Angiotensin I-converting enzyme localization on cultured fibroblasts by immunofluorescence

Angiotensin I-converting enzyme is responsible for the activation of angiotensin I and the inactivation of bradykinin. It has been localized by immunofluorescence on the endothelium of a variety of tissues and has been considered to be a specific marker for endothelial cells in culture. The present paper demonstrates, by immunofluorescence, the presence of angiotensin I-converting enzyme in monolayer cultures of fibroblasts derived from adult rat lung, bovine calf pulmonary artery, and human foreskin (CF-3 cells). Fluorescent localization of angiotensin I-converting enzyme was observed over the cytoplasm of adult rat lung and bovine calf pulmonary artery fibroblasts and as distinct areas overlying the nuclei of human foreskin fibroblasts. Determination of angiotensin I-converting enzyme activity by fluorimetric assay in parallel studies confirmed the presence of angiotensin I-converting enzyme activity in cultured fibroblasts. Immunofluorescent studies with antibody to Factor VIII demonstrated the presence of Factor VIII on cultured endothelial cells but not on fibroblasts. These results indicate that angiotensin I-converting enzyme is not confined to endothelial cells, and thus may not serve as a specific marker for endothelial cells in culture. Factor VIII may be a more specific marker for these cells.     

Angiotensin I-converting enzyme localization on cultured fibroblasts by immunofluorescence

Summary Angiotensin I-converting enzyme is responsible for the activation of angiotensin I and the inactivation of bradykinin. It has been localized by immunofluorescence on the endothelium of a variety of tissues and has been considered to be a specific marker for endothelial cells in culture. The present paper demonstrates, by immunofluorescence, the presence of angiotensin I-converting enzyme in monolayer cultures of fibroblasts derived from adult rat lung, bovine calf pulmonary artery, and human foreskin (CF-3 cells). Fluorescent localization of angiotensin I-converting enzyme was observed over the cytoplasm of adult rat lung and bovine calf pulmonary artery fibroblasts and as distinct areas overlying the nuclei of human foreskin fibroblasts. Determination of angiotensin I-converting enzyme activity by fluorimetric assay in parallel studies confirmed the presence of angiotensin I-converting enzyme activity in cultured fibroblasts. Immunofluorescent studies with antibody to Factor VIII demonstrated the presence of Factor VIII on cultured endothelial cells but not on fibroblasts. These results indicate that angiotensin I-converting enzyme is not confined to endothelial cells, and thus may not serve as a specific marker for endothelial cells in culture. Factor VIII may be a more specific marker for these cells. Presented in part at the 31st Annual Meeting of the Histochemical Society, April 11–15, 1980, New Orleans, Louisiana. Wendy Baur and Ms. Jane Aghajanian for expert assistance in the preparation of the cell cultures. This work was supported by Research Grant HL 14456 and Training Grant HL 07053 from the National Institutes of Health, Bethesda, MD.     

Dexamethasone selectively regulates the activity of enzymatic markers of cerebral endothelial cell lines

Summary Two endothelial cell lines were derived from grafts of the central nervous system using retrovirus mediated gene transfer to introduce the polyoma middle-T oncogene into fetal rat brain endothelial cells and transplantation of these cells into adult rat brain. In this report, we further characterize these cells and the effect of dexamethasone on the expression of specific enzymatic markers. These cells take up acetylated low density lipoprotein, leucine, and glucose, and express Factor VIII-related antigen, angiotensin converting enzyme, alkaline phosphatase, gamma-glutamyltranspeptidase, and as yet undescribed aminopeptidase A and B-like enzymes. When grown on semi-permeable membranes, these transformed cells do not spontaneously retain small hydrophilic molecules. In culture, one of the lines (EC 193) forms a confluent monolayer of spindle-shaped cells homogenously expressing gamma-glutamyltranspeptidase at a level comparable to primary cells. The other cell line (EC 219) grows as clusters of elongated cells, and gamma-glutamyltranspeptidase activity is expressed mainly in cells forming the clusters. This clustered pattern changes to a confluent one after culture on type-I collagen. Dexamethasone increases angiotensin-converting enzyme activity, and decreases the expression of gamma-glutamyltranspeptidase and aminopeptidase A, whereas the aminopeptidase B activity is little modified. Inhibition of aminopeptidase A activity by amastatin, potentiates angiotensin II effects on DNA synthesis. These results indicate that retrovirally transformed brain endothelial cells are a useful model for studying the blood-brain barrier in vitro and that dexamethasone, an agent with the potential to reduce brain edema, directly affects some blood-brain barrier properties in these endothelial cell lines.     

