Differential transient and long-term expression of DNA sequences introduced into T-lymphocyte lines

Interleukin 2 transcription factors as molecular targets of cAMP inhibition: delayed inhibition kinetics and combinatorial transcription roles

Elevation of cAMP can cause gene-specific inhibition of interleukin 2 (IL-2) expression. To investigate the mechanism of this effect, we have combined electrophoretic mobility shift assays and in vivo genomic footprinting to assess both the availability of putative IL-2 transcription factors in forskolin-treated cells and the functional capacity of these factors to engage their sites in vivo. All observed effects of forskolin depended upon protein kinase A, for they were blocked by introduction of a dominant negative mutant subunit of protein kinase A. In the EL4.E1 cell line, we report specific inhibitory effects of cAMP elevation both on NF-kappa B/Rel family factors binding at -200 bp, and on a novel, biochemically distinct "TGGGC" factor binding at -225 bp with respect to the IL-2 transcriptional start site. Neither NF-AT nor AP-1 binding activities are detectably inhibited in gel mobility shift assays. Elevation of cAMP inhibits NF-kappa B activity with delayed kinetics in association with a delayed inhibition of IL-2 RNA accumulation. Activation of cells in the presence of forskolin prevents the maintenance of stable protein-DNA interactions in vivo, not only at the NF-kappa B and TGGGC sites of the IL-2 enhancer, but also at the NF-AT, AP-1, and other sites. This result, and similar results in cyclosporin A-treated cells, imply that individual IL-2 transcription factors cannot stably bind their target sequences in vivo without coengagement of all other distinct factors at neighboring sites. It is proposed that nonhierarchical, cooperative enhancement of binding is a structural basis of combinatorial transcription factor action at the IL-2 locus.

A transcriptional switch between the Pig-1 and Sgs-4 genes of Drosophila melanogaster

Pig-1 and Sgs-4 are a pair of closely linked and divergently transcribed Drosophila melanogaster genes, which are both expressed in larval salivary glands but at different times during development. While Sgs-4 is expressed at high levels only at the end of the third instar, Pig-1 exhibits a major peak of expression during late second and early third instar. Thus, Pig-1 expression declines as Sgs-4 expression is induced. In this paper, we show that three adjacent elements located within the short region between these genes can account for the switch from Pig-1 to Sgs-4 expression. A 170-bp segment acts as an enhancer to direct Sgs-4 expression in late-third-instar salivary glands. A 64-bp sequence located just upstream from the enhancer can modify its temporal specificity so that it works throughout the third instar. Expression induced at mid-third instar by a combination of these two elements can be repressed by a negative regulatory sequence located still further upstream. We present evidence suggesting that the changing interactions between these regulatory elements and the Sgs-4 and Pig-1 promoters lead to the correct pattern of expression of the two genes.

A transcriptional switch between the Pig-1 and Sgs-4 genes of Drosophila melanogaster

Pig-1 and Sgs-4 are a pair of closely linked and divergently transcribed Drosophila melanogaster genes, which are both expressed in larval salivary glands but at different times during development. While Sgs-4 is expressed at high levels only at the end of the third instar, Pig-1 exhibits a major peak of expression during late second and early third instar. Thus, Pig-1 expression declines as Sgs-4 expression is induced. In this paper, we show that three adjacent elements located within the short region between these genes can account for the switch from Pig-1 to Sgs-4 expression. A 170-bp segment acts as an enhancer to direct Sgs-4 expression in late-third-instar salivary glands. A 64-bp sequence located just upstream from the enhancer can modify its temporal specificity so that it works throughout the third instar. Expression induced at mid-third instar by a combination of these two elements can be repressed by a negative regulatory sequence located still further upstream. We present evidence suggesting that the changing interactions between these regulatory elements and the Sgs-4 and Pig-1 promoters lead to the correct pattern of expression of the two genes.

