE TECHNOLOGY The biological cell is the basic sub-unit of any living system,the simplest unit that can exist as an independent living system(2).An individual cell has the ability to rep- licate,to differentiate,to migrate,to communicate,and to perform a whole host of other functions.In effect,this biological cell is a very social animal.It also is an amazingly complex system,the understanding of which challenges our engineering abilities to the utmost.In doing research on cellular and sub-cellular processes,not only can engineers contribute to the basic understanding of the biology of cells,but our engineering skills can be expanded,perhaps far more than we suspect
530 R.M. Nerem tation and imaging, has grown substantially, and although, in general, one would not characterize this as a high technology industry, there clearly are areas of commercial application where the limits of our knowledge are being pushed. In the last two decades the number of individuals who have identified themselves as biomedical engineers, or bioengineers, has dramatically increased, and the number of programs at academic institutions, and the number of students participating in these programs, continues to grow. Finally, both the engineer and the physical scientist are increasingly being recognized as important members of the interdisciplinary teams, and their approaches as necessary for research and development at the forefront of the biomedical field. With this evolution comes a different type of change-one in which the spectrum of activity has been extended from the organ and tissue level to include cellular phenomena. Just as medicine in general has moved to focus on cell and molecular biology, so have engineers participating in biomedical research and in health care technology applications. This has given rise to what some might call a sub-specialty of biomedical engineering, i.e., the field of cellular engineering. However, this emerging area of cellular engineering is not just a sub-specialty, because it transcends the normal borders of biomedical engineering, reaching beyond to provide a bridge to the field of biochemical engineering or bioprocess engineering. Just what is cellular engineering? As defined here it is the application of the principles and methods of engineering to problems in cell and molecular biology of both a basic and applied nature. To elaborate on this, let us examine the various elements of this definition. The application of the principles and methods of engineering means to include the ability to quantify information and to establish interrelationships, as well as to model biological, chemical, and physical phenomena. In applying this to problems in cell and molecular biology, it is both cellular and sub-cellular processes which are of interest. Finally, in including applications of both a basic and applied nature, the spectrum of activities ranges from very basic research, i.e., investigations in cell biology conducted with an engineering perspective, to the development and manufacturing of products based on this science, i.e., the commercialization of the technology arising out of the science of cell and molecular biology. The development of the technology necessary to grow cells in the laboratory has been critical in the advancement of cellular engineering. In fact, if there is a subtitle to the theme of this presentation, it might be "sex and the single cell." This is because the act of cell replication is an underlying topic, a thread, interwoven throughout the remainder of this text. In the next section we will briefly examine cell culture technology, i.e., the technology of growing cells in the laboratory. CELL CULTURE TECHNOLOGY The biological cell is the basic sub-unit of any living system, the simplest unit that can exist as an independent living system (2). An individual cell has the ability to replicate, to differentiate, to migrate, to communicate, and to perform a whole host of other functions. In effect, this biological cell is a very social animal. It also is an amazingly complex system, the understanding of which challenges our engineering abilities to the utmost. In doing research on cellular and sub-cellular processes, not only can engineers contribute to the basic understanding of the biology of cells, but our engineering skills can be expanded, perhaps far more than we suspect
Cellular Engineering 531 Although research designed to investigate the properties and behavior of biologi- cal cells goes back many centuries,it is in the last fifty years that the level of activ- ity in this field has,literally,exploded.An important part of this increase was due to the advent of cell culture,i.e.,the ability to grow living cells in the artificial en- vironment of a laboratory (19).In this out of body experience,cells not only survive, but can multiply and even express differentiated properties.Through the use of cell culture technology,i.e.,the ability to grow cells under controlled laboratory condi- tions,new opportunities for basic research have opened up.Furthermore,this tech- nology has led to commercial applications which will be discussed in the second half of this text. Modern cell culture dates back to the beginning of this century.A particularly im- portant contribution was that of Alexis Carrel,a French scientist working at the Rockefeller Research Institute in New York(27).On January 17,1912 he placed a tiny slice of heart muscle taken from a chick embryo in a culture medium.This culture continued for thirty-four years,until two years after Carrel's own death.Along the way,the heart muscle cells expired and what continued to propagate were fibroblasts. Still,Carrel's historic chick-cell culture flourished for more than thirty years.The fact that it was the fibroblasts,not the heart muscle cells,which flourished for thirty-four years,points out one fact about cell culture;some cells are not as easy to grow as oth- ers.Cells which are not easy to grow are called recalcitrant cells. At times,one is interested in using primary cells,i.e.,cells prepared directly from the tissues of an organism.In other cases,cells are passaged through subculturing, and in this way are maintained for extensive periods of time.Most cells in culture have a limited lifespan,i.e.,after a finite number of divisions in culture,they die. However,occasionally,cells become immortal and can be propagated indefinitely as a cell line.Such cells are often referred to as transformed.Whereas cells cultured from tissue are anchorage-dependent,i.e.,adherence to a surface is required for survival, transformed cells often grow in suspension.Whatever the case,a wide variety of cells is now available for basic research and for more applied studies. Because of the complex nutritional requirements of mammalian cells,the medium in which the cells are cultured is important if a specific cell type is to be successfully grown.Traditionally,what has been used in