Relationship between trinitrophenol and H-2 antigens on trinitrophenyl-modified spleen cells. III. Quantitative aspects of trinitrophenol binding on cells treated with trinitrobenzene sulfonic acid

Splenic lymphoblasts or normal spleen cells were treated with varying concentrations of TNBS in order to assess whether cell membrane H-2 molecules were derivatized with TNP. Cells treated with high concentrations of TNBS had their cell membrane H-2 molecules derivatized and functioned antigenically as inhibitors in a cold target TNP-CML competition assay. In contrast, cells derivatized with lower concentrations of TNBS had a significant proportion of their membrane proteins derivatized with TNP but did not have their H-2 molecules derivatized. These latter cells were unable to block anti-TNP cytotoxic effector cells in the competition assay. When cells were treated with 3H-TNBS, it was observed that TNP couples to cell membrane H-2, Ia and Ig molecules, and an estimate of the number of TNP molecules bound per cell at varying concentrations of TNBS was determined. The data obtained are consistent with there being a requirement for TNP to directly derivatize H-2 molecules on the cell membrane in order to create antigenic determinants that can be recognized by cytotoxic anti-TNP effector cells. As an alternative, there may be a requirement for the presence of a high density of TNP molecules per cell rather than direct H-2 derivatization by TNP in order to account for activity.     

Loss of fibronectin among the selective surface protein changes associated with p-formaldehyde promoted membrane vesiculation

p-formaldehyde-promoted membrane vesiculation of preiodinated cultures lead to the release of surface proteins of about 68 kd in normal and transformed rat liver epithelial cells and rat fibroblasts, concurrent with a marked decrease in surface-associated fibronectin. Membrane vesiculation in the presence of soybean trypsin inhibitor permitted the detection of an 18 kd surface protein in membrane vesicles and the increased expression of a 70 kd external component in normal but not in transformed epithelial cells.Our results show that the membrane vesiculation process is associated both with the release and selective degradation of specific cell surface proteins in a process which may involve surface protease activation. Our data also suggest the potential of the chemical vesiculation process as a probe to monitor differences in surface topography between different cell types.     

Immunochemical purification of probe-labeled plasma membrane proteins: An approach to the molecular anatomy of the cell surface

The probe 2,4,6-trinitrobenzene sodium sulfonate may be used under appropriate conditions for selective labelling of plasma membrane proteins exposed at the outer cell surface. Labeled proteins, solubilized by detergents, can be purified by reverse immunoadsorption using antiprobe antibodies covalently linked to Sepharose 4B. This method has been applied to an investigation of the outer cell surface structure of chicken embryo and hamster fibroblasts. Coelectrophoresis in sodium dodecyl sulfate-polyacrylamide gels of probe-labeled membrane proteins purified from baby hamster kidney fibroblasts have shown that 7 major protein groups of different molecular weight are exposed on both control and Rous sarcoma or polyoma virus-transformed cells. Moreover, the transformed cells display a nonvirion component of 80–100 k daltons that is not labeled by the probe in normal cells. In fibroblasts transformed by a temperature sensitive Rous sarcoma virus mutant, that transforms at 37°C but not at 41°C, the expression of this component is related to the expression of the transformed phenotype.     

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

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 (TNBS). Immunoprecipitation and subsequent two-dimensional, sodium-dodecyl sulfate, polyacrylamide gel electrophoresis (SDS-PAGE) of proteins from derivatized cells that had been labelled previously with [3H]D-glucosamine or [3H]L-leucine showed that TNBS reacted with most of the high molecular weight (HMW) acidic glycoproteins that became labelled with iodine by the lactoperoxidase technique and that bind the lectin, wheat germ agglutinin (WGA). After warming the cells to allow endocytosis to proceed, molecules haptenized with trinitrophenol (TNP) groups were followed radiochemically by means of [125I]anti-DNP antibodies. The half-life for internalization of proteins tagged with either [125I]anti-DNP IgG or Fab averaged about 5 min. A similar result was obtained when a monoclonal antibody directed against a single plasma membrane glycoprotein was used, or when the rate of surface loss of TNP groups unoccupied by antibodies was measured. Within 15 min at 37 degrees C, a steady-state between surface and cytoplasmic label was reached, with about 65% of the hapten located internally. Recycling of internalized TNP groups back to the cell surface also occurred rapidly (t 1/2 approximately 5 min). Most of the intracellular radioactivity was associated with a membrane fraction of density similar to that of the plasma membrane. Over a 4-h period, there was no significant entry of labeled molecules into lysosomes. By contrast, the fluid-phase marker, horseradish peroxidase, became associated with the lysosomes within 1 h. Our results are consistent with the view that the majority of plasma membrane glycoproteins are continuously being internalized and recycled at a high rate.     

