Adsorption of Mycoplasma pneumoniae to Neuraminic Acid Receptors of Various Cells and Possible Role in Virulence

Monkey, rat, and chicken tracheal epithelial cells, as well as monkey, rat, guinea pig, and chicken erythrocytes, adsorbed firmly to colonies of Mycoplasma pneumoniae and M. gallisepticum. Colonies of M. pulmonis also adsorbed erythrocytes but with less avidity than M. pneumoniae or M. gallisepticum; unlike the latter organisms, M. pulmonis did not adsorb tracheal epithelial cells. Colonies of M. orale type 1 and M. orale type 3 adsorbed only chicken red cells. Other mycoplasma species tested, including four of human origin and one of animal origin, did not adsorb red cells or epithelial cells. M. pneumoniae and M. gallisepticum appeared to attach to erythrocytes or tracheal epithelial cells by neuraminic acid receptors on these cells, whereas M. orale types 1 and 3 and M. pulmonis seemed to utilize another type or other types of receptors. Pretreatment of red cells or tracheal epithelial cells with receptor-destroying enzyme, neuraminidase, or influenza B virus removed the adsorption receptors for M. pneumoniae. Similarly, pretreatment of M. pneumoniae colonies with neuraminic acid-containing materials prevented adsorption of erythrocytes or respiratory tract cells. The adsorption sites on M. pneumoniae were specifically blocked by homologous but not heterologous antisera. This property made it possible to study the nature of the mycoplasma adsorption sites by testing the capacity of different fractions of the organism to block the action of adsorption-inhibiting antibodies. Such studies suggested that the mycoplasma binding sites were probably lipid or lipoprotein in nature. The glycerophospholipid hapten was implicated as one such site, since this serologically active hapten blocked the action of hemadsorption-inhibiting antibodies in M. pneumoniae rabbit antiserum. The affinity of M. pneumoniae for respiratory tract epithelium, unique among the mycoplasmas that infect man, may play a role in virulence, since this type of attachment provides an unusual opportunity for peroxide, secreted by the organism, to attack the tissue cell membrane without being rapidly destroyed by catalase or peroxidase present in extracellular body fluids.     

Promotion of Mycoplasma pulmonis growth in rat tracheal organ cultures by ammonium chloride

During exacerbation of respiratory mycoplasmosis in rats by environmental ammonia, numbers of Mycoplasma pulmonis organisms in the respiratory tract are increased. To test whether or not exposure of respiratory epithelium to ammonia in vitro promotes growth of the organism, rat tracheal organ cultures were treated with 50 mM ammonium chloride, inoculated with M. pulmonis, and quantitatively cultured. After 48 hours, treated tracheas harbored almost 10 times more M. pulmonis colony-forming units than control tracheas. Cellular lesions in the epithelium of treated tracheas resembled those in the nasal passages of rats exposed to gaseous ammonia. To determine whether or not growth-modifying factors were released from tracheal epithelium exposed to ammonium chloride, M. pulmonis growth was assessed in medium collected from ammonium chloride-treated and control tracheas. Growth in medium from treated tracheas was greater than that in medium from untreated tracheas.     

Detection of Mycoplasma pulmonis by fluorogenic nuclease polymerase chain reaction analysis

Mycoplasma pulmonis induces persistent infections in laboratory mice and rats and can contaminate biological materials. We developed a fluorogenic nuclease polymerase chain reaction (fnPCR) assay to detect M. pulmonis specifically. Primer and probe sequences for the assay were targeted to 16S rRNA sequences specific to M. pulmonis. The assay consistently detected the equivalent of fewer than 10 copies of template DNA. When evaluated against a panel of 24 species of bacteria, the M. pulmonis assay detected only M. pulmonis isolates. Evaluation of 10-fold serial dilutions of cultured M. pulmonis showed that the M. pulmonis fnPCR assay and culture on Dutch agar had comparable sensitivity in detecting viable M. pulmonis organisms, whereas the mouse antibody production test displayed positive serologic results at dilutions higher than those in which viable organisms could be detected. Finally, the M. pulmonis fnPCR assay was able to detect M. pulmonis DNA in nasopharyngeal wash fluid and trachea, lung, and uterus tissue collected from mice naturally infected with M. pulmonis but did not detect the organism in similar samples collected from uninfected, negative control mice. The M. pulmonis fnPCR assay provides a high-throughput, PCR-based method to detect M. pulmonis in infected rodents and contaminated biological materials.     

