The CD20/alphaCD20 'suicide' system: novel vectors with improved safety and expression profiles and efficient elimination of CD20-transgenic T cells.

Adoptive transfer of T lymphocytes is an attractive strategy for many experimental treatment strategies for cancer. Unfortunately, manipulated T cells could be responsible for serious adverse events. Retroviral CD20-transduced T cells may be able to control these unwanted effects. CD20-positive cells are sensitive to rituximab (RTX), a monoclonal antibody specific for CD20. This permits their selective elimination in vivo in case of adverse events. To this end, a system is required that permits efficient and safe transduction of donor T cells and effective elimination of CD20-positive T cells. We constructed different CD20-encoding retroviral vectors and investigated the impact of inclusion of the woodchuck post-transcriptional regulatory element (WPRE) and the chicken hypersensitivity site 4 insulator elements on the levels, homogeneity and stability of CD20 expression. Importantly, inclusion of either WPRE or insulator elements in the retroviral vector resulted in a dramatic improvement in the stability of CD20 expression. The insulator element also led to a much more homogeneous level of CD20 expression. We also show the efficient elimination of the CD20-transgenic T cells via RTX by different effector mechanisms. In conclusion, we have constructed CD20-encoding retroviral vectors with improved efficiency and safety profiles, which can be used as a suicide strategy.

T cell receptor-transgenic primary T cells as a tool for discovery of leukaemia-associated antigens.

Identification of a broad array of leukaemia-associated antigens is a crucial step towards immunotherapy of haematological malignancies. However, it is frequently hampered by the decrease of proliferative potential and functional activity of T cell clones used for screening procedures. Transfer of the genes encoding the T cell receptor (TCR) alpha and beta chains of leukaemia-specific clones into primary T cells may help to circumvent this obstacle. In this study, transfer of two minor histocompatibility antigen (minor H antigen)-specific TCRs was performed and the feasibility of the use of TCR-transgenic T cells for identification of minor H antigens through cDNA library screening was investigated. We found that TCR-transgenic cells acquired the specificity of the original clones and matched their sensitivity. Moreover, the higher scale of cytokine-production by TCR-transgenic T cells permits the detection of either small amounts of antigen-positive cells or cells expressing low amounts of an antigen. When applied in equal numbers, TCR-transgenic T cells and the original T cell clones produced similar results in the screening of a cDNA library. However, the use of increased numbers of TCR-transgenic T cells allowed detection of minute amounts of antigen, barely discernible by the T cell clone. In conclusion, TCR-transfer generates a large amount of functional antigen-specific cells suitable for screening of cDNA expression libraries for identification of cognate antigens.

Human primary T lymphocytes have a low capacity to amplify MLV-based amphotropic RCR and the virions produced are largely noninfectious.

The presence of replication-competent retrovirus (RCR) in retroviral-based gene therapy products poses a potential safety risk for patients. Therefore, RCR testing of clinical gene therapy products and monitoring of patients enrolled in gene therapy trials is required to assure viral safety. The requirement to test ex vivo-transduced cells originates from the presumed amplification of adventitious RCR during the transduction procedure. However, data on the capacity of different cell types to do so are lacking. In this study, we sought to analyze the amplification potential of primary human T lymphocytes after infection with amphotropic MLV-based RCR. The total number of viral particles produced after 1 or 2 weeks was measured by a quantitative 4070A env-specific RT-PCR assay. The fraction of infectious replication-competent viral particles was analyzed in the PG-4 S+L- assay. From this study, we conclude that the total number of viral particles RCR produced by T lymphocytes is 2-4 logs lower than the number produced by NIH-3T3 cells. Surprisingly, less than 1% of the viral particles produced by primary T lymphocytes appeared to be infectious, while nearly all virions produced by NIH-3T3 were. We conclude that primary human T lymphocytes are low producers of MLV-based amphotropic RCR.

Human chromosome 21 determines growth factor dependence in human/mouse B-cell hybridomas.