The preparation and use of pyridoxal [32P]phosphate as a labeling reagent for proteins on the outer surface of membranes.

Pyridoxal [32P] phosphate was prepared using [gamma-32P] ATP, pyridoxal, and pyridoxine kinase purified from Escherichia coli B. The pyridoxal [32P] phosphate obtained had a specific activity of at least 1 Ci/mmol. This reagent was used to label intact influenza virus, red blood cells, and both normal and transformed chick embryo fibroblasts. The cell or virus to be labeled was incubated with pyridoxal [32P] phosphate. The Schiff base formed between pyridoxal [32P] phosphate and protein amino groups was reduced with NaBH4. The distribution of pyridoxal [32P] phosphate in cell membrane or virus envelope proteins was visualized by autoradiography of the proteins separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis. The labeling of the proteins of both influenza and chick cells appeared to be limited exclusively to those on the external surface of the virus or plasma membrane. With intact red blood cells the major portion of the probe was bound by external proteins, but a small amount of label was found associated with the internal proteins spectrin and hemoglobin.     

Effects of cocultivation with transformed cells on surface proteins of normal cells.

Virally transformed fibroblasts do not have on their surface a major protein (large external transformation-sensitive, LETS) which is present in normal cells. Cocultivation of the transformed cells with normal cells whose surface proteins have been prelabelled induces an accelerated release of the LETS protein from the normal cells. We have investigated various conditions which affect this phenomenon. Our results show that alteration of cell surface proteins by cocultivation with the transformed cells is time and dose-dependent and requires cell contact. Serum was depleted at least 99% of plasminogen by affinity chromatography and used in the cocultivation experiments. It was found that activation of plasminogen was not required for the accelerated turnover of the LETS protein. Other diffusible proteases are also unlikely to be involved. The possibility that transformed cells have a membrane bound activity is discussed. The role of plasminogen activation was also tested for its relevance in transformation related proteolysis, growth and morphology of cells.     

Effects of cocultivation with transformed cells on surface proteins of normal cells

Virally transformed fibroblasts do not have on their surface a major protein (large external transformation-sensitive, LETS) which is present in normal cells. Cocultivation of the transformation cells with normal cells whose surface proteins have been prelabelled induces an accelerated release of the LETS protein from the normal cells. We have investigated various conditions which affect this phenomenon. Our results show that alteration of cell surface proteins by cocultivation with the transformed cells is time and dose-dependent and requires cell contact. Serum was depleted at least 99% of plasminogen by affinity chromatography and used in the cocultivation experiments. It was found that activation of plasminogen was not required for the accelerated turnover of the LETS protein. Other diffusible proteases are also unlikely to be involved. The possibility that transformed cells have a membrane bound activity is discussed. The role of plasminogen activation was also tested for its relevance in transformation related proteolysis, growth and morphology of cells.     

Different cyclic changes in the surface membrane of normal and malignant transformed cells

Transformed fibroblasts in interphase and normal fibroblasts in mitosis were agglutinated by Con A and the lectin from wheat germ, whereas normal fibroblasts in interphase and transformed fibroblasts in mitosis were not agglutinated by these lectins. The percentage of fluorescent cells at non-saturation concentrations of fluorescent ConA was also higher with transformed interphase and normal mitotic cells, than with normal interphase and transformed mitotic cells. Under the same conditions, a similar number of radioactively labeled ConA molecules were bound to normal and transformed cells in interphase and mitosis. Our results indicate different cyclic changes in the surface membrane of normal and transformed fibroblasts, so that regarding interaction with these lectins, normal mitotic cells resemble transformed interphase cells and transformed mitotic resemble normal interphase cells. The data suggest that there is a reversed cyclic change in the mobility of specific surface membrane sites in normal and transformed cells.     

Membrane phase state and the rearrangement of hematopoietic cell surface receptors.

