Nerve growth cones isolated from fetal rat brain. IV. Preparation of a membrane subfraction and identification of a membrane glycoprotein expressed on sprouting neurons

This study describes the preparation of a membrane subfraction from isolated nerve growth cone particles (GCPs) (see Pfenninger, K. H., L. Ellis, M. P. Johnson, L. B. Friedman, and S. Somlo, 1983, Cell, 35:573- 584) and the identification in this fraction of a glycoprotein expressed during neurite growth. While approximately 40 major polypeptides are visible in Coomassie Blue-stained SDS polyacrylamide gels of pelleted (partially disrupted) GCPs, a salt-washed membrane fraction prepared from lysed, detergent-permeabilized GCPs contains only 14% of this protein and has an unusually simple polypeptide pattern of seven major bands. Monoclonal antibodies have been generated to GCP membranes isolated from fetal rat brain. These antibodies have been screened differentially with synaptosomes from adult rat brain in order to identify those which recognize antigens expressed selectively during neurite growth. One such antibody (termed 5B4) recognizes a developmentally regulated membrane glycoprotein that is enriched in GCP membranes and expressed in fetal neurons sprouting in vitro. The 5B4 antigen in fetal brain migrates in SDS polyacrylamide gels as a diffuse band of approximately 185-255 kD, is rich in sialic acid, and consists of a small family of isoelectric variants. Freezing-thawing and neuraminidase digestion result in the cleavage of the native antigen into two new species migrating diffusely around 200 and 160 kD. Prolonged neuraminidase digestion sharpens these bands at about 180 and 135 kD, respectively. In the mature brain, antibody 5B4 recognizes a sparse polypeptide migrating at approximately 140 kD. As shown in the following paper (Wallis, I., L. Ellis, K. Suh, and K. H. Pfenninger, 1985, J. Cell Biol., 101:1990-1998), the fetal antigen is specifically associated with regions of neuronal sprouting and, therefore, can be used as a molecular marker of neurite growth.     

Coordination of growth and differentiation in the fetal lung

The male fetal lung begins to synthesize surfactant later in gestation than the female. This delay appears to be caused by androgens. We hypothesized that male fetal lung differentiation is delayed as a consequence of an extended phase of growth which is elicited by androgens. We observed that in vivo fetal lung protein synthesis relative to DNA synthesis peaked earlier in gestation in the female fetal lung and that this event was synchronous with the onset of differentiation. Pregnant rats were treated with dihydrotestosterone (DHT) during pregnancy, and fetal lung growth parameters were measured. Lung wet weight, dry weight, and DNA and protein concentrations were significantly elevated by DHT treatment. Type II cells and fibroblasts were isolated from lungs of DHT-treated fetuses. The number of total cells recovered was increased by 30%; the number of type II cells recovered was increased by 87%; and the number of fibroblasts recovered was increased by 42%. The type II cells which were recovered exhibited increased incorporation of [3H]thymidine into DNA and a reduced ratio of radiolabeled protein to radiolabeled DNA compared to that of cells from control lungs. Further studies were done in vitro with fibroblasts and type II cells isolated from untreated fetal rat lungs. Treatment of the fibroblasts with DHT during culture caused an increase in thymidine incorporation into DNA. This effect was not blocked by simultaneous treatment with cortisol, which normally causes reduced DNA synthesis and induces fibroblast differentiation. Treatment of the type II cells with DHT in culture caused a dose-dependent increase in cell number but a decrease in synthesis of disaturated phosphatidylcholine. These studies provide more direct evidence of the interrelationships between the control of growth and the control of differentiation in the fetal lung. DHT, a signal which delays the onset of expression of differentiation, also induces growth. We conclude that the controls of growth and of differentiation of the fetal lung are reciprocally linked.     