In vitro transfection of fresh thymocytes and T cells shows subset-specific expression of viral promoters

We describe conditions under which exogenous DNA templates can be introduced for transient expression into primary murine T lymphocytes. T cells at various stages of development, including concanavalin A-activated splenic T cells, immature pre-T cells, and even small cortical thymocytes, could be successfully transfected. A variety of model DNA constructs were compared in which different viral promoter regions were used to drive expression of the chloramphenicol acetyltransferase (CAT) reporter gene. All showed enhanced expression in cells that had been acutely stimulated with the Ca2+ ionophore A23187 and phorbol ester as chemical proxies for T-cell receptor-mediated signals. In addition, splenocytes but not thymocytes required prior treatment with a mitogen and interleukin-2 in order to express these constructs, implying that even postmitotic thymocytes may be held in a quasiactivated state. A most striking result was the finding that the viral regulatory sequences in the Rous sarcoma virus long terminal repeat and the simian virus 40 early region were subject to sharply differential regulation, with a rank order that changed depending on the developmental stage of the T cells. The most immature thymic blasts and several lymphoma cell lines expressed the pRSV-Cat and pSV2-Cat constructs similarly, but cortical thymocytes exhibited a strong preference for pSV2-Cat. Splenic concanavalin A-stimulated blasts, on the other hand, slightly preferred pRSV-Cat, a tendency which became exaggerated in factor-dependent T-cell lines. The ratio of pRSV-Cat to pSV2-Cat expression varied according to cell type by as much as 500-fold. These results argue against a trivial linkage of promoter preference to cell cycle status but instead provide evidence that activation of T cells at distinct stages of differentiation results in the expression of different ensembles of nuclear regulatory proteins. In contrast to the simian virus 40 and Rous sarcoma virus promoter regions, the long terminal repeats of the retroviruses mink cell focus-forming virus and Akv were expressed well in all primary T-lineage cells. Thus, they represent excellent model promoters for engineering developmental stage-independent expression of exogenous genes in murine T cells.

In vitro transfection of fresh thymocytes and T cells shows subset-specific expression of viral promoters

We describe conditions under which exogenous DNA templates can be introduced for transient expression into primary murine T lymphocytes. T cells at various stages of development, including concanavalin A-activated splenic T cells, immature pre-T cells, and even small cortical thymocytes, could be successfully transfected. A variety of model DNA constructs were compared in which different viral promoter regions were used to drive expression of the chloramphenicol acetyltransferase (CAT) reporter gene. All showed enhanced expression in cells that had been acutely stimulated with the Ca2+ ionophore A23187 and phorbol ester as chemical proxies for T-cell receptor-mediated signals. In addition, splenocytes but not thymocytes required prior treatment with a mitogen and interleukin-2 in order to express these constructs, implying that even postmitotic thymocytes may be held in a quasiactivated state. A most striking result was the finding that the viral regulatory sequences in the Rous sarcoma virus long terminal repeat and the simian virus 40 early region were subject to sharply differential regulation, with a rank order that changed depending on the developmental stage of the T cells. The most immature thymic blasts and several lymphoma cell lines expressed the pRSV-Cat and pSV2-Cat constructs similarly, but cortical thymocytes exhibited a strong preference for pSV2-Cat. Splenic concanavalin A-stimulated blasts, on the other hand, slightly preferred pRSV-Cat, a tendency which became exaggerated in factor-dependent T-cell lines. The ratio of pRSV-Cat to pSV2-Cat expression varied according to cell type by as much as 500-fold. These results argue against a trivial linkage of promoter preference to cell cycle status but instead provide evidence that activation of T cells at distinct stages of differentiation results in the expression of different ensembles of nuclear regulatory proteins. In contrast to the simian virus 40 and Rous sarcoma virus promoter regions, the long terminal repeats of the retroviruses mink cell focus-forming virus and Akv were expressed well in all primary T-lineage cells. Thus, they represent excellent model promoters for engineering developmental stage-independent expression of exogenous genes in murine T cells.

