Isolation of cellular DNA sequences that allow expression of adjacent genes

Gene trap mutagenesis in mice: new perspectives and tools in cancer research

The complete human DNA sequence of the human genome was published in 2004 and we entered the postgenomic era. However, many studies showed that gene function is much more complex than we expected, and that mutation of disease genes does not give any clue for molecular mechanisms for disease development. Since the first report on gene knockout mice in 1989, knockout mice have been shown to be a powerful tool for functional genomics and for the dissection of developmental processes in human diseases. In accordance with this successful application of knockout mice, three major mouse knockout programs are now underway worldwide, to mutate all protein-encoding genes in mouse embryonic stem cells using a combination of gene trapping and gene targeting. We developed the exchangeable gene trap method suitable for large scale mutagenesis in mice. In this method we can produce null mutation and post-insertional modification, enabling replacement of the marker gene with a gene of interest and conditional knockout. We herein discuss the effect of this gene-driven type approach for cancer research, especially for finding the genes that are related to cancer, but are paid little attention in hypothesis-driven cancer research.

Promoter Crosstalk Effects on Gene Expression

Closely spaced transcription signals formally assigned to different, neighboring genes can functionally interact both in their authentic genomic context as well as in engineered transgene constellations. To describe these promoter crosstalk effects quantitatively and qualitatively, we used various combinations of inducible and constitutive expression signals linked in cis. Our results demonstrate that such interactions can be bidirectional, making it difficult to unambiguously assign a particular promoter element exclusively to a specific gene. We show that especially for inducible promoters, crosstalk effects can cause a substantial distortion in the expression of proximal genes, challenging established strategies in generating transgenic animal models and tissue culture systems. Furthermore, these findings provide guidelines for the design of transgenic transcription units that, while refractory to interactions with their chromosomal environment, leave the expression programs of neighboring genes largely untouched.

The Surf-6 gene of the mouse surfeit locus encodes a novel nucleolar protein

The Surfeit locus contains the tightest cluster of mammalian genes so far described. The five Surfeit genes (Surf-1 to -5) that have been previously isolated and characterized do not share any DNA or amino acid sequence homology. These Surfeit genes appear to be housekeeping genes, with the Surf-3 gene encoding the 1.7a ribosomal protein and the Surf-4 gene encoding an integral membrane protein most likely associated with the endoplasmic reticulum. In this work, we have isolated the Surf-6 gene, a sixth member of the Surfeit locus. The Surf-6 gene contains four exons spanning a genomic region of 14 kb and specifies a mRNA of 2,571 bases. Surf-6 has features common to housekeeping genes because its transcript is present in every tissue tested, its 5' end is associated with a CpG-rich island, and its promoter does not contain a canonical TATA box. The Surf-6 long open reading frame encodes a novel highly basic polypeptide of 355 amino acids (28% Arg and Lys). By immunofluorescence and immunoblot analyses, the Surf-6 protein has been found to be located in the nucleolus and by immunocytochemical microscopy to be localized predominantly in the nucleolar granular component, a structure that is involved in ribosome maturation. These results indicate that the novel Surf-6 gene is involved in a nucleolar function.

Neuron specificity of the neurofilament light promoter in transgenic mice requires the presence of DNA unwinding elements

Three reporter genes, the chloramphenicol acetyltransferase (CAT), the lacZ, and the intronless NF-L DNA, were used to test the activity of the proximal promoter region (-292 bp) of the human neurofilament light (hNF-L) gene in transgenic mice. Surprisingly, the hNF-L/CAT construct was highly sensitive to position effect, and its expression was found at low levels in several tissues of adult transgenic mice (Beaudet, L., Charron, G., Houle, D., Tretjakoff, I. Peterson, A., and Julien, J.-P. (1992) Gene (Amst.) 116, 205-214). In contrast, the hNF-L/lacZ or the hNF-L/intronless constructs were expressed exclusively in the nervous system during embryonic development and in adult animals. The DNA sequences analysis of the different reporter genes revealed the presence of matrix attachment regions (MARs) within the 3'-untranslated regions of all three transgenes. DNA unwinding elements were found within the MARs of lacZ and hNF-L gene constructs but not in the CAT gene construct. When this element was removed from the lacZ construct, expression of the hNF-L/lacZ transgene became susceptible to position effect and was no longer tissue-specific. These results indicate that DNA unwinding elements are essential for position effect independence conferred by MARs to the hNF-L basal promoter.