most cases is a basal medium supple- mented with serum;this has usually been fetal calf,newborn calf,horse,or human serum at concentrations from 2 percent to 20 percent or greater.For many types of cells,the addition of serum,which provides growth factors,hormones,transferrin (an iron-binding protein),selenium(a trace element necessary for the growth of hu- man cells),and other required nutrients,is essential if the cells are to grow However,serum is a complex fluid and can be variable;thus each lot must be tested prior to use.There are also certain drawbacks to serum,e.g.,in some cases this growth promoter can even be toxic to cells.Because of this,there have been a num- ber of attempts to develop serum substitute products which can replace all or part of the serum required in a medium.In general,transformed cell lines have simpler me- dia requirements than untransformed cell lines,and thus are more likely to grow in serum-free culture.However,even though there are increasing reports of success in using low-serum or serum-free media,there still are many problems to be solved.For example,different cell lines of the same cell type may have different media require- ments,particularly with serum-free media.The fact is that,in spite of the progress which has been made,serum has so far defied all attempts at simulation
Cellular Engineering 531 Although research designed to investigate the properties and behavior of biological cells goes back many centuries, it is in the last fifty years that the level of activity in this field has, literally, exploded. An important part of this increase was due to the advent of cell culture, i.e., the ability to grow living cells in the artificial environment of a laboratory (19). In this out of body experience, cells not only survive, but can multiply and even express differentiated properties. Through the use of cell culture technology, i.e., the ability to grow cells under controlled laboratory conditions, new opportunities for basic research have opened up. Furthermore, this technology has led to commercial applications which will be discussed in the second half of this text. Modern cell culture dates back to the beginning of this century. A particularly important contribution was that of Alexis Carrel, a French scientist working at the Rockefeller Research Institute in New York (27). On January 17, 1912 he placed a tiny slice of heart muscle taken from a chick embryo in a culture medium. This culture continued for thirty-four years, until two years after Carrel's own death. Along the way, the heart muscle cells expired and what continued to propagate were fibroblasts. Still, Carrel's historic chick-cell culture flourished for more than thirty years. The fact that it was the fibroblasts, not the heart muscle cells, which flourished for thirty-four years, points out one fact about cell culture; some cells are not as easy to grow as others. Cells which are not easy to grow are called recalcitrant cells. At times, one is interested in using primary cells, i.e., cells prepared directly from the tissues of an organism. In other cases, cells are passaged through subculturing, and in this way are maintained for extensive periods of time. Most cells in culture have a limited lifespan, i.e., after a finite number of divisions in culture, they die. However, occasionally, cells become immortal and can be propagated indefinitely as a cell line. Such cells are often referred to as transformed. Whereas cells cultured from tissue are anchorage-dependent, i.e., adherence to a surface is required for survival, transformed cells often grow in suspension. Whatever the case, a wide variety of cells is now available for basic research and for more applied studies. Because of the complex nutritional requirements of mammalian cells, the medium in which the cells are cultured is important if a specific cell type is to be successfully grown. Traditionally, what has been used in most cases is a basal medium supplemented with serum; this has usually been fetal calf, newborn calf, horse, or human serum at concentrations from 2 percent to 20 percent or greater. For many types of cells, the addition of serum, which provides growth factors, hormones, transferrin (an iron-binding protein), selenium (a trace element necessary for the growth of human cells), and other required nutrients, is essential if the cells are to grow. However, serum is a complex fluid and can be variable; thus each lot must be tested prior to use. There are also certain drawbacks to serum, e.g., in some cases this growth promoter can even be toxic to cells. Because of this, there have been a number of attempts to develop serum substitute products which can replace all or part of the serum required in a medium. In general, transformed cell lines have simpler media requirements than untransformed cell lines, and thus are more likely to grow in serum-free culture. However, even though there are increasing reports of success in using low-serum or serum-free media, there still are many problems to be solved. For example, different cell lines of the same cell type may have different media requirements, particularly with serum-free media. The fact is that, in spite of the progress which has been made, serum has so far defied all attempts at simulation
532 R.M.Nerem For anchorage-dependent cells,another important factor in cell culture is the sur- face on which the cells are grown.The characteristics of this surface,e.g.,its micro- structure and surface chemistry,can induce changes in the cell.In fact,it is the interaction of a cell with its environment which may determine the structure and func- tion of a cell.This environment includes the medium,the surface to which the cells are adherent,and other factors,e.g.,the presence of flow. Cell culture is important because in many cases cellular engineering efforts are cen- tered around its use.At Georgia Tech this focus is the Bioengineering Center's Mam- malian Cell Culture Laboratory.Associated with this facility are 5 faculty and nearly 20 graduate students.The laboratory's research projects cover the spectrum from ba- sic to applied research,and the types of cells used in this laboratory can serve as an example of the wide range available for research today.Included are a variety of vas- cular endothelial cells (bovine aortic human dermal microvascular,and human um- bilical vein)and smooth muscle cells (bovine,and rat aortic)which are used in our studies of vascular biology and in research related to the development of tissue- engineered vascular prostheses.In addition,we use a number of different cell lines in our research.This includes 3T3 mouse fibroblasts,Bowes melanoma cells,BTC3 mouse pancreatic cells,and two lines of the AtT-20 mouse pituitary cell,one genet- ically engineered to secrete insulin,and the other human growth hormone.These are used in projects related to either tissue engineering or bioprocess engineering. Cell culture technology can thus provide the foundation for a wide array of activ- ities.As illustrated in Fig.1,this not only includes basic cell biology research,but also such applications as tissue engineering and bioprocess engineering.These will be dis- cussed later. Cell Culture Technology Cell Bioprocess Biology Engineering Tissue Other Engineering Applications FIGURE 1.The development of cell culture technology has not only resulted in an acceleration of basic research in cell and molecular biology,it also has led to the use of living cells in commercial product applications,e.g.,tissue engineering and bioprocess engineering
