High cell density culture of a recombinant Escherichia coli producing penicillin acylase in a membrane cell recycle fermentor

A recombinant Escherichia coli HB101(pPAKS2) producing penicillin acylase was cultured in a membrane cell recycle fermentor. The strain was very stable throughout the whole experiment. The main inhibitory by-product was acetic acid, and cell growth ceased when its concentration was above 14 g/L Cell density could be increased up to 145 g/L dry weight without experiencing by-product inhibition by regulating glucose concentration in the fermentor and by using total membrane recycle. Acetic acid formation was negligible not only when cells were cultured in medium containing no glucose but also when glucose was limited. Dissolved oxygen control as well as glucose limitation was an indispensable condition for minimizing acetic acid formation when the medium contained glucose. Low concentrations of accumulated acetic acid were reused when glucose was limited. Use of highly concentrated medium reduced the membrane surface area required for cell recycle greatly. The recycle fermentor could be operated in various operational modes including partial bleed and repeated recycle culture to give high productivity. Productivity of a repeated recycle system was over 10 times higher than that of a simple batch system.     

3T3-L1 adipocytes promote the growth of mammary epithelium

Murine mammary epithelium grows in association with predominantly adipocyte stroma in vivo. To investigate potential growth-promoting effects of adipocytes on mammary epithelium, we developed a co-culture system of mammary epithelium and adipocytes by taking advantage of the 3T3-L1 cell line. These cells undergo adipocyte differentiation when the culture reaches confluence and growth ceases. Mid-pregnant murine mammary epithelium was plated on lethally irradiated feeder layers of 3T3-L1 adipocytes, undifferentiated 3T3-L1 cells, 3T3-C2 fibroblasts (a subclone of 3T3 cells that does not undergo adipocyte differentiation), or tissue culture plastic. Mammary epithelial colony size on adipocyte feeder layers was 2-fold larger than colonies on 3T3-C2 cells and 4-fold larger than colonies on tissue culture plastic. Measurement of tritiated thymidine [3H]TdR incorporation and labelling index in mammary cells was significantly higher on adipocytes than on other feeder layers or plastic. There was a 6-fold increase in mammary cell number after 5 days in culture when mammary epithelium was plated on substrate-attached material ('extracellular matrix') derived from 3T3-L1 cells and a 4-fold increase in cell number when plated on plastic in conditioned medium derived from 3T3-L1 adipocytes compared with growth on plastic in unconditioned medium. We conclude that interaction of mammary epithelium with adipocytes results in a marked increase in proliferation of mammary epithelium and that extracellular components may mediate this effect.     

Polyvinyl formal surface promotes continuous growth of Vero cells in protein-free medium.

The effect of polyvinyl formal (PVF) culture surface on the growth of 10 mammalian continuous cell lines, including Swiss 3T6, NCTC clone 929 L, BHK-21 clone 13, CHO-K1, PK 15, A 431, HeLa, MDCK, LLC-MK2 and Vero in protein-free 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F12 supplemented with trace elements and L-ascorbic acid 2-phosphate, was investigated. Most of the cell lines showed only some initial proliferation on PVF similar to the polystyrene (PS) surface of commercially available culture flasks. In contrast, proliferation of monkey kidney cell line Vero was by far greater on PVF than on PS or poly-D-lysine treated culture surface. In addition, Vero cells on PVF could be subcultured in the protein-free medium without any significant decrease of growth rate in successive passages. These results showed that PVF provides a culture surface which selectively promotes continuous growth of Vero cells in protein-free, chemically defined medium.     

