Cell surface glycoproteins of CHO cells. I. Internalization and rapid recycling

Anticoagulant, anti-aggregation and antithrombotic effects of a novel hexapeptide

OBJECTIVES

Hexapeptide is a novel synthetic oligopeptide with a structure similar to that of eptifibatide. This study was designed to investigate the anticoagulant, anti-aggregation, disaggregation and anti-thrombogenesis effects of hexapeptide.

METHODS

The effects of antiplatelet aggregation induced by adenosine diphosphate (ADP), arachidonic acid (AA) and thrombin, and the effect of disaggregation of platelet aggregation induced by ADP were determined. The anticoagulation indexes were determined by different kits.

KEY FINDINGS

Hexapeptide 1 × 10(-5) -1 × 10(-4) m could significantly prolong rabbit blood clotting time, thrombin time, prothrombin time and activated partial thromboplastin enzyme time, and reduce the length, wet weight, dry weight and the index of thrombus in a concentration-dependent manner. Hexapeptide 1 × 10(-4) m decreased platelet adhesion rate by 40.2%. The platelet aggregation inhibition of hexapeptide in dogs and humans was more obvious than in rabbits and rats. The aggregation inhibition rate of 1 × 10(-5) m hexapeptide in dogs, rabbits, rats and humans induced by ADP was 93.9 ± 1.3%, 66.2 ± 1.4%, 76.1 ± 3.2% and 99.8 ± 0.2%, respectively; the 50% inhibitory concentration (IC50) of hexapeptide was 7.24 × 10(-8), 3.24 × 10(-6), 6.61 × 10(-6) and 8.91 × 10(-8) m, respectively. For the aggregation inhibition rate of hexapeptide in dogs, rabbits and humans induced by AA, the IC50 was 1.29 × 10(-9), 1.32 × 10(-6) and 9.33 × 10(-8) m, respectively; the IC50 of aggregation inhibition rates induced by thrombin was 2.88 × 10(-6), >1 × 10(-5) and 4.17 × 10(-6) m, respectively. The disaggregation rate of 1 × 10(-4) m hexapeptide in dog induced by ADP was 68.8 ± 7.4%.

CONCLUSIONS

Hexapeptide has anticoagulant, antiplatelet aggregation, disaggregation and antithrombotic effects in vitro.

Intracellular receptor/ligand sorting based on endosomal retention components

Endocytosed molecules are sorted in endosomes to different cellular destinations (e.g., to lysosomes or to the plasma membrane). Diverse endosomal sorting results have been reported for different ligands and receptors in a variety of cell types, but the general principles governing these sorting outcomes are not well understood. For example, we observed a wide range of sorting outcomes with the epidermal growth factor (EGF)/receptor system in fibroblasts using several members of the EGF family and site-directed ligand and receptor mutants. In this article we describe a mechanistic mathematical model of endosomal sorting based on the hypothesis that receptors may be selectively retained by the endosomal sorting apparatus and that this process may be modulated by receptor occupancy. Our results show that this single mechanism can account for the wide variety of observed sorting outcomes. By providing a conceptual framework for understanding endosomal sorting, this model not only helps interpret our experimental results for the EGF/receptor system, but also provides some insight into the principles governing sorting. For example, the model predicts that the influence of selective endosomal retention of receptor/ligand complexes is seen in deviations of ligand sorting outcomes from pure fluid phase sorting behavior. Furthermore, the model suggests that selective endosomal retention of complexes within endosomes gives rise to three sorting regimes characterized by distinguishable qualitative trends in the dependence of ligand sorting fractions on intracellular ligand concentrations.