532 R.M. Nerem For anchorage-dependent cells, another important factor in cell culture is the surface on which the cells are grown. The characteristics of this surface, e.g., its microstructure and surface chemistry, can induce changes in the cell. In fact, it is the interaction of a cell with its environment which may determine the structure and function of a cell. This environment includes the medium, the surface to which the cells are adherent, and other factors, e.g., the presence of flow. Cell culture is important because in many cases cellular engineering efforts are centered around its use. At Georgia Tech this focus is the Bioengineering Center's Mammalian Cell Culture Laboratory. Associated with this facility are 5 faculty and nearly 20 graduate students. The laboratory's research projects cover the spectrum from basic to applied research, and the types of cells used in this laboratory can serve as an example of the wide range available for research today. Included are a variety of vascular endothelial cells (bovine aortic human dermal microvascular, and human umbilical vein) and smooth muscle cells (bovine, and rat aortic) which are used in our studies of vascular biology and in research related to the development of tissueengineered vascular prostheses. In addition, we use a number of different cell lines in our research. This includes 3T3 mouse fibroblasts, Bowes melanoma cells, ~3TC3 mouse pancreatic cells, and two lines of the ART-20 mouse pituitary cell, one genetically engineered to secrete insulin, and the other human growth hormone. These are used in projects related to either tissue engineering or bioprocess engineering. Cell culture technology can thus provide the foundation for a wide array of activities. As illustrated in Fig. 1, this not only includes basic cell biology research, but also such applications as tissue engineering and bioprocess engineering. These will be discussed later. Cell Culture Technology Cell Biology Bioprocess Engineering Tissue Engineering Other Applications FIGURE 1. The development of cell culture technology has not only resulted in an acceleration of basic research in cell and molecular biology, it also has led to the use of living cells in commercial product applications, e.g., tissue engineering and bioprocess engineering
Cellular Engineering 533 As described earlier,there are many factors which influence our ability to grow cells,however,to a large degree these are at best only partially understood.The re- sult is that,although "the cultural revolution has begun"as advertised a few years ago by Invitron,a St.Louis-based company,cell culture technology remains more an art than a science.Still,the technology used in culturing mammalian cells has become a cornerstone for the development of cell and molecular biology as a scientific dis- cipline and in the commercial applications arising out of this basic research. BASIC RESEARCH IN CELL BIOLOGY As noted in the previous section,the advent of cell culture technology has helped to dramatically accelerate advances in cell biology.This is true of the entire spectrum of research on cellular and sub-cellular phenomena.The engineer,through the vari- ety of ways in which the principles and methods of engineering can be applied,has been a participant in this area.Another key factor which links engineering to cell bi- ology is its relationship to the biophysics of a cell,i.e.,the role of physical mecha- nisms and the influence of physical factors on cellular behavior. Because most biomedical researchers,e.g.,MDs and life scientists,have a train- ing which,in general,is biochemistry based,and which does not emphasize physics, these researchers have tended to focus more on the biochemistry of a cell and not on the biophysics of a cell.However,physical factors have been demonstrated to be im- portant,and one example of this is in the influence of mechanical stresses and the re- sulting mechanics of deformation.Engineers involved in such studies of biomechanics have contributed and continue to contribute to our understanding of organ physiol- ogy and tissue behavior.They are now applying their knowledge to the investigation of the mechanical nature of much of cellular phenomena and the application of the principles of mechanics in order to understand the structure and function of cells(38). This type of cellular biomechanical phenomena can be illustrated by the process of cell division.In examining this process for an eukaryotic cell,one must consider the entire reproductive cycle of the cell(2).This cell cycle includes a number of sep- arate phases.The actual process of cell division is called M phase (M=mitosis),and the next cycle starts with the G phase(G gap)which is the period of time between the end of M phase and the beginning of DNA synthesis.The period of DNA syn- thesis is called S phase(S synthesis),and it ends when the DNA content of the nu- cleus has doubled and the chromosomes have replicated.The cell then enters the G2 phase,which may be viewed as preparatory for the actual process of cell division.The G2 phase is followed by M phase which in itself is composed of two specific events, mitosis and cytokinesis.Mitosis involves the splitting of the content of the nucleus, which causes a variety of intracellular movements,of mechanical nature,to take place during the different mitotic phases.Also,during cytokinesis,when the cell divides its cytoplasm,there is a very distinct mechanical event when the contractile ring,which has formed from cytoskeletal components,cleaves the cell into two daughter cells. These,of course,are not purely mechanical events.They are more accurately called mechano-chemical phenomena,and there is in fact an increasing recognition of the importance of such phenomena,and the strong coupling between structure and func- tion in an eukaryotic cell. The study of the influence of hemodynamics on vascular biology/pathobiology, and as a factor in the localization of atherosclerosis (45),is one area of biomedical