Serum-free ex vivo expansion of CD34(+) hematopoietic progenitor cells

In an effort to obtain defined culture conditions for ex vivo expansion of hematopoietic stem and progenitor cells which avoid the supplementation of serum, we cultured human CD34(+) hematopoietic progenitor cells in a chemically defined, serum-free medium in the presence of hematopoietic growth factors (HGFs), stem cell factor (SCF), interleukin (IL)-1beta, IL-3, IL-6, and erythropoietin (EPO). A medium, SFM-1, was prepared according to a protocol previously optimized for semisolid progenitor cell assays containing Iscove's Modified Dulbecco's Medium (IMDM) plus cholesterol, bovine serum albumin, transferrin, nucleotides and nucleosides, insulin, and beta-mercaptoethanol. In static cultures seeded with CD34(+)-enriched progenitor cells isolated from human peripheral blood, a mean 76.6-fold expansion of total nucleated cells and a mean 4.6-fold expansion of colony-forming cells (CFC) was recorded after 14 days. Morphological analysis of the expanded cells revealed formation of myeloid, erythroid, and megakaryocytic cells. Flow cytometric analysis indicated that CD34(+) antigen expressing cells were maintained to a limited degree only, and cell populations expressing surface markers for myeloid (CD33, CD14, and CD15) and megakaryocytic (CD41a) lineages predominated. Within SFM-1, bovine serum albumin (BSA), cholesterin, and transferrin represented the most critical components needed for efficient total cell and CFC expansion. Addition of autologous patient plasma (APP) or fetal calf serum (FCS) to SFM-1 resulted in inferior cell amplification and CFC formation compared to controls in SFM-1, indicating that the components used in SFM-1 could replace exogenous serum. Four commercially available serum-free media resulted in either comparable or lower total cell and CFC yields as SFM-1. The transplantation potential of CD34(+) cells after culture in SFM-1 was assayed using limiting dilution analysis on preformed irradiated bone marrow stroma and revealed maintenance of long-term bone marrow culture initiating cell (LTCIC) levels during the culture period. These data indicate that HGF-supported multilineage ex vivo expansion of human CD34(+) hematopoietic progenitor cells is feasible using an IMDM-based culture medium which contains a restricted number of additives, resulting in analogous or improved yields of both primitive and differentiated cells compared to previously established protocols. We suggest that this culture protocol is of advantage when working with pharmaceutical-grade preparations under serum-free conditions.     

效应面法优化CHO细胞培养基中几种主要成分的含量

目的:考察培养基中葡萄糖、谷氨酰胺、血清、碳酸氢钠含量对CHO细胞生长繁殖的影响。方法:在CHO细胞培养基中添加不同成分的葡萄糖、谷氨酰胺、血清、碳酸氢钠,通过单因素实验结果结合Box-Behnken效应面法,根据二次回归模型的分析结果,以细胞表达蛋白体外活性为指标进行实验,考察培养基中葡萄糖、谷氨酰胺、血清、碳酸氢钠含量对细胞生长繁殖的影响。结果:根据回归方程分析结果,作出相应的曲面图和等高线图,优选出培养基中各组分的最佳配比为:葡萄糖2.54 g/L、谷氨酰胺0.59 g/L、血清8.3%,碳酸氢钠2.96 g/L。结论:Box-Behnken实验设计法用于细胞培养过程中考察培养基中各组分的优选是可行的,数学模型的预测值与实验观察值相符。通过对CHO细胞培养基成分的优化,使CHO细胞蛋白表达量高,有利于提高产品质量和降低生产成本。     

Difference in growth factor requirements of rat 3Y1 cells among growth in mass culture, clonal growth in low density culture, and stimulation to enter S phase in resting culture

A semiserum-free medium was developed for monolayer culture of rat 3Y1 fibroblastic cells. The main components of the developed medium added to Dulbecco's modified Eagle's medium (DMEM) were insulin, transferrin, epidermal growth factor, poly-D-lysine, bovine albumin, oleic acid, and bovine alpha-globulin. In this medium, 3Y1 cells grew in mass culture at much the same rate as in DMEM supplemented with 10% fetal bovine serum (FBS), and colonies, albeit of smaller sizes, did form. Virally transformed derivatives of 3Y1 (simian virus 40-3Y1, polyoma virus-3Y1 and adenovirus type 12-3Y1) also formed colonies in the semiserum-free medium. When trypsinized 3Y1 cells were seeded with the medium lacking alpha-globulin, neither growth in the mass culture nor clonal growth in the low density culture (clonal growth) occurred. In this case, cell spreading was inhibited by albumin, and this inhibition was overcome by adding alpha-globulin or treating dishes with serum. When albumin was excluded from the semiserum-free medium, clonal growth did not occur, whereas growth in mass culture and stimulation of DNA synthesis in the resting mass culture (stimulation of DNA synthesis) were not so drastically affected. When oleic acid was removed, growth in mass culture was inhibited considerably, but no considerable effect was seen on clonal growth or on stimulation of DNA synthesis. In the absence of insulin, stimulation of DNA synthesis was inhibited more markedly than when other components were removed, but such was not the case with growth in mass culture and clonal growth.     

