In vivo transposition of Minos, a Drosophila mobile element, in mammalian tissues.

Transposable elements have been used widely in the past 20 years for gene transfer and insertional mutagenesis in Drosophila. Transposon-based technology for gene manipulation and genomic analysis currently is being adopted for vertebrates. We tested the ability of Minos, a DNA transposon from Drosophila hydei, to transpose in mouse tissues. Two transgenic mouse lines were crossed, one expressing Minos transposase in lymphocytes under the control of the CD2 promoter/locus control region and another carrying a nonautonomous Minos transposon. Only mice containing both transgenes show excision of the transposon and transposition into new chromosomal sites in thymus and spleen cells. In addition, expression of Minos transposase in embryonic fibroblast cell lines derived from a transposon-carrying transgenic mouse resulted in excision of the transposon. These results are a first step toward a reversible insertional mutagenesis system in the mouse, opening the way to develop powerful technologies for functional genomic analysis in mammals.

Identification of thymus specific and developmentally regulated genes by an improved version of the mRNA differential display technique.

During embryogenesis in mouse, the thymus is seeded by waves of hematopoietic stem cells that provide the first peripheral T lymphocytes after birth. It is known that embryo thymocytes and adult thymocytes have different phenotypic and functional features. The identification of genes expressed in the thymus only during embryogenesis would help to understand the molecular basis underlying these characteristics. We used the mRNA differential display technique to compare gene expression between thymus and kidney from embryo (171/2 days) and adult mice. This technique is the method of choice for comparing gene expression because it is able to display rapidly and simultaneously the mRNA complement from several different types of cells. The major drawback of the method is that it leads to the cloning of many false positives and therefore needs a high throughput method to screen for the truly differentially expressed cDNAs. We combined advantages from previously described methods in order to develop a new version of the mRNA differential display technique that is fast, cheap, and reliable. Instead of oligo dT priming, we used random hexameres for the reverse transcription of total RNA and 10-mer primers for the amplification of internal parts of the cDNAs. We obtained reproducible and clean patterns of discrete bands. We were able to easily identify DNAs differentially amplified between embryo and adult tissues (embryo specific; E 58.73), between thymus and kidney (thymus specific; Thy 52.54), or between embryo and adult thymus (embryo thymus specific; E Thy 58.73) cDNA fragments. After reamplification, cloning, and sequencing of these DNA fragments, it appeared that in most cases, one band corresponded to a single DNA sequence. On a northern blot, each of these candidate genes recognized a transcript that is differentially expressed as expected. Thus, we report an optimized, reproducible, and fast mRNA differential display method that overcomes the usual problems met with the originally described technique or its reported modifications.

Induction of insulitis by glutamic acid decarboxylase peptide-specific and HLA-DQ8-restricted CD4(+) T cells from human DQ transgenic mice.

Insulin-dependent diabetes mellitus in humans is linked with specific HLA class II genes, e.g., HLA-DQA1*0301/ DQB1*0302 (DQ8). To investigate the roles of HLA-DQ8 molecules and glutamic acid decarboxylase (GAD) in disease development, we generated DQ8(+)/I-Abo transgenic mice expressing functional HLA-DQ8 molecules and devoid of endogenous mouse class II. DQ8(+)/I-Abo mice produced antigen-specific antibodies and formed germinal centers after immunization with GAD65 peptides. Two GAD peptide-specific (247-266 and 509-528), DQ8 restricted Th1 CD4(+) T cell lines, were generated from immunized DQ8(+)/I-Abo mice. They induced severe insulitis after adoptive transfer into transgene positive (but not negative) mice who were treated with a very low dose of streptozotocin that alone caused no apparent islet pathology. In addition to CD4, islet mRNA from these mice also showed expression of CD8, IFNgamma, TNFalpha, Fas, and Fas ligand. Our data suggest that a mild islet insult in the presence of HLA-DQ8 bearing antigen-presenting cells promotes infiltration of GAD peptide reactive T cells into the islet.

Susceptibility and resistance to antigen-induced apoptosis in the thymus of transgenic mice.

