Two hybridomas producing monoclonal antibodies reactive with natural killer cells were selected after fusion of 129 anti-C57BL/6 immune spleen cells with P3X63-Ag8.653 myeloma cells. Treatment of normal or stimulated cells with the 4LO3311 or the 4LO439 mAb and rabbit complement inhibited natural killer and antibody-dependent cellular cytotoxicities, whereas cell lysis mediated by natural cytotoxic cells, cytotoxic T lymphocytes, or activated macrophages was unaffected. Lymphokine-activated killer activity was reduced after complement-mediated treatment of interleukin-2-stimulated spleen cells with the 4LO3311 mAb but not after treatment with the 4LO439 mAb. Similar treatment of spleen cells with either mAb had no effect on the mitogen-induced proliferation of T and B lymphocytes and did not alter the frequency of antibody plaque-forming cells in immune spleen cell suspensions. The 4LO3311 and 4LO439 mAbs thus appear to be specific for NK cells and their progeny. Flow cytometry analysis confirmed that 4LO3311+ and 4LO439+ cells are phenotypically identical to NK-1.1+ cells. The epitope recognized by the 4LO3311 mAb has the same strain distribution as the NK-2.1 alloantigen previously detected with NZB anti-BALB/c antiserum, whereas the 4LO439 mAb appears to identify a new NK cell marker exclusively expressed in mice of C57BL lineage. The relationship of the molecules detected with either the 4LO3311 or the 4LO439 mAb to polymorphic antigens of the Ly series is discussed.

An indirect immunoperoxidase micro-assay, using a continuous cell line derived from ovine kidney cells (OK) and a previously characterized monoclonal antibody (7C2), specific for an exposed and highly conserved epitope of the fusion protein of different strains of RS virus, was used advantageously to rapidly titrate bovine, caprine and human strains of RSV by either quantal (TCID50) or plaque forming assays. Virus titers, obtained in less than 36 h, were in agreement with those obtained by the conventional plaque assays which required an incubation period of 4 days or more. This assay is also applicable to micro-neutralization of fusion inhibition assays for testing serum or screening monoclonal antibodies

Intracellular virus-induced polypeptides from 3 cytopathogenic and 2 non-cytophthogenic bovine viral diarrhea (BVD) virus reference strains were analyzed by radioimmunoprecipitation and polyacrylamide gel electrophoresis, using a specific bovine multivalent antiserum and a neutralizing BVD-virus monoclonal antibody. Electrophoretic patterns of major proteins demonstrate extensive variation between strains. Most notably, a major 80 000 (80K) polypeptide was present in all cytopathogenic strains but absent in both non-cytopathogenic strains. Furthermore, a neutralizing monoclonal antibody produced against the NADL strain immunoprecipitated a 53K glycoprotein indicating that this protein carries an important neutralization epitope that is not present in all strains tested.

Cytopathic and noncytopathic reference strains as well as Canadian field isolates of bovine viral diarrhea virus were analyzed by neutralization and immunofluorescence tests using a bovine viral diarrhea virus-specific neutralizing monoclonal antibody. Results on reference strains indicated three major antigenic groups: I) NADL-like, II) New York 1-like and III) Oregon C24V-like. Field isolates could be segregated into groups I and II and none could be typed into the group III. It appears that most bovine viral diarrhea virus strains share a common antigen which carries a major neutralization epitope. These characteristics would make this monoclonal antibody a useful reagent for taxonomic and epizootiological studies.

Hybridomas are as susceptible to mycoplasma contamination as animal cell cultures. Two in vitro methods and one in vivo passage into mice were compared for efficiency of curing hybridomas from mycoplasma contaminations. Four contaminated clones were treated with the combined action of 5-bromouracil (5-BrUra) and the Hoechst 33258 followed by photosensitization. The other in vitro method involved the use of BM-cycline. The success of overcoming the mycoplasma contamination was dependent on the level of the initial contamination of the individual hybridoma. BM-cycline was more efficient than the photosensitization method. For the most contaminated hybridomas, 10 successive treatments with 5-BrUra were necessary as compared to six treatments with BM-cycline. Moreover, the use of BM-cycline reduced the contamination by as much as 50% after the first treatment. After twenty passages following the curing of hybridomas with BM-cycline, cells were stable and retained their specificity and secretion of their respective immunoglobulins. Whereas, treatment with 5-bromouracil, recurrence of contamination was observed in one of the four hybridomas after 10 passages following treatment. Decontamination after one passage in peritoneal cavity in mice was not always sufficient since one of four hybridomas remained contaminated. BM-cycline appears to be the method of choice since it is more efficient, less time consuming, simpler and less expensive. Mycoplasma strains that could be identified were of bovine origin: Mycoplasma arginini and Acholeplasma laidlawii.