Multifactorial interaction of growth factors on Penaeus monodon lymphoid cells and the impact of IGFs in DNA synthesis and metabolic activity in vitro

Development of continuous cell lines from shrimp is essential to investigate viral pathogens. Unfortunately, there is no valid cell line developed from crustaceans in general and shrimps in particular to address this issue. Lack of information on the requirements of cells in vitro limits the success of developing a cell line, where the microenvironment of a cell culture, provided by the growth medium, is of prime importance. Screening and optimization of growth medium components based on statistical experimental designs have been widely used for improving the efficacy of cell culture media. Accordingly, we applied Plackett–Burman design and response surface methodology to study multifactorial interactions between the growth factors in shrimp cell culture medium and to identify the most important ones for growth of lymphoid cell culture from Penaeus monodon. The statistical screening and optimization indicated that insulin like growth factor-I (IGF-I) and insulin like growth factor-II (IGF-II) at concentrations of 100 and 150 ng ml⁻¹, respectively, could significantly influence the metabolic activity and DNA synthesis of the lymphoid cells. An increase of 53 % metabolic activity and 24.8 % DNA synthesis could be obtained, which suggested that IGF-I and IGF-II had critical roles in metabolic activity and DNA synthesis of shrimp lymphoid cells.

Electronic supplementary material

The online version of this article (doi:10.1007/s10616-014-9697-0) contains supplementary material, which is available to authorized users.     

Growth of human normal erythroid progenitors in liquid culture: a comparison with colony growth in semisolid culture

Most studies of erythropoiesis in vitro have employed cloning methods in semisolid medium. We have recently described a two-step liquid culture procedure that supports the proliferation and differentiation of human erythroid progenitors. In the present study, we have modified the procedure to allow large-scale cultures of erythroid cells derived from normal donors. The culture is divided into two phases. In the first phase, which is erythropoietin (Epo) independent, the early erythroid progenitors multiply and differentiate. In the second, Epo-dependent phase, they mature into orthochromatic normoblasts and enucleated erythrocytes. Using this procedure, erythroid cell yield reached 7.5 x 10(6)/ml and a total of 7 x 10(8) cells could be harvested per blood unit. A comparison of the growth of erythroid cells in liquid culture to their colony growth in semisolid culture indicated that cell growth was superior: 1) in liquid culture in terms of cell yield per originally cultured mononuclear cell, 2) per ml culture and per culture surface area and in the purity of the resultant erythroid cell population. In addition, it permits easier manipulation of the culture condition and components and sampling of greater than 1 x 10(7) cells at each maturation stage subsequent to the proerythroblast stage. This liquid culture procedure might provide an important experimental tool for studying erythroid cell development.     

效应面法优化CHO细胞培养基中几种丰要成分的含量

目的：考察培养基中葡萄糖、谷氨酰胺、血清、碳酸氢钠含量对CHO细胞生长繁殖的影响。方法：在CHO细胞培养基中添加不同成分的葡萄糖、谷氨酰胺、血清、碳酸氢钠,通过单因素实验结果结合Box-Behnken效应面法,根据二次回归模型的分析结果,以细胞表达蛋白体外活性为指标进行实验,考察培养基中葡萄糖、谷氨酰胺、血清、碳酸氢钠含量对细胞生长繁殖的影响。结果：根据回归方程分析结果,作出相应的曲面图和等高线图,优选出培养基中各组分的最佳配比为：葡萄糖2．54g／L、谷氨酰胺O．59g／L、血清8．3％,碳酸氢钠2．96g／L。结论：Box—Behnken实验设计法用于细胞培养过程中考察培养基中各组分的优选是可行的,数学模型的预测值与实验观察值相符。通过对CHO细胞培养基成分的优化,使CHO细胞蛋白表达量高,有利于提高产品质量和降低生产成本。     