Injection of TCR transgenic mice with antigenic peptide results in the deletion of immature thymocytes expressing the transgenic TCR. We have analyzed this process in mice transgenic for a TCR (F5) that recognizes a peptide from the influenza nucleoprotein (NP68). To determine whether deletion of immature thymocytes is the result of specific recognition of the antigenic peptide by the thymocytes or mature T cell activation, bone marrow chimeric mice were generated using a mixture of cells from F5 transgenic and nontransgenic mice. Injection of these mice with antigenic peptide leads to the preferential depletion of F5 transgenic thymocytes, whereas nontransgenic thymocytes remain largely unaffected. Furthermore, exposure of F5 fetal thymic lobes to peptide leads to thymocyte deletion even though no mature single positive T cells are present at this stage. These data suggest that Ag-induced death of immature thymocytes is due to peptide-specific recognition, although activated mature T cells appear to potentiate such deletion. Further administration of antigenic peptide to F5 mice results in the appearance of double-positive thymocytes that are resistant to Ag or anti-CD3-induced apoptosis. These data suggest a change in the ability of the cells to signal through the TCR-CD3 complex, resembling the state of anergy induced in peripheral T cells following chronic exposure to Ag.

Transgene-encoded human CD2 acts in a dominant negative fashion to modify thymocyte selection signals in mice.

CD2 is a cell surface glycoprotein present on all T cells which has been shown to function as an adhesion and signaling molecule. Expressed early in T cell development, human CD2 (HCD2) has been suggested to play a role during thymopoiesis. However, the relevance of CD2 in T cell development has been called into question recently, as neither disruption of the CD2 gene nor anti-CD2 antibody treatment of fetal thymic organ cultures in mouse were shown to have any discernible consequences. We have expressed HCD2 at high levels in transgenic mice and found a profound effect of the transgene on thymocyte differentiation. Transgenic thymuses are considerably reduced in cell number as a consequence of increased apoptosis of double-positive (DP) thymocytes in the cortex. The remaining DP cells have up-regulated levels of T cell receptor (TCR) and are resistant to apoptosis mediated by administration of antigen. These effects are dependent on the cytoplasmic domain of HCD2, as mice expressing comparable levels of a tailless HCD2 transgene have a normal phenotype. The HCD2 cytoplasmic domain contains several regions of identity with mouse CD2 and can interact effciently with mouse intracellular signaling machinery. These results suggest there is considerable cross-talk between CD2 and TCR on developing thymocytes with consequences for the stimulation threshold of mature T cells.

Locus control region function and heterochromatin-induced position effect variegation.

Human CD2 locus control region (LCR) sequences are shown here to be essential for establishing an open chromatin configuration. Transgenic mice carrying an hCD2 mini-gene attached only to the 3' CD2 transcriptional enhancer exhibited variegated expression when the transgene integrated in the centromere. In contrast, mice carrying a transgene with additional 3' sequences showed no variegation even when the latter integrated in centromeric positions. This result suggests that LCRs operate by ensuring an open chromatin configuration and that a short region, with no enhancer activity, functions in the establishment, maintenance, or both of an open chromatin domain.

Tolerance in TCR/cognate antigen double-transgenic mice mediated by incomplete thymic deletion and peripheral receptor downregulation.

Influenza nucleoprotein (NP)-specific T-cell receptor transgenic mice (F5) were crossed with transgenic mice expressing the cognate antigenic protein under the control of the H-2Kb promoter. Double-transgenic mice show negative selection of thymocytes at the CD4+8+TCRlo to CD4+8+TCRhi transition stage. A few CD8+ T cells, however, escape clonal deletion, and in the peripheral lymphoid organs of these mice, they exhibit low levels of the transgenic receptor and upregulated levels of the CD44 memory marker. Such cells do not proliferate upon exposure to antigen stimulation in vivo or ex vivo, however, they can develop low but detectable levels of antigen-specific cytotoxic function after stimulation in vitro in the presence of IL-2.