Mechanistic study of the uptake of wheat germ agglutinin-conjugated PLGA nanoparticles by A549 cells

The purpose of this study was to evaluate the extent and mechanism of uptake of wheat germ agglutinin-conjugated PLGA nanoparticles by A549 cells. PLGA nanoparticles of 150 nm were prepared by a solvent diffusion method and covalently conjugated to FITC-WGA (fWGA) or FITC-bovine serum albumin (fBSA) by a two-step carbodiimide method. Uptake of fWGA-PLGA and fBSA-PLGA nanoparticles by confluent A549 cells was quantified by fluorometry. A549 cellular uptake of fWGA-PLGA nanoparticles at 2 h, 37 degrees C was 5.02-fold that of fBSA-PLGA nanoparticles at a loading concentration of 2.65 mg/mL. The difference in uptake between the two types of nanoparticles was increased to 7.84-fold at a higher loading concentration of 5.3 mg/mL, but was reduced to 2.07-fold by lowering the uptake temperature to 4 degrees C. Coincubation with 5 mg/mL of unlabeled WGA negated the differential uptake of fWGA-PLGA nanoparticles at 4 degrees C, suggesting that the nanoparticles interacted with a specific WGA-binding receptor on the cell membrane. Internalization of the fWGA-PLGA nanoparticles by the A549 cells was confirmed by confocal microscopy. Filipin (1 microg/mL), a known inhibitor of caveolae, reduced the 1-h uptake of the nanoparticles by 75%. Surface modification of PLGA nanoparticles with WGA significantly enhanced its endocytosis by A549 cells by a receptor-mediated, caveola-dependent pathway.

Cell membrane modification for rapid display of proteins as a novel means of immunomodulation: FasL-decorated cells prevent islet graft rejection

Long-term display of exogenous proteins on the cell surface may have important research and therapeutic implications. We report a novel method for the cell-surface display of proteins that involves generation of a chimeric protein with core streptavidin, biotinylation of cells, and "decoration" with the protein. A chimeric protein with the extracellular portions of FasL (SA-FasL) was efficiently displayed on the cell surface within 2 hr without detectable cellular toxicity. Biotin and SA-FasL persisted on the cell surface for weeks in vitro and in vivo. Immunomodulation with SA-FasL-decorated splenocytes effectively blocked alloreactive responses in naive and presensitized rodents and prevented the rejection of allogeneic pancreatic islets. This approach may serve as an alternative to gene transfer-based expression with broad research and therapeutic applications.

Monomorphic molecules function as additional recognition structures on haptenated target cells for HLA-A1-restricted, hapten-specific CTL

Hapten-specific T cells have been shown to recognize haptenated peptides with high avidity and, in some instances, with promiscuous MHC restriction. In this study, the impact of Ag density on MHC restriction of a CTL response specific to the trinitrophenyl (TNP) hapten was investigated. In this study, we demonstrate a novel recognition mechanism used by TNP-specific CD8(+) CTL in the presence of high Ag doses. Although low levels of TNP epitopes on target cells allowed for HLA-A1-restricted CTL activity only, entirely MHC-independent target cell recognition became operative at high TNP loading. In both cases, recognition was mediated by the TCR. This MHC-independent recognition is target cell type restricted and critically involves in our model direct recognition of the ectonucleotidase family surface molecule CD39 by the CTL.

Labeling of the Internal Pool of GP IIb-IIIa in Platelets by c7E3 Fab Fragments (abciximab): Flow and Endocytic Mechanisms Contribute to the Transport

Abciximab is a new antiplatelet therapeutic in ischemic cardiovascular disease. The drug, chimeric Fab fragments of a murine monoclonal antibody (MoAb) (c7E3), blocks GP IIb-IIIa function. However, its capacity to reach all receptor pools in platelets is unknown. Electron microscopy and immunogold labeling were used to localize abciximab in platelets of patients receiving the drug for up to 24 hours. Studies on frozen-thin sections showed that c7E3 Fab, in addition to the surface pool, also labeled the surface-connected canalicular system (SCCS) and -granules. Analysis of gold particle distribution showed that intraplatelet labeling was not accumulative and in equilibrium with the surface pool. After short-term incubations of platelets with c7E3 Fab in vitro, gold particles were often seen in lines within thin elements of the SCCS, some of which appeared in contact with -granules. Little labeling was associated with Glanzmann’s thrombasthenia platelets, confirming that the channels contained bound and not free c7E3 Fab. Endocytosis of abciximab in clathrin-containing vesicles was visualized by double staining and constitutes an alternative mechanism of transport. The remaining free pool of GP IIb-IIIa was evaluated with the MoAb AP-2; flow cytometry showed it to be about 9% on the surface of nonstimulated platelets but 33% on thrombin-activated platelets. The ability of drugs to block all pools of GP IIb-IIIa and then to be associated with secretion-dependent residual aggregation must be considered when evaluating their efficiency in a clinical context.