Bacterial shape maintenance: an evaluation of various models

In this article, we examine a large number of combinations of growth models, with separate attention to cell volume, cylindrical surface-area, polar caps, nascent poles, onset of constriction, precision of cell division and interdivision-time dispersion, for Escherichia coli cells growing in steady state at various doubling times. Our main conclusion is striking, and quite general: exponential cylindrical surface-area growth is not possible, irrespective of the behaviour of cell volume, the polar regions, the nascent poles, or any other feature of cell growth-such cells never reach steady state. The same is true of linear cylindrical surface-area growth, regardless of when during the cell cycle the doubling in growth rate takes place. Only after the introduction of feedback into the surface-area growth law, do the cultures attain steady state, all of them. The other components of the models contribute only marginally to the properties of the steady state. Thus, whether the feedback applies just to the cylindrical portion of the cell or to its entire surface area affects only the coefficient of variation of cell radius and the radius-volume correlation. The dynamics of old-pole maintenance, constant area or constant shape, influences the radius-length and radius-volume correlations and, to a much lesser extent, the coefficients of variation of cell radius and length; how the nascent poles grow, whether linearly or exponentially, does not seem to matter at all. The absolute dimensions of the cells are set by the growth rate of the culture and have almost no effect when the feedback is taken to apply to the entire cell surface area; when it is limited to the cylindrical portion of the cell, however, both radius-length and radius-volume correlations increase with increasing doubling time. Comparison with published values was inconclusive. The nature of cell surface-area growth has therefore been settled, but whether the volume increases by simple-exponential or by pseudo-exponential growth, or whether the old poles maintain a constant shape or a constant area during the cell cycle, can be determined only with more precise experimental data. The form of nascent-pole growth is not resolvable by present techniques.     

Digital image analysis of growth and starvation responses of a surface-colonizing Acinetobacter sp.

Surface growth of an Acinetobacter sp. cultivated under several nutrient regimens was examined by using continuous-flow slide culture, phase-contrast microscopy, scanning confocal laser microscopy, and computer image analysis. Irrigation of attached coccoid stationary-phase Acinetobacter sp. cells with high-nutrient medium resulted in a transition from coccoid to bacillar morphology. Digital image analysis revealed that this transition was biphasic. During phase I, both the length and the width of cells increased. In contrast, cell width remained constant during phase II, while both cell length and cell area increased at a rate greater than in phase I. Cells were capable of growth and division without morphological transition when irrigated with a low-nutrient medium. Rod-shaped cells reverted to cocci by reduction-division when irrigated with starvation medium. This resulted in conservation of cell area (biomass) with an increase in cell number. In addition, the changes in cell morphology were accompanied by changes in the stability of cell attachment. During phase I, coccoid cells remained firmly attached. Following transition in high-nutrient medium, bacillar cells displayed detachment, transient attachment, and drifting behaviors, resulting in a spreading colonization pattern. In contrast, cells irrigated with a low-nutrient medium remained firmly attached to the surface and eventually formed tightly packed microcolonies. It is hypothesized that the coccoid and bacillar Acinetobacter sp. morphotypes and associated behavior represent specialized physiological adaptations for attachment and colonization in low-nutrient systems (coccoid morphotype) or dispersion under high-nutrient conditions (bacillar morphotype).     

Control of cytolysis of BALB/c-3T3 cells by platelet-derived growth factor: a model system for analyzing cell death

In culture medium supplemented with 10% clotted blood serum, the saturation density of BALB/c-3T3 cells is determined jointly by cell replication and cell loss. By prelabelling cellular DNA with ³H-thymidine and also by time lapse photography, we studied cell loss independently of replication. Cell loss was accelerated when BALB/c-3T3 cells were transferred from serum-supplemented medium, which contains the platelet derived growth factor (PDGF), to medium supplemented with platelet-poor plasma which lacks it. Loss occurred via the disintegration of cell attached to the surface of the tissue culture dish. Cytolysis of individual cells occurred rapidly; less than 15 minutes transpired between the first indication of a perturbance (by phase contrast microscopy) and fragmentation of the cell cytoplasm. Kinetic analysis was consistent with random cell death rather than a fixed lifetime. The percentage of cells undergoing cytolysis was governed by the cell density; at high densities, such as are present in confluent cultures, a higher percentage of cell loss was noted than at low density. Cell death was antagonized by partially purified or electrophoretically homogenous preparations of-PDGF. Pure PDGF stimulated cell survivial at ng/ml in a concentration dependent fashion. The process of cell replication was not necessary for survival because PDGF prevented cytolysis in the presence of methotrexate, an inhibitor of DNA synthesis. A brief (4 hour) treatment with PDGF prevented cell death; such PDGF treated cells displayed increased survival after being taken up with trypsin and planted onto a fresh surface in plasma supplemented medium. Pituitary fibroblast growth factor, a functional analogue of PDGF for induc of DNA synthesis in BALB/c-3T3 cells, also functioned as an anticytolytic agent. By contrast, epidermal growth factor and insulin did not. Cytolysis of SV40-transformed cells occurred at a constitutively low rate and was insensitive to PDGF.     