Cellular Engineering 533 As described earlier, there are many factors which influence our ability to grow cells, however, to a large degree these are at best only partially understood. The result is that, although "the cultural revolution has begun" as advertised a few years ago by Invitron, a St. Louis-based company, cell culture technology remains more an art than a science. Still, the technology used in culturing mammalian cells has become a cornerstone for the development of cell and molecular biology as a scientific discipline and in the commercial applications arising out of this basic research. BASIC RESEARCH IN CELL BIOLOGY As noted in the previous section, the advent of cell culture technology has helped to dramatically accelerate advances in cell biology. This is true of the entire spectrum of research on cellular and sub-cellular phenomena. The engineer, through the variety of ways in which the principles and methods of engineering can be applied, has been a participant in this area. Another key factor which links engineering to cell biology is its relationship to the biophysics of a cell, i.e., the role of physical mechanisms and the influence of physical factors on cellular behavior. Because most biomedical researchers, e.g., MDs and life scientists, have a training which, in general, is biochemistry based, and which does not emphasize physics, these researchers have tended to focus more on the biochemistry of a cell and not on the biophysics of a cell. However, physical factors have been demonstrated to be important, and one example of this is in the influence of mechanical stresses and the resulting mechanics of deformation. Engineers involved in such studies of biomechanics have contributed and continue to contribute to our understanding of organ physiology and tissue behavior. They are now applying their knowledge to the investigation of the mechanical nature of much of cellular phenomena and the application of the principles of mechanics in order to understand the structure and function of cells (38). This type of cellular biomechanical phenomena can be illustrated by the process of cell division. In examining this process for an eukaryotic cell, one must consider the entire reproductive cycle of the cell (2). This cell cycle includes a number of separate phases. The actual process of cell division is called M phase (M = mitosis), and the next cycle starts with the G1 phase (G = gap) which is the period of time between the end of M phase and the beginning of DNA synthesis. The period of DNA synthesis is called S phase (S = synthesis), and it ends when the DNA content of the nucleus has doubled and the chromosomes have replicated. The cell then enters the G 2 phase, which may be viewed as preparatory for the actual process of cell division. The G2 phase is followed by M phase which in itself is composed of two specific events, mitosis and cytokinesis. Mitosis involves the splitting of the content of the nucleus, which causes a variety of intracellular movements, of mechanical nature, to take place during the different mitotic phases. Also, during cytokinesis, when the cell divides its cytoplasm, there is a very distinct mechanical event when the contractile ring, which has formed from cytoskeletal components, cleaves the cell into two daughter cells. These, of course, are not purely mechanical events. They are more accurately called mechano-chemical phenomena, and there is in fact an increasing recognition of the importance of such phenomena, and the strong coupling between structure and function in an eukaryotic cell. The study of the influence of hemodynamics on vascular biology/pathobiology, and as a factor in the localization of atherosclerosis (45), is one area of biomedical
534 R.M.Nerem research where the importance of physical factors as an effect on cellular behavior is recognized,and in which engineers have been involved.In the early days this in- volvement was due to the realization that the physical forces imposed by flowing blood might be an important influence on the biology of the arterial wall.In vivo studies demonstrated differences in endothelial cell shape and F-actin localization,the permeability of the endothelium,and the recruitment of monocytes into the arterial wall in regions of differing hemodynamic environment (35).Such studies also sug- gested a possible influence of hemodynamic environment on the endothelial cell turn- over rate (9),the process of cell replication whereby a confluent endothelial monolayer is maintained,as aging cells die.Unfortunately,it is very difficult to quan- tify in vivo the detailed characteristics associated with the flow of blood;thus these in vivo results,in terms of any demonstration of a hemodynamic effect,had to be viewed as qualitative in nature. For this reason,more than a decade ago a number of laboratories began to use cell culture in the study of flow effects on vascular biology.The first effort in studying cultured vascular endothelial cell monolayers in the presence of flow was that of Dewey et al.(17)who used a cone-plate viscometer to investigate shear stress effects. Other laboratories,including my own (29,30),have joined in the use of cell culture to study the effects of a cell's biomechanical environment.This has included studies on the influence of a fluid-imposed shear stress,on changes due to the direct effect of pressure,and on alterations associated with the cyclic stretching of cells grown on a compliant membrane(35).Recently a workshop on mechanical stress effects on vas- cular cells was held in Atlanta,and the abstract proceedings of this meeting are rep- resentative of the current state of knowledge(3).What is clear from this meeting is that,in response to its biomechanical environment,the vascular endothelial cell re- sponds by altering its cytoskeletal structure.With this alteration there is a change in cell shape and an alignment of the cell's major axis within the stress field,as illus- trated in Fig.2,and there are a variety of associated changes in cell function. As part of this there is an influence of mechanical stress on the cell's ability to rep- licate.As noted earlier,in vivo observations suggest an influence of a hemodynamic environment on endothelial cell turnover rates,and it is believed that higher rates oc- cur in regions characterized hemodynamically as low-shear stress in comparison to those characterized as high-shear stress(9).Whether or not it is in fact a shear-stress effect cannot be determined from in vivo experiments.Furthermore,it is well known that the turnover rate of an endothelial monolayer in culture is far higher than that observed in vivo in endothelium.Thus,in our own laboratory we have focused on studying the effects of laminar flow and the associated fluid-imposed shear stress on the process of cell replication in cultured endothelial monolayers.What we have found is that in the presence of shear stress,the overall proliferation rate of sub-con- fluent endothelial cells is altered(30).For endothelial cells cultured on a polyester sub- strate,e.g.,Thermanox or Mylar,and for low-shear stresses,e.g.,5 dynes/cm2 or less,any effect is absent as was noted in the earlier study of Dewey et al.(17).How- ever,for shear stresses in the range from 30 to 90 dynes/cm2,cell proliferation is sig- nificantly slowed,with the growth at 90 dynes/cm2 being almost totally arrested,at least for 24 hours. These earlier results were based on cell density data,and to confirm this 3H-thy- midine incorporation and autoradiography measurements were performed.These were conducted post-shear,i.e.,after removal of the coverslip from the flow cham-