Expression of inducible nitric oxide synthase in human umbilical vein endothelial cells during primary culture

The adaptive response of endothelial cells to stress may lead to the upregulation of nitric oxide (NO) production. Herein, we report inducible nitric oxide synthase (iNOS) induction in primary cultures of human umbilical vein endothelial cells (HUVEC). The enzyme expression was earlier observed in 12-h cultures, reaching maximal levels after 3 days and decreasing when cells become confluent. The time course of NO production by HUVEC paralleled iNOS expression during the whole culture period, indicating that enzyme was functionally active. Conversely, iNOS induction could not be further detected in HUVEC subcultures passed once from cells presenting maximal levels of iNOS expression in the primary culture. Induction of iNOS in HUVEC was not related to lipopolysaccharide contamination, since the enzyme expression was not affected in the presence of polymyxin B added to primary cultures. Further analysis showed that aminoguanidine, a specific iNOS inhibitor, did not affect cell proliferation, suggesting that the NO produced by HUVEC may not be directly related to cell growth. Platelet endothelial cell adhesion molecule-1 expression was upregulated during cell confluence, in contrast to the decrease of iNOS expression and activity. The data suggest that iNOS expression may be a molecular mechanism mediating the adaptive response of endothelial cells to culture environment.     

Growth-promoting actions of extracts from mouse submaxillary glands on human endothelial cells in culture.

Extracts of submaxillary glands from two different strains of inbred mice were mitogenic for human endothelial cells in culture. The mitogenic activity of extracts from glands of males of the SWR/J and C57BL/10J strains were equivalent, and the growth stimulating effect was unrelated to renin or esteroproteolytic activity. Mitogenic activity in extracts from SWR/J females was less than that from males, and extracts from C57BL/10J females were inactive. The polypeptide growth factors, epidermal (EGF) and fibroblast (FGF) growth factors, also stimulated replication of endothelial cells. Cells from either umbilical arteries or veins responded to submaxillary extracts, EGF, or FGF with a similar increase in cell number, increase in protein and enhanced uptake of 3H-thymidine. The proliferative response was associated with decreased activity of angiotensin I converting enzyme which is localized on the endothelial surface. Nerve growth factor (NGF) was not mitogenic for endothelial cells. Extracts of submaxillary glands from male mice of either strain contained approximately 20 times more EGF than extracts from females, as determined by immunodiffusion. Mitogenic activity of the extracts was completely inhibited by antiserum to EGF, suggesting that the active component of these preparations is EGF.     

Thrombin stimulates c-sis gene expression in microvascular endothelial cells

We have determined whether expression of the c-sis gene product, platelet-derived growth factor (PDGF), is regulated in cultured renal microvascular endothelial cells by factors to which vascular endothelial cells may be exposed at sites of perivascular cellular proliferation. Thrombin exposure increased endothelial cell levels of c-sis message by 3-5-fold over a time course that peaked at 4 h after exposure. Similarly, thrombin-exposed microvascular endothelial cells released increased amounts of PDGF activity into their media. The thrombin effect was not mediated through the proteolytic activity of thrombin, as proteolytically inactive thrombin stimulated the c-sis expression as well as native thrombin. This stimulation was mimicked by exposure of cells to biologically active phorbol esters, suggesting that thrombin action may be mediated through activation of kinase C (Ca2+/phospholipid-dependent enzyme). Thus, thrombin regulates the expression and release of PDGF activity from endothelial cells in culture and may act in vivo to stimulate mitogen release from endothelial cells, thereby inducing proliferation of perivascular cells.     

Effects of angiotensins on cellular hypertrophy and c-fos expression in cultured arterial smooth muscle cells.