Labeling of the Internal Pool of GP IIb-IIIa in Platelets by c7E3 Fab Fragments (abciximab): Flow and Endocytic Mechanisms Contribute to the Transport

Abciximab is a new antiplatelet therapeutic in ischemic cardiovascular disease. The drug, chimeric Fab fragments of a murine monoclonal antibody (MoAb) (c7E3), blocks GP IIb-IIIa function. However, its capacity to reach all receptor pools in platelets is unknown. Electron microscopy and immunogold labeling were used to localize abciximab in platelets of patients receiving the drug for up to 24 hours. Studies on frozen-thin sections showed that c7E3 Fab, in addition to the surface pool, also labeled the surface-connected canalicular system (SCCS) and -granules. Analysis of gold particle distribution showed that intraplatelet labeling was not accumulative and in equilibrium with the surface pool. After short-term incubations of platelets with c7E3 Fab in vitro, gold particles were often seen in lines within thin elements of the SCCS, some of which appeared in contact with -granules. Little labeling was associated with Glanzmann’s thrombasthenia platelets, confirming that the channels contained bound and not free c7E3 Fab. Endocytosis of abciximab in clathrin-containing vesicles was visualized by double staining and constitutes an alternative mechanism of transport. The remaining free pool of GP IIb-IIIa was evaluated with the MoAb AP-2; flow cytometry showed it to be about 9% on the surface of nonstimulated platelets but 33% on thrombin-activated platelets. The ability of drugs to block all pools of GP IIb-IIIa and then to be associated with secretion-dependent residual aggregation must be considered when evaluating their efficiency in a clinical context.

Degradation of mutant influenza virus hemagglutinins is influenced by cytoplasmic sequences independent of internalization signals

A mutant influenza virus hemagglutinin, HA+8, having a carboxyl-terminal extension of 8 amino acids that included 4 aromatic residues, was internalized within 2 min of arriving at the cell surface and was degraded quickly by a process that was inhibited by ammonium chloride. Through second-site mutagenesis, the internalization sequence of HA+8 was found to closely resemble the internalization signals of the transferrin receptor or large mannose 6-phosphate receptor. Comparison of the intracellular traffic of HA+8 and a series of other HA mutants that differed in their rates of internalization revealed a relation between the amount of the protein on the plasma membrane at steady state and the internalization rate that would be predicted if most of each protein recycled to the cell surface. However, there was no simple correlation between the internalization rate and the rate of degradation, indicating that transport to the compartment where degradation occurred was not simply a function of the concentration of the proteins in early endosomes. The internal populations of both HA+8, which was degraded with a t1/2 of 1.9 h, and HA-Y543, which was degraded with a t1/2 of 2.9 h, were found by cell fractionation and density-shift experiments to reside in early endosomes with little accumulation in lysosomes. A fluid-phase marker reached lysosomes 3-4-fold faster than these proteins were degraded. Degradation of these mutant HAs involved a rate-determining step in early endosomes that was sensitive to some feature of the protein that depended upon sequence differences in the cytoplasmic domain unrelated to the internalization signal.