Expression in vivo of CD45RA, CD45RB and CD44 on T cell receptor-transgenic CD8+ T cells following immunization.

We used mice transgenic for a major histocompatibility complex class I-restricted T cell receptor to study the changes of phenotype in vivo which follow priming by antigen of CD8 T cells. We show that following priming with peptide, CD44 on CD8 T cells is up-regulated. The change of phenotype was relatively stable, as primed CD8 cells isolated from thymectomized mice 6 weeks after priming still expressed increased levels of CD44. CD8 T cells in these mice are still responsive to peptide and could represent long-lived primed cells. No down-regulation in vivo of the CD45RA or CD45RB isoforms was found, indicating that there is a differential regulation of the expression of CD44 and CD45RB by activated CD8 transgenic T cells. These results contradict earlier studies in vitro which showed that CD8 T cells which have been primed earlier belong to the CD45RA- or CD45RB- subset.

Thymic selection and peptide-induced activation of T cell receptor-transgenic CD8 T cells in interleukin-2-deficient mice.

The requirement for interleukin-2 (IL-2) in repertoire selection and peripheral activation of CD8 T cells was tested in mice rendered IL-2 deficient by gene targeting and expressing a transgenic T cell receptor (TcR) (F5) specific for influenza nucleoprotein (NP) 366-374 + H-2Db. Positive selection of the transgenic F5 TcR into the CD8 compartment proceeded normally. Both in vivo and in vitro, the antigenic peptide induced depletion of immature thymocytes and proliferation of mature CD8 T cells regardless of the presence of an intact IL-2 gene. In contrast, cytotoxic T lymphocyte (CTL) activity was only generated by T cells from IL-2+ F5 transgenic mice. Exogenous IL-2 was able to fully restore the CTL response of IL-2-/- responder cells in vitro. Thus, both in vivo and in vitro, clonal expansion of CD8 T cells can proceed in the absence of IL-2, whereas in peptide-immunized F5 transgenic mice, induction of cytotoxic effector function is IL-2 dependent.

Functional commitment to helper T cell lineage precedes positive selection and is independent of T cell receptor MHC specificity.

Thymocyte differentiation proceeds from double positive CD4+CD8+ to single positive T cells. It has been proposed that this process occurs by an instructive or a stochastic mechanism. In this report, we show that in recombination-deficient mice (RAG-1-I-) constitutive expression of a CD8 transgene allows maturation of CD4+(CD8tg+) cells, which express mature levels of a transgenic class I-restricted T cell receptor, F5. Rescued F5+CD4+(CD8tg+) cells have equivalent levels of T cell receptor expression as CD8end+ cells, respond to cognate antigen and, upon stimulation, they exhibit a phenotype characteristic of CD4+ helper T cells. These data are consistent with a model of differentiation that predicts that thymocytes become functionally committed to a helper or cytotoxic lineage before the final step of positive selection and independently of MHC specificity of their T cell receptor.

Human CD4 restores normal T cell development and function in mice deficient in murine CD4.

The ability of a human coreceptor to function in mice was investigated by generating human CD4 (hCD4)-expressing transgenic mice on a mouse CD4-deficient (mCD4-/-) background. From developing thymocyte to matured T lymphocyte functions, hCD4 was shown to be physiologically active. By examining the expansion and deletion of specific V beta T cell families in mutated mice with and without hCD4, it was found that hCD4 can participate in positive and negative selection. Mature hCD4 single positive cells also were found in the periphery and they were shown to restore MHC class II-restricted alloreactive and antigen-specific T cell responses that were deficient in the mCD4 (-/-) mice. In addition, these hCD4 reconstituted mice can generate a secondary immunoglobulin G humoral response matching that of mCD4 wild-type mice. The fact that human CD4 is functional in mice and can be studied in the absence of murine CD4 should facilitate studies of human CD4 activity in general and human immunodeficiency virus 1 gp120-mediated pathogenesis in acquired immune deficiency syndrome specifically.

T cell activation results in physical modification of the mouse CD8 beta chain.