The effect of modifying the culture medium on cell polarity in a human colon carcinoma cell line

The human colonic cancer cell line HT29, when grown in DMEM, forms a morphologically unpolarized cell culture in which the cells are covered with irregular microvilli and devoid of belt zonula occludens type tight junctions. However, by modifying the culture medium and growing the cells in RPMI, a different morphology was obtained. A large number of intracellular luminae appeared and at late confluency 90–95% of cells exhibited an epical brush border after four subsequent passages. Junctional complexes and a well developed zonula occludens were revealed under the apical brush border. Immuno-electron microscopical localization of specific markers, sucrase isomaltase (SI), secretory components (SC) and β2 microglobulin (β2M) showed that SI was limited to the apical surface, whereas 2M and SC were located at the basolateral surfaces. These results indicate that modification of culture conditions affects the ability of HT29 cells to express epithelial cell polarity.     

Difference in growth factor requirements of rat 3Y1 cells among growth in mass culture,clonal growth in low density culture,and stimulation to enter S phase in resting culture

Summary A semiserum-free medium was developed for monolayer culture of rat 3Y1 fibroblastic cells. The main components of the developed medium added to Dulbecco's modified Eagle's medium (DMEM) were insulin transferrin epidermal growth factor, poly-d-lysine, bovine albumin, oleic acid, and bovine α-globulin. In this medium, 3Y1 cells grew in mass culture at much the same rate as in DMEM supplemented with 10% fetat bovine serum (FBS), and colonies, albeit of smaller sizes, diddform. Virally transformed derivatives of 3Y1 (simian virus 40-3Y1, polyoma virus-3Y1 and adenovirus type 12-3Y1) also formed colonies in the semiserum-free medium. When trypsinized 3Y1 cells were seeded with the medium lacking α-globulin, neither growth in the mass culture nor clonal growth in the low density culture (clonal growth) occurred. In this case, cell spreading was inhibited by albumin, and this inhibition was overcome by adding α-globulin or treating dishes with serum. When albumin was excluded from the semiserum-free medium, clonal growth did not occur, whereas growth in mass culture and stimulation of DNA synthesis in the resting mass culture (stimulation of DNA synthesis) were not so drastically affected. When oleic acid was removed, growth in mass culture was inhibited considerably, but no considerable effect was seen on clonal growth or on stimulation of DNA synthesis. In the absence of insulin, stimulation of DNA synthesis was inhibited more markedly than when other components were removed, but such was not the case with growth in mass culture and clonal growth.     

Growth and differentiation of epidermal cells from the rainbow trout established as explants and maintained in various media

Growth and a number of differentiated characteristics of cultured epidermal cells from the rainbow trout Oncorhynchus mykiss were compared using two commercially available serum–free media, a dermal substrate/serum free kit and a serum–containing medium which had been previously optimized for epidermal cell culture. Each medium supported short term growth over 15 days. Only the medium supplied for dermal substrate culture supported longer growth periods. This medium was supplied for use with a collagen/stromal substrate but gave good cultures even without the substrate. Differentiation, measured by examining mucous cells, cytokeratins, epidermal growth factor receptor, gap junction status and ultrastructure showed that serum–free media gave quantitatively and qualitatively superior expression and short term retention of differentiation over serum–containing medium. Epithelial cell growth with expression of differentiated characteristics can be maintained in primary culture in serum–free medium for at least as long as in serum–containing medium. This provides a useful technique for use when serum presence in medium is undesirable or proves toxic to the specialized cell type under investigation.     

Formulation of a basal medium for primary cell culture of the marine sponge Hymeniacidon perleve

Marine sponge cell culture is a potential route for the sustainable production of sponge-derived bioproducts. Development of a basal culture medium is a prerequisite for the attachment, spreading, and growth of sponge cells in vitro. With the limited knowledge available on nutrient requirements for sponge cells, a series of statistical experimental designs has been employed to screen and optimize the critical nutrient components including inorganic salts (ferric ion, zinc ion, silicate, and NaCl), amino acids (glycine, glutamine, and aspartic acid), sugars (glucose, sorbitol, and sodium pyruvate), vitamin C, and mammalian cell medium (DMEM and RPMI 1640) using MTT assay in 96-well plates. The marine sponge Hymeniacidon perleve was used as a model system. Plackett-Burman design was used for the initial screening, which identified the significant factors of ferric ion, NaCl, and vitamin C. These three factors were selected for further optimization by Uniform Design and Response Surface Methodology (RSM), respectively. A basal medium was finally established, which supported an over 100% increase in viability of sponge cells.     

