Isolation of a second recombinant human respiratory syncytial virus monoclonal antibody fragment (Fab RSVF2-5) that exhibits therapeutic efficacy in vivo.

A second human respiratory syncytial virus (RSV)-neutralizing monoclonal antibody was isolated and its binding site was identified. Fab F2-5 is a broadly reactive fusion (F) protein-specific recombinant Fab generated by antigen selection from a random combinatorial library displayed on the surface of filamentous phage. In an in vitro plaque-reduction test, the Fab RSVF2-5 neutralized the infectivity of a variety of field isolates representing viruses of both RSV subgroups A and B. The Fab recognized an antigenic determinant that differed from the only other human anti-F monoclonal antibody (RSV Fab 19) described thus far. A single dose of 4.0 mg of Fab RSVF2-5/kg of body weight administered by inhalation was sufficient to achieve a 2000-fold reduction in pulmonary virus titer in RSV-infected mice. The antigen-binding domain of Fab RSVF2-5 offers promise as part of a prophylactic regimen for RSV infection in humans.

Recombinant human Fab antibody fragments to HIV-1 Rev and Tat regulatory proteins: direct selection from a combinatorial phage display library.

A human Fab phage display library has been produced from peripheral blood lymphocytes of an individual who was asymptomatic after 10 years of infection with human immunodeficiency virus type-1 (HIV-1). The library was panned against the HIV-1 Rev and Tat regulatory proteins and several clones, producing Fab binding to these proteins, were isolated (3 to Rev and 4 to Tat) with binding constants varying from 10(-6)M to 10(-8)M. DNA sequencing demonstrated two unique anti-Rev Fab clones, but the four anti-Tat Fab comprised only two unique IgG1 heavy chain Fd fragments, illustrating redundancy of light chains. Peptide mapping of the epitopes recognized by these Fab indicated that three of the anti-Tat Fab were directed to the functional domain between amino acid residues 22-33 of the Tat molecule, and that binding was inhibited by reduction of this cysteine-rich region with dithiothreitol. The anti-Rev Fab were directed to sites adjacent to the Rev basic nucleolar localization sequence (residues 52-64) and to the Rev activation domain (residues 75-88). Binding constants were of a similar order to that of an anti-Rev single-chain Fv fragment (SFv) used successfully for intracellular immunization, and as such intracellular effects with the human anti-Tat and anti-Rev Fab are not precluded. These newly described human antibody fragments to HIV-1 regulatory proteins may be critical moieties for gene therapeutic protocols, to control HIV-1 replication in human cells.

Human monoclonal Fab fragments from a combinatorial library prepared from an individual with a low serum titer to a virus.

An IgG1k lambda Fab library was generated on the surface of phage beginning with bone marrow RNA from a healthy 22-year-old human donor. The donor had been immunized to measles in his early childhood but had only a low serum titer to a measles antigen preparation. The resulting library of approximately 10(7) clones was panned against the measles antigen preparation and three positive Fab-producing clones identified by ELISA. One of the Fabs was found to be specific to measles and to bind with high apparent affinity (10(8) M-1). The other two bind with lower affinity and show marked cross-reactivity with a number of other antigens. They possess heavy chains derived, with extensive somatic modification, from the single member gene family VH6. The study indicates that both high-affinity specific antibodies and lower affinity polyreactive antibodies can be derived from the library approach under appropriate conditions.

Molecular profile of an antibody response to HIV-1 as probed by combinatorial libraries.

A large number (33) of human Fab fragments reacting with HIV-1 surface glycoprotein gp120 have been generated by selection from a combinatorial IgG1 kappa library displayed on the surface of phage. The library was prepared from a long term asymptomatic HIV-seropositive donor. Analysis of the sequences from these Fabs shows the heavy chains can be placed in groups, many of which contain intraclonal variants, almost certainly corresponding to chains used in vivo. Further variants can be accessed via chain shuffling experiments in which a given light chain is recombined with a library of heavy chains. Heavy chain promiscuity, i.e. the ability of heavy chains to pair with different light chains with retention of antigen binding, is dependent on the particular heavy chain considered and probably excludes the identification of in vivo light chain partners. The antibodies examined here are primarily to the CD4 binding site on gp120 and broadly reflect the serum profile of the donor. The antibodies show evidence of extensive somatic modification indicative of an antigen-driven response. The heavy chain CDR3 regions of the antibodies show a remarkably conserved extended length. A number also show strong sequence conservation in CDR3 against a background of considerable diversity in the rest of the VH gene supporting a central role for this region in antigen recognition.