The T lymphocyte glycoprotein, CD8, is an essential component of the response of class I MHC-restricted T cells to Ag. CD8 is expressed on the surface of class I-restricted T cells as disulfide-bonded heterodimers and higher multimers of two distantly related polypeptides, alpha and beta. The CD8 alpha polypeptide, expressed in transfection studies as homodimers, is able to reproduce both the adhesive and stimulatory properties of CD8, leaving the function of the CD8 beta polypeptide unresolved. Herein we demonstrate that the CD8 beta polypeptide changes physically during T cell maturation and activation by reversibly altering its sialic acid content. These changes occur specifically on CD8 beta not -alpha, indicating that the primary role of the CD8 beta chain may be regulatory, influencing the physical structure of the CD8 complex, and suggesting a novel mechanism of controlling receptor/ligand interactions.

T cell unresponsiveness correlates with quantitative TCR levels in a transgenic model.

Two lines of transgenic mice were developed which differ in their level of expression of V beta 11. To determine the role of TCR density in tolerance induction, these mice were bred with I-E expressing mice and were investigated for tolerance induction. T cells expressing V beta 11 at high density are deleted by negative selection. T cells expressing < 10% of the normal TCR receptor density are not subject to negative selection and were not activated in vitro. There is a correlation in receptor density in both strains of mice with in vitro activation. These findings support the notion that there is a definable quantitative signal threshold which is critical for tolerance induction.

T cell deletion follows chronic antigen specific T cell activation in vivo.

Exposure of mice transgenic for a TCR (F5) to cognate peptide antigen results in thymic depletion of CD4+CD8+ cells and expansion and activation of peripheral CD8+ TCR(tg)+ T cells. In the thymus apoptotic DNA ladder is evident as early as 3 h after peptide injection. Long exposure of intact or thymectomized F5 TCR transgenic mice to peptide antigen leads to depletion of most of the peripheral CD8+ T cells bearing the F5 receptor, with the remaining cells having lower levels of transgenic TCR compared with non-treated animals. In the thymus of intact F5 TCR transgenic mice such continuous exposure to antigen results in the reappearance of CD4+CD8+ with lower levels of the transgenic receptor.

In vitro negative selection of alpha beta T cell receptor transgenic thymocytes by conditionally immortalized thymic cortical epithelial cell lines and dendritic cells.

We have established conditionally immortalized thymic cortical epithelial cell lines from transgenic mice carrying a temperature-sensitive SV40 large T antigen. One of these cell lines expresses cortical markers and produces IL-1 alpha, IL-6, IL-7, and TGF-beta 1. These cells express class I major histocompatibility complex (MHC) constitutively and class II MHC upon induction with IFN-gamma. The cells appear to have a normal class I antigen presenting pathway since messages for both peptide transporter genes (TAP1, TAP2) were detected. The ability of these cortical epithelial cells to present peptide antigen was compared to that of thymic dendritic cells. In suspension culture with alpha beta T cell receptor (TcR) transgenic thymocytes, these epithelial cells and dendritic cells (pre-pulsed with peptide cognate for the transgenic TcR) caused down-regulation of CD4, CD8, and TcR in an antigen dose-dependent and MHC-restricted manner. CD4dullCD8dull cells were taken as evidence for negative selection because these cells contained apoptotic DNA. Concentration of peptide required for negative selection of thymocytes was similar between dendritic cells and cortical epithelial cells. In contrast, alpha beta TcR transgenic spleen cells were activated only by dendritic cells but not by cortical epithelial cells.

Positive and negative selection in transgenic mice expressing a T-cell receptor specific for influenza nucleoprotein and endogenous superantigen.