An increase in cell size and protein content was observed when quiescent arterial smooth muscle cells in culture were incubated with either angiotensin II or III. These effects were inhibited by the specific angiotensin type-1 receptor antagonist losartan (DuP753) but not by CGP42112A. In parallel, a transient and dose-dependent induction of c-fos was demonstrated not only with angiotensins II and III but also with angiotensin I. Both angiotensins II and III exerted their maximal effect at 1 microM, while angiotensin I needed a tenfold-higher concentration to exert an identical effect. As for hypertrophy, losartan also inhibits angiotensin-induced c-fos expression, suggesting that this gene may be involved into the hypertrophic process. Angiotensin-I-mediated c-fos induction is partially inhibited by the angiotensin-converting enzyme inhibitors captopril and trandolaprilate; given that an angiotensin-converting enzyme activity was detected in these smooth muscle cell cultures, these results suggest that angiotensin-I-induced c-fos expression is mediated in part via angiotensin-I conversion to angiotensin II, but also by other unidentified pathway(s). Angiotensin I could essentially induce smooth muscle cell hypertrophy by indirect mechanisms, while angiotensins II and III act directly on smooth muscle cells.     

Paraquat depresses the activity of angiotensin converting enzyme expressed by human umbilical vein endothelial cells in culture

The activity of angiotensin converting enzyme (ACE) in cell lysate of cultured human umbilical vein endothelial cells (HUVEC) after a 24-hour incubation with 10(-3) and 10(-4)M of paraquat (PQ) was decreased. However, LDH released into the culture medium of HUVEC during the 24-hour incubation with PQ was not increased. Many investigators show that the change in serum ACE activity reflects the impairment of vascular endothelial cells. We showed in this report that ACE was decreased even at an early stage of endothelial injury induced by PQ, when LDH release is not yet increased.     

Localization of angiotensin converting enzyme (kininase II). II. Immunocytochemistry and immunofluorescence.

The cellular and subcellular sites of angiotensin converting enzyme (kininase II) in lung tissue and endothelial cells in culture were examined by immunocytochemical and immunofluorescence techniques. Converting enzyme is capable of inactivating bradykinin and of converting angiotensin I to its potent lower homolog, angiotensin II. Immunocytochemistry at the electron microscope level used goat anti- (pig lung and angiotensin converting enzyme) coupled to 11-MP (11-microperoxidase) via glutaraldehyde or to 8-MP (8-microperoxidase) via a bifunctional active ester, bis-succinyl succinate. The latter conjugate, which does not contain complex polymers, has been characterized in detail in terms of immunoreactivity and peroxidase activity.     

Fucosterol decreases angiotensin converting enzyme levels with reduction of glucocorticoid receptors in endothelial cells

The modulation of angiotensin converting enzyme (ACE) levels was studied using fucosterol, one of phytosterols, in cultured bovine carotid endothelial cells. Addition of fucosterol to the culture medium resulted in the decrease of ACE activity of endothelial cells; however, fucosterol did not directly inhibit ACE activity. Dexamethasone elevated the levels of ACE in normal cells, but this effect was not seen in the fucosterol-treated cells. Receptor assays showed that the amount of glucocorticoid receptors in fucosterol-treated cells decreased to an undetectable level. These results indicate that fucosterol lowers the ACE levels on the endothelial cells by inhibiting the synthesis of glucocorticoid receptors involved in the regulation of ACE levels.     

Isolation, characterisation, growth and culture of endothelial cells from the rat aorta

Rat aortic endothelial cells have been isolated by the explantation technique and grown in culture. They have been identified morphologically using standard staining techniques, biochemically by identification of angiotensin convertase and have been positively stained for Factor VIII-related antigen by immunofluorescence using both anti-human and anti-rat Factor VIII antibodies. The explantation technique is a successful alternative to enzyme digestion which is not applicable to rat aortic endothelial cells because of the nature of their attachment to the subendothelial layer.     

Primary culture of capillary endothelial cells from the spiral ligament and stria vascularis of bovine inner ear

Summary Methods for isolation and culture of microvascular endothelial cells of the inner ear were devised to provide an in-vitro system for studying endothelial functions in this tissue. Capillaries from the stria vascularis and spiral ligament were treated enzymatically to free them from surrounding tissue. Contamination by extraneous tissue was minimized by banding capillary segments in Percoll gradients and culture in plasma-derived serum on a fibronectin-coated substrate. Although only small amounts of inner ear tissue were available, tritiated thymidine autoradiography demonstrated that considerable growth in culture was possible. Addition of heparin and endothelial cell growth supplement to the medium enhanced proliferation. The endothelial origin of the cultured cells was confirmed by immunofluorescent demonstration of the presence of Factor VIII-related antigen and angiotensin-converting enzyme. In addition, tight junctions between cells were observed in both thin sections and platinum replicas obtained by freezefracture techniques. Endothelial cells from neither the stria vascularis nor the spiral ligament allowed passage of horseradish peroxidase across the monolayer during a 5-min period. However, endothelial cells from the stria vascularis exhibited a greater amount of pinocytotic activity than those of the spiral ligament, a difference that is also observed in vivo. Methods for expanding a small population of endothelial cells with retention of specialized properties into one of sufficient size for morphologic and biochemical studies have been demonstrated for the inner ear.     