Intracellular trafficking of epidermal growth factor family ligands is directly influenced by the pH sensitivity of the receptor/ligand interaction

Using members of the epidermal growth factor (EGF) family as well as site-directed recombinant human EGF mutants, we investigated how ligand binding properties influence endosomal sorting. Mouse EGF (mEGF), human EGF (hEGF), and transforming growth factor alpha (TGF alpha) bind to the human EGF receptor (EGFR) with similar affinities at pH 7.4. However, the binding properties of these ligands have substantially different pH sensitivities resulting in varying degrees of dissociation from the receptors at lower pH levels characteristic of endosomes. We employed a steady-state sorting assay to determine the fraction of ligand sorted to recycling versus degradation as a function of the number of intracellular ligand molecules in mouse B82 fibroblasts. mEGF, hEGF, and TGF alpha display significantly different steady-state endosomal sorting patterns which correspond to the extent of their dissociation at endosomal pH. Moreover, several recombinant hEGF mutants with differing affinities exhibit altered endosomal sorting compared to hEGF, demonstrating a similar direct relationship between ligand binding properties and endosomal sorting outcomes. Intracellular trafficking of the EGF ligands was also monitored by measuring the observed degradation rate constants. These likewise show marked differences that correlate with the differing pH sensitivities of the ligands' binding properties.

The effect of botulinum toxin type D on the triggered and constitutive exocytosis/endocytosis cycles in cultures of bovine adrenal medullary cells

The extracellular fluid phase marker, horseradish peroxidase, enters chromaffin cells when triggered to secrete catecholamine. This triggered uptake, like secretion, is abolished in cells pre-incubated with botulinum toxin. Endocytosis of horseradish peroxidase into unstimulated cells is unaffected by botulinum toxin but is inhibited when the temperature is reduced. Once internalised by the unstimulated cells, horseradish peroxidase is released back into the extracellular fluid, the rate of release being temperature sensitive but unaffected by carbamylcholine or botulinum toxin. These results suggest that triggered exocytosis is a necessary event to precede triggered endocytosis, and that botulinum toxin may affect only the triggered exocytosis/endocytosis cycle and not the constitutive cycle.

Adsorptive endocytosis and membrane recycling by cultured primary bovine brain microvessel endothelial cell monolayers

The dynamics of membrane recycling were examined in primary cultures of brain microvessel endothelial cells (BMECs). Because the BMEC surface was dominated by galactosylated glycoconjugates, ricin agglutinin (RCAI) was used as a tracer to follow the endocytosis and recycling of RCAI binding sites. These binding sites accounted for 75% of the iodinatable or most externally disposed plasma membrane proteins. Because greater than 90% of the RCAI that had bound to BMECs was removed by a brief, nontoxic treatment with galactose, the amounts and kinetics for internalization and efflux of [125I]RCAI were measured. Both endocytosis and efflux were energy dependent. By using pseudo-first-order kinetics, the t1/2 values for RCAI binding, internalization and efflux were 5, 18 and 13-14 min, respectively. By comparing efflux with and without galactose present, we found that 60% of the RCAI binding sites that had been internalized were returned to the cell surface and reinternalized. Quantifying the distribution of gold-RCAI following internalization showed kinetics consistent with that obtained using radiolabeled RCAI. Both horseradish peroxidase (HRP) and gold-conjugated RCAI that had bound BMEC at 4 degrees C became localized within more caveolae within 2.5 min of warming to 37 degrees C to permit endocytosis. With time, RCAI appeared within endosomes and tubules and vesicles of which some were located in the trans-Golgi network (TGN). The distribution of HRP-RCAI contrasted with that of free HRP, which was not routed to the TGN. The absence of RCAI conjugates in association with the basolateral membrane domain suggested the presence of functional tight junctions and maintenance of polarity throughout the duration of these experiments. These results showed that membrane recycling was more extensive and much slower than fluid-phase endocytosis in cultured BMECs. Moreover, we found that endocytosis of membrane by BMECs in culture was similar to that reported for brain endothelium in vivo in that a fraction of the cell surface membrane was routed to the TGN.

