To study the cellular infiltrate that occurs within the airways of infants with respiratory syncytial virus bronchiolitis, samples of airways secretions were obtained by bronchial lavage from the lower respiratory tract of infants ventilated for this condition and from the upper airway of non-intubated infants with this disorder using nasopharyngeal aspirates.Cytospin samples were prepared so that differential cell counts could be performed on the cells obtained and alkaline phosphatase-antialkaline phosphatase immunocytochemical analysis of lymphocyte subsets was carried out using a panel of monoclonal antibodies, which included anti-CD3, anti-CD4, anti-CD8, anti-CD19, and anti-TcR-y.Results from the lower and upper airways were similar. Large numbers of inflammatory cells were obtained, of which neutrophils accounted for a median of 93%/o in the upper airway and 76% in the lower airway. The numbers of CD8 positive cells detected were small and consistently less than CD4 positive cells, median CD4:CD8 ratios being 22*5:1 and 15:1 for the lower and upper airways. CD19 positive cells were rarely observed and no y8 positive lymphocytes were detected.These results indicate that neutrophils probably play a major part in causing symptoms in these infants. They do not support the concept that excessive lymphocyte mediated cytotoxic activity is principally responsible for the pathology in respiratory syncytial virus bronchiolitis. (Arch Dis Child 1994; 71: 428-432) The respiratory syncytial virus is unique in its ability to cause yearly epidemics of respiratory disease.1 2 As a result almost all infants will have been infected by the virus by the second year of life and, of these, 05-2% develop acute bronchiolitis severe enough to be admitted to hospital. Results from studies using a formalin inactivated virus3 and other more recent attempts to produce a vaccine suggest that it is unlikely that we will be able to effectively prevent this distressing illness until the immunological mechanisms involved in the disease process, including those aspects conferring protection, are clarified.

The Epstein-Barr virus (EBV) nuclear antigen EBNA-1 is essential for viral genome maintenance in vitro and may be the only EBV protein expressed by the majority of latently infected cells in vivo. EBNA-1 may therefore be critical to the evasion of host immunity which allows persistent infection. EBNA-1 includes a polymorphic internal repeat domain of unknown significance and unique regions which mediate all known functional activities and which have hitherto been assumed to be conserved between strains. Monoclonal antibodies were generated using a construct based on EBNA-1 of the prototype B95-8 strain, deleted for the repeat domain. These antibodies showed a limited profile of recognition of EBNA-1 in common laboratory EBV+ cell lines by immunoprecipitation and immunostaining. The observed antigenic heterogeneity also extended to spontaneously transformed B lymphoblastoid cell lines (LCLs) representing viral isolates circulating within US and UK populations. DNA fragments spanning the C-terminal unique domain of EBNA-1 from eleven spontaneous LCLs were amplified by polymerase chain reaction for sequencing, which directly demonstrated extensive and unexpected variability between diverse type 1 EBV isolates. The resulting polymorphism affects most of the putative MHC Class I binding epitopes which could be identified within this region using published sequence motifs, and influences MHC binding by variants of at least one such peptide in the processing mutant cell line T2. These findings could be related to the apparent lack of recognition of EBNA-1 by cytotoxic T lymphocytes.

Monoclonal anti-idiotypic antibodies generated against idiotypic immunoglobulin (Ig) of neoplastic B lymphocytes can be selected from growing hybridoma clones by their ability to recognize idiotypic but not normal IgM. This group of antibodies can be subdivided into those that bind to the target tumor cells in the presence of normal human serum (approximately 85% of the clones) and those in which binding is inhibited by serum (approximately 15%). The former appear to be specific for private idiotypic determinants whereas the latter recognize cross-reacting idiotypic determinants. Such cross-reactivity is reflected both in recognition of a small percentage of normal Ig and also in binding to other lymphomas. The anti-idiotypes specific for private determinants can be used for therapy, with only idiotypic Ig secreted by tumor cells able to block its access to cells. The cross- reacting anti-idiotypes will face in addition the barrier of the proportion of normal Ig with which it reacts. The attraction of using a single monoclonal reagent for more than one patient has led us to develop an assay that measures the level of such blocking and to propose that those recognizing less than 30 micrograms/mL of normal Ig could be placed in a panel for possible therapy for several patients; less restriction need apply to antibodies for monitoring tumor progress. The assay is described, and examples of such antibodies raised against lymphoma cells from two patients are given together with comparisons with them of anti-idiotypes specific for private determinants.