Selection of DNA clones with enhancer sequences

A method is described for selection of DNA clones that contain enhancer sequences that activate gene expression. An Escherichia coli-rodent cell shuttle vector, pPyE0, was used that contains polyoma viral DNA without the polyoma enhancer region. Replication of pPyE0 DNA in mouse cells is markedly reduced due to deletion of the polyoma enhancer region. Insertion of mouse genomic DNA fragments that contain putative enhancer sequences into pPyE0 adjacent to the polyoma origin of replication restored, to varying extents, the ability of the recombinant plasmid DNA to replicate in mouse cells. Recombinant plasmids that replicate well in mouse cells, therefore, are amplified selectively. Transfection of mouse neuroblastoma or fibroblast cells that constitutively synthesize polyoma large tumor antigen with a library of mouse genomic DNA fragments inserted in pPyE0 yielded many recombinant plasmids. DNA inserts from each of the 16 clones that were examined stimulated the expression of an enhancerless chloramphenicol acetyltransferase reporter gene. The DNA inserts from 4 clones that were studied resulted in 4- to 13-fold increases in chloramphenicol acetyltransferase mRNA in transfected mouse cells. Nucleotide sequence analysis led to the identification of 5 genomic DNA clones that were obtained by selection. All of the homologies found were to regions of DNA that are thought to be involved in the regulation of gene expression.

Conservation of the organization of five tightly clustered genes over 600 million years of divergent evolution

The organization of the mouse surfeit locus is unusual in that it contains six housekeeping genes (Surf-1-Surf-6), which are unrelated by sequence homology, in the tightest mammalian gene cluster thus far described. A maximum of only 73 base pairs separates any two of the four well-characterized genes, and two of the genes overlap at their 3' ends. The direction of transcription of each of the five surfeit genes, Surf-1-Surf-5, alternates with respect to that of its neighbor, suggesting cis-interaction or coregulation between the genes by mechanisms such as the sharing of regulatory elements and/or antisense regulation. The Surf-3 gene has been identified as encoding the ribosomal protein L7a (Rpl7a). We have used the high conservation of the Rpl7a gene to clone the chicken gene and surrounding genomic DNA. The tight clustering and juxtaposition of at least five of the surfeit genes (Surf-1-Surf-5) and their associated CpG-rich islands have been found to be conserved over the 600 million years of divergent evolution that separates birds and mammals. This strongly suggests that the surfeit locus represents a different form of gene cluster in which gene organization may play both a positive and negative regulatory role in gene expression possibly via cis-interactions between the closely spaced genes.

A gene trap approach in mouse embryonic stem cells: the lacZ reported is activated by splicing, reflects endogenous gene expression, and is mutagenic in mice

We have confirmed that the gene trap vector pGT4.5 creates spliced fusion transcripts with endogenous genes and prevents the synthesis of normal transcripts at the site of integration. cDNA was prepared to the lacZ fusion transcript in three ES cell lines to recover endogenous exon sequences upstream of lacZ. Each of the clones detected a unique-sized endogenous transcript, as well as the fusion transcript in the ES cell line from which the clone was derived. Sequence analysis of these clones and larger clones isolated from a random-primed cDNA library showed that the splice acceptor was used properly. For two insertions, the expression patterns of the lacZ reporter and the associated endogenous gene were compared in situ at three embryonic stages and were found to be similar. Three gene trap insertions were transmitted into the germ line, and abnormalities were observed with two of the three insertions in the homozygous state. RNA obtained from mice homozygous for the two mutant gene trap insertions was analyzed for normal endogenous transcripts and negligible amounts were detected, indicating that little splicing around the gene trap insertion occurred. This work demonstrates the capacity of the gene trap vector to generate lacZ fusion transcripts, to accurately report endogenous gene expression, and to mutate the endogenous gene at the site of integration.