Interleukin 6 (IL-6) serves as a growth factor for mouse plasmacytomas. As a model for IL-6-mediated growth of plasmacytomas, we study IL-6-dependent B-cell hybridomas, which can be generated through fusion of B lymphocytes with a plasmacytoma cell line, e.g., SP2/0. In the present report, we have investigated the peculiar behavior of B-cell hybridomas with respect to IL-6 dependence. We demonstrate that although newly generated hybridomas are IL-6 dependent, many hybridomas lose this dependency at frequencies as high as 50%, shortly after fusion. We speculated that the loss of IL-6-dependent growth is due to the well-known chromosomal instability of B-cell hybridomas. Consequently, loss of IL-6 dependence is the result of loss of a specific chromosome(s). This model implies the existence of an "IL-6 dependency" gene, the loss of which makes hybridomas capable of proliferating in the absence of IL-6. Because SP2/0 is IL-6 independent, the IL-6-dependent phenotype of B-cell hybridomas, and hence the IL-6 dependency gene, must be derived from the B lymphocyte. We have tested this model by generating human/mouse B-cell hybridomas through fusion of human B lymphocytes with SP2/0. We then analyzed the human chromosome content of 10 IL-6-dependent and 14 IL-6-independent subclones. From that analysis we concluded that the presence of human chromosome 21 correlated with IL-6 dependence. This correlation was confirmed by microcell fusion experiments in which a single copy of chromosome 21 was introduced into IL-6-independent hybridomas, resulting in reconstitution of the IL-6-dependent phenotype. We therefore conclude that chromosome 21 carries an IL-6 dependency gene.

Leucine-58 in the putative 5th helical region of human interleukin (IL)-6 is important for activation of the IL-6 signal transducer, gp130.

A model of the tertiary structure of human IL-6, derived from the crystal-structure of granulocyte-colony stimulating factor, reveals a 5th helical region in the loop between the first and second alpha-helix. To investigate the importance of this region for biological activity of IL-6, residues Glu-52, Ser-53, Ser-54, Lys-55, Glu-56, Leu-58, and Glu-60 were individually replaced by alanine. IL-6.Leu-58Ala displayed a 5-fold reduced biological activity on the IL-6 responsive human cell lines XG-1 and A375. This reduction in bioactivity was shown to be due to a decreased capacity of the mutant protein to trigger IL-6 receptor-alpha-chain-dependent binding to the IL-6 signal transducer, gp130.

Development of an interleukin (IL) 6 receptor antagonist that inhibits IL-6-dependent growth of human myeloma cells.

The pleiotropic cytokine interleukin 6 (IL-6) plays a role in the pathogenesis of various diseases, such as multiple myeloma, autoimmune and inflammatory diseases and osteoporosis. Therefore, specific inhibitors of IL-6 may have clinical applications. We previously succeeded in developing receptor antagonists of IL-6 that antagonized wild-type IL-6 activity on the human Epstein-Barr virus (EBV)-transformed B cell line CESS and the human hepatoma cell line HepG2. However, these proteins still had agonistic activity on the human myeloma cell line XG-1. We here report the construction of a novel mutant protein of IL-6 in which two different mutations are combined that individually disrupt the association of the IL-6/IL-6 receptor (R) alpha complex with the signaltransducing "beta" chain, gp130, but leave the binding of IL-6 to IL-6R alpha intact. The resulting mutant protein (with substitutions of residues Gln160 to Glu, Thr163 to Pro, and replacement of human residues Lys42-Ala57 with the corresponding residues of mouse IL-6) was inactive on XG-1 cells and weakly antagonized wild-type IL-6 activity on these cells. By introducing two additional substitutions (Phe171Leu, Ser177Arg), the affinity of the mutant protein for IL-6R alpha was increased fivefold, rendering it capable of completely inhibiting wild-type IL-6 activity on XG-1 cells. Moreover, this mutant also antagonized the activity of IL-6, but not that of leukemia inhibitory factor, oncostatin M, or GM-CSF on the human erythroleukemia cell line TF-1, demonstrating its specificity for IL-6. These data demonstrate the feasibility of developing specific IL-6R antagonists. The availability of such antagonists may offer an approach to specifically inhibit IL-6 activity in vivo.

Peripheral human CD5+ and CD5- B cells may express somatically mutated VH5- and VH6-encoded IgM receptors.