High level transient gene expression in human lymphoid cells by SV40 large T antigen boost

High level transient gene expression in lymphoid cells has always been challenging because of the difficulty to efficiently transfect such cells. This has precluded any attempt to clone cDNA encoding proteins by means of their specific biological function in lymphoid cells. We have developed a very efficient transient eukaryotic expression system analogous to the well-known expression system in COS cells. Firefly luciferase and human CD2 genes were used as reporter genes and cloned into the eukaryotic shuttle vector pCDM8 which contains the strong cytomegalovirus promoter and the SV40 origin of replication for autonomous plasmid replication in permissive host cells that express the large SV40 T Antigen. Co-transfection of the reporter plasmids together with an SV40 T Ag expressing plasmid resulted in the several fold amplification of either the Luc activity or the cell surface expression of the CD2 marker in a transient assay. The level of amplification was dependent on the strength of the promoter used to drive the SV40 T Ag expression and was correlated with the extent of autonomous replication of the reporter plasmid in transfected cells. This highly efficient transient gene expression by SV40 T Ag boost was suitable to several human cell lines, making this system of general interest for expression cloning strategies or other gene transfer application that need high level expression.

Factors influencing transient expression in cytotoxic T cells following DEAE dextran-mediated gene transfer

A number of transfection protocols have been tested for the introduction of exogenous DNA into cytotoxic T cells. These included electroporation, lipofection, calcium phosphate coprecipitation, polybrene-assisted gene transfer, and DEAE dextran-mediated transfer. Only the latter gave significant and reproducible transfection efficiencies coupled with low toxicity. The DEAE dextran protocol was optimized for the transfection of a transcription reporter construct pRSVcat into a cloned cytotoxic cell line. Among the parameters investigated were cell density, amount of input DNA, concentration of DEAE dextran, DNA adsorption time, temperature, use of permeabilization and expression facilitators, and recovery time. The optimized protocol was then used to demonstrate the presence of cis-acting regulatory regions in the 5'-flanking sequences of two cytotoxic cell-specific serine protease genes and, in addition, was shown to be applicable to other cloned T-cell lines.

Indiscriminate activity from the B19 parvovirus p6 promoter in nonpermissive cells

B19 parvovirus is absolutely tropic for human erythroid progenitor cells. Among the untested mechanisms underlying this tropism is the possibility of cell-specific positive regulation of the promoter in permissive cells. Using the bacterial chloramphenicol acetyltransferase and firefly luciferase reporter genes, we detected strong activity from the B19 P6 promoter in transfected nonpermissive cells. Very high-level expression was seen in a T lymphoblastoid cell line, CEM. No transcriptional enhancement occurred in an erythropoietin-dependent semipermissive cell line. A putative second B19 promoter at map unit 44 (P44) was nonfunctional and unable to confer tissue specificity. Thus, tropism is unlikely to be regulated at the level of transcriptional initiation from either the P6 or P44 promoter.

Interleukin-1 synergy with phosphoinositide pathway agonists for induction of interleukin-2 gene expression: molecular basis of costimulation

The macrophage-derived cytokine interleukin-1 (IL-1) can provide a second signal with antigen to elicit production of interleukin-2 (IL-2) by helper T cells. The pathway(s) involved remains controversial, with protein kinase C and cyclic AMP (cAMP) invoked as possible second messengers. In the murine thymoma EL4.E1, IL-1 could synergize with the phosphoinositide pathway, because the cells made higher levels of IL-2 in the presence of IL-1 than could be induced by phorbol ester plus calcium ionophore alone. IL-1 is unlikely to act through a sustained increase in cAMP in these cells because it did not raise cAMP levels detectably and because IL-1 and forskolin had opposite effects on IL-2 gene expression. Inducible expression of a transfected reporter gene linked to a cloned fragment of the murine IL-2 gene promoter was initially increased by IL-1 costimulation, implying that IL-1 can increase the rate of transcription of IL-2. The minimal promoter elements required for iL-1 responsiveness were located within 321 bp of the IL-2 RNA cap site, and further upstream sequences to -2800 did not modify this response. IL-1 costimulation resulted in enhanced activity of both an inducible NF-kappa B-like factor and one of two distinct AP-1-like factors that bind to IL-2 regulatory sequences. Neither was induced, however, by IL-1 alone. Another AP-1-like factor and NFAT-1, while inducible in other cell types, were expressed constitutively in the EL4.E1 cells and were unaffected by IL-1. These results are discussed in terms of the combinatorial logic of IL-2 gene expression.