Rapid endocytosis and recycling of wheat germ agglutinin binding sites on CHO cells: evidence for two compartments in a nondegradative pathway

The internalization and recycling of CHO cell plasma membrane components have been quantified by using iodinated wheat germ agglutinin (WGA) as an adsorptive tracer. Most of these binding sites are thought to be composed of a subpopulation of plasma membrane proteins called high-molecular-weight acidic glycoproteins (HMWAG). Greater than 90% of the WGA bound on the cell surface can be removed by brief treatment with N-acetylglucosamine (GlcNAc). At 37 degrees C, endocytosis of WGA that had been allowed to bind to the surface at 4 degrees C is curvilinear with an initial rapid phase occurring with a t1/2 of 6-8 min. Within 20 min, accumulation has slowed gradually to steady-state with 65% of the cell-associated WGA located intracellularly or resistant to removal by GlcNAc. These portions are unaffected by increasing the extracellular concentration of WGA from 0.003 microM to 2.8 microM. By using pulse-chase experiments, the observed decrease in rate of endocytosis is shown to be due to return of the WGA-HMWAG complexes to the cell surface. More than 60% of the WGA that had been internalized is recycled within 30 min, with a mean t1/2 of 17 min. Recycling involved at least two intracellular populations where a significant fraction (less than 30%) of the WGA-HMWAG complexes are lost gradually from the rapidly recycling pool. Most of the WGA-HMWAG complexes that had internalized are not delivered to the lysosome. These results demonstrate the magnitude of rapid and continuous recycling of WGA binding sites between the cell surface and endosomes in fibroblasts.

Endocytosis of wheat germ agglutinin binding sites from the cell surface into a tubular endosomal network

By using fluorescence and electron microscopy, the endocytic pathway encountered by cell surface components after they had bound wheat germ agglutinin (WGA) was visualized. The majority of these components are thought to consist of sialylated glycoproteins (HMWAG) that represent a subpopulation of the total cell surface proteins but most of the externally disposed plasma membrane proteins of the cell. Examination of semi-thin sections by medium- and high-voltage electron microscopy revealed the three-dimensional organization of vesicular and tubular endosomes. Binding of either fluorescein isothiocyanate-, horseradish peroxidase-, or ferritin-conjugated WGA to cells at 4 degrees C showed that the HMWAG were distributed uniformly over the cell surface. Warming of surface-labeled cells to 37 degrees C resulted in the endocytosis of WGA into peripheral endosomes via invagination of regions of both coated and uncoated membrane. The peripheral endosome appeared as isolated complexes comprising a vesicular element (300-400 nm diam.) surrounded by and continuous with tubular cisternae (45-60 nm diam.), which did not interconnect the endosomes. After 30 min or more label also became localized in a network of anastomosing tubules (45-60 nm diam.) that were located in the centrosomal region of the cell. Endocytosed WGA-HMWAG complexes did not become associated with cisternae of the Golgi apparatus, although tubular and vesicular endosomes were noted in the vicinity of the trans-Golgi region. The accumulation of WGA-HMWAG in the endosomes within the centrosomal region was inhibited when cells were incubated at 18 degrees C. None of these compartments contained acid phosphatase activity, a result that is consistent with other data that the HMWAG do not pass through lysosomes initially. The kinetics of labeling were consistent with the interpretation that recycling of most of the WGA binding surface glycoproteins occurred rapidly from early peripheral endosomes followed by the late trans-Golgi compartment. In conclusion, a portion of cell surface glycoproteins are routed to a complex arrangement of tubular and vesicular compartments following endocytosis that includes a putative post-endosomal, tubular reticulum that appears to be separate from the trans-most Golgi saccule.

