Purification and analysis of a human sarcoma associated antigen.

S1, a heterophile antigen present on human sarcoma cell lines in culture, has been previously defined by this laboratory [1,2]. This antigen is also present in guinea-pig kidney. Purification of the antigen to homogeneity has now been achieved by a combination of ammonium sulfate fractionation, DEAE-cellulose, sephadex, high pressure liquid chromotography and affinity chromotography. S1 is a monomeric protein of 70,000 Da, as indicated by the presence of a single band on SDS-PAGE. Amino acid analysis demonstrates the prevalence of glycine, lysine and glutamic acid. Aspartic acid was found to be the N-terminal residue with further sequence of glycine-valine-alanine-glutamic acid (gly-val-ala-glut).

[Action of adenosine on intestinal motility after experimental mesenteric ischemia in dogs]. / Action de l'adénosine sur la motricité intestinale après une ischémie mésentérique expérimentale chez le chien.

The purpose of the present work was to demonstrate beneficial action of adenosine on intestinal motility in the dog after experimental ischemia and to establish the role of A1- and A2-purinergic receptors. Adenosine was compared to an A1-agonist (N6-cyclohexyl-adenosine or CHA), an A2-agonist (5'-(N-ethyl) carboxamido-adenosine or NECA) and an inhibitor of adenosine cellular uptake (dipyridamole). Motility was analyzed by recording the electromyogram with intraparietal electrodes. Experimental ischemia was obtained by clamping the superior mesenteric artery during two hours. Under physiologic conditions, adenosine (1 mg.kg-1.i.v.) increased myoelectric complexes frequency by 49 p. 100 and the jejunal motility index by 46 p. 100. CHA (1 microgram.kg-1.i.v.) and dipyridamole (0.5 mg.kg-1.i.v.) increased myoelectric complexes frequency by 42 p. 100 and 67 p. 100 while NECA (1 microgram.kg-1.i.v.), increased the motility index by 30 p. 100. Adenosine and NECA increased mesenteric arterial blood flow by 35 p. 100 and 57 p. 100 respectively. After a two hours ischemia the return to normal electromyogram was 10.1 +/- 5.7 h. Adenosine (administered 10 min after the end of ischemia) reduced this recovery period to 3.3 +/- 0.8 h, NECA to 1.3 +/- 0.8 h and dipyridamole to 5.0 +/- 2.6 h. We conclude that, under physiological conditions, adenosine modifies jejunal motility directly via A1-receptors of smooth muscle and indirectly via A2-receptors of mesenteric vessels. On the other hand, after experimental ischemia, adenosine improves the post-ischemic restoration via vessels A2-receptors only.

The effect of insulin on human-mouse hybridoma formation.

To evaluate the effect of insulin on the formation of human-mouse hybridoma clones, P3U1 mouse plasmacytoma cells were fused with human lymph-node lymphocytes in the presence of polyethylene glycol. After fusion, cells were grown for 2 weeks in HAT medium supplemented with insulin (H1AT, 10(-1)-10(-5) units/ml) or in HAT medium alone. The addition of 10(-3) units/ml of insulin to HAT medium resulted in an over two fold increase in the number of clones formed and in the average colony size compared to growing the fused cells in HAT medium alone. In view of the recent increasing interest in human-mouse hybridoma fusions it is suggested that selective medium HAT should be supplemented by insulin (H1AT) to enhance the number of colonies formed and provide a more efficient way for stabilizing the newly formed hybrids.

Monoclonal antibody defining fibroblasts appearing in fetal and neoplastic tissues.

VIF3 is a hybridoma-derived mouse IgG monoclonal antibody (MoAb) generated in a fusion with the use of a malignant fibrous histiocytoma as the immunizing agent and shown to recognize a 70-kilodalton antigen expressed within connective tissues. Of 55 human tissue culture cell lines tested by indirect immunofluorescence, VIF3 was shown to bind to 20 of 35 (57%) sarcomas, 4 of 9 (44%) normal fibroblasts, and none of 11 carcinomas and other neoplasm-derived cell lines. A panel of over 259 human frozen tissue sections obtained from surgical pathology specimens, postmortem studies, and elective abortions was used to further determine the histopathologic specificity of VIF3. MoAb VIF3 was found to bind to 15 of 19 (79%) human sarcoma tissue sections tested. It was also shown to recognize an antigen expressed on a subset of fibroblasts dispersed within the stroma of carcinomas obtained from all 46 patients tested, as well as a subset of cells within 3 of 10 benign hyperplastic tissues (30%). VIF3-positive cells were detected in all 60 fetal tissues tested including amniotic membranes, placentas, and umbilical cords. In contrast, fibroblasts of normal adult tissues tested stained infrequently (22/97 or 23%) with this reagent. The results confirm that this MoAb is directed against a human connective tissue-specific marker. VIF3 detects a marker appearing on fetal fibroblasts that is typically not present in normal adult tissues, but reappears in association with neoplastic diseases. MoAb VIF3 therefore defines a fibroblast-oncofetal antigen and as such may serve as a probe for immunopathologic studies.