Fc gamma RII, but not erythropoietin or GM-CSF, mediates calcium mobilization in fetal hemopoietic blast cells.

A proportion of fetal liver hemopoietic blast cells express Fc gamma RII, and addition of the anti-Fc gamma RII monoclonal antibody CIKM5 induces a rise in calcium in these cells in suspension. Although these cells are thus capable of mobilizing intracellular calcium in response to surface receptor mediated events, neither granulocyte-macrophage colony-stimulating factor (GM-CSF) nor erythropoietin produced detectable changes in intracellular calcium ion concentration in these cells.

Differentiation-linked activation of the respiratory burst in a monocytic cell line (U937) via Fc gamma RII. A study of activation pathways and their regulation.

Activation of the respiratory burst in the monocytic cell line U937 by cross-linking human 40-kDa FcR for IgG (Fc gamma RII) with the IgG1 mAb, CIKM5, is dependent on the maturation state of the cell. Addition of anti-Fc gamma RII to undifferentiated cells does not activate the respiratory burst but differentiation with human rIFN-gamma (200 U/ml) for 13 to 15 days results in maximal stimulation by this agonist, with half-maximal responses in cells incubated for 10 to 12 days. During maturation the development of responsiveness to cross-linking Fc gamma RII occurs later than the development of responsiveness to the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (maximal responses at 7 to 9 days), or the chemotactic peptide FMLP (half-maximal responses at 7 to 9 days). The late development of maximal Fc gamma RII responses is not associated with either increased Fc gamma RII expression, enhanced calcium mobilization induced by anti-Fc gamma RII, changes in protein kinase C activity (PKC) or a switch in PKC isotype expression. Activation of the respiratory burst via Fc gamma RII may not be mediated by activation of PKC as the kinase inhibitors 1-(5-isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride and N-[2-(methylamino)ethyl]-5-isoquinolinesulfonamide dihydrochloride inhibited the Fc gamma RII response by less than 20% at concentrations which inhibit the 12-O-tetradecanoylphorbol-13-acetate-induced respiratory burst by more than 80%. IFN-gamma U937 cells did not metabolize incorporated arachidonate into eicosanoids when stimulated with anti-Fc gamma RII, suggesting that eicosanoids do not mediate activation of the respiratory burst, and this was confirmed by the lack of inhibition by the specific 5'-lipoxygenase and glutathione S-transferase inhibitor, piriprost, and the cyclo-oxygenase inhibitor, indomethacin. In addition there was no significant release of radiolabeled arachidonate in response to anti-Fc gamma RII. The response to anti-Fc gamma RII is inhibited by pertussis toxin, suggesting that signal transduction is via a GTP-binding protein. Agents that elevate intracellular cAMP increased the magnitude of the cAMP transients stimulated by anti-Fc gamma RII and also inhibited the respiratory burst. FMLP responses showed a similar pattern of sensitivity to this range of inhibitors, suggesting that both Fc gamma RII and FMLP receptor share common regulatory mechanisms. However, the termination of the respiratory burst activated via Fc gamma RII and FMLP receptor is independently regulated, in that after FMLP-induced activation there is no subsequent inhibition of the Fc gamma RII-mediated response and vice versa.

Fc gamma RII-mediated superoxide production by phagocytes is augmented by GM-CSF without a change in Fc gamma RII expression.