Transformed murine hematopoietic cells of several lineages bound the fluorescent membrane probe merocyanine 540, whereas their normal counterparts did not. Similar selective binding was reproduced in artificial liposomes which bound this probe above their phase transition temperature, but not below it. The regions of the membrane to which merocyanine 540 binds along with the receptors for the lectin concanavalin A, but not the receptors for the lectin wheat germ agglutinin, were rearranged during the course of induced differentiation of erythroleukemia cells. Based on these findings, we propose a model of hematopoietic cell surface differentiation in which proteins such as concanavalin A receptors, which are destined for removal from the plasma membrane, are specifically associated with disordered, liquid-like lipid domains which can be visualized with merocyanine 540. For the specific case of erythroid differentiation, these domains and their associated proteins are collected at the region of the membrane where nuclear extrusion occurs and are eliminated from the reticulocyte plasma membrane by the enucleation event.     

The preparation and use of pyridoxal [32P] phosphate as a labeling reagent for proteins on the outer surface of membranes

Pyridoxal [³²P] phosphate was prepared using [γ-³²P]ATP, pyridoxal, and pyridoxine kinase purified from Escherichia coli B. The pyridoxal [³²P] phosphate obtained had a specific activity of at least 1 Ci/mmol. This reagent was used to label intact influenza virus, red blood cells, and both normal and transformed chick embryo fibroblasts. The cell or virus to be labeled was incubated with pyridoxal [³²P] phosphate. The Schiff base formed between pyridoxal [³²P] phosphate and protein amino groups was reduced with NaBH₄. The distribution of pyridoxal [³²P] phosphate in cell membrane or virus envelope proteins was visualized by autoradiography of the proteins separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis.The labeling of the proteins of both influenza and chick cells appeared to be limited exclusively to those on the external surface of the virus or plasma membrane. With intact red blood cells the major portion of the probe was bound by external proteins, but a small amount of label was found associated with the internal proteins spectrin and hemoglobin.     

Probe for polyanionic regions on the cell surface

Summary The binding of polyuridylate to cells in the presence of proflavine may be used as a probe to provide relative estimates of exposed polyanionic regions on the external surface of the cell. This probe binds preferably to hydrophilic, polyanionic regions, and soluble polysaccharides containing either carboxylate or sulfate groups compete with the binding of this probe. Binding of the probe to protein and lipid regions is considerably weaker. Virustransformed human fibroblasts bind 10 times less of the probe than nontransformed cells when confluent monolayers are compared. However, as the cell density is decreased, the amount of probe bound per cell increases dramatically both for transformed as well as for normal cells. In fact, human fibroblasts (a) derived from normal donors, (b) from donors with different metabolic disorders, and (c) transformed by simian virus 40, all bind about the same amount of probe when compared at the same density. Populations of human fibroblasts aged in vitro, which contain high proportions of large cells and grow only to relatively low densities in monolayers, bind disproportionately large amounts of the complex.     

A comparison of membrane components of normal and transformed BALB/c cells.

Membrane glycolipids, glycoproteins, and surface proteins of normal and transformed BALB/c cell lines have been compared. Several virally and spontaneously transformed cell lines showed differences in membrane components compared to normal A31 cells. These differences consisted of increased amounts of simpler gangliosides, absence of the large external transformation sensitive (LETS) protein, and the appearance of a major new glycoprotein band of about 105 000 molecular weight. In contrast, the spontaneously transformed cell line that caused the fastest growing tumors in vivo and the most rapid animal death (3T12T) did not have these changes. A31 and 3T12T glycolipid profiles appear similar as did glycoproteins and cell surface proteins detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. When Pronase-generated glycopeptides were analyzed by Sephadex G-50 chromatography, and enrichment in faster-eluting species was seen in two killing tumor lines (c5T and 3T12T) compared to A31. Regressing tumor lines (MSC, c5) did not show this change. Isolated membrane glycoproteins yield glycopeptides of different sized after Pronase digestion. In addition, several 3T12T glycoproteins yield glycopeptides that are larger than those from the corresponding glycoproteins of A31 cells. It appears that glycopeptide alterations associated with transformation occur in several membrane glycoproteins.     