Tumor cells from patients with B cell neoplasms often secrete small amounts of free monoclonal light chains that can be found in the urine. Such tumor-derived light chains of the lambda type from a patient with typical chronic lymphocytic leukemia have been used to raise mouse monoclonal antibodies (MoAbs). A hybridoma-secreting antibody that recognized the idiotypic lambda chain but not normal lambda chains by a preliminary screen but which also reacted with idiotypic IgM from the patient's tumor cells was selected. This MoAb in fact recognized 1 in 20 X 10(3) molecules of pooled normal lambda chains, thus establishing its specificity for a private idiotypic determinant. It failed to give a detectable reaction with normal IgM, normal serum, or a panel of IgM paraproteins. The antibody bound to the patient's neoplastic B cells but not to normal tonsillar cells. The site of binding of the antibody to idiotypic IgM is clearly separate from that of another MoAb specific for idiotypic determinants on heavy plus light chains, since the two showed additive binding curves. The determinant also appeared to be less available in dimeric lambda chains than in monomeric lambda chains or in idiotypic IgM. Antibodies to idiotypic determinants on light chains show some technical advantages and should be useful for monitoring and possibly treating B cell tumors, either alone or together with the more conventional anti-idiotypic antibodies that usually recognize the heavy and light chain combination.

No abstract

Lymphokine-activated killer (LAK) cells were successfully generated in all cases from blood mononuclear cells obtained from six patients with lymphoma. The LAK cells from three of these patients and from five normal adult donors were tested for their effector abilities in antibody-dependent cellular cytotoxicity (ADCC) against guinea pig leukemic lymphocytes coated with various antiidiotype antibodies. Cells from all the donors behaved similarly. Mouse monoclonal antibodies of IgG1, IgG2a, and IgG2b isotypes invoked no ADCC. However, substantial ADCC was invoked by the chimeric antibody FabFc, in which Fab'gamma from mouse antiidiotype is thioether-bonded to human normal Fc gamma. Similar results were obtained on testing LAK cells from a normal donor against uncultured human lymphoma targets coated with native or chimeric antiidiotype. The ADCC invoked by the mouse-human chimeric antibodies appears to depend on the human Fc gamma they display and not on the univalency of the derivatives used. The findings imply that LAK technology could usefully augment serotherapy that uses antibody derivatives displaying human Fc gamma.

SUMMARYInfection of T lymphoblastoid CEM cells with the IIIB isolate of HIV-1 results in modulation ofthe expression of several cellular antigens in addition to the CD4 molecule. The intercellular adhesion receptor LFA-1 (CDI la/CD 18) and HLA-DR are markedly induced in the cytoplasm and at the cell surface, and the CD7 antigen is down-regulated, being virtually undetectable by sensitive immunocytochemical techniques in the infected cell population. These modulatory effects are to some degree dependent on the virus isolate examined, as the CBL-1 British isolate did not induce comparable phenotypic changes in the CEM cell line. Furthermore, these effects are not reproduced by recombinant gpl20 (IIIB isolate) or p24 added exogenously to uninfected CEM cells. The CD7 molecule appears to play a regulatory role in T cell proliferation, and the LFA-1 integrin molecule is involved in a wide range of immunologically important cell-cell interactions, as well as HIV-induced syncytium formation. The possible contributions of such effects to the pathogenesis of HI V infection are considered.

Lymphokine-activated killer (LAK) cells were successfully generated in all cases from blood mononuclear cells obtained from six patients with lymphoma. The LAK cells from three of these patients and from five normal adult donors were tested for their effector abilities in antibody-dependent cellular cytotoxicity (ADCC) against guinea pig leukemic lymphocytes coated with various antiidiotype antibodies. Cells from all the donors behaved similarly. Mouse monoclonal antibodies of IgG1, IgG2a, and IgG2b isotypes invoked no ADCC. However, substantial ADCC was invoked by the chimeric antibody FabFc, in which Fab'gamma from mouse antiidiotype is thioether-bonded to human normal Fc gamma. Similar results were obtained on testing LAK cells from a normal donor against uncultured human lymphoma targets coated with native or chimeric antiidiotype. The ADCC invoked by the mouse-human chimeric antibodies appears to depend on the human Fc gamma they display and not on the univalency of the derivatives used. The findings imply that LAK technology could usefully augment serotherapy that uses antibody derivatives displaying human Fc gamma.