Interleukin-1 synergy with phosphoinositide pathway agonists for induction of interleukin-2 gene expression: molecular basis of costimulation

The macrophage-derived cytokine interleukin-1 (IL-1) can provide a second signal with antigen to elicit production of interleukin-2 (IL-2) by helper T cells. The pathway(s) involved remains controversial, with protein kinase C and cyclic AMP (cAMP) invoked as possible second messengers. In the murine thymoma EL4.E1, IL-1 could synergize with the phosphoinositide pathway, because the cells made higher levels of IL-2 in the presence of IL-1 than could be induced by phorbol ester plus calcium ionophore alone. IL-1 is unlikely to act through a sustained increase in cAMP in these cells because it did not raise cAMP levels detectably and because IL-1 and forskolin had opposite effects on IL-2 gene expression. Inducible expression of a transfected reporter gene linked to a cloned fragment of the murine IL-2 gene promoter was initially increased by IL-1 costimulation, implying that IL-1 can increase the rate of transcription of IL-2. The minimal promoter elements required for iL-1 responsiveness were located within 321 bp of the IL-2 RNA cap site, and further upstream sequences to -2800 did not modify this response. IL-1 costimulation resulted in enhanced activity of both an inducible NF-kappa B-like factor and one of two distinct AP-1-like factors that bind to IL-2 regulatory sequences. Neither was induced, however, by IL-1 alone. Another AP-1-like factor and NFAT-1, while inducible in other cell types, were expressed constitutively in the EL4.E1 cells and were unaffected by IL-1. These results are discussed in terms of the combinatorial logic of IL-2 gene expression.

Regulatory anatomy of the murine interleukin-2 gene

We have cloned the mouse IL2 gene and sequenced 2800 bp of 5' flanking DNA. Comparison to the previously reported human sequence revealed extensive identity (approximately 86%) between the two genes from +1 to -580 with additional small islands of homology further upstream. Proximal sites which have been shown to be important in regulation of the human IL2 gene are well conserved in sequence and location. Transfection experiments using hybrid gene constructs containing varying lengths of the mouse 5' flanking DNA linked to a CAT reporter gene have demonstrated the presence of several novel positive and negative regulatory elements. One negative regulatory region lying between -750 and -1000 consists primarily of alternating purines and pyrimidines and is absent from the human gene. The conserved region from -321 and -578, an upstream segment from -1219 to -1332, and another region of approximately 450 bp from -1449 to -1890, which contained a well-conserved sequence of 60 bp, were each associated with enhanced levels of expression. We found no evidence for intragenic or downstream enhancer elements in this gene. All the elements identified affect only the magnitude of the inducible response, for no region when deleted had the effect of altering either the need for induction, the kinetics of stimulation, or the cell-type specificity of expression. Deletion studies suggest a strong requirement for NFAT binding even in the presence of extensive 5' flanking sequence. Therefore we conclude that IL2 gene expression is controlled primarily through a central TH1-specific signaling pathway, which acts through proximal elements, while distal cis-elements exert a secondary modulating effect.

Energy dependence of DNA-mediated gene transfer and expression

When cells are transfected with calcium phosphate-precipitated DNA, most of the DNA is internalized through active endocytosis but a small pool is postulated to be internalized by passive diffusion. The relative importance of these two DNA pools in gene transfer and expression has never been examined. We now show that mouse Ltk- fibroblasts internalize greater than 90% of radioactively labeled, calcium phosphate-precipitated DNA by a temperature- and energy-dependent process but a small pool of 2-3% of radioactively labeled DNA may be internalized even at 4 degrees C or under energy depletion conditions. The contribution of these two pools of DNA to gene expression was examined by transfecting Ltk- cells with DNA from a plasmid encoding marker for both transient expression (human growth hormone) and stable integration (neomycin resistance). When cells were depleted of ATP during transfection, transient expression of human growth hormone was reduced to only 2% of control values at day 4 posttransfection. Under the same condition, stable expression of neomycin resistance also was reduced to less than 1% of the control. Therefore, these data showed that only the pool of calcium phosphate-precipitated DNA internalized through an energy-dependent process is ultimately responsible for transient and stable expression of exogenous genes.