Angiotensin-converting enzyme: accumulation in medium from cultured endothelial cells.

The activity of angiotensin converting enzyme has been measured in endothelial cells cultured from hog aorta, and in serum-free culture medium taken from both endothelial cells and smooth muscle cells. Endothelial cells maintained in medium containing 20% fetal calf serum contained 43 pmol/min/10⁶ cells of converting enzyme activity; freshly collected cells contained 518 pmol/min/10⁶ cells. Endothelial cells held in serum-free medium release this activity into the medium in amounts up to 40 times that associated with the cells; at the same time the activity associated with the cells rises 2 fold. The rise in cell-associated activity and the appearance of activity in the medium are both blocked by cycloheximide. These observations provide direct evidence that endothelial cells can produce excess angiotensin-converting enzyme and release it in active form, and thus serve as the source of circulating converting enzyme activity.     

Partial expression of catecholaminergic traits in cholinergic chick ciliary ganglia: Studies in vivo and in vitro

We have previously demonstrated that at embryonic Day (E) 8, some cells of the chick ciliary ganglion (CG) contain the catecholaminergic (CA) enzyme tyrosine hydroxylase (TH), but not phenylethanolamine-N-methyltransferase (PNMT); and that in culture essentially all cells express both enzymes. In the present study, we sought to determine, first, whether the expression of adrenergic traits in the CG in vivo is transient or permanent in the CG. To do so, CGs were removed from E5 to postnatal Day 5, fixed, and processed for the immunocytochemical localization of the CA enzymes: TH, L-amino acid decarboxylase (AADC), and PNMT. At all stages examined, some CG neurons expressed TH immunoreactivity (TH-IR) and all contained AADC-IR. However, none stained with PNMT antibodies, indicating that these cells stably express some, but not all, of the CA enzymes. Second, we examined whether CG neurons in culture expressed other CA markers. CG neurons did not contain detectable levels of TH enzyme activity nor did they transport and store exogenously supplied monoamines. These results indicate that some but not all traits necessary for adrenergic function are present in CG neurons in vitro. Third, we sought to establish whether CA expression in CG neurons is affected by modification in culture conditions. Cultures of CG neurons continued to express TH-IR even when grown in the presence of either 50% HCM or 20 mM KCl for 5 days. Finally, the expression of the cholinergic enzyme, choline acetyltransferase (CAT) was assessed in CG cultures by biochemical assay. CAT activity increased five-fold between 5 and 17 days in vitro, irrespective of the presence of TH-IR in 100% of the CG neurons of sister cultures. These data suggest that at least a subpopulation of CG neurons express both TH and CAT in culture. We conclude that the postmitotic neurons of the CG are able to express some but not all of the traits characteristic of a CA phenotype while maintaining cholinergic expression. These findings suggest that (1) the appearance of the full complement of adrenergic properties is not coordinated and may be regulated by different environmental cues and (2) parasympathetic neurons can express both adrenergic and cholinergic traits simultaneously.     

Endothelium-independent conversion of angiotensin I by vascular smooth muscle cells