Fluid-phase endocytosis by primary cultures of bovine brain microvessel endothelial cell monolayers

Blood-brain barrier (BBB) fluid-phase endocytosis was examined in primary cultures of bovine brain microvessel endothelial cell (BMEC) monolayers. By fluorescence spectroscopy. Lucifer yellow (LY, a fluorescent, soluble molecule used as a marker for pinocytosis) accumulation by BMEC was observed to be linear over a concentration range of 0.05 to 1.0 mg/ml. Time-dependent uptake of LY exhibited curvilinear kinetics composed of an initially rapid uptake rate of 1338 ng of LY/mg protein per hour at 0.5 mg/ml LY. Within 20 min, the rate of LY accumulation slowed to a steady-state rate of 23 ng of LY/mg protein per hour. Accumulation of LY was inhibited in the presence of metabolic inhibitors, potassium cyanide or 2-deoxyglucose, and was decreased, but not completely inhibited, at 4 degrees. Pulse-chase experiments revealed that efflux of LY was very rapid with at least 80% of the accumulated LY being lost within 2 min and was not sensitive to low temperature. Only 3-5% of the LY initially accumulated by BMEC remained cell-associated after a 30-min chase. The calculated turnover of the endocytic compartment's total volume (per hour) is 8- to 20-fold less than values for fibroblasts and macrophages, respectively. We have interpreted these data to suggest that the efflux of most of the LY involves loss from a rapidly recycled compartment of finite volume, possibly caveolae, that had sequestered marker during accumulation and suggest that these results are consistent with the present understanding of BBB pinocytosis in vivo.

Iterative fractionation of recycling receptors from lysosomally destined ligands in an early sorting endosome

To study the fusion and separation of endocytic compartments, we have used digital image analysis to quantify the accumulation of fluorescent ligands in endosomes during continuous endocytosis for periods of 1-20 min. Fluorescently labeled transferrin (Tf) and low density lipoproteins (LDL) were used as markers of recycling receptors and lysosomally directed ligands respectively. By measuring the intensity of individual endosomes, we found that the amount of LDL per endosome increases 30-40-fold between 1 and 10 min and then plateaus. In contrast, the amount of Tf per endosome reaches a steady state within 2 min at a level that is only three to four times that at 1 min. We used pulse-chase double label methods to demonstrate that Tf cycles through the compartment in which the LDL accumulates. When both Tf and LDL are added to cells simultaneously for 2 min, nearly all endosomes contain both labels. With 2-4 min further incubation in the absence of external ligands, LDL-containing compartments become depleted of Tf as Tf is directed to para-Golgi recycling endosomes. However, if Tf is added to the medium 2-4 min after a pulse with LDL, most of the LDL-containing endosomes become labeled with Tf. The data indicate that at least 30-40 endocytic vesicles containing both Tf and LDL fuse with an endosomal compartment over a period of 5-10 min. LDL accumulates within this compartment and Tf is simultaneously removed. Simple mathematical models suggest that this type of iterative fractionation can lead to very high efficiency sorting.

Kinetic and morphological evidence for endocytosis of mammalian cell integrin receptors by using an anti-fibronectin receptor beta subunit monoclonal antibody

Monoclonal antibody (mAb) 7E2.2, which recognizes the beta subunit of the hamster fibronectin receptor (FnR) (Brown, P.J., and Juliano, R. L. (1988) Exp. Cell Res. 177. 303), was used to examine the distribution of and to quantify the internalization of the FnR and possibly related integrins on adherent fibroblasts. Purified 7E2.2 IgG was iodinated and used in binding and internalization studies. Binding to Chinese hamster ovary cells was saturable with a Km of 0.3 nM and an estimated total number of cell surface beta subunits at 2 x 10(5) per cell. The FnR colocalized with fibronectin at cell adhesion contact sites and also was distributed evenly over the dorsal cell surface as discrete clusters. By using a direct immunocolloidal gold approach, the FnR was not associated with coated pits at 4 degrees C until internalization followed warming of the labeled cells to 37 degrees C. A proportion of the FnRs were endocytosed with a half-time of 6.5 min and, consistent with clathrin-mediated uptake, this was sensitive to hypertonic conditions. Receptor-immunocomplexes rapidly became localized within coated pits, small diameter tubules, and peripheral endosomes but the majority remained at the cell surface. At subsaturating concentrations of bound 7E2.2, approximately one-fourth of the total cell receptor population resided intracellularly at any one moment following steady-state; however, appreciable degradation of the iodinated mAb was not detected following accumulation for 4 h at 37 degrees C. These data showed that at least a portion of the FnR are endocytosed via a receptor-mediated pathway and suggested that these receptors do not immediately enter a degradative compartment.