534 R.M. Nerem research where the importance of physical factors as an effect on cellular behavior is recognized, and in which engineers have been involved. In the early days this involvement was due to the realization that the physical forces imposed by flowing blood might be an important influence on the biology of the arterial wall. In vivo studies demonstrated differences in endothelial cell shape and F-actin localization, the permeability of the endothelium, and the recruitment of monocytes into the arterial wall in regions of differing hemodynamic environment (35). Such studies also suggested a possible influence of hemodynamic environment on the endothelial cell turnover rate (9), the process of cell replication whereby a confluent endothelial monolayer is maintained, as aging cells die. Unfortunately, it is very difficult to quantify in vivo the detailed characteristics associated with the flow of blood; thus these in vivo results, in terms of any demonstration of a hemodynamic effect, had to be viewed as qualitative in nature. For this reason, more than a decade ago a number of laboratories began to use cell culture in the study of flow effects on vascular biology. The first effort in studying cultured vascular endothelial cell monolayers in the presence of flow was that of Dewey et aL (17) who used a cone-plate viscometer to investigate shear stress effects. Other laboratories, including my own (29,30), have joined in the use of cell culture to study the effects of a cell's biomechanical environment. This has included studies on the influence of a fluid-imposed shear stress, on changes due to the direct effect of pressure, and on alterations associated with the cyclic stretching of cells grown on a compliant membrane (35). Recently a workshop on mechanical stress effects on vascular cells was held in Atlanta, and the abstract proceedings of this meeting are representative of the current state of knowledge (3). What is clear from this meeting is that, in response to its biomechanical environment, the vascular endothelial cell responds by altering its cytoskeletal structure. With this alteration there is a change in cell shape and an alignment of the cell's major axis within the stress field, as illustrated in Fig. 2, and there are a variety of associated changes in cell function. As part of this there is an influence of mechanical stress on the cell's ability to replicate. As noted earlier, in vivo observations suggest an influence of a hemodynamic environment on endothelial cell turnover rates, and it is believed that higher rates occur in regions characterized hemodynamically as low-shear stress in comparison to those characterized as high-shear stress (9). Whether or not it is in fact a shear-stress effect cannot be determined from in vivo experiments. Furthermore, it is well known that the turnover rate of an endothelial monolayer in culture is far higher than that observed in vivo in endothelium. Thus, in our own laboratory we have focused on studying the effects of laminar flow and the associated fluid-imposed shear stress on the process of cell replication in cultured endothelial monolayers. What we have found is that in the presence of shear stress, the overall proliferation rate of sub-confluent endothelial cells is altered (30). For endothelial cells cultured on a polyester substrate, e.g., Thermanox or Mylar, and for low-shear stresses, e.g., 5 dynes/cm 2 or less, any effect is absent as was noted in the earlier study of Dewey et al. (17). However, for shear stresses in the range from 30 to 90 dynes/cm 2, cell proliferation is significantly slowed, with the growth at 90 dynes/cm 2 being almost totally arrested, at least for 24 hours. These earlier results were based on cell density data, and to confirm this 3H-thymidine incorporation and autoradiography measurements were performed. These were conducted post-shear, i.e., after removal of the coverslip from the flow cham-
Cellular Engineering 535 (A) (B) FIGURE 2.Photomicrographs of cultured BAEC grown on Thermanox under control conditions (a)and under a laminar steady state flow (shear stress =85 dynes/cm2)for 24 hours (b);flow from left to right.Reproduced from reference 28.published by permission of ASME
Cellular Engineering 535 (A) (B) FIGURE 2. Photomicrographs of cultured BAEC grown on Thermanox under control conditions (a) and under a laminar steady state flow (shear stress = 85 dynes/cm 2) for 24 hours (b), flow from left to right, Reproduced from reference 28. published by permission of ASME
536 R.M.Nerem ber,and these results suggest that the inhibition of cell proliferation by shear stress is at the level of DNA synthesis,and not due to endothelial cells rounding up during mitosis and being removed by flow(30).Furthermore,to determine whether the ef- fect of shear stress on endothelial cell growth was related to an alteration in cell cy- cle kinetics,both cell size and DNA content were analyzed by flow cytometry (33). For sub-confluent cells and following exposure to a shear stress of 30 dynes/cm2 for 24 hours,the percentage of sub-confluent cells in S phase was reduced significantly, with there simultaneously being an increase in the percentage of cells in Go/G phase.When these experiments were repeated with confluent cells,similar results were obtained.Thus,shear stress appears to inhibit the transition of endothelial cells from the Go/G phase to S phase in the cell cycle. In separate experiments,cells in both sheared and static cultures were videotaped at 400x magnification in order to visualize and quantify the process of mitosis and cytokinesis(46).Whereas in static culture a cell rounds up as part of the process of division,elongated endothelial cells in the presence of flow preserve their elongated, flattened shape over the entire duration of cell division.In some cases,the temporal nature of the intracellular mitotic events was observed to be altered.Thus,the influ- ence of flow is not only to hold up entry into S phase,but also to affect the cell di- vision process itself. Finally,alterations in the endothelial cell's growth state were further studied by analysis of the cell cycle-dependent accumulation of the messenger RNA(mRNA)for the proto-oncogene c-myc(8).Both control and shear stress treated endothelial cells were exposed to the potent agonist a-thrombin,and total cellular RNA was prepared and analyzed on northern blots.Using a probe for the proto-oncogene c-myc,we ob- served that c-myc levels at one hour increased 28 fold in