Shark cytotoxic macrophages interact with target membrane amino groups

The types of target structures recognized by cytotoxic macrophages have been described for various microorganisms, but have not been defined for tumor cells. Tumoricidal macrophages are selective in their destructive mechanisms, sparing normal cells while directing their lytic machinery toward neoplastic targets. The cytotoxic activity of macrophages from a primitive vertebrate, the nurse shark, closely resembles the activity of mammalian tumoricidal macrophages. Host defense mechanisms of these animals appear to rely on antigen nonspecific cellular effector systems, and it has been postulated that macrophage-mediated cytotoxicity plays a dominant role in protection during periods of decreased environmental temperatures when lymphocyte responses of poikilothermic vertebrates are compromised. Similar to mammalian tumoricidal macrophages shark macrophages display selective recognition of target cells. Previous studies showed that TNP modification of targets was protective, preventing recognition by the shark spontaneously cytotoxic macrophage. Additionally, it was shown that cytotoxic activity was inhibited in a dose dependent fashion by the addition of excess unlabeled targets. In the present study, similar inhibition experiments with hapten-modified targets have been used to determine the nature of the target structures recognized by the shark cytotoxic macrophage. Cold targets modified with haptens which react covalently with free amino groups on cell membranes, trinitrobenzene sulfonic acid (TNBS) and flourescein isothiocyanate (FITC), are not recognized by the cytotoxic macrophage. The relative amount of membrane bound TNP was correlated with inhibition of cytotoxicity. Conversely, target cells modified with sulfhydryl reacting reagents, N-iodoacetyl-N'-(5-sulfonic-1-naphthyl) ethylene diamine and dithionicotinic acid, are recognized similarly to untreated targets. Moreover, TNP-containing lipids, permitted to diffuse into target membranes without covalent binding, do not alter target recognition, indicating that TNP itself has no effect on macrophage:target interaction. From these data, it is concluded that the shark cytotoxic macrophage interacts with membrane bound amino, but not sulfhydryl groups. The ability to distinguish between membrane structures may have appeared early in evolution as a means of preserving self cells while retaining protective nonspecific cytotoxic mechanisms.     

Relationship between trinitrophenyl and H-2 antigens on trinitrophenyl-modified spleen cells. I. H-2 antigens on cells treated with trinitrobenzene sulfonic acid are derivatized.

Spleen cells were treated with TNBS in order to determine if cell surface H-2 antigens are derivatized with TNP. By labeling the cell membrane of the TNP-modified cells with 125I, followed by detergent lysis and immune precipitation with anti-TNP, it was determined that no H-2 antigenic activity remained in the supernatant. Further, by the use of an antibody-induced antigen redistribution assay it was found that previous exposure to TNP-modified cells to anti-TNP in the absence of complement rendered these cells resistant to lysis by anti-H-2 in the presence of complement. Together these data indicate that at the concentration of TNBS used for modification, H-2 antigens are derivatized with TNP. However, in addition to H-2, other proteins including immunoglobulin were also derivatized with TNP. Anti-TNP cytotoxic effector cells were blocked from their cytotoxic activity by anti-TNP antiserum. These data indicate that TNP directly couples to H-2 antigens on the cell surface of TNP-modified cells and that TNP is associated with the antigenic determinant that the cytotoxic T cell recognizes.     

The study of rous sarcoma virus-transformed baby hamster kidney cells using fluorescent probes.

The fluorescent probes pyrene, pyrene butyric acid and N-phenyl 1-naphthylamine have been used to investigate the changes that accompany in vitro transformation of a baby hamster kidney cell line using Rous sarcoma virus. The fluorescent probes which reside in the membrane were used to compare the changes in microviscosity and polarity of the membranes of normal cells with two transformed cell lines. The spectrofluorimetric data indicate that following transformation the probe N-phenyl 1-naphthylamine resides in a more polar environment. However, using the probe pyrene, the yield of excimer indicates decreased mobility of this probe in the membrane of transformed cells. The data also indicate differences between the two transformed cell lines. Laser photolysis was used to study the lifetime of the pyrene probes and the quenching of the pyrene fluorescence in the membrane by several different quenching molecules. The data indicate differences between the three cell lines and suggest that transformation decreases movement within the membrane.     

Affinity-purified antigen-specific products produced by T cells share epitopes recognized by heterologous antisera raised against several different antigen-specific products from T cells

Heterologous antisera to murine or rat T-cell antigen-binding molecules (T-ABM) were raised in rabbits or sheep. The T-ABM used for immunization were purified by affinity for antigen and did not bear known immunoglobulin isotypes. T-ABM and anti-T-ABM were raised in three separate laboratories. Antisera to T-ABM were exchanged and tested for binding to T-ABM in three separate laboratories. Thus antisera to at least three distinct T-ABM were tested directly for binding to T-ABM or by adsorption of biological activity. Rabbit antisera to murine trinitrophenol (TNP)-specific T-ABM or rat AgB-specific T-ABM bound both murine or rat T-ABM, indicating evolutionary conservation of T-ABM. Similar results were found with sheep antisera to murine T-ABM. In addition, all heterologous anti-T-ABM antisera used bound murine T-ABM specific for TNP, 4-hydroxy-3-nitrophenyl acetate (NP), SRBC, or T-cell membrane proteins with similar structure. Thus, there is a commonality of antigenic determinants between various T-ABM and T-cell membrane homologues which may be T-cell surface receptors for foreign antigen.     