Comparison of effects of epidermal and insulin-like growth factors,gastrin releasing peptide and retinoic acid on fetal lung cell growth and maturation in vitro

The role in cell multiplication and maturation of several factors present in the late fetal lung was explored on isolated fetal rat pulmonary fibroblasts and alveolar epithelial type II cells cultivated in serum-free medium. The low degree of reciprocal contamination of each cell population was assessed by immunocytochemistry. Epidermal Growth Factor (EGF) stimulated thymidine incorporation and DNA accumulation in both cell types. In type II cells, it increased labeled-choline incorporation into surfactant phosphatidylcholine (PC), consistently with previous data obtained with lung explant cultures, but not into non-surfactant PC. Insulin-like growth factor (IGF)-I slightly stimulated DNA accumulation in fibroblasts although it did not significantly stimulate thymidine incorporation, contrary to IGF-II which presented a dose-dependent stimulating activity of thymidine incorporation. Neither IGF-I nor IGF-II stimulated type II cell growth. IGFs thus appear to primarily control the growth of lung mesenchyme. In type II cells, they stimulated the most non-surfactant PC biosynthesis. Gastrin releasing peptide (GRP) which was recently reported to promote fetal lung growth in vivo and to stimulate surfactant biosynthesis in lung organ culture revealed as a growth factor for type II cells only, at concentrations below 10 ⁻⁹ M. At concentration 10 ⁻⁸ M, although it did not affect DNA synthesis, GRP tended to increase surfactant and non-surfactant-PC biosynthesis. Retinoic acid inhibited thymidine incorporation into type II cells on a dose-dependent manner but nevertheless enhanced surfactant-PC biosynthesis to a similar extent as EGF. It is suggested that retinoic acid may represent a differentiation or maturation factor for the alveolar epithelium.     

Comparison of effects of epidermal and insulin-like growth factors, gastrin releasing peptide and retinoic acid on fetal lung cell growth and maturation in vitro.

The role in cell multiplication and maturation of several factors present in the late fetal lung was explored on isolated fetal rat pulmonary fibroblasts and alveolar epithelial type II cells cultivated in serum-free medium. The low degree of reciprocal contamination of each cell population was assessed by immunocytochemistry. Epidermal Growth Factor (EGF) stimulated thymidine incorporation and DNA accumulation in both cell types. In type II cells, it increased labeled-choline incorporation into surfactant phosphatidylcholine (PC), consistently with previous data obtained with lung explant cultures, but not into non-surfactant PC. Insulin-like growth factor (IGF)-I slightly stimulated DNA accumulation in fibroblasts although it did not significantly stimulate thymidine incorporation, contrary to IGF-II which presented a dose-dependent stimulating activity of thymidine incorporation. Neither IGF-I nor IGF-II stimulated type II cell growth. IGFs thus appear to primarily control the growth of lung mesenchyme. In type II cells, they stimulated the most non-surfactant PC biosynthesis. Gastrin releasing peptide (GRP) which was recently reported to promote fetal lung growth in vivo and to stimulate surfactant biosynthesis in lung organ culture revealed as a growth factor for type II cells only, at concentrations below 10(-9) M. At concentration 10(-8) M, although it did not affect DNA synthesis, GRP tended to increase surfactant and non-surfactant-PC biosynthesis. Retinoic acid inhibited thymidine incorporation into type II cells on a dose-dependent manner but nevertheless enhanced surfactant-PC biosynthesis to a similar extent as EGF. It is suggested that retinoic acid may represent a differentiation or maturation factor for the alveolar epithelium.     