Freshly purified neutrophils and monocytes respond to multiple cross-linking of Fc gamma RII with the IgG1 monoclonal antibody, CIKM5, with a rapid rise in Ca(2+)i, but not with a respiratory burst, although superoxide is generated by these cells when stimulated with the chemotactic peptide, FMLP, or phorbol ester (TPA). Incubation in vitro for 30-60 min at 37 degrees C in medium + 0.1% FCS had no effect on the neutrophil superoxide response to CIKM5 but induced a weak monocyte response in 11/13 experiments. However, incubation with rhGM-CSF (10 ng/ml) under similar conditions induced a neutrophil respiratory burst in response to cross-linking Fc gamma RII in 12/14 experiments and enhanced the monocyte response by 181%. GM-CSF also enhanced the response of neutrophils and monocytes to FMLP by 308% and 165%, respectively. The response to TPA was not significantly enhanced by GM-CSF. rhIFN-gamma (100 mu/ml) was ineffective as a priming agent for all agonists tested in short-term incubations but augmented the monocyte response to CIKM5 after 5 d exposure in vitro. Whilst GM-CSF induced neutrophil superoxide production in response to cross-linking Fc gamma RII, there was no concomitant change in Fc gamma RII expression either in in vitro studies of neutrophils from healthy individuals or in in vivo studies of patients receiving GM-CSF. Stimulation of unprimed neutrophils with CIKM5 induced a rapid transient increase in intracellular calcium levels to 181% of resting levels. However, incubation with GM-CSF did not further augment the calcium transients above the stimulated level. The mechanism by which GM-CSF induces an enhanced respiratory burst in response to cross-linking of Fc gamma RII remains to be elucidated, but is not related to receptor expression or increases in receptor mediated calcium mobilization.

Phenotypic characterization of non-haemopoietic small cell tumours of childhood with monoclonal antibodies to leucocytes, epithelial cells and cytoskeletal proteins.

Recently, great interest has been shown in the histological identification of small cell tumours of childhood--nephroblastoma (Wilms' tumour), neuroblastoma, rhabdomyosarcoma and Ewing's sarcoma--using immunohistochemical methods. However, several antigens operationally specific for leucocyte typing in blood and marrow are also expressed on cells of epithelial and neural origin. We undertook phenotypic characterization of 17 non-haemopoietic small cell tumours of childhood using a panel of 30 monoclonal antibodies to leucocyte, epithelial and cytoskeletal antigens using a sensitive alkaline phosphatase-anti-alkaline phosphatase technique on cryostat sections of fresh tumour. Our results demonstrated frequent expression of the leucocyte-associated antigens CD10 (CALLA), CD9 (p24) and CDw32 (FcRII) in these small cell tumours and occasional expression of MHC class II (HLA-DR) and HNK-1 antigens. However, the leucocyte-associated antigens CD45 (leucocyte common), CD22 (pan B-cell), CD11b (C3bi receptor), CD15 (Lewisx) or CDw42 (platelet gp Ib) were not detected on any tumour. Aberrant expression of desmin, neurofilament and UJ13A antigen was found in nephroblastoma and of epithelial-associated markers (CIBr17 and 43-9F) in neuroblastoma. Our results also demonstrated broad reactivity in frozen section with two monoclonal antibodies specific for melanoma (NKI/C-3) or epithelial cells (OM-1) in paraffin sections. Hence, it is necessary to include monoclonal antibodies to CD45 and pan-epithelial antigens, e.g. LP34 (cytokeratin) or HEA125 for the precise immunohistochemical identification of small round cell malignancies of childhood.

Mechanism of human monocyte activation via the 40-kDa Fc receptor for IgG.

It is shown that a mAb specific for the human 40-kDa FcR (FcRII) leads to activation of human monocytic cells but that extensive cross-linking of the receptor is required. Calcium mobilization can be induced in immature monocytic cells (undifferentiated U937 cells) and peripheral blood monocytes with an intact IgG1 anti-FcRII antibody (CIKM5) but not by F(ab')2 fragments of this antibody. The intact antibody can bind in a tripartite manner by its two F(ab') sites and its Fc-binding site whereas the F(ab')2 fragments of this antibody can only bind in a divalent fashion. A rise in intracellular free calcium ion concentration occurs when F(ab')2 fragments are cross-linked with F(ab')2 anti-mouse Ig indicating that more extensive cross-linking of FcRII is required rather than an obligatory requirement for an Fc-FcRII interaction. Calcium mobilization in response to intact or cross-linked F(ab')2 fragments of CIKM5 is associated with superoxide production only in IFN-gamma-primed peripheral blood monocytes and IFN-gamma differentiated U937 cells indicating that the activation signal produced via FcRII is inadequate to fully stimulate non-"primed" cells. A second mAb reactive with FcRII (2E1) does not cause calcium mobilization in monocytes or U937 cells, and partially blocks the effects of CIKM5. 2E1 also blocks CIKM5 superoxide production in IFN-gamma-primed monocytes and differentiated U937 cells. This may be explained in part by the fact that 2E1 is an IgG2a antibody and can only participate in bipartite binding with FcRII. When 2E1 is cross-linked with F(ab')2 anti-mouse Ig there is a small calcium response. This does not cause superoxide generation in IFN-primed monocytes but does do so in IFN-gamma differentiated U937 cells. FcRII is also expressed on granulocytes and some B cells but the effects of cross-linking the receptor on these cells differ from those seen in monocytes.

