c-fos expression interferes with thymus development in transgenic mice

To study the function of the proto-oncogene c-fos in hematopoietic tissues, transgenic mice were generated that express c-fos from the H2-Kb promoter in several organs. These H2-c-fos mice have enlarged spleens and hyperplastic thymuses containing an increased number of thymic epithelial cells. The exogenous c-fos expression specifically affects T cell development in the thymus, thereby increasing the fraction of mature thymocytes. Results obtained with bone marrow radiation chimeras suggest that the altered distribution of T cell subsets is not a direct effect of c-fos expression within the T cell lineage. No changes in the proportion of hematopoietic cell lineages are seen in the spleen, and these mice do not develop lymphoid malignancies. B and T cell function, however, is impaired, and H2-c-fos mice are immune deficient. It appears that c-fos specifically stimulates the proliferation of thymic epithelial cells, and may thus indirectly affect T cell development.     

Inducible transformation of cells from transgenic mice expressing SV40 under lac operon control.

If it were possible to clone in vitro cells of any type, at any stage of differentiation, from an extensively characterized animal such as the mouse, many areas of cell biology would benefit. Indeed, it would be even more helpful if these cells could subsequently be restored to their normal in vivo phenotype whenever required. Here, we describe a step on the pathway to such an idealized "clonable" mouse. In principle, it seeks to link a "universal" transforming agent to a regulatory system that is relatively simple, yet quite foreign to the mouse. A plasmid containing the bacterial lac operator/promoter region linked to the SV40 large T antigen and a vector containing the lac repressor that can be expressed in mammalian cells were coinjected into fertilized mouse oocytes utilizing the standard techniques for generating transgenic mice. Two progeny were obtained that express large T antigen in the presence, but not the absence, of the nonmetabolizable lac inducer, isopropyl-beta-thio-D-galactoside. This report characterizes fibroblast cell lines established from these transgenics that are readily transformed in vitro with isopropyl-beta-thio-D-galactoside. A significant proportion of the cells are restored to their "normal" (nontransformed phenotype) when isopropyl-beta-thio-D-galactoside is removed.     

Langerhans cells are strongly reduced in the skin of transgenic mice overexpressing follistatin in the epidermis

Activins are members of the transforming growth factor-beta (TGF-beta) family and are important for skin morphogenesis and wound healing. TGF-beta1 is necessary for the population of the epidermis with Langerhans cells (LC). However, a role for activin in LC biology is not known. To address this question, we analyzed skin from transgenic mice overexpressing the activin antagonist follistatin in the epidermis. Using immunofluorescence, we observed a striking decrease in the number of LC in the epidermis of transgenic mice in comparison to wild-type mice. Nevertheless, these LC expressed normal levels of major histocompatibility complex (MHC)-class II and Langerin/ CD207 in situ. In explant cultures of whole ear skin the number of dendritic cells (DC), which migrated into the culture medium, was reduced. This reduction was even more pronounced in cultures of epidermal sheets. Virtually all emigrated cutaneous DC displayed typical morphology with cytoplasmic "veils", showed translocation of MHC-class II to the surface membrane, and expressed the maturation marker 2A1. Thus, cutaneous DC from transgenic mice seemed to mature normally. These results demonstrate that overexpression of follistatin in the epidermis affects LC trafficking but not maturation and suggest a novel role of the follistatin-binding partner activin in LC biology.     

Characterization of target cells for MCF viruses in AKR mice

The recombinant (MCF) class of murine leukemia virus appears to play an important role in lymphomagenesis in AKR and other mice. Although much effort has been extended in characterizing MCF viruses, relatively little is known about the cells they infect. I examined what cells were targets in AKR mice for both lymphomagenic and nonlymphomagenic MCF viruses. Lymphomagenic MCF viruses of thymic origin (AKR-247 and C58L1) were found to infect and replicate selectively in immature lymphocytes only present in thymic cortex, whereas nonlymphomagenic MCF viruses of splenic origin (C58v-1-C77 and C58v-2-C45) selectively infected and replicated in cells that appeared to B lymphocytes. Virus-binding studies suggested that neither T- nor B-lymphocyte tropisms were determined by selective attachment of virus to the respective cells. These findings demonstrate that in contrast with ecotropic viruses, which can infect many types of cells in the mouse, specific cellular tropisms can exist for MCF viruses, and that MCF infection, and therefore oncogenicity, is closely linked to cellular differentiation.     