A high-affinity monoclonal antibody (GIF-1) to human gamma-interferon: neutralization of interferon mediated inhibition of retrovirus production and 2'-5' (A) synthetase induction.

An hybridoma clone secreting an IgG1 monoclonal antibody (GIF-1) specific for human gamma-interferon (HuIFN-gamma) has been generated using HAT medium supplemented with insulin (HIAT) at the initial stage of cell fusion. This antibody is capable of neutralizing the antiviral activity of HuIFN-gamma, the ability of HuIFN-gamma to inhibit retroviral replication in RD-114 cells, and the ability of HuIFN-gamma to induce the 2'-5' oligoadenylate (A) synthetase in RD-114 and HeLa cells. Eluate from an immunoaffinity column containing GIF-1 yielded two protein bands of molecular weight of 20 and 25 kd when subjected to SDS-PAGE.

The use of insulin-enriched culture medium and human-human hybridoma formation: a preliminary study.

Human lymphoblastoid cell line WI-L2-729-HFZ was fused with human lymph-node lymphocytes in one fusion and with human spleen cells in another fusion to generate human-human hybridomas. In both, increasing doses of insulin were added to the HAT medium immediately after the PEG-mediated cell fusion (10(-1)-10(-5) IU/ml) and the number of clones formed was determined 3 weeks later. 10(-3) IU/ml of insulin resulted in a 2- to 5-fold increase in the number of clones generated compared to the control plates. In view of the known difficulties in generating high-yield human-human hybridoma fusions, it is suggested that the use of insulin-supplemented HAT medium may provide a more efficient way in obtaining such clones.

Modulation of hybridoma formation by dexamethasone.

In an effort to determine the effect of dexamethasone on hybridoma formation, spleen cells from BALB/c mice hyperimmunized with sheep red blood cells (SRBC) were fused with mouse plasmacytoma cells (P3U1) in the presence of polyethylene glycol (PEG). Dexamethasone was added in decreasing doses (10(-3) to 10(-9) mM) to the hypoxanthine-aminopterin-thymide (HAT) medium immediately after the PEG-mediated cell fusion. 10(-3) mM of this steroid was found to inhibit markedly the number and size of hybridoma clones generated, while 10(-5) mM dexamethasone was shown to enhance hybridoma formation. The effect of 10(-3) mM dexamethasone was most pronounced when added immediately after fusion. When this dose was given 48 or 120 h after cell fusion, the extent of the inhibitory effect was less pronounced. High concentration of dexamethasone may also inhibit monoclonal antibody production by hybridomas once generated. An increase in the number of clones formed was observed when 10(-5) mM dexamethasone was added to HAT medium as well as an increase in the average colony size. Large clones were also observed with lower dexamethasone doses ranging from 10(-7) to 10(-9) mM. Possible mechanisms on the effect of dexamethasone on hybridoma formation are discussed.

Detection of retroviral particles in hybridomas secreting monoclonal antibodies.

Electronmicroscopy of hybridoma clones derived by fusing BALB/c mouse spleen cells with P3U1 mouse plasmacytoma cells to generate monoclonal antibodies against human sarcoma antigens, revealed the presence of large number of viral particles. These particles were also seen budding from the cell surfaces. The intracytoplasmic particles were intracisternal and resembled type-A oncornavirus, while the budding and extracellular forms, with a centrally located nucleoid, resembled mature type-C oncornaviruses. Cells of the parental P3U1 palsmacytoma cell line and of the NS-1 myeloma cell line contained morphologically identical viral structures. The scientific and medical communities engaged in hybridoma research should be alert to the possible presence of viruses in hybridomas and their products. The question is raised as to whether it is safe to use mouse monoclonal antibodies for clinical purposes, both diagnostic and therapeutic.

The interaction of Kaposi's sarcoma with monoclonal antibodies to human sarcoma and connective tissue differentiation antigens.