Insulin processing in primary endosomes is not responsible for insulin resistance observed in parametrial adipocytes from lactating rats

The fate of [125I insulin and the insulin receptor after internalization was characterized in parametrial adipocytes from virgin rats. Parallel experiments were carried out on parametrial adipocytes from 2-4-day lactating rats, which are insulin resistant. Similar results were obtained in adipocytes from either group of animals. Insulin caused 10% of the plasma membrane insulin receptor to be translocated to a compartment resistant to extracellular trypsin. The intracellularly located insulin receptor rapidly recycled to the plasma membrane at 37 degrees C. An endosomal compartment involved in both the endocytosis and subsequent recycling of [125I]insulin and the insulin receptor to the plasma membrane was identified on sucrose density floatation gradients. [125I]Insulin internalized at 37 degrees C accumulated in a fraction of modal density 1.12 g/ml. Crosslinking experiments revealed the presence of intact [125I]insulin-insulin receptor complexes in endosomes. After a pulse with [125I]insulin, 55-60% of the 125I radioactivity recovered in the endosome compartment was intact [125I]insulin. The remainder was composed of low molecular weight degradation products. Endosomal 125I radioactivity was rapidly retroendocytosed to the medium with a mean half-life of 6 min. These results suggest: (1) [125I]insulin and the insulin receptor are internalized by parametrial adipocytes into an early endosomal compartment (primary endosomes), from which the receptor, intact [125I]insulin, and [125I]tyrosine are returned to the cell surface; and (2) the damping of the insulin signal observed in parametrial adipocytes from lactating rats is not expressed at the level of altered endocytotic processing of [125I]insulin and the insulin receptor.

Cell surface glycoproteins of CHO cells. II. Surface distribution and pathway of internalization

The surface distribution and pathway for internalization of the major cell surface proteins of Chinese hamster ovary (CHO) cells have been investigated after reacting cells at 4 degrees C with the membrane-impermeant reagent trinitrobenzenesulfonate. Molecules, haptenized with trinitrophenol groups, the majority of which are in a group of high molecular weight acidic glycoproteins (HMWAG), were labelled at 4 degrees C with anti-dinitrophenol immunoglobulins coupled to fluorescein isothiocyanate (FITC), horseradish peroxidase, or colloidal gold and either immediately fixed for mapping their distribution or followed intracellularly after warming to allow endocytosis to proceed. The distribution of label on the CHO cell surface was non-random with a large proportion arranged in clusters from 100 to 300 nm in diameter. Antibody label was concentrated heavily on microvilli, and about 10% of the molecules were always associated with clathrin-coated pits. Upon warming the cells to 37 degrees C, HMWAG were internalized immediately into smooth-membraned tubules (less than 80 nm luminal diameter) that appeared to connect with vesicles (less than 300 nm luminal diameter) located in the cortical cytoplasm. By 60 min, labelled antibody was located within larger vesicles (greater than 300 nm luminal diameter) that had a morphology characteristic of multivesicular bodies and not lysosomes. There was no evidence for entry of labelled molecules into either electron-dense, secondary lysosomes or into the Golgi cisternae, suggesting that neither compartment is involved in the major pathway of cell surface endocytosis. Our results are consistent with the view that the majority of plasma membrane protein are internalized as small discrete domains by a pathway very similar to that described by others for adsorptive endocytosis.