Nuclear GTP-binding proteins of Swiss 3T3 cells

The GTP-binding proteins of Swiss 3T3 cell nuclei were analyzed by filter binding assay and UV cross-linking analysis. The results showed the presence of multiple GTP-binding proteins in the nuclei. Scatchard analysis revealed that the Kd value for GTP binding to high-affinity components was 69 nM, that to low-affinity components being 2.7 microM. The GTP-binding activities of some nuclear proteins were found to change significantly in response to the growth conditions of the cells. During culture of cells in medium without serum, the GTP-binding activity of a 140 kDa protein clearly decreased, whereas that of a 40 kDa protein increased.     

Unit Cell Hypothesis for Streptococcus faecalis

The mass doubling times of exponential-phase cultures of Streptococcus faecalis were varied from 30 to 110 min by omitting glutamine from a defined growth medium and providing different concentrations of glutamate (ranging from 300 to 14 μg/ml). After Formalin fixation, cells were dried by the critical point method, and carbon-platinum replicas were prepared. The surface area and volume of cell poles seen in these replicas were estimated by a computer-assisted, three-dimensional reconstruction technique. It was found that the amount of surface area and volume of poles seen in these replicas were independent of the growth rate of culture from which the samples were taken. These observations were consistent with the unit cell model hypothesis of Donachie and Begg, in which a small number of surface sites would produce a constant amount of new cell surface regardless of the mass doubling time of the culture. However, measurements of the thickness of the cell wall taken from thin sections of the same cells showed that the cell wall increased in thickness as a function of the increase in cellular peptidoglycan content which occurs when the growth rate of this organism is slowed down by a decrease in glutamate concentration. Thus, it would seem that although the size of polar shells made by S. faecalis is invariant with growth rate, the amount of wall precursors used to construct these shells is not.     

Kinetic studies of polyhydroxybutyrate granule formation in Wautersia eutropha H16 by transmission electron microscopy

Wautersia eutropha, formerly known as Ralstonia eutropha, a gram-negative bacterium, accumulates polyhydroxybutyrate (PHB) as insoluble granules inside the cell when nutrients other than carbon are limited. In this paper, we report findings from kinetic studies of granule formation and degradation in W. eutropha H16 obtained using transmission electron microscopy (TEM). In nitrogen-limited growth medium, the phenotype of the cells at the early stages of granule formation was revealed for the first time. At the center of the cells, dark-stained "mediation elements" with small granules attached were observed. These mediation elements are proposed to serve as nucleation sites for granule initiation. TEM images also revealed that when W. eutropha cells were introduced into nitrogen-limited medium from nutrient-rich medium, the cell size increased two- to threefold, and the cells underwent additional volume changes during growth. Unbiased stereology was used to analyze the two-dimensional TEM images, from which the average volume of a W. eutropha H16 cell and the total surface area of granules per cell in nutrient-rich and PHB production media were obtained. These parameters were essential in the calculation of the concentration of proteins involved in PHB formation and utilization and their changes with time. The extent of protein coverage of the granule surface area is presented in the accompanying paper.     

Isolation and partial characterization of endothelial cell extracellular complexes

Human endothelial cells release components into the growth medium that stimulate cell-substratum adhesion. Several macromolecular components were isolated by ultracentrifugation of the endothelial cell conditioned medium. The components were heterogeneous, consisting of several sizes when examined by sedimentation velocity and gel filtration. When the extracellular components were evaluated by electron microscopy, structurally discrete particles were observed. The extracellular components and the complexes mediated cell-substratum adhesion to both human umbilical and arterial endothelial cells. The majority of the extracellular components that promote endothelial cell adhesion were pelleted by ultracentrifugation. Although the complexes contained fibronectin, antibodies to fibronectin did not inhibit cell adhesion to the complexes. Significant inhibition of endothelial cell adhesion was observed in the presence of heparin and heparan sulfate. The supernatant fraction following ultracentrifugation of the growth medium contained a component that suppressed endothelial cell adhesion to culture dishes coated with fibronectin, type I collagen, and endothelial cell complexes. SDS-polyacrylamide gel electrophoresis indicated that the complexes contained several components, and the majority of the large-molecular-weight components were pelleted by ultracentrifugation. The conditioned medium from human endothelial cells contains specific complexes that promote cell-substratum adhesion and components that suppress cell-substratum adhesion.