Inhibition of Sindbis Virus Release by Media of Low Ionic Strength: an Electron Microscope Study

Release of Sindbis virus from infected cells is inhibited by lowering the ionic strength of the medium. To determine the nature of the inhibited step, we examined, by electron microscopy, both freeze-etched and thin-sectioned preparations which had been fixed with either glutaraldehyde or formaldehyde. Inhibitory medium had two different effects on Sindbis virus release: virus budding was partially inhibited, and those virions which did mature were precipitated on the surface of the cell. Freeze-etched, inhibited cells showed very few viral buds. After shift to normal medium, the number of budding virions increased dramatically, far exceeding the quantity found in normal controls. Thus, low ionic strength medium clearly inhibited an early stage of virus maturation. The results were the same regardless of the fixative. Thin sections of glutaraldehyde-fixed, inhibited cells contained large extracellular aggregates of mature virus which were not present in similar, formaldehyde-fixed preparations. Fixation of radioactively-labeled, inhibited cultures revealed that approximately half of the virus that could be released from inhibited cells by raising the ionic strength of the medium could also be released by formaldehyde, but not by glutaraldehyde. This fraction probably represents mature virus attached to the cell surface by the ionic conditions.     

The effect of thiol-active compounds and sterols on the membrane-associated hemolysin of Mycoplasma pulmonis

Abstract Previous studies had shown that Mycoplasma pulmonis contained a bovine serum albumin-dependent, membrane-associated hemolysin. Biochemical analyses were performed to further characterize this activity. The membrane-associated hemolytic activity could be activated by dithiothreitol and β-mercaptoethanol, and inactivated by oxidizing compounds, a sulfhydryl inhibitor and heat treatment. Cholesterol and other sterols were inhibitory in a stereo-specific manner, but they did not interfere with adherence of M. pulmonis to red blood cells. These results indicated that once attached, the M. pulmonis hemolysin recognized cholesterol in the opposing membrane leading to red cell lysis. Because of the unique location of this toxin and its sensitivity to cholesterol, the mycoplasma membrane hemolysins may belong to a unique class of bacterial toxins.     

Parathyroid cell variants may be provoked during immersion fixation

Parathyroid glands of cattle, dogs, cats, mice and rats were immersed in glutaraldehyde or mixtures consisting of glutaraldehyde, formaldehyde and acrolein in either Na-phosphate, Na/K-phosphate or Na-cacodylate buffer, and postfixed with OsO4 in the same buffers or, alternatively, in s-collidine. Excellent preservation of bovine, feline and murine parathyroid glands was achieved with fixation mixtures containing 1% glutaraldehyde, 1.5-2% formaldehyde and 2.5-5% acrolein in 0.1 M Na-cacodylate with or without Ca2+ and Mg2+, Na-phosphate or Na/K-phosphate at 4 degrees C followed by postfixation with 1% OsO4 in the same buffers or in s-collidine containing sucrose, Ca2+ and Mg2+. This procedure largely abolished the occurrence of parathyroid cell variants. Bovine parathyroid glands were also satisfactorily preserved with 1% glutaraldehyde and 2% formaldehyde whereas 1% glutaraldehyde and 2.5 or 5% acrolein, lower or higher buffer osmolarity, or immersion at room temperature led to vacuolization of RER and to breakdown of membranes. In contrast, all fixation protocols led to the formation of dark and light cell variants and to multinucleated syncytial cells in dog and rat parathyroids. The results thus show that parathyroid cell variants arise during immersion fixation and that aldehydes, buffers and temperature are important factors for provoking parathyroid cell variants.     

Mycoplasma pneumoniae attachment to glutaraldehyde-treated human WiDr cell cultures

Attachment of Mycoplasma pneumoniae to host cells initiates disease, and the attachment components may represent important protective immunogens for preventing disease. We have studied the mechanisms of attachment using in vitro cell culture systems and selected pathogenic and nonpathogenic strains of M. pneumoniae. Attachment of the pathogenic strains M129 and PI-1428 was several fold greater than attachment of the nonpathogenic strain, and attachment of strains M129 and PI-1428 was reduced by 21 to 63% when human WiDr cell monolayers were exposed to neuraminidase, supporting the concept that M. pneumoniae attaches to mammalian cells by a neuraminidase-sensitive glycoconjugate. While attachment of the two pathogenic strains was markedly reduced by treating the WiDr cells with glutaraldehyde, glutaraldehyde treatment produced minimal effects on the attachment of the nonpathogenic strain B176. Glutaraldehyde treatment also altered the temperature dependence of attachment by the pathogenic strains. Because glutaraldehyde-treated WiDr cell monolayers showed little difference in attachment between pathogenic and nonpathogenic strains, glutaraldehyde-treated cells are not appropriate cell substrates for studying M. pneumoniae attachment mechanisms or identifying immunogens for vaccine development.     