Hemagglutination activity, structural protein profiles and neutralization assays were used in a comparative study of bovine herpesvirus 1 strains from the U.S.A., Canada, Great Britain, Denmark and Malaysia with equine, feline and human herpesviruses in order to further characterize the bovine herpesvirus 1 hemagglutinin. Bovine herpesvirus 1 strains of different geographical origins all showed hemagglutinating activity for mouse erythrocytes; furthermore, feline herpesvirus 1 was also shown to hemagglutinate mouse erythrocytes. Analyses of partly purified viruses showed that a distinctive and specific polypeptides profile is associated with each species of herpesviruses used in our study; strains of bovine herpesvirus 1 from North America, Europe and Southeast Asia however, presented a remarkable similarity as to their electrophoretic protein patterns. A protein similar to the 97-kDa bovine viral hemagglutinin was not identified with the hemagglutinating feline herpesvirus. An important neutralization epitope on the bovine viral hemagglutinin was also not found on feline, equine and human herpesviruses but was identified on all bovine strains tested from North America, Europe and Southeast Asia stressing the importance of the bovine hemagglutinin for eventual prophylactic purposes.

Preparation of monoclonal antibodies to bovine virus diarrhea virus (BVDV) yielded some hybridoma cells that secreted monoclonal antibodies against the Madin-Darby bovine kidney cells. The anti-cellular monoclonal antibodies reacted with other bovine cells (bovine turbinate and testicle) but not with cell lines derived from other animal species. Subclones derived from one hybridoma partially blocked the infectivity of BVDV, possibly through the binding of the monoclonal antibodies with an epitope close to the receptor site of BVDV and not by way of steric hindrance. Unexpectedly, these same subclones completely blocked the infectivity of bovine enterovirus-3 (BEV-3) strain 240A and partially blocked the infectivity of BEV-2 and BEV-3 (ATCC strain) but not that of other serotypes. Other subclones derived from two other hybridomas, although cell membrane specific, did not have a protective activity against BEV or BVDV.

The NADL strain of bovine viral diarrhea virus (BVDV) was concentrated by hollow fiber ultrafiltration or polyethylene glycol and purified by centrifugation through sucrose or potassium tartrate gradients. The protein content of polyethylene glycol concentrates was much lower than that of ultrafiltration concentrates. Conversely, recoveries of infectivity were greater using polyethylene glycol (100%) as compared to ultrafiltration (50%). Sucrose or potassium tartrate density gradients appeared comparable in purification of BVD virus. Peak infectivity fractions in both gradients corresponded quite well, having densities of 1.12-1.14 g/cm3, and showed a 150-fold reduction of protein when compared to crude viral supernate. Further examination by negative stain electron microscopy revealed integral pleomorphic, roughly spherical particles in both purified virus preparations. Small knob-like projections could be seen on viral particles.

We have generated three hybridoma-producing monoclonal antibodies (MAs) that show a different spectrum of reactivity to human mammary tissues. Two of these antibodies, 1F10B4 and 1F10G2, recognize a cytoplasmic determinant highly expressed in most of the primary and metastatic breast carcinomas studied, and weakly (or not at all) in normal breast and nonbreast tissues. 3C6F9 detected a surface determinant common to both normal and neoplastic mammary epithelium. Five hundred hybridomas were obtained from the fusion of NS-1 myeloma cells with spleen cells of mice hyperimmunized with the well-characterized human breast carcinoma cell line BT-20. After the initial screenings and clonings, three monoclonal antibodies (1F10B4, 1F10G2, and 3C6F9) showing a restricted range of reactivity were selected for further investigation. These three antibodies recognized a panel of neoplastic mammary cell lines; however, the degree of reactivity could not be correlated to any of the various characteristics of these epithelial cell lines. Moreover, immunofluorescence analysis of acetone-fixed cryostat section showed that 1F10B4 and 1F10G2 recognize the vast majority of the 37 primary and metastatic breast cancers tested, binding strongly to 47% and 67% of them respectively. Only one of the primary carcinomas was not recognized by 1F10B4. On the other hand, these two MAs reacted weakly or not at all with normal breast and nonbreast tissues showing only few focal reactivities with the luminal pole of some ducts of the breast; very weak staining in renal tubular epithelial cells, in few keratinocytes and epithelial cells lining some sebaceous glands in the skin; and a moderate staining in biliary ducts of the liver. All mesenchymal structures including smooth and striated muscle tissues, lymph nodes, and connective tissue were negative. On the other hand, 3C6F9 recognized a more limited number of human mammary tumors and reacted with normal ductal epithelium in the breast and with nonbreast tissues. Because of their wide spectrum of reactivity with breast cancer cells and restricted recognition of normal mammary tissues, their cytoplasmic localization, and their heterogeneous distribution within a single neoplasm, 1F10B4 and 1F10G2 are now being used to characterize antigenic phenotypes of tumor-associated antigens in retrospective studies performed on conventional formalin-fixed, paraffin-embedded human mammary carcinomas.