Differences in density of Concanavalin A-binding sites due to differences in surface morphology of suspended normal and transformed 3T3 fibroblasts.

Calculations of the density of Concanavalin A (Con A)-binding sites on normal and transformed fibroblasts have, as yet, been based on the unproven assumption that suspended cells are smooth spheres. We studied the surface morphology of suspended normal and transformed fibroblasts with scanning and transmission electron microscopes, and found a large difference in surface morphology between suspended normal and transformed 3T3 cells. When this difference in surface morphology was taken into account, the estimated cell surface area of normal 3T3 cells was approximately seven times larger than that of transformed 3T3 cells. Since equal numbers of 3H-Con A molecules are bound on normal and transformed cells, the density of Con A-binding sites is approximately seven times greater on transformed than on normal 3T3 cells. The difference in density of Con A-binding sites between normal and transformed fibroblasts might be sufficient to explain the difference in agglutination response, as originally suggested by Burger, and may also be the cause of the different degrees of clustering of Con A-binding sites on the plasma membrane of these cells.     

Differential antigen-presenting cell requirements of hapten-specific T cell lines restricted to class II determinants

We used a panel of class II-restricted T cell lines (TCL), generated against trinitrophenyl (TNP)-modified autologous peripheral blood mononuclear cells (PBMC), to examine the antigen-presenting functions of various PBMC-derived class II-positive cell types, including adherent cells, B + null cells, and activated T cells. However, activated T cells and transformed or activated B cells differed in their ability to present TNP to the TCL; TNP-modified activated lymphocytes stimulated only a subset of the class II-restricted TCL that responded to class II-positive resting cells. Moreover, certain antigen-specific TCL distinguished between antigen presented on activated T cells and transformed B cells. The differences in stimulatory capacity for particular TCL did not appear to reflect differences in the expression of class II molecules or in the ability of these cells to deliver hormonal signals or process antigen. Instead, the data suggest that differences in the ability of the cells to recognize antigen on the surface of different class II-positive cells may be a function of a secondary cell surface interaction.     

Cell surface changes during muscle differentiation in vitro: a study with the probe 2,4,6-trinitrobenzene sulphonate.

Cell surface changes during muscle differentiation in vitro, were investigated using the non permeant probe 2,4,6-trinitrobenzene sulphonate (TNBS) in order to label the aminogroups of proteins exposed on the outer surface of the plasma membrane. Surface proteins of chick myotubes and 'mature' unfused myoblasts (myoblasts grown for 7 days in a calcium-depleted medium) were found to bind an equal amount of probe, which is twice the amount bound by surface proteins in 'immature' myoblasts (1--2 days of culture) and fibroblasts. This indicates that a 'remodelling' of the plasma membrane outer surface takes place in the course of muscle cell differentiation even in the absence of cell fusion. Moreover, the total amount of TNBS bound to the surface was 4--5 times greater in myotubes than in unfused myoblasts. This appears to result from the surface expansion which occurs in myotubes during the development of the T tubule system.     

High frequency of natural autoantibodies in normal newborn mice

Spleen cells from 6-day-old nonimmunized BALB/c and BALB.B10 mice were fused with the nonsecreting hybridoma cell line Sp2/0. Three hundred and eighty-four immunoglobulin-secreting hybrids were screened for antibody activity against mouse actin, tubulin, and myosin, and against TNP, peroxidase, renin, DNA, and neurofilaments. At least 24 hybridomas in the collection (6.25%) exhibited antibody activity against this panel of antigens. Ten of these hybrids were cloned, were propagated, and the corresponding monoclonal IgM protein was isolated from ascitic fluids and was further characterized. At least four groups of antibody specificities were identified: 1) one clone reacting with TNP only; 2) one clone reacting with both actin and tubulin; 3) two clones which bound to both TNP and actin; and 4) a fourth group, comprising the six other clones, which all exhibited widespread reactivity and bound to actin, tubulin, myosin, and TNP. These results indicate: 1) B cell clones directed against self antigens are activated in the internal environment and are recovered consequently by somatic cell hybridization; 2) the widespread antibody specificities found for these newborn mouse antibodies are very similar to those previously characterized with human natural antibodies and human monoclonal Ig; and 3) the frequency of B cells binding to cytoskeletal proteins and TNP is very high (at least 6.25%).