Previous studies have indicated that a sizable fraction of adult human peripheral blood B cells may express IgM receptors encoded by somatically mutated V regions. From these studies it was uniquely associated with the peripheral blood B cell compartment, was uniquely associated with the peripheral blood B cell compartment, associated with particular VH gene segments and/or B cell subpopulations. We have addressed these issues by analyzing > 80 VH5 and VH6-encoded mu transcripts from unseparated peripheral blood, tonsil, and spleen B cells, as well as from B cells separated on the basis of CD5 Ag expression. The results demonstrate that somatically mutated VH5 and VH6 regions are ubiquitously expressed in IgM-bearing B cells in all peripheral adult human lymphoid organs, and that the occurrence of somatic mutations does not segregate with either CD5+ or CD5- B cell populations. The distribution and nature of mutations, as well as the occurrence of clonally related but divergent transcripts suggests that at least some of the mutations were selected by Ag.

Molecular analysis of VH and VL regions expressed in IgG-bearing chronic lymphocytic leukemia (CLL): further evidence that CLL is a heterogeneous group of tumors.

We report the heavy (H) and light (L) chain variable (V) region sequences of cDNAs encoding the Ig receptor of two cases of CD5+ IgG-bearing CLL P87 and P103. In both CLL cases the H chain was encoded by members of the VH3 gene family. The L chain expressed by P87 belonged to the V lambda IV subgroup, whereas P103 used a member of the V kappa III subgroup. The VH3.P87 gene differed by only three nucleotides from 38P1, a VH3 gene previously cloned from a fetal liver cDNA library. Nucleotide sequence analysis demonstrated that the V kappa III.P103 gene differed by seven nucleotides from its most homologous germline counterpart, the Humkv325 gene, a highly conserved gene frequently expressed in IgM-bearing CLL. The nucleotide sequences of VH3.P103 and V lambda IV.P87 could not be reliably matched with reported germline V genes. The analysis of multiple independently obtained VH and VL cDNA clones from each tumor showed a lack of intraclonal diversification. The data show that V regions expressed in isotype-switched CD5+ CLL may be either in/near germline configuration or somatically mutated. Furthermore, these tumors, like their IgM-bearing counterparts, do not seem to undergo intraclonal diversification.

The majority of human tonsillar CD5+ B cells express somatically mutated V kappa 4 genes.

We have fractionated human tonsillar B cells on the basis of CD5 expression and determined the nucleotide sequences of immunoglobin light chain variable (V) regions encoded by the single member of the V chi 4 gene family in both CD5+ and CD5- populations. The majority of cDNA from both CD5+ and CD5- B cells populations harbored somatic mutations. Thus, human tonsillar CD5+ B cells, unlike their murine counterparts, are capable of activating their somatic hypermutation mechanism, resulting in the accumulation of somatic mutation in the VL regions.

Deletion mapping of Ig VH gene segments expressed in human CD5 B cell lines. JH proximity is not the sole determinant of the restricted fetal VH gene repertoire.

VH gene segments expressed in a panel of monoclonal human CD5 B cell lines have been positioned on the IgH locus by deletion mapping. The analysis yielded a relative order of VH fragments of the VH2, VH4, VH5, and VH6 gene families that was consistent with, and provided a further refinement of existing maps of the human IgH locus. We demonstrate that four of six VH gene segments expressed in the CD5 B cell lines map > 500 kb from the cluster of JH segments. Two of the gene segments, positioned at approximately 850 kb (58p2) and approximately 500 kb (1-9III) from the JH segments, respectively, belong to the previously identified small cohort of second trimester fetal VH gene segments. The data show that JH proximity is not the sole determinant of restricted VH gene utilization in early human ontogeny.

Molecular approaches to the study of human B-cell and (auto)antibody repertoire generation and selection.

We have shown that the restricted repertoire of VH genes expressed in second trimester human fetal liver is not solely determined by JH proximity. Furthermore, by following the fate of two VH gene segments in different B-cell repertoires, we have provided evidence that multiple factors contribute to the frequency with which individual VH genes are utilized. We found that the repertoire of adult blood IgM-bearing B cells contains a high proportion of B lymphocytes that express extensively mutated VH genes. Finally, we show that somatically-mutated variants of particular VH and VL genes that, in germline configuration, are frequently found in the early B-cell repertoire and in natural autoantibodies, encode pathogenic IgG autoantibodies characteristic of human SLE. These VH and VL genes harbor all the characteristics of an antigen-driven B-cell activation and selection process.