Variation in enzymatic transient gene expression assays

We examined causes for high variability in data from enzymatic transient gene expression assays. Our results strongly suggest that variation in transfection efficiency is the major cause of data variation and can seriously compromise valid interpretation of data. We compared averaging data from multiple transfections and cotransfection of a second reporter gene as methods for correcting for variation in transfection efficiency. We found that transfection efficiency can be so highly variable that neither method necessarily overcomes the resulting bias in data. Depending upon the degree in variation in transfection efficiency, a combination of the two methods may be advisable. The need to normalize data for transfection efficiency is dependent upon the difference in strengths of promoters being tested and the relative variability of the transfection method used. We also show that the level of reporter gene expression between transfection experiments performed on different days can vary by more than 10-fold.

Structure and expression of the Lyt-3a gene of C.AKR mice

The mouse Lyt-3a gene, which encodes the Lyt-3.1 T-cell surface alloantigen of the C.AKR strain, has been cloned, and the nucleotide sequence of its exons and more than 2 kb of 5' flanking sequence have been determined. The gene extends over approximately 16 kb of DNA and consists of six exons encoding leader, leader plus V-like domain, membrane-proximal, transmembrane, and cytoplasmic domains. The only difference between the coding region of the Lyt-3a gene and the cDNA sequences reported for Lyt-3b (Nakauchi et al. 1987. Panaccio et al. 1987) is at position 77 of the mature protein where Lyt-3a encodes serine and Lyt-3b encodes arginine. This substitution must therefore be the basis for the serological distinction between the Lyt-3.1 and Lyt-3.2 alloantigens. Potential TATA and CAAT sequences, two Sp1 protein binding sites, two extended repeats of the dinucleotide, CA, a number of short inverted repeats, and an inverted segment of the mouse B1 repetitive sequence are found 5' to the Lyt-3a gene. Two consensus poly-A addition signals and a complete copy of the mouse B1 sequence are found 3' to the gene. Both B1-related regions are flanked by short direct repeats suggesting that they arose by an insertional mechanism. Cotransfection of the Lyt-3a gene together with a cloned Lyt-2a gene resulted in expression of both Lyt-2 and Lyt-3.1 on the surface of Ltk- and BW5147 cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Electroporation of mammalian cells with a firefly luciferase expression plasmid: kinetics of transient expression differ markedly among cell types

Electroporation in combination with the luciferase reporter system provides a rapid and sensitive means of determining transient expression from plasmid constructs in various mammalian cells. Conditions are described allowing efficient electroporation of several cell types with a commercial electroporation device. Measurements on extracts prepared at various times after electroporation showed a rapid rise in luciferase activity for up to 12 hr, which was followed, in certain cell types, by a rapid decline. With such cells, determinations made after 48-72 hr, as is conventional, would greatly underestimate expression of the luciferase reporter gene. Therefore, to observe optimum activity, the kinetics of expression must be determined for each cell type.

Human P45IA1 upstream regulatory sequences expressing the chloramphenicol acetyltransferase gene. Effect of Ha-MSV enhancer and comparison of transient with stable transformation assays

Upstream sequences of the human P450IA1 gene were inserted into a promoterless expression vector (pSVO-cat) containing the chloramphenicol acetyltransferase (CAT) gene, with and without the Harvey murine sarcoma virus (Ha-MSV) core enhancer, and either plasmid was transfected into human breast carcinoma MCF-7 and MDA-231 and mouse hepatoma Hepa-1 cell lines. In most instances constitutive and inducible CAT activities in the transient CAT expression assay were similar (within 3-fold) to those in the stable transformation CAT assay (selection of G418-resistant colonies following co-transfection with pSV2-neo). In the case of Ha-MSV-containing constructs stably integrated in the two human breast cancer lines, however, CAT expression was more than two orders of magnitude greater than that transiently expressed in these cells. Since the major difference between these two assays is plasmid copy number, these data suggest the presence of limiting amounts of tissue-specific positive-control enhancer-binding factor(s) in the breast carcinoma cell lines.