Four monoclonal antibodies (McAbs) previously generated against human soft tissue sarcomas and reacting with connective tissue differentiation antigens were evaluated for their interaction with tissues obtained from patients with classic Kaposi's sarcoma. Biopsy was performed on active neoplastic lesions from the skin of 26 patients, frozen sections were prepared, and the binding of the McAbs was tested using the indirect immunofluorescence assay. Clinically uninvolved skin from the same patients as well as skin and muscle from eight non-cancer patients were treated similarly and served as controls. McAbs IXG11, 23H7, IIIE5, and 15G5 interacted strongly with the Kaposi's sarcoma lesions and weakly with the uninvolved skin in 22 of 26 (84%), 23 of 26 (88%), 12 of 14 (85%), and 1 of 6 (16%) of the patients, respectively. IXG11, 23H7, and IIIE5 interacted weakly with the skin of seven of eight non-cancer patients. McAb 15G5 was found to bind strongly to tumor lesions, to the respective uninvolved skin in four of five Kaposi's sarcoma patients, and also to skin and connective tissues of muscle from non-cancer patients. The mode of interaction was morphologically different for each McAb. It is suggested that McAbs IXG11, 23H7 and IIIE5 identify markers whose expression is markedly increased in Kaposi's sarcoma lesions as compared with uninvolved skin of the same patients. These markers may serve as immunologic probes for the investigation of this neoplastic process.

Clonal stability and heterogeneity of hybridomas: analysis by multiparameter flow cytometry.

Flow cytometry of cellular DNA and RNA content was employed to determine DNA ploidy, proliferation, and RNA content of hybridoma cultures shortly after cell fusion and sequentially thereafter. The parental mouse myeloma cell line P3UI was characterized by tetraploid DNA content, S-phase 50%, and high RNA, the parental mouse spleen cells by diploid DNA content, low proliferation, and low RNA content. Hybridoma cultures studied as early as 21 days after fusion were found to contain the sum of the parental cells' DNA content (hexaploid), or if less, more than that of the myeloma parental cells. Only one clone of 35 tested was shown to be hexaploid and the rest hypertetraploid or hyperpentaploid. Hybrid cell cultures were frequently found to contain a variable mixture of unfused parental cells. The high proliferation of hybridoma cells determined by flow cytometry indicates that these cells would eventually overgrow the parental cells. Flow cytometry also enabled an accurate estimation of the effect of various doses of dexamethasone added to HAT medium immediately after cell fusion on hybridoma formation. Cultures treated with 10(-5) mM of the hormone had a higher DNA ploidy than cultures grown in the presence of 10(-3) mM dexamethasone. No parental cells were observed in the hybridoma cultures studied with this hormone. Sequential DNA/RNA measurements of hybridoma clones showed a decrease in DNA ploidy over time with high dexamethasone doses and a minimal increase or no change with low hormone dose. Flow cytometry is suggested to be a useful technique for evaluating the effects of various agents on DNA ploidy and proliferation and on stability of fused cells.

Monoclonal antibodies to human sarcoma and connective tissue differentiation antigens.

The use of monoclonal antibodies to distinguish human sarcoma from carcinoma cells has been explored. Spleen cells from a BALB/c mouse immunized with a human malignant fibrohistiocytoma were fused with cells of the mouse P3U1 plasmacytoma cell line. Antibodies were then screened for reactivity against human sarcoma and carcinoma cells growing in culture. This work has yielded 2 immunoglobulin G monoclonal antibodies VIE4 and VIF3 which, respectively, reacted with 85% (17 of 20) and 90% (18 of 20) of sarcoma lines tested but with none of eight carcinoma cell line preparations. Reactivity against normal fibroblasts was also demonstrated. By immunofluorescence, the antigens detected by the two antibodies appear to have distinctive intracellular distributions. Immunoprecipitation with VIF3 has shown that it is detecting a protein with a molecular weight of 70,000. When tested against pathological frozen tissue sections, VIF3 reacted with four of 11 and VIE4 with three of 11 human sarcomas but with none of ten carcinomas tested. VIF3 occasionally bound to normal adult connective tissues, whereas no such reactivity was seen with VIE4. These antibodies appear to be directed to fibroblastic markers associated with sarcomas and connective tissue differentiation antigens.

Natural and recombinant Escherichia coli-derived interferon-gamma differ in their reactivity with monoclonal antibody.

Monoclonal antibody GIF-1 was found to neutralize human natural immune interferon (IFN-gamma), but not Escherichia coli-derived recombinant IFN-gamma. In addition, GIF-1 antibody failed to immunoprecipitate 125I-labeled recombinant IFN-gamma, whereas it precipitated natural IFN-gamma in a concentration-dependent manner. The lack of recognition of recombinant IFN-gamma by antibody GIF-1 may not be due to the absence of the oligosaccharide moiety in the molecules of recombinant IFN-gamma alone, because removal of carbohydrate from natural IFN-gamma by treatment with a mixture of glycosidases did not alter the selective binding of antibody, i.e., deglycosylated and untreated natural IFN-gamma were equally neutralized and immunoprecipitated by GIF-1 antibody. In addition, a minor monomeric component of natural IFN-gamma with the m.w. of 15,500, which apparently lacks carbohydrate, was also recognized by antibody GIF-1. These results suggest that the discriminative recognition of natural and recombinant IFN-gamma by monoclonal antibody GIF-1 may be due to a conformational difference at or near the active regions of natural and recombinant human IFN-gamma molecules.