Sequence analysis of small subunit ribosomal RNA genes and its use for detection and identification of Leishmania parasites.

The sequence of the most variable part of the small subunit ribosomal RNA (SSU rRNA) gene, comprising 800 bases, was analysed for 9 Leishmania taxa and compared with those of Trypanosoma brucei, Trypanosoma cruzi and Crithidia fasciculata. Considerable differences were observed between the sequence of the Leishmania taxa on the one hand and those of Crithidia and Trypanosoma on the other. Amongst the Leishmania taxa only a few point mutations were found, all located within 2 sequence blocks in the central part of the SSU rRNA gene, which are unique for Kinetoplastida. These unique sequences were used for the development of kinetoplastid-specific probes and a Leishmania-specific PCR assay of high sensitivity (less than 10 parasites could be detected). Based on the observed point-mutations an identification of the Leishmania parasites, according to complex, could be achieved by direct sequencing, restriction fragment analysis or single-stranded conformation polymorphism of the PCR-generated fragments.

Expression of members of the immunoglobulin VH3 gene families is not restricted at the level of individual genes in human chronic lymphocytic leukemia.

Previous studies on the repertoire of Ig VH genes utilized in the malignant cells of chronic lymphocytic leukemia (CLL) have suggested a non-random expression pattern. In particular, individual genes from the VH1, VH5, and VH6 gene families have been frequently found in CLL. With regard to other VH gene families, including the large VH3 family which is expressed in greater than 50% of CLL cases, it is unknown whether CLL cells utilize either a broad or a restricted repertoire of VH genes. In the present paper, we analyzed the VH genes expressed in a collection of 11 CLL cases. The results of these experiments demonstrate that there is no apparent restriction in the usage of individual members of the VH3 gene families in CLL.

Antibody specificity and immunoglobulin VH gene utilization of human monoclonal CD5+ B cell lines.

Human B lymphocytes that bear the CD5 antigen are relatively abundant in early ontogeny and comprise a small fraction of the B cell population in adults. The CD5 B cell subset has attracted much attention because of its possible involvement in autoimmune disease and certain B cell malignancies. To begin to understand the role of CD5 B cells in disease processes, we have generated a panel of ten human monoclonal B cell lines selected for expression of the CD5 antigen. These cell lines were obtained by Epstein-Barr virus transformation of B lymphocytes isolated from the spleen, liver and bone marrow of a 19-week-old fetus, from cord blood and from peripheral blood of healthy volunteers. In addition, one cell line was isolated from the spleen of a patient with chronic lymphocytic leukemia. Here, we describe the antibody and immunoglobulin VH gene repertoire of this panel of CD5 B cell lines. The results of these experiments show that (a) some but not all CD5 B cell lines secrete polyreactive antibodies that bind to a variety of self- and xenoantigens and (b) members of the small VH4, VH5 and VH6 gene families are overrepresented in this panel of cell lines. Nucleotide sequence analysis revealed the expression of VH gene elements that have been previously reported in the preimmune B cell repertoire, in CD5 B cell tumors and in polyreactive antibodies.

Down modulation of fibronectin messenger RNA in metastasizing rat prostatic cancer cells revealed by differential hybridization analysis.

To identify genes whose expression is down modulated in the process of metastasis, gene expression was analyzed in cell lines derived from Dunning R-3327 rat prostatic tumor sublines. A complementary DNA (cDNA) library from the anaplastic nonmetastasizing subline AT-1 was used for a differential hybridization analysis, using probes derived from mRNAs of the AT-1 and the metastasizing MAT-LyLu subline. In this way 14 cDNA clones were isolated representing 6 differentially expressed genes. The expression levels in a panel of tumor sublines measured with these cDNA clones were tested for correlation with the anaplastic non-metastasizing phenotype. One cDNA clone, designated pSE-1, whose expression was high in all tested sublines with that phenotype, appeared to represent the gene for fibronectin. To further investigate the down modulation of this gene, we studied its expression in AT-2 (anaplastic, nonmetastasizing tumor) and lines derived therefrom that exhibited a high metastatic potential after transfection with the v-Ha-ras oncogene. In the genetically manipulated metastasizing tumor sublines, fibronectin mRNA levels were approximately 4- to 8-fold lowered compared to the nonmetastasizing parental AT-2 line.