endothelial cells in static cul- ture,but only 4 fold in endothelial cells which had been exposed to flow.Further experiments with other a-thrombin signal transduction events will need to be per- formed in order to delineate the mechanism involved;however,it is clear that the in- fluence of flow and the associated shear-stress level is manifest at the gene expression level. An obviously critical question is how does an endothelial cell recognize a mechan- ical stimulus,and having done so,how does it transduce that signal into the types of change in cell replication which we have observed?The intracellular events that reg- ulate the endothelial cell's growth program in response to shear stress are not known However,in many cell types,it is thought that the signaling mechanisms important to alterations in cell shape and cytoskeletal structure also are involved in growth con- trol.If this is true in regard to endothelial cell growth,then the recognition is perhaps a membrane event,possibly linked to the control of Ca++metabolism.There are re- ports of a transient elevation in intracellular calcium in response to flow (4),and we have seen this recently in our own laboratory (21).An important biochemical path- way controlling Ca+mobilization is the phospholipase C-mediated hydrolysis of polyphosphoinositides.In this pathway,activation of a cell surface receptor stimu- lates a phosphoinositide-specific phospholipase C to hydrolyze phosphatidylinositol 4,5 bisphosphate (PIP2)to form inositol 1,4,5,trisphosphate (IP3),a potent mobilizer of intracellular Ca++,and diacylglycerol,a stimulator of the Ca++/phos- pholipid-dependent enzyme,protein kinase C.Recently,it has been demonstrated that there is shear-stress stimulation of this pathway,with a peak in IP;being measured shortly after the onset of flow(37).Furthermore,studies in our laboratory have im-
536 R.M. Nerem ber, and these results suggest that the inhibition of cell proliferation by shear stress is at the level of DNA synthesis, and not due to endothelial cells rounding up during mitosis and being removed by flow (30). Furthermore, to determine whether the effect of shear stress on endothelial cell growth was related to an alteration in cell cycle kinetics, both cell size and DNA content were analyzed by flow cytometry (33). For sub-confluent cells and following exposure to a shear stress of 30 dynes/cm 2 for 24 hours, the percentage of sub-confluent cells in S phase was reduced significantly, with there simultaneously being an increase in the percentage of cells in G0/G1 phase. When these experiments were repeated with confluent cells, similar results were obtained. Thus, shear stress appears to inhibit the transition of endothelial cells from the Go/GI phase to S phase in the cell cycle. In separate experiments, cells in both sheared and static cultures were videotaped at 400• magnification in order to visualize and quantify the process of mitosis and cytokinesis (46). Whereas in static culture a cell rounds up as part of the process of division, elongated endothelial cells in the presence of flow preserve their elongated, flattened shape over the entire duration of cell division. In some cases, the temporal nature of the intracellular mitotic events was observed to be altered. Thus, the influence of flow is not only to hold up entry into S phase, but also to affect the cell division process itself. Finally, alterations in the endothelial cell's growth state were further studied by analysis of the cell cycle-dependent accumulation of the messenger RNA (mRNA) for the proto-oncogene c-myc (8). Both control and shear stress treated endothelial cells were exposed to the potent agonist c~-thrombin, and total cellular RNA was prepared and analyzed on northern blots. Using a probe for the proto-oncogene c-myc, we observed that c-myc levels at one hour increased 28 fold in endothelial cells in static culture, but only 4 fold in endothelial cells which had been exposed to flow. Further experiments with other a-thrombin signal transduction events will need to be performed in order to delineate the mechanism involved; however, it is clear that the influence of flow and the associated shear-stress level is manifest at the gene expression level. An obviously critical question is how does an endothelial cell recognize a mechanical stimulus, and having done so, how does it transduce that signal into the types of change in cell replication which we have observed? The intracellular events that regulate the endothelial cell's growth program in response to shear stress are not known. However, in many cell types, it is thought that the signaling mechanisms important to alterations in cell shape and cytoskeletal structure also are involved in growth control. If this is true in regard to endothelial cell growth, then the recognition is perhaps a membrane event, possibly linked to the control of Ca ++ metabolism. There are reports of a transient elevation in intracellular calcium in response to flow (4), and we have seen this recently in our own laboratory (21). An important biochemical pathway controlling Ca ++ mobilization is the phospholipase C-mediated hydrolysis of polyphosphoinositides. In this pathway, activation of a cell surface receptor stimulates a phosphoinositide-specific phospholipase C to hydrolyze phosphatidylinositol 4,5 bisphosphate (PIP2) to form inositol 1, 4, 5, trisphosphate (IP3), a potent mobilizer of intracellular Ca ++, and diacylglycerol, a stimulator of the Ca++/phos - pholipid-dependent enzyme, protein kinase C. Recently, it has been demonstrated that there is shear-stress stimulation of this pathway, with a peak in IP3 being measured shortly after the onset of flow (37). Furthermore, studies in our laboratory have im-
Cellular Engineering 537 plicated protein kinase C(PKC)as part of the signaling pathway which links a shear- stress-related mechanical signal to the intracellular events underlying alterations in cell morphology(22).This second messenger may also be involved in the control of the endothelial cell's growth program. It should be emphasized that in vivo endothelial cells reside in a flow environment, and thus to study vascular endothelial biology in static culture is at best a simulation of a region of flow stasis and at worst artifactual.Furthermore,endothelial cells re- spond differently to differing flow environments,and one should never collectively talk of flow as a single stimulus.Just as there are different chemical agonists,each with their own separate effect,there also are different types of flow agonists,i.e.,a variety of types of flow environments,each of which will have their own agonist ef- fect(23).Thus,as important as cell culture studies have been to the study of vascu- lar biology,it is clear that there is much more which needs to be done if we are to engineer the cell culture environment so as to make it truly simulate physiologic con- ditions.This includes