Regulated plasmalemmal expansion in nerve growth cones

To study the mechanisms underlying plasmalemmal expansion in the nerve growth cone, a cell-free assay was developed to quantify membrane addition, using ligand binding and sealed growth cone particles isolated by subcellular fractionation from fetal rat brain. Exposed versus total binding sites of 125I-wheat germ agglutinin were measured in the absence or presence of saponin, respectively, after incubation with various agents. Ca2(+)-ionophore A23187 in the presence of Ca2+ increases the number of binding sites (Bmax) but does not change their affinity (KD), indicating that new receptors appear on the plasma membrane. Similarly, membrane depolarization by high K+ or veratridine significantly induces, in a Ca2(+)-dependent manner, the externalization of lectin binding sites from an internal pool. Morphometric analysis of isolated growth cones indicates that A23187 and high K+ treatment cause a significant reduction in a specific cytoplasmic membrane compartment, thus confirming the lectin labeling results and identifying the plasmalemmal precursor. The isolated growth cones take up gamma-amino-butyric acid and serotonin, but show no evidence for Ca2(+)-dependent transmitter release so that transmitter exocytosis is dissociated from plasmalemmal expansion. The data demonstrate that plasmalemmal expansion in the growth cone is a regulated process and identify an internal pool of precursor membrane.     

Membrane glycoproteins of the nerve growth cone: diversity and growth regulation of oligosaccharides

A subcellular fraction prepared from fetal rat brain and enriched in growth cone membranes is analyzed for its lectin-binding proteins. Growth-associated glycoproteins are identified by comparing the growth cone glycoproteins with those of synaptosomes. Protein was resolved in one- or two-dimensional gels, electroblotted, and blots probed with radioiodinated concanavalin A, wheat germ agglutinin, and Ricinus communis agglutinins I and II. In one-dimensional gels, each lectin recognizes approximately 20 polypeptides (with substantial overlap) most of which migrate diffusely and have relatively high molecular masses (range 30-200 kD). The seven major Coomassie-staining proteins of the membrane fraction (34-52 kD) are not the major lectin-binding proteins. In two-dimensional gels, the lectin-binding proteins are either streaked across the pH gradient or exist as multiple spots, indicating broad charge heterogeneity. Seven wheat germ agglutinin- and Ricinus communis agglutinin II-binding glycoproteins are present in greater abundance in growth cone fractions compared with synaptosomes. Most notably, an acidic, sialic acid-rich protein (27-30 kD, pI 4.0; termed gp27-30) is most abundant at postnatal day 4, but absent from adult brain. The protein's very acidic isoelectric point is due, at least in part, to its high sialic acid content. Growth regulation of specific protein-linked oligosaccharides suggests that they play a special role in growth cone function. In addition, the great diversity of growth cone glycoproteins from whole brain suggests glycoprotein heterogeneity among growth cones from different neuron types.     

Functional roles for the cytoplasmic domain of the type III transforming growth factor beta receptor in regulating transforming growth factor beta signaling

Transforming growth factor beta (TGF-beta) signals through three high affinity cell surface receptors, TGF-beta type I, type II, and type III receptors. The type III receptor, also known as betaglycan, binds to the type II receptor and is thought to act solely by "presenting" the TGF-beta ligand to the type II receptor. The short cytoplasmic domain of the type III receptor is thought to have no role in TGF-beta signaling because deletion of this domain has no effect on association with the type II receptor, or with the presentation role of the type III receptor. Here we demonstrate that the cytoplasmic domains of the type III and type II receptors interact specifically in a manner dependent on the kinase activity of the type II receptor and the ability of the type II receptor to autophosphorylate. This interaction results in the phosphorylation of the cytoplasmic domain of the type III receptor by the type II receptor. The type III receptor with the cytoplasmic domain deleted is able to bind TGF-beta, to bind the type II receptor, and to enhance TGF-beta binding to the type II receptor but is unable to enhance TGF-beta2 signaling, determining that the cytoplasmic domain is essential for some functions of the type III receptor. The type III receptor functions by selectively binding the autophosphorylated type II receptor via its cytoplasmic domain, thus promoting the preferential formation of a complex between the autophosphorylated type II receptor and the type I receptor and then dissociating from this active signaling complex. These studies, for the first time, elucidate important functional roles of the cytoplasmic domain of the type III receptor and demonstrate that these roles are essential for regulating TGF-beta signaling.     