Retinoic acid modulates attachment of mouse fibroblasts to laminin substrates

The effect of retinoic acid treatment on cell attachment to plastic substrates precoated with fibronectin, gelatin, laminin, and type IV collagen was investigated. Both retinoic acid-treated and control cells attached efficiently to fibronectin or gelatin substrates without any significant difference. In contrast, retinoic acid-treated cells attached to laminin or type IV collagen substrates, while control cells showed little or no attachment. The minimal effective concentration of retinoic acid for pretreatment to yield a significant increase in the attachment assay was higher than 10(-8) M. The attachment of retinoic acid-treated cells to laminin substrates reached a maximum at 60 min, while that to type IV collagen substrates had a time lag and did not reach a maximum by 60 min. The effect of retinoic acid treatment reached a maximum at 2 days and was partly reversible. These results suggest that retinoic acid may increase NIH/3T3 cell adhesion through an effect on laminin receptors. Other mouse fibroblast lines, 3T3-Swiss, 3T6-Swiss, Balb/3T3, and Balb/3T12-3 (spontaneously transformed Balb/3T3), responded to retinoic acid treatment in a manner similar to that of NIH/3T3 cells. However, the virus-transformed Balb/3T3 lines, SV-T2 and M-MSV, showed significant attachment to laminin substrates without retinoic acid treatment, and retinoic acid did not affect or slightly decreased the cell attachment to laminin substrates.     

Leishmania mexicana mexicana: Attachment and Uptake of Promastigotes to and by Macrophages In Vitro1

Promastigotes of Leishmania mexicana mexicana attach to mouse macrophages in vitro in the absence of serum by a wheat germ agglutinin-like ligand on the surface of the promastigote that binds to the N-acetyl glucosamine moiety of a receptor on the surface of the macrophage. The binding is temperature dependent, and the macrophage receptor is trypsin, cytochalasin B, and glutaraldehyde sensitive. The promastigote ligand is proteolytic enzyme and glutaraldehyde insensitive. Uptake follows attachment and is assisted or inhibited as for attachment. Treatment of promastigotes with proteolytic enzymes uncovers a receptor for a serum component that binds strongly to a mouse macrophage receptor in vitro. The strain of mice donating the macrophages had little effect upon attachment and uptake except that A strain mouse macrophages attached fewer promastigotes in 10 min than those of outbred mice, but took up as many promastigotes over 90 min as those of outbred mice. Low responder Biozzi mouse macrophages took up more promastigotes than high responder Biozzi mouse macrophages. Normal unheated human, rabbit, and guinea pig sera lysed promastigotes and so inhibited their attachment to macrophages in vitro. Unheated immune serum showed an enhanced inhibition of attachment. Heated normal serum allowed attachment and uptake, while promastigotes treated with heated immune serum showed enhanced attachment to and uptake by macrophages. Treatment of macrophages in vitro with immune serum enhanced their ability to attach promastigotes and to engulf them. Repeated 90-min exposures of a population of promastigotes to uptake by mouse macrophages in vitro did not deplete the population of any sub-population more likely to be taken by macrophages. The first sub-population to be taken up survived better in macrophages over 24 h than subsequently engulfed sub-populations.     

Effects of laminin, fibronectin and type IV collagen on liver cell cultures

The effects on mouse liver cells of laminin, fibronectin and type IV collagen, all of which are the main matrix of the basement membrane, were studied. Laminin, a glycoprotein isolated from cultures of rat yolk sac carcinoma cells, promoted the attachment of mouse fetal liver cells to laminin-coated dishes, but did not have a strong influence upon the attachment of normal adult liver cells. On the other hand, fibronectin which was purified from mouse plasma promoted the attachment of adult liver cells but not that of fetal liver cells. The number of neonatal liver cells attached to the surfaces coated was intermediate between those of fetal and adult liver cells in each matrix. DNA synthesis and cell proliferation during the culture of full-term fetal liver cells in laminin-coated dishes were higher than those in fibronectin- or type IV collagen-coated dishes. The amount of alpha-fetoprotein secreted in the laminin-coated dishes was more than in other groups. No differences in secretion of albumin into media, however, were observed in either group. These results suggest that laminin may be necessary for cell growth, tissue organization and cell differentiation during the normal development of liver in vivo.     