the use of realistic flow environments,but other factors,e.g., the medium,extracellular matrix components,and the presence of neighboring cells, will also need to be included. TISSUE ENGINEERING Tissue engineering,still in its infancy,is an activity within the field of medical and biological engineering which predates its name.The term originated in 1987 at a bi- oengineering panel meeting held at the National Science Foundation.In early 1988 the first tissue engineering meeting was held at Lake Tahoe,California.At this meet- ing a working definition was formulated (39): Tissue engineering is the application of the principles and methods of engineering and the life sciences toward the fundamental understanding of structure-function relationships in normal and pathological mammalian tissues and the development of biological substitutes to restore,maintain,or improve functions. Contained in the above is the essence of tissue engineering,i.e.,the use of living cells, together with extracellular components,either natural or synthetic,in the develop- ment of implantable parts or devices for the restoration or replacement of function. An excellent example of tissue engineering,one which demonstrates the importance of cell culture,is the development of artificial skin (32,40).The use of the term ar- tificial here must be qualified since many of the approaches,in using living cells and matrix molecules,are quite natural.In at least one case,dermal and epidermal cells, together with extracellular matrix and nutrients,are grown in culture to produce a skin which in effect is a living equivalent of that found normally on the body(6,7). Another type of artificial skin graft involves a highly porous collagen matrix which serves as a template for the graft(44).When the graft is attached to a wound,fibro- blasts migrate to it from surrounding tissue and permeate the collagen sponge.These fibroblast cells produce new collagen,the original matrix is slowly degraded,and epi- dermal cells from the edges of the wound grow inwards and cover the graft area. There are still other entries into this market.Currently there are four different companies working to develop a tissue-engineered artificial skin.This skin,as the first product developed with the technology of tissue engineering,is an important develop- ment.However,there are a number of other applications of this emerging technology
Cellular Engineering 53 7 plicated protein kinase C (PKC) as part of the signaling pathway which links a shearstress-related mechanical signal to the intracellular events underlying alterations in cell morphology (22). This second messenger may also be involved in the control of the endothelial cell's growth program. It should be emphasized that in vivo endothelial cells reside in a flow environment, and thus to study vascular endothelial biology in static culture is at best a simulation of a region of flow stasis and at worst artifactual. Furthermore, endothelial cells respond differently to differing flow environments, and one should never collectively talk of flow as a single stimulus. Just as there are different chemical agonists, each with their own separate effect, there also are different types of flow agonists, i.e., a variety of types of flow environments, each of which will have their own agonist effect (23). Thus, as important as cell culture studies have been to the study of vascular biology, it is clear that there is much more which needs to be done if we are to engineer the cell culture environment so as to make it truly simulate physiologic conditions. This includes the use of realistic flow environments, but other factors, e.g., the medium, extracellular matrix components, and the presence of neighboring cells, will also need to be included. TISSUE ENGINEERING Tissue engineering, still in its infancy, is an activity within the field of medical and biological engineering which predates its name. The term originated in 1987 at a bioengineering panel meeting held at the National Science Foundation. In early 1988 the first tissue engineering meeting was held at Lake Tahoe, California. At this meeting a working definition was formulated (39): Tissue engineering is theapplication of the principles and methods of engineering and the life sciences toward the fundamental understanding of structure-function relationships in normal and pathological mammalian tissues and the development of biological substitutes to restore, maintain, or improve functions. Contained in the above is the essence of tissue engineering, i.e., the use of living cells, together with extracellular components, either natural or synthetic, in the development of implantable parts or devices for the restoration or replacement of function. An excellent example of tissue engineering, one which demonstrates the importance of cell culture, is the development of artificial skin (32,40). The use of the term artificial here must be qualified since many of the approaches, in using living cells and matrix molecules, are quite natural. In at least one case, dermal and epidermal cells, together with extracellular matrix and nutrients, are grown in culture to produce a skin which in effect is a living equivalent of that found normally on the body (6,7). Another type of artificial skin graft involves a highly porous collagen matrix which serves as a template for the graft (44). When the graft is attached to a wound, fibroblasts migrate to it from surrounding tissue and permeate the collagen sponge. These fibroblast cells produce new collagen, the original matrix is slowly degraded, and epidermal cells from the edges of the wound grow inwards and cover the graft area. There are still other entries into this market. Currently there are four different companies working to develop a tissue-engineered artificial skin. This skin, as the first product developed with the technology of tissue engineering, is an important development. However, there are a number of other applications of this emerging technology
