Phenotypic Mapping of Human Mesothelial Cells

James A. Ross,1 Ian Ansell,1 J. Thomas Hjelle,3 James D. Anderson,2 Marcia A. Miller-Hjelle,3 James W. Dobbie2

In recent years it has become clear that the mesothelium plays a prominent homeostatic role in the peritoneum, and can be profoundly altered in disease and during peritoneal dialysis. The cell-surface phenotype of the mesothelial cell has not been thoroughly investigated. This study begins to identify cell surface molecules which may be important in mesothelial functions such as adhesion and interaction with cells of the immune system. The expression of adhesion structures on mesothelial cells such as CD44, the ß integrin chain CD29, the ß3 integrin chain CD61 and a chains CD49a (a1), CD49b (a2), CD49c (a3), CD49e (a5), and CD51 (av) is described. In addition, a wide range of novel molecules including CD90, CD105, CD140b, CD142, CD147, CD151, CD157, CD165, and CD166 are identified. The role and function of such molecules in mesothelial biology and their significance for peritoneal dialysis is discussed.

Key words

Mesothelium, surface, antigen, adhesion, phenotype

From:

1Molecular Immunology Group, 2Mesothelial and Peritoneal Research Centre, Lister Research Laboratories, University Department of Surgery, Royal Infirmary, Edinburgh, United Kingdom; 3Department of Biomedical and Therapeutic Sciences, University of Illinois College of Medicine at Peoria, Illinois, USA.

Introduction


Mesothelium is a tissue of ancient lineage, being a direct descendant of the lining of the coelomic cavity of lower animals (1). The coelomic cavity in its most rudimentary form separates tubular viscera from the body wall. The prime function of this space is to permit the unrestricted movement of internal organs, e.g., intermittent, integrated sequential contraction and relaxation; peristalsis, or rhythmic contraction and relaxation of structures, and ventricular beating. This cavity is also used in annelid worms for effecting locomotion and is the precursor of synovial-lined joints. In order to subserve all of these related functions the cavity has from an early stage developed the property of self-lubrication (2,3). It would appear that the ability of the cavity to maintain nonstick surfaces with very low coefficients of friction is an example of a highly conserved evolutionary development. As species have evolved the cavity has been used successively as a convenient conduit for fluids distributing oxygen, nutrients, cells, waste products, and reproduction (2). Following the first histological description by Von Recklinghausen in 1863 of the endothelioid covering of the peritoneum (4), the limited resolution of the light microscope prevented any deeper probing of the structure of the thin cellular monolayer we now call the mesothelium, and there was little interest until the advent of continuous ambulatory peritoneal dialysis (CAPD) as a life-maintaining therapy. This provoked research aimed at elucidating the ultrastructure of mesothelium, and studies that have shown that peritoneal lamellar bodies contain a protein similar to, if not identical to, surfactant protein A (SP-A), which is a product of type 2 pneumocytes and an essential factor in dispersal and function of pulmonary surfactant. It is now believed that mesothelium is the main member of a family of tissues of ancient lineage, the principal function of which is the formation and secretion of lamellar bodies whose phospholipid lamellae subserve surfactant, water-repellent, and lubricant functions in various regions of the organism. Recognition of the importance of this biological system has opened up new possibilities in exploration of the physiology, pharmacology, and pathology of the peritoneum.

In recent years it has become clear that the mesothelium plays a prominent homeostatic role in the peritoneum, and can be profoundly altered in disease and during peritoneal dialysis. Increasing evidence suggests that the mesothelial cell contributes to the control of immune cells in both the normal and inflamed peritoneal cavity. The human mesothelial cell has been shown to be capable of producing the pro-inflammatory cytokines IL-6 and IL-8 in response to an appropriate stimulus (5). In addition, the expression of specialized adhesion molecules such as VCAM-1 and ICAM-1, which together with chemotactic factors such as IL-8 are involved in the recruitment and passage of neutrophils and other leukocytes, suggest a prominent role for the mesothelium in the generation and control of an inflammatory response. The cell-surface phenotype of the mesothelial cell has not been thoroughly investigated, and the effects of CAPD, such as glycation of surface molecules, and surgical procedures on the expression and integrity of the mesothelial cell surface, remain uninvestigated. This study begins to identify cell-surface molecules which may be important in mesothelial functions, especially adhesion and interaction with cells of the immune system.