Herpes simplex virus-1 utilizes the host actin cytoskeleton for its release from axonal growth cones

Herpes simplex virus type 1 (HSV-1) has evolved mechanisms to exploit the host cytoskeleton during entry, replication and exit from cells. In this study, we determined the role of actin and the molecular motor proteins, myosin II and myosin V, in the transport and release of HSV-1 from axon termini, or growth cones. Using compartmentalized neuronal devices, we showed that inhibition of actin polymerization, but not actin branching, significantly reduced the release of HSV-1 from axons. Furthermore, we showed that inhibition of myosin V, but not myosin II, also significantly reduced the release of HSV-1 from axons. Using confocal and electron microscopy, we determined that viral components are transported along axons to growth cones, despite actin or myosin inhibition. Overall, our study supports the role of actin in virus release from axonal growth cones and suggests myosin V as a likely candidate involved in this process.     

Intracellular regulators of neuronal sprouting: calmodulin-binding proteins of nerve growth cones

The focus of this study is a quantitative biochemical analysis of the calcium-dependent interactions of calmodulin with a nerve growth cone preparation from fetal rat brain (Pfenninger, K. H., L. Ellis, M. P. Johnson, L. B. Freidman, and S. Somlo, 1983, Cell 35:573-584). The presence of calmodulin as an integral component of this preparation is demonstrated, and quantitative binding studies are presented. The binding of 125I-calmodulin to nerve growth cone material is shown to be highly specific, calcium dependent, and saturable at nanomolar calmodulin concentrations. Additionally, the growth cones' binding components appear to be membrane proteins. The individual molecular mass species of growth cone proteins displaying calcium-dependent calmodulin binding are also detailed and presented in comparison with those of synaptosomes. This analysis reveals differences between the calmodulin binding proteins of the growth cone preparation and the synaptosome fraction, suggesting the presence in growth cones of a specialized set of components which may be involved in regulatory mechanisms controlling neuritic sprouting.     

Paracrinology of growth regulation

Embryonic and fetal growth is dependent on genetic factors and epigenetic factors such as peptide growth factors. We describe here the interactions of several peptide growth factors during the growth and function of two cell types, growth plate chondrocytes from the ovine fetus and astroglial cells from the newborn rat cerebral cortex. Isolated chondrocytes released two endogenous growth factors, basic fibroblast growth factor (bFGF) and insulin-like growth factor II (IGF II). Although the latter was released in greater abundance, as detected by radioimmunoassay, exogenous bFGF was more than a thousand fold more active as a mitogen. Insulin was also able to increase chondrocyte replication at physiological concentrations, and bFGF, insulin and IGFs were additive in their effects on DNA and protein synthesis. Transforming growth factor beta (TGF beta), which is abundant in bone, had little effect on chondrocyte DNA or total protein synthesis alone, but blocked the stimulatory actions of insulin and IGFs on these parameters. However, TGF beta when alone or in combination caused an increase in the collagen: non collagenous protein ratio of new proteins synthesized by chondrocytes. Adult rat brain is a rich source of IGF II, and both IGF I and II are present during neurogenesis and gliagenesis in the fetal and neonatal rat respectively. We have cultured astroglial cells isolated from neonatal rat cerebral cortex to examine the production and interaction of peptide growth factors during their growth. Isolated astroglial cells contained mRNAs encoding both IGF I and II but abundance was not regulated by other hormones or growth factors. Using affinity cross-linking we found that cultured cells also released two species of IGF binding protein (IGF-BP) of 33 kDa and 38 kDa. Northern blot analysis using homologous cDNA probes showed that astroglial cells expressed IGF-BP2 and BP3, but little BP1. Both IGF I and II were mitogenic for astroglial cells, as was insulin at physiologic concentrations. Exogenous IGF-BP2 was able to modulate the mitogenic actions of exogenous IGF I. These two very different cell models show many similarities of endogenous growth control. Both release IGFs and IGF-BPs which regulate mitogenic rate. In addition, in both insulin functions as a growth factor at physiologic concentrations. These findings suggest common principles governing embryonic and fetal growth and development. Studies have shown that fetal and neonatal growth is independent of regulation by classic hormones (e.g. growth hormones) synthesized by the mother or the fetus. It is believed that embryonic and fetal growth is controlled by two major mechanisms, namely, (i) the genetic factors as determined by the embryonic and fetal genome, and (ii) the epigenetic and environmental factors that alter the expression of the embryonic or fetal genome.(ABSTRACT TRUNCATED AT 400 WORDS)     