The bidirectional promoter of the divergently transcribed mouse Surf-1 and Surf-2 genes

The ubiquitously expressed mouse Surf-1 and Surf-2 genes are divergently transcribed, and their heterogeneous start sites are separated by up to a maximum of only 73 bp. By using in vitro DNase I, dimethyl sulfate methylation, and gel retardation assays, we have identified five putative promoter control elements between and around the Surf-1 and Surf-2 start sites. The effects of each site on the regulation of Surf-1 and Surf-2 transcription have been studied in vivo, and four sites were found to be functional promoter elements. A novel binding site is required for efficient use of the intermediate but not the major start site of Surf-1. Three elements function in a bidirectional manner and are shared for efficient and accurate expression of both Surf-1 and Surf-2. One is an UEF (USF, MLTF) binding site which had a small effect on the use of the intermediate start sites of Surf-1 and also affected the major start sites of Surf-2. Another has sequence homology to the RPG alpha binding site associated with some ribosomal protein gene promoters and is required for efficient expression of the major but not intermediate start sites of Surf-1 and all start sites of Surf-2. The third, an RPG alpha-like site, is used for all start sites of both Surf-1 and Surf-2. Dissection of this cellular promoter region showed that different binding sites affect the use of different start sites and revealed a complex interaction between multiple elements that constitute a bona fide bidirectional promoter.

The bidirectional promoter of the divergently transcribed mouse Surf-1 and Surf-2 genes

The ubiquitously expressed mouse Surf-1 and Surf-2 genes are divergently transcribed, and their heterogeneous start sites are separated by up to a maximum of only 73 bp. By using in vitro DNase I, dimethyl sulfate methylation, and gel retardation assays, we have identified five putative promoter control elements between and around the Surf-1 and Surf-2 start sites. The effects of each site on the regulation of Surf-1 and Surf-2 transcription have been studied in vivo, and four sites were found to be functional promoter elements. A novel binding site is required for efficient use of the intermediate but not the major start site of Surf-1. Three elements function in a bidirectional manner and are shared for efficient and accurate expression of both Surf-1 and Surf-2. One is an UEF (USF, MLTF) binding site which had a small effect on the use of the intermediate start sites of Surf-1 and also affected the major start sites of Surf-2. Another has sequence homology to the RPG alpha binding site associated with some ribosomal protein gene promoters and is required for efficient expression of the major but not intermediate start sites of Surf-1 and all start sites of Surf-2. The third, an RPG alpha-like site, is used for all start sites of both Surf-1 and Surf-2. Dissection of this cellular promoter region showed that different binding sites affect the use of different start sites and revealed a complex interaction between multiple elements that constitute a bona fide bidirectional promoter.

A promoter trap in embryonic stem (ES) cells selects for integration of DNA into CpG islands

An analysis of several G418-resistant ES cell lines produced by electroporation of a promoterless neo gene (NASTI), shows an enrichment for integrations within, or adjacent to, CpG islands. A detailed analysis of two of the cell lines reveals short regions of homology between the genomic target DNA and the construct ends, and that recombination may be mediated by DNA Topoisomerase I. The DNA flanking the insert detects transcription of endogenous genes, and in one cell line divergent transcripts are detected. This use of ES cells should provide an effective and efficient means of creating insertional mutations in mice.

Entrapment Vectors: A New Tool for Mammalian Genetics

The use of mouse embryonic stem (ES) cells has facilitated the creation of mouse strains carrying desired genetic alterations. Recently, "entrapment" vectors have been developed to identify and mutate genes active during embryogenesis. These vectors in combination with ES cells may provide an efficient screening strategy to find genes that control mouse development.