The adherence to eukaryotic cells of Escherichia coli, Neisseria gonorrhoeae, Pseudomonas aeruginosa, Staphylococcus aureus, Staphylococcus epidermidis and the yeast Candida albicans was studied by light microscopy with an in vitro micromethod involving different cell lines. The method is inexpensive, consumes little time and material, and is reproducible. It was used to show that the gram-positive Cowan I strain of S. aureus, which naturally forms protein A on its surface, adheres in much larger numbers to human lung fibroblasts than the protein A-free Wood 46 strain, the strain of S. epidermidis, and the encapsulated Smith strain. The presence of a capsule on the latter strain apparently prevented its attachment to the fibroblasts. Among the gram-negative species studied, a piliated clinical isolate of N. gonorrhoeae, displaying the opaque colonial phenotype, adhered in larger numbers than another isolate lacking pili and displaying the transparent phenotype. E. coli K12 attached slightly to the cell line, whereas P. aeruginosa adhered to it moderately. One strain of C. albicans tested did not attach in any detectable numbers. No clear correlation between bacterial cell surface hydrophobicity, as evaluated by the hexadecane assay, and adherence to eukaryotic cells could be demonstrated for these microorganisms. With our method, bacterial attachment proceeded best at 37 degrees C and did not require more than 1 h of contact with the cell monolayer. The method described revealed differences in the adherence to eukaryotic cells, not only among species, but also between strains of the same species.

The relative importance of three major proteins of influenza virus in the mechanism of induction of cell-mediated cytotoxicity (natural killer cell activity) was assessed by an overnight chromium-51-release assay using radiolabeled K-562 cells as target cells and Ficoll-Hypaque-purified peripheral-blood lymphocytes as effector cells. Incubation of peripheral-blood lymphocytes with influenza virus (whether type-A or type-B) showed that intact and formalin-inactivated influenza virus enhanced cell-mediated cytotoxicity equally. The stimulation by intact or inactivated virions was comparable to that induced by the two major internal mediators of positive natural killer cell regulation, namely interferon and interleukin-2. This virus-induced cell-mediated cytotoxicity, which was mediated by human natural-killer 1+ cells, could be blocked only with monoclonal antibodies to the hemagglutinin and not with antinucleoprotein or antimatrix protein monoclonal antibodies, results indicating that the hemagglutinin of influenza virus is a potent mediator of natural killer cell stimulation in vitro.

Immunoprecipitation of [3 H]amino acid labelled virus with monoclonal or human convalescent rubella sera and subsequent analysis by electrophoresis and fluorography, revealed three structural proteins of rubella virus: VP3 : 59,000; VP2 : 44,800; and VP1 : 33,000.

The X-31 strain of influenza A (H3N2) virus has been covalently bound to CNBr activated agarose for the separation of anti-haemagglutinin antibodies. The virus immunoadsorbent was used repeatedly under high ionic strength alkali buffer and acid conditions without altering appreciably its antibody binding capacity. Sequential elution of bound anti-haemagglutinin antibodies with increasing concentrations of sodium iodide has enabled the physical separation of antibody populations with low and high avidity for the virus immunoadsorbent. In haemagglutination inhibition (h1) assays, the less avid population reacted only with the homologous X-31 virus, wheras the more avid antibody population reacted both with the homologous and the related cross-reactive A/England/42/72 (H3N2) strains. Sequential elution under acid conditions did not completely remove the bound anti-haemagglutinin antibodies and those eluted retained little of their anti-haemagglutinin activity. From a practical point of view, given a specific antiserum, it is feasible to use whole virus as an immunoadsorbent for the purpose of isolating populations of antibodies of different avidities and cross-reactivities. Furthermore, sodium iodide as an eluting agent has proved most effective in recovery of active and stable antibodies from the agarosebound virus.