Insulin-like growth factor II binding and action in human fetal fibroblasts

To investigate the role of insulin-like growth factor II (IGF-II) in human prenatal growth, IGF-II binding and biological action were studied in four lines of fetal and three lines of postnatal human fibroblasts. Specific binding of IGF-II was similar in both groups: 15.7% and 14.9% for fetal and postnatal fibroblasts, respectively. This was 5-10 times the amount of IGF-I binding found in these cells. IGF-I and IGF-II caused dose-dependent increases in [14C]aminoisobutyric acid (AIB) uptake. IGF-II was sevenfold less potent than IGF-I in stimulating this metabolic response in both fetal and postnatal fibroblasts. The maximal effect of IGF-II in stimulating [14C]AIB uptake approach that of IGF-I. Similar results were obtained when IGF-I and IGF-II stimulation of [3H]thymidine incorporation was compared in fetal and postnatal fibroblasts. Incubation in the presence of alpha IR-3, a monoclonal antibody to the type I IGF receptor, inhibited the ability of both IGF-I and IGF-II to stimulate [14C]AIB uptake and [3H]thymidine incorporation in fetal and postnatal cells. A monoclonal antibody to the insulin receptor did not affect IGF action. These data indicate that IGF-II is a potent metabolic and mitogenic stimulus for human fetal fibroblasts. However, despite the presence of abundant type II IGF receptors on both fetal and postnatal human fibroblasts, IGF-II stimulation of amino acid transport and DNA synthesis appears to be mediated through the type I rather than through its own type II IGF receptor.     

Effects of oligohydramnios on lung growth and maturation in the fetal rat

Oligohydramnios (OH) retards fetal lung growth by producing less lung distension than normal. To examine effects of decreased distension on fetal lung development, we produced OH in rats by puncture of uterus and fetal membranes at 16 days of gestation; fetuses were delivered at 21 or 22 days of gestation. Controls were position-matched littermates in the opposite uterine horn. OH lungs had lower weights and less DNA, protein, and water, but no differences in saturated phosphatidylcholine, surfactant proteins (SP)-A and -B, and mRNA for SP-A, -B, -C, and -D. To evaluate effects on epithelial differentiation, we used RTI(40) and RTII(70), proteins specific in lung to luminal surfaces of alveolar type I and II cells, respectively. At 22 days of gestation, OH lungs had less RTI(40) mRNA (P < 0.05) and protein (P < 0.001), but RTII(70) did not differ from controls. With OH, type I cells (in proportion to type II cells) covered less distal air space perimeter (P < 0.01). We conclude that OH, which retards lung growth, has little effect on surfactant and impedes formation of type I cells relative to type II cells.     