Effect of target gene CpG content on spontaneous mutation in transgenic mice

Transgenic mutation assays utilizing bacterial target genes display a high frequency of spontaneous mutation at CpG sequences. This is believed to result from the fact that: (1) the prokaryotic genes currently being used as transgenic mutation targets have a high CpG content and (2) these sequences are methylated by mammalian cells to produce 5-methylcytosine (5MC), a known promutagenic base. To study the effect of CpG content on the frequency and type of spontaneous mutation, we have synthesized an analogue of the bacterial lacI target gene (mrkII) that contains a reduced number of CpG sequences. This gene was inserted into a lambda vector and used to construct trangenic mice that undergo vector rescue from genomic DNA upon in vitro packaging. Results on spontaneous mutation frequency and spectrum have been collected and compared to those observed at the lacI gene in Big Blue™ transgenic mice. Spontaneous mutations at the mrkII gene occurred at a frequency in the mid-10⁻⁵ range and were predominantly base pair substitutions, similar to results seen in Big Blue™. However, mrkII mutations were distributed toward the carboxyl end of the gene instead of the bias toward the amino terminus seen in lacI. Unexpectedly, 23% of the spontaneous mrkII mutations were GC → AT transitions at CpG sequences (compared to 32% in lacI), despite the reduction in CpG number from 95 in lacI to only 13 in mrkII. Nine of the CpG bases undergoing transition mutations in mrkII have not been recorded previously as spontaneous sites in Big Blue™. Therefore, substantial reduction of the number of CpG sequences in the lacI transgene did not significantly reduce the rate of spontaneous mutation or alter the contribution of CpG-related events. This suggests that other factors are also operating to establish frequency and composition of spontaneous mutations in transgenic targets.     

Uteroglobin promoter-targeted c-MYC expression in transgenic mice cause hyperplasia of Clara cells and malignant transformation of T-lymphoblasts and tubular epithelial cells

To investigate the influence of the proto-oncogene c-MYC on tumor development in different epithelial tissues which secrete Clara Cell Secretory Protein (uteroglobin, UG), transgenic mouse lines were established expressing the human c-MYC proto-oncogene under the control of the rabbit UG-promoter. These mice expressed the c-MYC transgene in Clara cells and other UG expressing tissues like uterus and prostate. In the bronchioalveolar epithelium of the lung hyperplasias developed originating from Clara cells. Surprisingly, transgenics most frequently developed T-lymphoblastic lymphomas, a polycystic kidney phenotype and renal cell carcinoma derived from tubular epithelial cells, which are both tissues that had so far not been known to express UG. Immunohistological studies in UG/MYC transgenics and in a transgenic line (UG/eGFP) expressing Green Fluorescent Protein confirmed that the uteroglobin promoter is not only active in Clara cells, but also in tubular epithelial cells of the kidney and in lymphatic tissue. The UG/MYC transgenics will be useful to investigate the biochemical mechanisms underlying the development of carcinomas and the oncogenic properties of c-MYC in epithelial cells of various tissues.     

Papillomavirus-like particles are an effective platform for amyloid-beta immunization in rabbits and transgenic mice

Immunization with amyloid-beta (Abeta) prevents the deposition of Abeta in the brain and memory deficits in transgenic mouse models of Alzheimer's disease (AD), opening the possibility for immunotherapy of AD in humans. Unfortunately, the first human trial of Abeta vaccination was complicated, in a small number of vaccinees, by cell-mediated meningoencephalitis. To develop an Abeta vaccine that lacks the potential to induce autoimmune encephalitis, we have generated papillomavirus-like particles (VLP) that display 1-9 aa of Abeta protein repetitively on the viral capsid surface (Abeta-VLP). This Abeta peptide was chosen because it contains a functional B cell epitope, but lacks known T cell epitopes. Rabbit and mouse vaccinations with Abeta-VLP were well tolerated and induced high-titer autoAb against Abeta, that inhibited effectively assembly of Abeta(1-42) peptides into neurotoxic fibrils in vitro. Following Abeta-VLP immunizations of APP/presenilin 1 transgenic mice, a model for human AD, we observed trends for reduced Abeta deposits in the brain and increased numbers of activated microglia. Furthermore, Abeta-VLP vaccinated mice also showed increased levels of Abeta in plasma, suggesting efflux from the brain into the vascular compartment. These results indicate that the Abeta-VLP vaccine induces an effective humoral immune response to Abeta and may thus form a basis to develop a safe and efficient immunotherapy for human AD.     