Male germ cell specific sulfogalactoglycerolipid is recognized and degraded by mycoplasmas associated with male infertility

Sulfogalactosylglycerolipid (SGG) is the major mammalian male germ cell glycolipid and has been implicated in sperm/egg binding. Mycoplasma pulmonis, a species of Mollicutes, is associated with male infertility in rodents. Purified SGG incubated in the presence of M. pulmonis was enzymatically degraded by both desulfation and deacylation. Desulfation occurred primarily at alkaline pH, and deacylation also increased with increased pH, indicating that these represent novel enzymatic activities. Digestion was facilitated, but not dependent on, the presence of detergent. Rat spermatozoa exposed to M. pulmonis showed a reduction in SGG content which was particularly marked for cauda (mature) spermatozoa. With the aid of tlc overlay binding procedure, intact M. pulmonis were found to bind specifically to sulfated glycolipids and thus SGG may provide the cell membrane receptor for this organism. The topology of mycoplasma binding to rat sperm was consistent with the known topology of sperm SGG. The reduced binding (and subsequent digestion) of caput spermatozoan SGG correlates with the membrane colocalization of SGG and its endogenous binding protein at this stage. Separation of SGG and its binding protein during epididymal sperm maturation appears to facilitate M. pulmonis binding to and digestion of cauda sperm SGG. The binding and degradation of the sperm SGG by M. pulmonis may play a role in the induction of infertility which follows infection with these organisms by interfering in sperm/egg receptor recognition.     

Characterization of the ftsZ gene from Mycoplasma pulmonis, an organism lacking a cell wall.

The ftsZ gene is required for cell division in Escherichia coli and Bacillus subtilis. In these organisms, FtsZ is located in a ring at the leading edge of the septum. This ring is thought to be responsible for invagination of the septum, either causing invagination of the cytoplasmic membrane or activating septum-specific peptidoglycan biosynthesis. In this paper, we report that the cell division gene ftsZ is present in two mycoplasma species, Mycoplasma pulmonis and Acholeplasma laidlawii, which are eubacterial organisms lacking a cell wall. Sequencing of the ftsZ homolog from M. pulmonis revealed that it was highly homologous to other known FtsZ proteins. The M. pulmonis ftsZ gene was overexpressed, and the purified M. pulmonis FtsZ bound GTP. Using antisera raised against this purified protein, we could demonstrate that it was expressed in M. pulmonis. Expression of the M. pulmonis ftsZ gene in E. coli inhibited cell division, leading to filamentation, which could be suppressed by increasing expression of the E. coli ftsZ gene. The implications of these results for the role of ftsZ in cell division are discussed.     

Role of Pasteurella pneumotropica and Mycoplasma pulmonis in Murine Pneumonia

Pneumonia caused by Mycoplasma pulmonis and Pasteurella pneumotropica was studied in conventional, specific pathogen-free (SPF), and germ-free mice. When P. pneumotropica was serially passed in conventional mice, M. pulmonis, as well as P. pneumotropica, was recovered from mice with gross lesions. When M. pulmonis was serially passed in conventional mice, both organisms were recovered. SPF mice given a nasal instillation of M. pulmonis alone, P. pneumotropica alone, or a combination of the two developed pneumonia when both organisms were present. These findings suggested that both organisms contribute to typical murine pneumonia. That M. pulmonis might be an L form of P. pneumotropica was suggested because some SPF mice inoculated with either organism yielded both on culture. This possibility was investigated with mole per cent guanine plus cytosine (GC) content and nucleic acid hybridization techniques. The GC content of P. pneumotropica is 42.2 mole per cent and that of M. pulmonis is 28.6 mole per cent. No specific hybrids between deoxyribonucleic acid (DNA) from M. pulmonis and DNA from P. pneumotropica were detected. This and the wide disparity in GC content showed that M. pulmonis is not an L form of P. pneumotropica. In germ-free mice, intranasal instillation with either organism alone produced pneumonia. The lesions produced when each organism was inoculated independently were characterized by areas of consolidation with perivascular and peribronchial lymphocytic infiltration. Qualitatively, the lesions produced when both organisms were inoculated simultaneously more closely resembled those seen in naturally occurring murine pneumonia. Statistical analysis indicated that the quantitative effect of the two organisms was additive. The indirect fluorescent antibody technique was used to locate organisms in lung tissue sections. M. pulmonis localized in the bronchial epithelium and P. pneumotropica localized in the alveolar lesions.     