Synthesis of surfactant components by cultured type II cells from human lung

We examined the effect of monolayer culture on surfactant phospholipids and proteins of type II cells isolated from human adult and fetal lung. Type II cells were prepared from cultured explants of fetal lung (16-24 weeks gestation) and from adult surgical specimens. Cells were maintained for up to 6 days on plastic tissue culture dishes. Although incorporation of [methyl-3H]choline into phosphatidylcholine (PC) by fetal cells was similar on day 1 and day 5 of culture, saturation of PC fell from 35 to 26%. In addition, there was decreased distribution of labeled acetate into PC, whereas distribution into other phospholipids increased or did not change. The decrease in saturation of newly synthesized PC was not altered by triiodothyronine (T3) and dexamethasone treatment or by culture as mixed type II cell/fibroblast monolayers. The content of surfactant protein SP-A (28-36 kDa) in fetal cells, as measured by ELISA and immunofluorescence microscopy, rose during the first day and then fell to undetectable levels by the fifth. Synthesis of SP-A, as measured by [35S]methionine labeling and immunoprecipitation, was detectable on day 1 but not thereafter. Levels of mRNAs for SP-A and for the two lipophilic surfactant proteins SP-B (18 kDa) and SP-C (5 kDa) fell with half-times of maximally 24 h. In contrast, total protein synthesis measured by [35S]methionine incorporation increased and then plateaued. In adult cells, the content of SP-A and its mRNA decreased during culture, with time-courses similar to those for fetal cells. We conclude that in monolayer culture on plastic culture dishes, human type II cells lose their ability to synthesize both phospholipids and proteins of surfactant. The control of type II cell differentiation under these conditions appears to be at a pretranslational level.     

Vinculin is one of the major endogenous substrates for intrinsic tyrosine kinases in neuronal growth cones isolated from fetal rat brain

Neuronal growth cones, the motile tips of growing neuronal processes, are responsible for the exact guidance of extending neurites. To elucidate the mechanisms of their biochemical signal transduction in growth cones, the growth-cone-enriched fraction was isolated biochemically from fetal rat brain and the endogenous protein phosphorylation in the fraction was analyzed under the conditions where tyrosine residues were preferentially phosphorylated. One of the major phosphoproteins was a 130-kDa slightly acidic protein which reacted with antiphosphotyrosine antibody. Its phosphoryl residues were alkali-stable. Thus, the 130-kDa protein was concluded to be susceptible to tyrosine phosphorylation. This protein was a component of cytoskeletal proteins thought to be associated indirectly with membranes. All the behavior of the 130-kDa protein was compatible with the properties of vinculin, a component of focal contacts which are responsible for the stable or motile adhesion between cells or between a cell and the substratum. Immunochemical analyses showed that the 130-kDa protein was specifically recognized by anti-vinculin antibody. Therefore, the 130-kDa protein was concluded to be vinculin. Tyrosine phosphorylation of the protein appeared to be relatively more pronounced in the growth-cone-enriched fraction than in adult synaptosomes. The results suggest that tyrosine phosphorylation of vinculin may be regulated developmentally and it may be involved in the functions of growth cones.     

Cell-specific and developmental expression of phospholipase C-gamma and diacylglycerol in fetal lung

Epidermal growth factor (EGF) receptor (EGFR) regulates development of cell-cell communication in fetal lung, but the signal transduction mechanisms involved are unknown. We hypothesized that, in late-gestation fetal rat lung, phospholipase C-gamma (PLC-gamma) expression and activation by EGF is cell specific and developmentally regulated. PLC-gamma immunolocalized to cuboidal epithelium and mesenchymal clusters underlying developing saccules. PLC-gamma protein increased from day 17 to day 19 and then decreased. In cultured fetal lung fibroblasts, EGF stimulated PLC-gamma phosphorylation 2.6-fold (day 17), 10.8-fold (day 19), and 4.2-fold (day 21). EGF stimulated (3)H-labeled diacylglycerol production in fibroblasts (beginning on day 18 in female and on day 19 in male rats), but not in type II cells at any time during gestation. EGFR blockade abrogated the observed stimulation of PLC-gamma phosphorylation by EGF. In conclusion, PLC-gamma expression and activation by EGF in fetal lung are cell specific, corresponding to the development of EGFR expression. EGF induces diacylglycerol production in a cell- and gestation-specific manner. PLC-gamma activation by EGFR in fetal lung fibroblasts may be involved in EGF control of lung development.     