538 R.M.Nerem One of the more promising areas of tissue engineering is in the development of bio- logical substitutes based on the encapsulation of cultured cells.One of the important areas for use of encapsulated cell technology is in the development of bioartificial or- gans,e.g.,an artificial pancreas(20).Pancreas transplants have failed to be useful because of immunological rejection.An alternative approach is the development of implantable insulin pumps;however,these are not without problems either.This has led to the interest in developing a bioartificial pancreas (12).Although there are sev- eral possible designs for a bioartificial pancreas,one involves the use of microencap- sulated islet cells (31).In such a device,the islet cells,which secrete insulin,are surrounded by a semi-permeable membrane.This membrane must be permeable to the transport of insulin so it can be passed into the blood stream.The cells also will need nutrients,oxygen,and other molecules necessary for the maintenance of met- abolic function.However,the membrane must,in addition,protect the islet cells from bacteria,lymphocytes,and other proteins responsible for immune rejection.Impor- tant in the design of a bioartificial pancreas is its ability to respond rapidly to changes in glucose level.Equally necessary is the long term survival of the islet cells and the secretion of the insulin.As part of this,it also is important that islet cell function not change with a buildup of hormones.Although,in most of these areas,microencap- sulated islet cells function quite well,one problem is that the insulin production rate is on the low side. There are other applications of encapsulated cell technology in tissue engineering. Cima et al.(11)have shown that both liver and cartilage cells can be transplanted suc- cessfully,at least in small animals,using cells which are encapsulated in a degrada- ble polymer substrate.Neurological deficits also can be treated by transplantation within the brain of polymer encapsulated cells which release the missing neurotrans- mitter,and Aebischer et al.(1)have investigated the ability of encapsulated dopamine secreting cells to reverse experimental Parkinson's disease.Finally a major tissue- engineering market,where encapsulated cell technology may have application,is in the development of blood substitutes,i.e.,an artificial blood (36).Current efforts are based on the chemical cross-linking of hemoglobin,in many cases bovine derived,but there are potential problems associated with both incomplete cross-linking and the in- troduction of the foreign,bovine proteins into the body.In the future the use of mi- croencapsulation,together with stem-cell culture and controlled hematopoiesis, should prove important. Another application of tissue engineering is in the development of an artificial blood vessel for use in the bypass and replacement of diseased arteries (24,42).A number of groups have been interested in applying cell culture technology to the de- velopment of such tissue-engineered vascular prostheses.Much of this effort has been focused on hybrid vascular grafts,i.e.,a graft constructed out of synthetic material such as dacron or polytetrafluoroethylene(PTFE),but seeded with cultured endothe- lial cells prior to implantation,in order to provide a natural interface with flowing blood (47).Although initial results in terms of increased graft patency are promising, it is clear that this only partially simulates an actual,living blood vessel. Others have attempted to use the co-culture of endothelial cells and smooth mus- cle cells in the construction of an artificial blood vessel.Most notable is the effort of Weinberg et al.(42)who constructed an artificial blood vessel using bovine aortic en- dothelial cells,bovine aortic smooth muscle cells and advential fibroblasts.This was composed of three layers,one of endothelial cells,one of smooth muscle cells together
538 R.M. Nerem One of the more promising areas of tissue engineering is in the development of biological substitutes based on the encapsulation of cultured cells. One of the important areas for use of encapsulated cell technology is in the development of bioartificial organs, e.g., an artificial pancreas (20). Pancreas transplants have failed to be useful because of immunological rejection. An alternative approach is the development of implantable insulin pumps; however, these are not without problems either. This has led to the interest in developing a bioartificial pancreas (12). Although there are several possible designs for a bioartificial pancreas, one involves the use of microencapsulated islet cells (31). In such a device, the islet cells, which secrete insulin, are surrounded by a semi-permeable membrane. This membrane must be permeable to the transport of insulin so it can be passed into the blood stream. The cells also will need nutrients, oxygen, and other molecules necessary for the maintenance of metabolic function. However, the membrane must, in addition, protect the islet cells from bacteria, lymphocytes, and other proteins responsible for immune rejection. Important in the design of a bioartificial pancreas is its ability to respond rapidly to changes in glucose level. Equally necessary is the long term survival of the islet cells and the secretion of the insulin. As part of this, it also is important that islet cell function not change with a buildup of hormones. Although, in most of these areas, microencapsulated islet cells function quite well, one problem is that the insulin production rate is on the low side. There are other applications of encapsulated cell technology in tissue engineering. Cima et al. (11) have shown that both liver and cartilage ceils can be transplanted successfully, at least in small animals, using cells which are encapsulated in a degradable polymer substrate. Neurological deficits also can be treated by transplantation within the brain of polymer encapsulated cells which release the missing neurotransmitter, and Aebischer et al. (1) have investigated the ability of encapsulated dopamine secreting cells to reverse experimental Parkinson's disease. Finally a major tissueengineering market, where encapsulated cell technology may have application, is in the development of blood substitutes, i.e., an artificial blood (36). Current efforts are based on the chemical cross-linking of hemoglobin, in many cases bovine derived, but there are potential problems associated with both incomplete cross-linking and the introduction of the foreign, bovine proteins into the body. In the future the use of microencapsulation, together with stem-cell culture and controlled hematopoiesis, should prove important. Another application of tissue engineering is in the development of an artificial blood vessel for use in the bypass and replacement of diseased arteries (24,42). A number of groups have been interested in applying cell culture technology to the development of such tissue-engineered vascular prostheses. Much of this effort has been focused on hybrid vascular grafts, i.e., a graft constructed out of synthetic material such as dacron or polytetrafluoroethylene (PTFE), but seeded with cultured endothelial cells prior to implantation, in order to provide a natural interface with flowing blood (47). Although initial results in terms of increased graft patency are promising, it is clear that this only partially simulates an actual, living blood vessel. Others have attempted to use the co-culture of endothelial cells and smooth muscle cells in the construction of an artificial blood vessel. Most notable is the effort of Weinberg et al. (42) who constructed an artificial blood vessel using bovine aortic endothelial cells, bovine aortic smooth muscle cells and advential fibroblasts. This was composed of three layers, one of endothelial cells, one of smooth muscle cells together