Modulation by centrifugation of cell susceptibility to chlamydial infection.

Enhancement of chlamydial infection of cell monolayers by centrifugation was shown to depend on induced cell surface changes. Evidence for this came from analysis of two forms of organism attachment which take place during centrifugation. In 'productive binding', organisms attached to cells and then entered and infected them. In 'unproductive binding', organisms became attached to cells but were not ingested. These organisms could be stripped from the cells by treatment with trypsin and could then infect fresh monolayers. Measurement of attachment kinetics during centrifugation showed that cells passed through three different susceptibility states. Only productive binding occurred in the first 20 min; cells then entered a refractory state during which no attachment took place At about 45 min, attachment recommenced but this allowed only unproductive binding. Induced movement of cell surface structures may enhance infection by promoting specific or non-specific interactions. Failure of ingestion may result from insufficient cell 'receptors' for circumferential binding of the whole chlamydial surface so that engulfment cannot take place.     

Dextran T 500--induction of spreading in Ehrlich ascites tumour cells on glass surface

The experiments were carried out in order to find factors which could induce attachment and spreading of Ehrlich ascites tumour (EAT) cells on solid substrata. In normal culture media, serum-free as well as serum-containing, these cells did not spread and very weakly attached onto glass. It was found that after coating the cell surfaces with dextran T 500 the EAT cells strongly attached and spread extensively on glass. This spreading could be inhibited or reversed by washing out the dextran or adding calf serum. Dextran T 500 caused rapid spreading also in chick embryo fibroblasts and mouse lymphocytes. Some aspects of these results in connection with contemporary views concerning the processes of cell attachment and spreading are discussed.     

Hapten-specific T lymphocyte activation by glutaraldehyde-treated macrophages: an argument against antigen processing by macrophages.

In this communication the effects of glutaraldehyde treatment of trinitrophenyl-(TNP) modified macrophages on their ability to stimulate TNP-specific guinea pig T lymphocyte proliferation were studied. TNP-modified macrophages briefly treated with glutaraldehyde retained much of their ability to stimulate TNP-primed T cells. In contrast, similar treatment of allogeneic macrophages or soluble protein antigen-pulsed syngeneic macrophages completely eliminated their ability to stimulate a mixed leukocyte reaction or protein antigen-specific proliferation, respectively. TNP-modification did not appear to interfere with glutaraldehyde reactivity since macrophages treated with glutaraldehyde before or after TNP-modification stimulated equivalent T cell responses. However, glutaraldehyde treatment of TNP-modified macrophages that had been cultured overnight dramatically reduced their ability to stimulate TNP-specific T cells. Glutaraldehyde-treated TNP-modified macrophages also expressed the same genetic restrictions of T cell activation as untreated stimulators. Thus, T cells primed with syngeneic TNP-modified macrophages were restimulated only by glutaraldehyde-treated TNP-modified syngeneic, but not by allogeneic, macrophages. These results are discussed with respect to the nature of the TNP-specific immunogen recognized by T cells.     

Analysis of alpha 1 beta 1, alpha 2 beta 1 and alpha 3 beta 1 integrins in cell--collagen interactions: identification of conformation dependent alpha 1 beta 1 binding sites in collagen type I.

Integrins can mediate the attachment of cells to collagen type I. In the present study we have investigated the possible differences in collagen type I recognition sites for the alpha 1 beta 1 and alpha 2 beta 1 integrins. Different cyanogen bromide (CB) fragments of the alpha 1 (I) collagen chain were used in cell attachment experiments with three rat cell types, defined with regard to expression of collagen binding integrins. Primary rat hepatocytes expressed alpha 1 beta 1, primary rat cardiac fibroblasts alpha 1 beta 1 and alpha 2 beta 1, and Rat-1 cells only alpha 2 beta 1. All three cell types expressed alpha 3 beta 1 but this integrin did not bind to collagen--Sepharose or to immobilized collagen type I in a radioreceptor assay. Hepatocytes and cardiac fibroblasts attached to substrata coated with alpha 1(I)CB3 and alpha 1(I)CB8; Rat-1 cells attached to alpha 1(I)CB3 but only poorly to alpha 1(I)CB8-coated substrata. Cardiac fibroblasts and Rat-1 cells spread and formed beta 1-integrin-containing focal adhesions when grown on substrata coated with native collagen or alpha 1(I)CB3; focal adhesions were also detected in cardiac fibroblasts cultured on alpha 1(I)CB8. The rat alpha 1 specific monoclonal antibody 3A3 completely inhibited hepatocyte attachment to alpha 1(I)CB3 and alpha 1(I)CB8, as well as the attachment of cardiac fibroblasts to alpha 1(I)CB8, but only partially inhibited the attachment of cardiac fibroblasts to alpha 1(I)CB3. 3A3 IgG did not inhibit the attachment of Rat-1 cells to collagen type I or to alpha 1(I)CB3.(ABSTRACT TRUNCATED AT 250 WORDS)     