Materials and methods


This study was undertaken in accordance with the Declaration of Helsinki (1989). Ethical permission was granted by the Lothian Health Board Ethical Committee, and written, informed consent was obtained from each patient. Small biopsy specimens of human omentum were removed at the time of elective colorectal surgery and were used for the establishment of mesothelial cell lines. Primary cultures were prepared by a short incubation in trypsin-EDTA in order to avoid fibroblast contamination. The cells were maintained in 75-cm2 flasks (Costar, U.K.) with Dulbecco's Modified Eagle's Medium (DMEM; Gibco-BRL, U.K.) plus 10% fetal calf serum. Culture media were supplemented with 50 IU/mL penicillin, 50 µg/mL streptomycin, and 2 mM glutamine (all from Life Technologies, U.K.). Cell culture flasks were incubated at 37°C in a 5% CO2-in-air atmosphere, and the medium changed every 3 to 4 days. Primary cultures were detached by trypsin-EDTA, and washed in complete medium before being plated on. All mesothelial cell lines were characterized by the expression of a panel of cell-surface markers, and were used for experiments between the second and fourth passage. Primary antibodies were obtained from the Sixth International Workshop on Human Leucocyte Differentiation Antigens (6), and the second antibody was F(ab')2 fragment sheep anti-mouse-Ig fluoroisothiocyanate conjugate (FITC; Sigma, U.K.). Flow cytometry was carried out using an Epics XL-MCL Flow Cytometer (Coulter Electronics).

Results


This study characterizes the expression of cell surface molecules on primary cultures of human peritoneal mesothelial cells (Table I). Most of the molecules investigated are adhesion molecules, or molecules generally expressed on platelets or endothelium. The mean equivalent of soluble fluorescence (MESF) provides an indication of the density of the molecule on the cell surface. The standard form of CD44 (Figure 1, Figure 2), was detected with a panel of CD44 antibodies and was expressed at a very high level on mesothelial cells. The ß1 integrin chain CD29 and a chains CD49b (a2), CD49c (a3), and CD51 (av) (Figure 2) were also present at high to moderate density on the surface of mesothelial cells. Tissue factor (CD142) was detected with a large panel of antibodies and the molecule was present at a moderate density (Table I). CD47, CD151 (PETA-3), CD90 (Thy-1), CD9 (Figure 2), CD63 (LIMP), and a number of newly characterized molecules such as the AD2 antigen (CD165) and CD157 (BST-1) were expressed at a moderate to high level on mesothelial cells. The platelet-derived growth factor receptor beta (PDGF-RB, CD140b) was also expressed at high density. A number of molecules were expressed at low level including CD166 (activated leucocyte cell adhesion molecule (ALCAM, CD6-ligand), the integrin a chains CD49a and CD49e, the integrin ß3 chain (CD61), CD105 (endoglin), and CD106 (vascular cell adhesion molecule, VCAM-1).

Discussion


Figure 1: The standard form of CD44 is highly expressed on mesothelium. Perhaps the best- studied aspect of CD44 function is its ability to bind components of the extracellular matrix, primarily hyaluronic acid (HA). Also known as hyaluronan or hyaluronate, HA has roles in wound healing, selective adhesion, cell proliferation, cell locomotion, and regulation of immune function in inflammation. Few, if any, successful therapeutic modalities are entirely free from serious pathological sequelae, and the effects of CAPD, such as glycation of surface molecules, and of surgical procedures on the expression and integrity of the mesothelial cell surface molecules remain uninvestigated. Figure 2: Flow-cytometric profile of fluorescence intensity (x axis) against relative cell count (y axis) for isotype-matched control antibody (solid line) and (a) CD44 (G44-26); (b) the integrin a2 chain CD49b (Gi9); (c) the integrin a3 chain CD49c (ASC-1); (d) the integrin av chain CD51 (CLB-M9); (e) CD47 (Bric-126); (f) CD151 (11B1.G4; (g) CD90 (V45); and (h) CD9 (CLB-throb/8) (dotted lines).