A transgenic mouse was generated expressing on most (> 80%) of thymocytes and peripheral T cells a T-cell receptor isolated from a cytotoxic T-cell clone (F5). This clone is CD8+ and recognizes alpha alpha 366-374 of the nucleoprotein (NP 366-374) of influenza virus (A/NT/60/68), in the context of Class I MHC Db (Townsend et al., 1986). The receptor utilizes the V beta 11 and V alpha 4 gene segments for the beta chain and alpha chain, respectively (Palmer et al., 1989). The usage of V beta 11 makes this TcR reactive to Class II IE molecules and an endogenous ligand recently identified as a product of the endogenous mammary tumour viruses (Mtv) 8, 9, and 11 (Dyson et al., 1991). Here we report the development of F5 transgenic T cells and their function in mice of the appropriate MHC (C57BL/10 H-2b, IE-) or in mice expressing Class II MHC IE (e.g., CBA/Ca H-2k and BALB/c H-2d) and the endogenous Mtv ligands. Positive selection of CD8+ T cells expressing the V beta 11 is seen in C57BL/10 transgenic mice (H-2b). Peripheral T cells from these mice are capable of killing target cells in an antigen-dependent manner after a period of in vitro culture with IL-2. In the presence of Class II MHC IE molecules and the endogenous Mtv ligand, most of the single-positive cells carrying the transgenic T-cell receptor are absent in the thymus. Unexpectedly, CD8+ peripheral T-cells in these (H-2k or H-2d) F5 mice are predominantly V beta 11 positive and also have the capacity to kill targets in an antigen-dependent manner. This is true even following backcrossing of the F5 TcR transgene to H-2d scid/scid mice, in which functional rearrangement of endogenous TcR alpha- and beta-chain genes is impaired.

Thymic depletion and peripheral activation of class I major histocompatibility complex-restricted T cells by soluble peptide in T-cell receptor transgenic mice.

Injection of mice transgenic for a class I major histocompatibility complex-restricted T-cell receptor with a soluble peptide antigen from influenza virus nucleoprotein results in clonal depletion of double-positive immature thymocytes in the thymus and activation of mature T cells in the periphery, accompanied by a transient up-regulation of the T-cell receptor and CD3 and CD8 coreceptor molecules.

A self-reactive T cell population that is not subject to negative selection.

In male mice expressing a transgenic alpha beta TCR which recognizes a male antigen (HY), T cells which do not express normal levels of CD8 escape thymic deletion and appear in the periphery. These consist of two distinct populations, one which lacks expression of both CD4 and CD8, and one with low levels of CD8. Neither population has anti-HY reactivity, consistent with the known requirement of this TCR for CD8. We now describe the consequences of expression of both the anti-HY TCR transgene and a constitutive CD8.1 transgene on T cells of male mice. Peripheral T cells in these male 'double transgenic' mice express both the anti-HY TCR and normal levels of CD8, and can proliferate to male antigen in vitro. These cells do not express the endogenous allele of CD8 (CD8.2), suggesting that the increase in CD8 levels due to the CD8.1 transgene leads to the deletion of the CD8.2low population. In contrast, the CD8.1 transgene does not lead to the deletion of the CD8.2- population. This implies that, unlike the majority of alpha beta T cells, TCR+CD4-CD8- cells in TCR transgenic mice are not subject to deletion.

Deletion analysis of the human CD2 gene locus control region in transgenic mice.

Genomic sequences located at the 3' flanking region of the human CD2 gene confer high level tissue-specific, position-independent expression of the gene when introduced in the germ line of mice. In order to further characterize these sequences a range of deletions, from the 3' end were produced and transgenic mice were generated with the human CD2 (hCD2) gene linked to these deleted fragments. This allowed us to establish the minimum sequences necessary for the copy-dependent transgene expression. 2.1 kb or 1.5 kb of 3' flanking sequences linked to a hCD2 mini-gene is sufficient to allow T-cell specific, copy-dependent, integration-independent expression in transgenic mice. 1.1 kb of 3' sequences results in the gene being expressed in a T-cell specific manner, but copy-dependent, integration-independent expression was not observed in a small number of transgenic animals. 0.2 or 0.5 kb of 3' flanking sequences were insufficient to allow expression above the level previously found with a human CD2 gene which lacked 3' flanking sequences. We conclude that the Locus Control Region (LCR) effect is caused by 1.5 kb of flanking sequences immediately 3' to the polyadenylation signal of the gene.