Cre reconstitution allows for DNA recombination selectively in dual-marker-expressing cells in transgenic mice

Cre/LoxP-based DNA recombination has been used to introduce desired DNA rearrangements in various organisms, having for example, greatly assisted genetic analyses in mice. For most applications, single gene promoters are used to drive Cre production for conditional gene activation/inactivation or lineage-tracing experiments. Such a manipulation introduces Cre in all cells in which the utilized promoter is active. To overcome the limited selectivity of single promoters for cell-type-specific recombination, we have explored the ‘dual promoter combinatorial control’ of Cre activity, so that Cre activity could be restricted to cells that express dual protein markers. We efficiently reconstituted Cre activity from two modified, inactive Cre fragments. Cre re-association was greatly enhanced by fusing the Cre fragments separately to peptides that can form a tight antiparallel leucine zipper. The co-expressed Cre fusion fragments showed substantial activity in cultured cells. As proof of principle of the utility of this technique in vivo for manipulating genes specifically in dual-marker-positive cells, we expressed each inactive Cre fragments in transgenic mice via individual promoters. Result showed the effective reconstitution of Cre activates LoxP recombination in the co-expressing cells.     

Optimal artificial mini-introns for transgenic expression in the cells of mice and hamsters

Introns can frequently enhance transgene expression, and sometimes they are absolutely substantial. Based on an analysis of murine genes, in which mRNA does not have alternative splicing, a universal design of the efficiently spliced artificial introns of small sizes has been proposed. These introns are shown to be efficiently spliced in CHO cells from hamster ovaries. The proposed strategy can be used to include introns in cDNA, which would elevate the production of recombinant proteins in cell culture, as well as in transgenic animals.     

Haploid male germ cells show no susceptibility to transformation by simian virus 40 large tumour antigen in transgenic mice.

Cell-type specific tumorigenesis can be induced in transgenic mice by the directed expression of simian virus 40 (SV 40) large tumour antigen (TAg). In an attempt to determine the susceptibility of haploid male germ cells to neoplastic transformation by this oncogene, transgenic mice were generated that harboured a chimeric gene composed of the SV40 T antigen genes fused to the 2.3-kb 5' flanking sequences of the rat proacrosin gene. It was previously shown that this regulatory sequence is able specifically to direct the expression of CAT reporter gene in male germ cells with the onset of translation in early haploid male germ cells. The transgene showed regulated expression in male germ cells. Although T antigen immunostaining was detected specifically in spermatids, no testicular pathology was observed. This indicates that spermatids show no susceptibility to transformation by oncogene TAg. However, in about 10% of animals of two independent transgenic lines, we could find non-testicular tumours in abdomen with a sarcoma-like structure in advanced age which showed SV40 TAg expression.     

Gene-trap-based target site for cre-mediated transgenic insertion

There is an increasing need for tissue-specific gene expression regulatory elements to study normal and disease development in the mouse. However, the cloning and characterization of these elements are time-consuming and costly. Thus, there is a particular need to be able to identify gene expression patterns without having to clone the promoter elements. Gene-trap strategies identify expression patterns assigned for endogenous genes using reporters, such as LacZ (Gossler et al., 1989; Skarnes, 1990) or green fluorescent protein (GFP) (Ishida and Leder, 1999; Zheng and Hughes, 1999). The gene-trap vector randomly inserts into the genome and "steals" regulatory elements for the reporter. Here we describe an improved gene-trap strategy, which allows an efficient Cre recombinase-mediated insertion of any transgene into the trapped loci as a post-integrational modification and links the expression of the transgene to that of the reporter.     