Immunologic reactions of rabbit anti-Mycoplasma arthritidis serum with in vitro cultivated rat synovial cells

Summary The pathogenesis of the intra-articular, arthritic-inflammatory reaction caused byMycoplasma arthritidis in susceptible rats and mice is poorly understood. To investigate this problem, synovial cells from normal Sprague-Dawley rats were cultured and studied in vitro. These cells continued to produce hyaluronic acid as measured by viscosity and chemical assays. Normal synovial cells were treated with rabbit serum specimens taken before and after immunization withM. arthritidis. Cytotoxicity assays indicated that the cells were killed in the presence of rabbit anti-M. arthritidis serum but not with preimmunization serum specimens. The anti-M. arthritidis serum was not cytotoxic to monolayer cultures of HEp-2, Vero, or L-cells. Antiserum produced in response toM. fermentans, M. hominis, andM. pulmonis did not produce a cytotoxic effect on the cultured synovial cells. From immunofluorescence studies it was demonstrated that the interactions occurred between the rabbit anti-M. arthritidis serum and synovial cell surface antigens. Extreme precautions were taken to prevent mycoplasmal contamination of rats and the synovial cells in culture. These observations would appear to support previous reports implicating mycoplasmas as biological triggering mechanisms of autoimmune reactions. This research was supported in part by funds from the National Science Foundation, Grant DPP72-05787, and the U.S. Veterans Administration.     

Susceptibility of newly established mouse strain MPS to Mycoplasma pneumoniae infection

A newly established mouse strain, MPS, which is more sensitive to Mycoplasma pulmonis than ICR, ddY and other mouse strains was examined for its susceptibility to Mycoplasma pneumoniae. In experimental infections with M. pneumoniae, it was observed that M. pneumoniae attached to tracheas of MPS mice, and M. pneumoniae cells were isolated from tracheas and lungs of MPS mice even after four weeks of infection, while no mycoplasmas were isolated from ICR and ddY mice after one week of infection. Specific antibodies against M. pneumoniae were also observed by the Western blotting in the sera of MPS mice infected with M. pneumoniae. Although any lung lesion could not be observed in this work, this newly established mouse strain MPS may be useful for experiments of M. pneumoniae infection, especially for the analysis of strain differences in susceptibility to M. pneumoniae infection.     

Growth of endothelial and HeLa cells on a new multipurpose microcarrier that is positive,negative or collagen coated

A new cell culture microcarrier that can be covalently bonded by cell attachment proteins and can be thin-sectioned for electron microscopy was synthesized. It was easily made by sulfonating cross-linked polystyrene beads for a negative surface charge followed by covalent attachment of polyethylenimine for a positive charge. Cell attachment proteins, e.g. collagen, was covalently bonded directly to the microcarrier using a carbodiimide or after activating the microcarrier surface with glutaraldehyde. HeLa-S3 cells attached, spread and grew to confluence more efficiently on the positive microcarriers and those coated with collagen than on the negative ones. Endothelial cells grew best on those with a negative surface charge. The nature of the microcarrier surface was not the only aspect involved in cell adhesion but also the type of serum proteins adsorbed. Qualitatively different proteins coated the microcarriers depending upon whether the carrier was negative, positive or coated with collagen. Comparison of various types of available microcarriers indicated that the modified cross-linked polystyrene beads used here were best for transmission and scanning electron microscopy. Endothelial cells grown on the microcarriers had the same ultrastructure as cells grown in monolayers in culture dishes. Of a variety of microcarriers tested the modified cross-linked polystyrene beads were the only ones that could be used for both ultrastructural and biochemical techniques.