The surface of the mesothelial cell has been largely ignored despite the cell being a key player in the successful use of CAPD. Many of the molecules on the surface of the mesothelial surface may play an important role in preventing damage and adhesions, and in interacting with cells of the immune system. CD44 (7), for example, is the principle cell-surface receptor for hyaluronan, which is a major constituent of serosal cavities, where it functions as a lubricant between opposing surfaces. The role of hyaluronan as a surface protectant or barrier in preventing mesothelial damage and exfoliation has led to recommendations that it be used as a constituent of peritoneal dialysate. The demonstration of a high density of cell- surface CD44 on mesothelial cells in this study supports the concept of therapeutic benefit being derived from the inclusion of hyaluronan as an additive in dialysates.

The ß1 integrin (CD29), the ß3 integrin a chain (CD61), the chains CD49a, CD49b, CD49c, CD49e and CD51, and the av chain, are all involved in cell-cell and cell-matrix interactions. CD49a forms part of the receptor for collagen and laminin and is inducible by lipopolysaccharide in monocytes. It is the sole collagen receptor in liver and smooth muscle, and is thought to be responsible for the stability of such tissues. Whether a similar function pertains in the mesothelial lining is not known. The CD49b/ß1 receptor is present on numerous cells, including endothelial cells, and mediates cell adhesion to collagen type I, II, III and IV, and to laminin. It is the receptor for echovirus 1 and the binding of CD49b to collagen is known to regulate the production of matrix metalloproteinase-1 and collagen type I. Its regulatory role in mesothelial cells has not been explored. CD49c (a3) is present on nearly all adherent cell lines which have been examined. The a3/ß1 adhesion receptor includes in its ligands laminin-5 (epiligrin / nicein / kallinin), fibronectin, collagen, entactin, and invasin. Anti-a3 antibodies have been shown to crosslink a3/ß1 and to induce tyrosine phosphorylation of focal adhesion kinase suggesting an involvement in signal transduction (8). This has not yet been demonstrated in mesothelial cells. The a5 molecule (CD49e) forms part of the receptor for fibronectin and binding through the a5/ß1 receptor has been shown to be important for cell survival and the induction of apoptosis (9). The integrity of these molecules will clearly be important in the biological function of mesothelial cells and may be affected by long-term exposure to dextrose dialysate. As observed in the tissue pathology of matrix ground substance in diabetes, glycation of these molecules on the mesothelial cell would seriously impact on cell-cell and cell-matrix interactions.

A number of these molecules have not previously been identified in mesothelial cells but have been described on platelets. Platelet-endothelial cell tetra-span antigen (PETA-3, CD151) was originally identified as a novel human platelet-surface glycoprotein. Although this glycoprotein is present in low abundance on the platelet surface the PETA-3 antibody 14A2.H1 stimulates platelet aggregation and mediator release. A cDNA clone encoding PETA-3 has been isolated and a single PETA-3 RNA transcript (1.6 kb) has been detected in an endothelial cell line. The role of this molecule in mesothelium is unknown. CD63 is also a member of the tetraspan family of transmembrane proteins and is heavily glycosylated. CD63 has been used to demonstrate activation on platelets but its function is unknown although it has been shown to be associated with the CD29/CD49c and CD29/CD49f integrins by immunoprecipitation studies (10). CD47 has broad tissue distribution. Its cellular function is as an adhesion receptor and a receptor for thrombospondin. The presence on mesothelium of molecules normally expressed on platelets or involved in platelet aggregation may be related to the exudation of fibrin and other molecules, and to a lesser extent, of platelets, which is a reflex response of peritoneum to inflammatory stimuli whereby temporary adhesions and loculation of organisms prevent rapid spread through the voluminous cavity.