Isolation of alveolar type II cells from fetal rat lung by differential adherence in monolayer culture

Type II alveolar epithelial cells were isolated from fetal rat lung by differential adherence in monolayer culture. The preparation had a high degree of purity, as assessed by phase contrast microscopy and immunocytochemistry. Purity, based on reactivity with specific anti-adult lung serum (SAALS), which recognizes only type II cells, was 91% for cells isolated from 19-day fetal lungs and 79% for cells isolated from 21-day fetal lungs. The lower purity of type II cells in cultures derived from 1-day postnatal rat lungs (51% cells reactive with SAALS) is probably due to a lower tendency of the type II cells from neonatal rats to adhere to culture dishes than of type II cells from fetal rats. Type II cells isolated from 21-day fetal lungs contained a higher percentage phosphatidylglycerol and incorporated [Me-3H]choline faster into phosphatidylcholine (PC) than type II cells isolated from 19-day fetal lungs. Moreover, in cell preparations derived from lungs at fetal day 21, a higher percentage of epithelial cells contained lamellar bodies than in preparations derived from lungs at fetal day 19. The observation of these differences in the stage of maturation indicates that these differences, which are typical features of the original material, are not obliterated by differentiation during the culture. Type II cells isolated according to the present procedure were capable of synthesizing PC with a high percentage of the disaturated species. This method for the isolation of fetal type II cells may be a useful tool in studies concerning surfactant synthesis and its regulation in the fetal lung.     

Similarities in structure and function of bovine serum erythrotropin and human insulin-like growth factor II: two fetal erythroid cell stimulating factors

Serum erythrotropin (ET) was isolated from fetal bovine serum. Partial sequence analysis of the N-terminal portion of the peptide indicated that the first 20 amino acids were practically identical to those found in human insulin-like growth factor II (IGF II). The effect of IGF II on [3H] thymidine incorporation in cell cultures of fetal bovine liver was similar to the effect of ET. Both factors acted synergistically with erythropoietin but not with platelet derived growth factor. The stimulation of thymidine incorporation by ET and IGF II on cell cultures of fetal liver erythroid cells was at least 15 times higher than their effects on cultures of fetal calf intestine, lung and kidney cells.     

The types II and III transforming growth factor-beta receptors form homo-oligomers

Affinity-labeling experiments have detected hetero-oligomers of the types I, II, and III transforming growth factor beta (TGF-beta) receptors which mediate intracellular signaling by TGF-beta, but the oligomeric state of the individual receptor types remains unknown. Here we use two types of experiments to show that a major portion of the receptor types II and III forms homo-oligomers both in the absence and presence of TGF-beta. Both experiments used COS-7 cells co-transfected with combinations of these receptors carrying different epitope tags at their extracellular termini. In immunoprecipitation experiments, radiolabeled TGF-beta was bound and cross-linked to cells co-expressing two differently tagged type II receptors. Sequential immunoprecipitations using anti-epitope monoclonal antibodies showed that type II TGF-beta receptors form homo-oligomers. In cells co- expressing epitope-tagged types II and III receptors, a low level of co- precipitation of the ligand-labeled receptors was observed, indicating that some hetero-oligomers of the types II and III receptors exist in the presence of ligand. Antibody-mediated cross-linking studies based on double-labeling immunofluorescence explored co-patching of the receptors at the cell surface on live cells. In cells co-expressing two differently tagged type II receptors or two differently tagged type III receptors, forcing one receptor into micropatches by IgG induced co- patching of the receptor carrying the other tag, labeled by noncross- linking monovalent Fab'. These studies showed that homo-oligomers of the types II and III receptors exist on the cell surface in the absence or presence of TGF-beta 1 or -beta 2. In cells co-expressing types II and III receptors, the amount of heterocomplexes at the cell surface was too low to be detected in the immunofluorescence co-patching experiments, confirming that hetero-oligomers of the types II and III receptors are minor and probably transient species.