The addition of insulin to HAT medium (HIAT) enhances hybridoma formation.

To assess the effects of insulin on the formation of hybridomas, sheep red blood cell (SRBC) immunized spleen cells from BALB/c mice were fused with P3U1 mouse myeloma cells. After fusion, cells were grown for 2 weeks in HAT medium containing insulin (HIAT) (doses ranging between 10(-1) to 10(-9) units/ml) or HAT medium alone. The number of hybridoma colonies was found to be significantly increased in the presence of HIAT medium compared to HAT alone. In addition, the average size of the hybridoma clones was at least doubled and the cumulative colony size index per plate increased several folds. A significant rise in the number of wells containing clones secreting anti-SRBC monoclonal antibodies was again shown in the presence of HIAT compared to HAT medium alone. The maximal effect of insulin on the above biological parameters ranged between 10(-1) and 10(-4) units/ml. It is concluded that the addition of insulin to HAT medium (HIAT) enhances hybridoma formation and thus, its more frequent use may considerably expediate ongoing research efforts on the production of monoclonal antibodies.

Enhancement of hybridoma formation by addition of insulin to HAT medium (HIAT).

The effects of insulin on the formation of hybridoma clones following fusion experiments with SRBC immunized BALB/c mouse spleen cells and P3U1 mouse plasmacytoma cells were evaluated. The addition of insulin to HAT medium (HIAT) resulted in significant increases in the number and size of hybridoma colonies generated. The total number of anti-SRBC antibody-secreting clones also increased as much as sevenfold using insulin-supplemented medium compared to HAT alone. In view of the increasing interest in hybridoma technology for monoclonal antibody production, the use of insulin-supplemented medium (HIAT) may significantly expedite ongoing work by providing a more efficient method for the establishment of stable clones.

Comparative analysis of the influences of human gamma, alpha and beta interferons on human multipotential (CFU-GEMM), erythroid (BFU-E) and granulocyte-macrophage (CFU-GM) progenitor cells.

Preparations of human interferon (HuIFN) immune (gamma) (2 X 10(7) units/mg protein), HuIFN leukocyte (alpha) (1.4 X 10(8) units/mg protein) and HuIFN fibroblasts (beta) (10(6) U/mg protein) were assessed for their influence on colony formation of human hematopoietic progenitor cells: colony forming unit-granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM), burst forming unit-erythroid (BFU-E), day 7 colony forming unit granulocyte-macrophage (CFU-GM) and day 14 CFU-GM. Colony formation by CFU-GEMM and BFU-E was suppressed equally by the three preparations of HuIFN, but colony formation by CFU-GM was suppressed differentially. CFU-GM were, on the whole, more responsive to HuIFN gamma than HuIFN alpha, and HuIFN beta was least effective. HuIFN alpha, but not HuIFN gamma or HuIFN beta, suppressed colony formation from CFU-GM without also suppressing the total number of colonies plus clusters. This was due to an increase in the numbers of clusters formed in the presence of HuIFN alpha. The suppressive influence on colonies from CFU-GM by the preparations of HuIFN and the enhancement of clusters by HuIFN alpha was apparently equal for colonies and clusters of neutrophils, eosinophils, macrophages and neutrophils plus macrophages. The suppressive effects of HuIFN gamma were inactivated by a monoclonal antibody to HuIFN gamma and the suppressive and enhancing effects of HuIFN alpha were inactivated with a heteroantiserum to HuIFN alpha. Depletion of monocytes, T lymphocytes and B lymphocytes from the target bone marrow cells had no influence on the effects of the preparations of HuIFN. These results demonstrate that the effects of HuIFN gamma and HuIFN alpha are due to the HuIFN themselves and that these actions on the hematopoietic progenitor cells are probably not mediated through monocytes and/or lymphocytes.

An enzyme-linked immunosorbent assay (ELISA) for the detection of monoclonal antibodies recognizing surface antigens expressed on viable cells.

A solid-phase enzyme-linked immunosorbent assay (ELISA) has been developed for detecting monoclonal antibodies binding to surface antigens expressed on viable adherent cells of tumor cell lines. This assay utilizes a sheep anti-mouse IgG to which a beta-galactosidase is linked. It is highly sensitive and permits quantification of IgG monoclonal antibody levels. In studies of monoclonal antibodies prepared against human tumors, the ELISA assay revealed the presence of antigens which were not seen using acetone-fixed cell immunofluorescence methods. This assay is safe, rapid, low cost, and gives reproducible quantitative results. As such it should prove useful to laboratories engaged in the study of antigens expressed on cell membranes.