Isolation of cellular promoters by using a retrovirus promoter trap

A retrovirus vector has been used to isolate transcriptional promoters from mammalian cells. The virus contains a selectable gene encoding histidinol dehydrogenase (his) in the U3 region of the 3' long terminal repeat (LTR). When the virus is passaged, duplication of LTRs places his sequences just 30 nucleotides from the adjacent cellular DNA. As a result, selection for histidinol resistance generates cell clones in which his is expressed on transcripts initiating in the flanking cellular DNA. Upstream cellular sequences, cloned after amplification by polymerase chain reaction, hybridized to RNA from uninfected cells, indicating that the adjacent promoters were transcriptionally active prior to virus integration. Two cloned transcribed flanking sequences also contained highly active transcriptional promoters, as estimated by their ability to activate expression of a linked reporter gene. Thus, U3His vectors provide a rapid and efficient means to isolate promoters active in different cell types. Moreover, by selecting for cell clones containing proviruses integrated in expressed genes, the virus may make an effective insertional mutagen.

One of the tightly clustered genes of the mouse surfeit locus is a highly expressed member of a multigene family whose other members are predominantly processed pseudogenes

The mouse surfeit locus is unusual in that it contains a number of closely clustered genes (Surf-1, -2, and -4) that alternate in their direction of transcription (T. Williams, J. Yon, C. Huxley, and M. Fried, Proc. Natl. Acad. Sci. USA 85:3527-3530, 1988). The heterogeneous 5' ends of Surf-1 and Surf-2 are separated by 15 to 73 base pairs (bp), and the 3' ends of Surf-2 and Surf-4 overlap by 133 bp (T. Williams and M. Fried, Mol. Cell. Biol. 6:4558-4569, 1986; T. Williams and M. Fried, Nature (London) 322:275-279, 1986). A fourth gene in this locus, Surf-3, which is a member of a multigene family, has been identified. The poly(A) addition site of Surf-3 lies only 70 bp from the poly(A) addition site of Surf-1. Transcription of Surf-3 has been studied in the absence of the other members of its multigene family after transfection of a cloned genomic mouse DNA fragment, containing the Surf-3 gene, into heterologous monkey cells. Surf-3 specifies a highly expressed 1.0-kilobase mRNA that contains a long open reading frame of 266 amino acids, which would encode a highly basic polypeptide (23% Arg plus Lys). The other members of the Surf-3 multigene family are predominantly, if not entirely, intronless pseudogenes with the hallmarks of being generated by reverse transcription. The role of the very tight clustering on regulation of expression of the genes in the surfeit locus is discussed.

One of the tightly clustered genes of the mouse surfeit locus is a highly expressed member of a multigene family whose other members are predominantly processed pseudogenes

The mouse surfeit locus is unusual in that it contains a number of closely clustered genes (Surf-1, -2, and -4) that alternate in their direction of transcription (T. Williams, J. Yon, C. Huxley, and M. Fried, Proc. Natl. Acad. Sci. USA 85:3527-3530, 1988). The heterogeneous 5' ends of Surf-1 and Surf-2 are separated by 15 to 73 base pairs (bp), and the 3' ends of Surf-2 and Surf-4 overlap by 133 bp (T. Williams and M. Fried, Mol. Cell. Biol. 6:4558-4569, 1986; T. Williams and M. Fried, Nature (London) 322:275-279, 1986). A fourth gene in this locus, Surf-3, which is a member of a multigene family, has been identified. The poly(A) addition site of Surf-3 lies only 70 bp from the poly(A) addition site of Surf-1. Transcription of Surf-3 has been studied in the absence of the other members of its multigene family after transfection of a cloned genomic mouse DNA fragment, containing the Surf-3 gene, into heterologous monkey cells. Surf-3 specifies a highly expressed 1.0-kilobase mRNA that contains a long open reading frame of 266 amino acids, which would encode a highly basic polypeptide (23% Arg plus Lys). The other members of the Surf-3 multigene family are predominantly, if not entirely, intronless pseudogenes with the hallmarks of being generated by reverse transcription. The role of the very tight clustering on regulation of expression of the genes in the surfeit locus is discussed.