Expression of threonine-biosynthetic genes in mammalian cells and transgenic mice

Threonine is a nutritionally essential amino acid (EAA) for the growth and development of humans and other nonruminant animals and must be provided in diets to sustain life. The aim of this study was to synthesize threonine in mammalian cells through transgenic techniques. To achieve this goal, we combined the genes involved in bacterial threonine biosynthesis pathways into a single open reading frame separated by self-cleaving peptides (2A) and then linked it into a transposon system (piggyBac). The plasmids pEF1a-IRES-GFP-E2F-his and pEF1a-IRES-GFP-M2F-his expressed Escherichia coli homoserine kinase and threonine synthase efficiently in mouse cells and enabled cells to synthesize threonine from homoserine. This biosynthetic pathway occurred with a low level of efficiency in transgenic mice. Three transgenic mice were identified by Southern blot from 72 newborn mice, raising the possibility that a high level of expression of these genes in mouse embryos might be lethal. The results indicated that it is feasible to synthesize threonine in animal cells using genetic engineering technology. Further work is required to improve the efficiency of this method for introducing genes into mammals. We propose that the transgenic technology provides a promising means to enhance the synthesis of nutritionally EAAs in farm animals and to eliminate or reduce supplementation of these nutrients in diets for livestock, poultry and fish.     

Visualization and ablation of phenylethanolamineN-methyltransferase producing cells in transgenic mice

We cloned and sequenced the mouse phenylethanolamineN-methyltransferase (PNMT) gene which encodes the enzyme that catalyses the conversion of norepinephrine to epinephrine. The ability of various length sequences flanking the mouse or human PNMT genes to direct expression of reporter genes in transgenic mice was examined. We show that 9 kb of 5 flanking sequences from the cloned mouse PNMT gene can direct expression of theEscherichia coli -galactosidase (lacZ) gene to predicted regions of the adrenal, eye can direct in the adult transgenic mouse. The transgene was also expressed during development, in the myelencephalon, adrenal medulla and dorsal root ganglia. PNMT-producing cells were ablated by expression of the diphtheria toxin (DT-A) gene driven by the human PNMT promoter, resulting in abnormalities in the adrenal medulla, eye and testis. The hPNMT_{8 kb}-DT-A line presents a model with which to examine the developmental ramifications of deletion of PNMT-producing cell populations from the adrenal medulla and retina.     

Generation of transgenic mice overexpressing EfnB2 in endothelial cells

Genetic studies have shown that ephrin-B2 and its cognate EphB4 receptor are necessary for normal embryonic angiogenesis. Moreover, there is overwhelming evidence that ephrin-B2 is involved in tumor vascularization, yet its role in adult angiogenesis has been difficult to track genetically. Here, we report the generation of transgenic mice that over-express EfnB2 specifically in endothelial cells (ECs). We show that exogenous expression of EfnB2 under the control of the Tie2 promoter/enhancer regions in ECs does not affect viability or growth of the transgenic animals. We further show that targeted expression of EfnB2 in ECs is not sufficient to rescue severe cardiovascular defects at mid-gestation stages but rescues early embryonic lethality associated with loss-of-function mutation in EfnB2. This mouse model will be useful to study the role of ephrin-B2 in physiological and pathological angiogenesis.     

Differentiation patterns of CD4/CD8 thymocyte subsets in cocultures of fetal thymus and lymphohemopoietic cells from c-fos transgenic and normal mice.

This study examined the involvement of c-fos protooncogene in thymocyte development from lymphohemopoietic T cell progenitors, within the thymic microenvironment. We first analyzed the thymocytes developing in vitro in the fetal thymus from the c-fos transgenic mice and found a high proportion of CD4+ single positive (SP) cells. We then seeded either fetal liver or bone marrow (BM) cells from normal donors onto lymphocyte-depleted fetal thymus explants of c-fos transgenic mice. The results showed an increased proportion of mature CD4+ SP and decreased CD4+CD8+ double positive (DP) cells. A similar pattern of CD4/CD8 thymocyte subsets was observed when either thymus or BM cells from c-fos transgenic mice developed within a normal thymic stroma. The kinetics of thymocyte development in organ culture (from Days 3 to 11) suggested that the SP cells obtained under these conditions may have bypassed the CD4+CD8+ DP phase. It appears that the altered pattern of thymocyte development manifested in adult c-fos transgenic mice can be induced by the early embryonic thymic stroma, and may also involve cells in the lymphohemopoietic tissues.