The conversion of angiotensin I (AT-I) to angiotensin II (AT-II) by angiotensin I-converting enzyme (ACE) is a key step in the action of angiotensins. ACE is constitutively expressed in endothelial cells, but can also be detected at low levels in smooth muscle cells (SMC). Furthermore, in rats the ACE activity can be induced in SMC in vivo by experimental hypertension or vascular injury and in vivo by corticoid treatment. This study was therefore undertaken to evaluate the conversion of AT-I and its subsequent effects in SMC in basal conditions and after stimulation by dexamethasone. Using rat and human SMC, showed that dexamethasone induced ACE expression and that this enzyme was functional, leading to AT-II-dependent intracellular signaling. A fourfold increase in phospholipase C activity in response to AT-I was observed in dexamethasone-activated SMC compared with quiescent SMC. This effect of dexamethasone on signal transduction is dependent on ACE activity, whereas AT-II receptor parameters remain unchanged. The action of AT-I was blocked by an AT1 receptor antagonist, suggesting that it was mediated by AT-II. Similarly, dexamethasone-induced ACE expression was present in human SMC, and calcium signaling was mobilized in response to AT-I in activated human cells. Experiments performed with cocultures of endothelial cells and SMC in a Transwell system showed that the response to AT-I was limited to the compartment where AT-I was localized, suggesting that AT-I does not pass through the endothelial cell barrier to interact with underlying SMC. Our data suggest that in rat, as in human SMC, the conversion of AT-I into AT-II and the signal transduction in response to AT-I are ACE expression-dependent. In addition, the present findings show that this SMC response to AT-I is endothelium-independent, supporting the idea of a local generation of AT-II in the vascular wall.     

New substrates for the radioassay of angiotensin converting enzyme of endothelial cells in culture.

To develop means of measuring angiotensin converting enzyme of endothelial cells in culture, we have synthesized benzoyl-Phe-Ala-Pro-OH (I), benzoyl-Pro-Phe-Arg-OH (II) and benzoyl-Gly-His-Leu-OH (III), each bearing a 3H-atom on the para-position of its benzoyl moiety. All three of the acylated tripeptides are substrates for the enzyme. Substrate I exhibits the lowest Km (12.5 micrometer) and yields the most sensitive assay: the enzyme of 10(6) cells can be measured in a 30 min incubation at 37 degrees C. Radiolabelled reaction product is separated from substrate by extraction of acidified reaction mixture with an organic solvent, and the rate of formation of product can be quantified by liquid scintillation counting of the organic phase. Substrate III can also be used to measure angiotensin converting enzyme of cells but requires longer incubations (180--240 min) and high salt concentrations (0.75 M Na2SO4). Substrate II is not specific: it is hydrolyzed by more than one enzyme of endothelial cells.     

Angiotensin converting enzyme in cultured endothelial cells and growth medium. Relationships to enzyme from kidney and plasma

We have previously reported that cultured cells from swine aorta possess angiotensin converting enzyme (peptidyldipeptide hydrolase, EC 3.4.15.1) and release it into serum-free culture medium. The present work compares enzyme from these two sources, and from swine kidney and serum, with respect to antibody and lectin binding. Purified enzyme from swine kidney, and the activity in swine serum, cultured endothelial cells and culture medium bind similarly to rabbit antibodies prepared against the kidney converting enzyme. Enzyme from each of these sources was allowed to bind to an immobilized lectin (Ricinus communis), which binds to terminal galactose residues of glycoproteins. Increasing concentrations of galactose were used to remove enzyme from the lectin column and the distribution of enzyme activity in the galactose eluates was determined. The elution pattern was similar for kidney and endothelial cell enzyme, and different from the pattern found for both serum and medium enzymes. Neuraminidase treatment of either serum or medium enzyme altered the distribution of activity eluted to that found for endothelial cell or kidney enzymes. The effects of neuraminidase suggest that the difference in lectin binding between cell and medium enzyme reflects differences in the number of terminal sialic acid residues that cover galactose residues.     

Primary culture of microvascular endothelial cells from bovine retina

Summary To provide an in vitro system for studying retinal capillary function we have developed methods for isolation and culture of microvascular endothelial cells from retina. Retinal microvessels were isolated by homogenization of the retina and collection of the microvessels onto nylon mesh. Treatment of the isolated microvessels with collagenase and dispase followed by Percoll gradient centrifugation yielded endothelial cells that were largely free of pericytes. A homogeneous population of endothelial cells that were capable of at least six population doublings was obtained by plating onto a fibronectin coated substrate in plasma derived serum. The endothelial origin of these cells was confirmed by the presence of Factor VIII antigen, angiotensin converting enzyme activity, numerous tight junctions, and a cell surface that did not bind platelets. A second cell type, which did not exhibit these cell markers and which is presumably the intramural pericyte, was obtained when the isolated microvessels were plated on tissue culture grade plastic in fetal bovine serum. Supported by Research Grants 5R01-EY03772 and 5R01-ES02380 from the U.S. Public Health Service (G. W. G.) and Established Investigator Award 31-107 from the American Heart Association (A. L. B.).