The molecule BST-1 (CD157) has cyclic ADP-ribose cyclase and hydrolase activity, and appears to facilitate preB-cell growth. The molecule is expressed in umbilical vein endothelial cells but is only present at a very low level in a variety of hematopoietic cell lines. Cyclic ADP-ribose is a stimulator of Ca2+ release from the intracellular Ca2+ pool, and has emerged as a potential regulator of insulin secretion in pancreatic beta cells. BST-1 is a glycosyl-phosphatidylinositol (GPI)-anchored surface molecule that exhibits homology with CD38. The presence of this molecule in the peritoneal cell may be of interest in terms of future gene therapy strategies for diabetes. As a stimulator of Ca2+ release from the intracellular pool, BST-1 is possibly of considerable importance in the control of secretion of lamellar bodies and surfactant protein A by mesothelium (11).

Neurothelin/Basigin (CD147) is a developmentally regulated immunoglobulin-like surface glycoprotein which has previously been detected on blood-brain barrier endothelium, epithelial tissue barriers, and neurons. CD90 (Thy-1) is expressed at high levels on mesothelial cells. This molecule is expressed on high endothelial cells but not on normal endothelia and its expression has been reported in various stromal cell lines although the function of the molecule in the human immune system remains undefined.

CD166 (CD6-ligand) interactions have been implicated in the regulation of T-cell adhesion and activation. CD6 is a member of the scavenger receptor family, whereas its human ligand (ALCAM) belongs to the immunoglobulin superfamily. A number of homologs of CD166 have been identified including BEN in the chicken, and neurolin in the zebrafish. The presence of the CD6-ligand molecule on mesothelial cells may be of interest with regard to inflammatory conditions of the peritoneal cavity. CD165 (AD2) is another molecule present on mesothelial cells which has been implicated in the interaction of T-cells with epithelial cells. Clearly, there are many molecules on the surface of the mesothelial cell which may be important in adhesion, barrier function, and in the interaction between mesothelium and cells of the immune system. This study begins to reveal the complex molecular phenotype of the mesothelial cell, and suggests that thorough investigation of the function of these molecules on the mesothelial cell surface would be useful in understanding the normal and abnormal biology of the mesothelium.

References


  1. Dobbie JW. From philosopher to fish: The comparative anatomy of the peritoneal cavity as an excretory organ and its significance for peritoneal dialysis in man. Perit Dial Int 1988; 8:3-6.
  2. Dobbie JW. New concepts in molecular biology and ultrastructural pathology of the peritoneum: their significance for peritoneal dialysis. Am J Kidney Dis 1990; 2:97-109.
  3. George CJ, Ellis AE, Bruno DW. On the remembrance of the abdominal pores in rainbow trout Salmo gairdneri Richardson, and other salmonid spp. J Fish Biol 1982; 21:643-47.
  4. Von Recklinghausen, F. Zur fettresorption. Virchows Arch Path Anat 1863; 26:172.
  5. Topley N, Mackenzie RK, Williams JD. Macrophages and mesothelial cells in bacterial peritonitis. Immunobiology 1996; 195:563-73.
  6. Leucocyte Typing VI. Kishimoto T, Kikutani H, von dem Borne AEGK, Goyert S, Mason DY, Miyasaka M, Moretta L, Okomura K, Shaw S, Springer TA, Sugamura K,Zola H, eds. Garland Publishing, New York and London, 1997.
  7. Borland G, Ross JA, Guy K. Forms and functions of CD44. Immunology 1998; 93:139-48.
  8. Kornberg LJ, Earp HS, Turner CE, et al. Signal transduction by integrins: increased protein tyrosine phosphorylation caused by clustering of beta 1 integrins. Proc Natl Acad Sci 1991; 88:8392-6.
  9. Meredith JE, Fazeli B, Schwarz MA. The extracellular matrix as a cell survival factor. Mol Biol Cell 1993; 3:953-61.
  10. Berditchevski F, Bazzoni G, Hemler ME. Specific association of CD63 with the VLA-3 and VLA-6 integrins. J Biol Chem 1995; 270:17784-90.
  11. Dobbie JW. Surfactant protein A and lamellar bodies: A homologous secretory function of peritoneum, synovium and lung. Perit Dial Int 1996; 16:574-81.
Corresponding author:
James A. Ross, PhD, Lister Research Laboratories, University Department of Surgery, Royal Infirmary, Edinburgh EH3 9YW, UK.