Prospective study of childhood acute lymphocytic leukemia: hematologic, immunologic, and cytogenetic correlations.

We studied the karyotype in 81 consecutively diagnosed children with acute lymphocytic leukemia (ALL) treated at one institution on a randomized treatment protocol. In 75 patients (93%), a morphological cytogenetic result was obtained, and 57 (65%) were successfully G-banded. Of the 75 patients, 46 (61%) showed abnormal chromosomes, mainly hyperdiploidy and pseudodiploidy, and 29 had no detectable abnormality. Our findings confirmed that the karyotype has prognostic significance. Duration of complete remission was 93% at 42 months for patients with high hyperdiploidy (greater than 50). For patients with an apparently normal karyotype, it was 58%; and for patients with structural abnormalities it was 15%. The significance of these findings was confirmed by multivariate analysis, which showed age and karyotype to be the most important determinants of duration of remission.

Diagnosis of granulocytic sarcoma facilitated by monoclonal antibodies.

A 36 year old woman presented with a nasopharyngeal tumour which was diagnosed and treated as diffuse large cell lymphoma. Twelve mth later the patient developed acute myeloid leukemia. At this stage, the original biopsies were reviewed and considered in retrospect to be granulocytic sarcoma on the basis of staining for chloracetate esterase and lysozyme. She achieved and maintained marrow and peripheral blood remission with chemotherapy, but developed several cutaneous nodules and 2 breast lumps. One breast lump was excised and was found, by the use of monoclonal antibodies, to carry myeloid markers. Thus monoclonal antibodies provided additional confirmatory evidence for the diagnosis of granulocytic sarcoma.

The phenotype of mantle zone lymphoma studied by monoclonal antibodies.

The clinical, pathological and immunological features of a case of mantle zone lymphoma are described. The patient presented at the age of 16 with a history of painless enlargement of the inguinal lymph nodes, biopsy of which revealed a nodular small cell lymphoma. During the course of 11 yr he was treated with total nodal irradiation, splenectomy and combination chemotherapy at different times. A recent lymph node biopsy reviewed along with the previous node biopsies was diagnosed as mantle zone lymphoma. At this stage, the immunological studies showed that the neoplastic lymphoid cells had characteristic markers of mantle zone lymphocytes. He is asymptomatic with mild generalized lymphadenopathy 11 yr after the initial diagnosis. This case illustrates the diagnostic and therapeutic problems which may be encountered. Detailed immunological marker studies with an extended panel of monoclonal antibodies are described.

Monoclonal anti-T-cell antibodies react with circulating myeloid leukemia cells and normal tissue macrophages.

Rabbit and monoclonal antibodies to human myeloid leukemia cells, monocytic leukemia cells and human thymocytes have shown the existence of common T-cell/myeloid/monocyte antigens. For this reason, the specificity of a series of monoclonal antibodies to human T-cells (OKT 1, 3, 4, 5, 6, 8, 9, 10; and NA1/34) was tested by immunofluorescence (cytofluorograph) and complement-mediated cytotoxicity against human myeloid leukemia and normal blood cells and leukemic cell lines. In addition, an immunohistological analysis of the specificity of OKT4, 9.3, Leu 3a, OKT3 and NA1/34 antibodies was performed using normal lymphoid tissues and a sensitive immunoperoxidase technique. Normal human peripheral blood mononuclear cells reacted with OKT3 ("pan T-cell", mean 54%), OKT4 ("helper T-cell", mean 35%) and OKT 5/8 ("suppressor T-cell", mean 18%) as previously reported. However, OKT3 reacted with the cell lines K562 (myeloid), RC2a and THP-1 (monocytoid) and U937 (macrophage) as well as with cells from 9/65 myeloid leukemia patients. OKT4 reacted with the cell lines HL60 (promyelocyte), RC2a and U937 and also with cells from 6/60 myeloid leukemia patients. OKT5 reacted with the cell lines K562 and THP-1. OKT1 ("pan T-cell") reacted with THP-1 and with myeloid and monocytic leukemia samples (5/32) as did OKT6 ("cortical thymocyte") (3/32). OKT10 ("common thymocyte") reacted with a range of leukemia cell lines (B-cell, pre- B-cell and macrophage) as well as 7/21 myeloid leukemia samples. In tissue sections Leu 3a, (9.3 and OKT4 to a lesser extent), stained paracortical lymphocytes, plus subcapsular and medullary macrophages, and dendritic cells present within the paracortex.(ABSTRACT TRUNCATED AT 250 WORDS)

Diagnostic information derived from immunotyping 1000 patients with leukemia and lymphoma.

Immunotyping analysis has been performed on cells from 1000 patients with leukemia or lymphoma using 14 markers of cell lineage or differentiation stage over a 4 yr period. Results showed considerable heterogeneity of cell type among these groups of malignant diseases not readily apparent by morphology and histochemistry. Immunotyping contributed additional diagnostic information in 30% of patients and should be a routine procedure in 8 disease categories. These are: acute leukemia, cell type not determined; acute lymphoblastic leukemia; lymphocytosis of undetermined origin; chronic myeloid leukemia-terminal blast crisis; chronic lymphocytic leukemia; malignant lymphoma-leukemic phase; Sézary syndrome, mycosis fungoides and chronic T cell leukemia; malignant lymphoma and lymphadenopathy- ? lymphoma. Immunotyping provided information on cell lineage and differentiation stage of major leukemic cell populations. Abnormal monoclonal proliferations of B lymphocytes and the presence of primitive cells amongst normally mature tissue cells were identified. Disturbances in normal lymphoid and monocytic cell populations in blood, marrow or tissues could also be demonstrated. Many of the reagents used in this period are now replaced by monoclonal antibody reagents to human lineage and differentiation antigens. These are expected to increase diagnostic usefulness of these techniques.

A rapid and simple radiometric assay for thymidine phosphorylase of human peripheral blood cells.

A radiometric method is described for measuring thymidine phosphorylase activity in human peripheral blood cells. The substrates [14C]thymidine or [14C]thymine are converted to the base or deoxynucleoside, respectively, and two alternative chromatographic methods to isolate the products of the reaction have been employed. With the described methods the specific activity for thymidine phosphorylase in human lymphocytes is 0.21 +/- 0.08; monocytes 0.21 +/- 0.16 and granulocytes 0.17 +/- 0.02 mu mol.h-1.mg-1 protein. For human T- or null lymphoblasts, thymidine phosphorylase activity was found to be approximately 10% of that of B lymphoblasts.

Classification of childhood acute lymphocytic leukaemia using rabbit antisera to leukaemia cells and lymphoblastoid cell lines.

Leukaemic cells from seventeen untreated acute lymphocytic leukaemia (A.L.L.) patients have been typed at presentation with heteroantisera prepared in rabbits by complement mediated cytotoxicity. Reactivity with antisera has suggested a preliminary grouping of patients, in this study, into Null, pre-T (post-thymic T-cell precursor) and T-cell subtypes. The leukaemic cells have also been classified with the established markers, E-rosette receptor, surface immunoglobulin and C3 receptor as well as total white cell count at presentation. Anti NALM-1 (Null-lymphoblastoid cell line) serum reacted with cells from all childhood A.L.L. patients tested but could be made specific for cALL antigen of the common Null form of A.L.L. by further absorption or dilution. Antisera to membrane fractions from cells of high white cell count Null-A.L.L. patients reacted with cells of these patients as well as T-A.L.L. patients. Anti MOLT-4 (adult T-A.L.L. derived lymphoblastoid cell line) serum reacted only with cells of T-A.L.L. patients.