The mouse surfeit locus contains a very tight cluster of four "housekeeping" genes that is conserved through evolution

The four identified alternating transcription units (Surf-1 to Surf-4) in the mouse surfeit locus are very tightly clustered, no two neighboring units being separated by more than 73 base pairs and the Surf-2 and Surf-4 transcription units overlapping by 133 base pairs at their 3' ends. All four surfeit genes, which are unrelated by sequence similarity, were found to have the properties of "housekeeping" genes, being expressed in a variety of differentiated mouse cell lines and containing unmethylated CpG-rich islands in the vicinity of their 5' ends. The unusual organization of the four surfeit genes was found not to be unique to the mouse: the same juxtaposition of the genes was found to be conserved in a number of different mammals, including humans. The four human surfeit genes were also found to be transcriptionally active.

Isolation of tobacco DNA segments with plant promoter activity

We constructed a promoter probe vector, pGVL120, to isolate plant DNA segments with promoter activity in tobacco. Plant nuclear DNA Sau3A fragments were inserted in front of the npt-II sequence, and a mixture of recombinant plasmids was mobilized to Agrobacterium sp. and used to transform tobacco protoplasts. By kanamycin selection, transformed plant cell lines containing NPT-II T-DNAs were isolated. Eight of these cell lines were regenerated and analyzed for the levels of NPT-II activity in stem, root, midrib, and leaf. These levels demonstrated novel regulation patterns in each isolate. One cell line, T20, was analyzed in detail and found to contain four different T-DNAs. One of the recloned T-DNAs, T20-2, contains an insert of 401 base pairs in front of the NPT-II sequence, and by reintroducing this T-DNA into plant cells we could demonstrate that this insert provides a promoter sequence. The NPT-II enzyme activity under the control of the P20 promoter is especially high in stem and root, but low in leaf and callus, both in the originally isolated T20 plant and in independently isolated transformants with the T20-2 T-DNA.

Random isolation of gene activator elements from the human genome

Long-range-acting gene activator elements were randomly isolated from the human genome by functional selection. HeLa cells were transfected with an enhancer trap, a plasmid containing an enhancerless xanthine-guanosine phosphoribosyltransferase (gpt) gene transcribed from the simian virus 40 early promoter, and stably transformed GPT+ cells were selected. From several transformants, human DNA sequences flanking the enhancer trap were cloned. Two gene activators (GA1 and GA2) were found in the cloned human DNAs. GA1 and GA2 showed strong enhancer activity both in a stable transformation assay and in a transient expression assay. They had functional properties similar to those of other known enhancers: GA1 and GA2 activated the expression of a linked gene over distances of at least 5 kilobases both upstream and downstream in an orientation-independent fashion. GA1 may be required for the initial establishment of gene activation but was not essential for the maintenance of active expression. GA1 and GA2 were active not only in HeLa cells but also in other types of human cells, such as neuroblastoma cells. This indicates a limited but relatively broad cell type specificity. The HeLa genome contains multiple copies of GA1, while GA2 exists once in the genome.

Activation of an enhancerless gene by chromosomal integration

Expression of enhancerless (E-) and enhancer-containing (E+) genes that are chromosomally integrated was examined. An E- plasmid (pE-cat) containing a chloramphenicol acetyltransferase (cat) gene linked to the simian virus 40 (SV40) early promoter or its E+ counterpart plasmid (pE+-cat) containing the SV40 enhancer was cotransfected into thymidine kinase (TK)-deficient L cells with a cloned tk gene. A number of TK+ transformants were isolated, and expression of the cointegrated cat gene in these cell lines was quantitatively determined by the assay of CAT activity. The results indicated unexpectedly that the E- cat gene was as actively expressed as the E+ cat gene. Analysis of CAT mRNA by primer extension indicated that the E- cat gene, as well as the E+ cat gene, was transcribed from the "native" initiation site contained in the SV40 early promoter region. The active expression of the E- cat gene was maintained in secondary TK+ transformants that arose by transfection with genomic DNA from the primary transformant. These results suggest that expression of the integrated E- cat gene is activated by endogenous enhancer elements.

The MES-1 murine enhancer element is closely associated with the heterogeneous 5' ends of two divergent transcription units

The location in the mouse genome of the 149-base pair MES-1 element, previously isolated by its ability to restore expression to an enhancerless selectable gene, was analyzed. The active moiety of the single-copy MES-1 element is located between the 5' ends of two divergent transcription units, SURF-1 and SURF-2, both of which specify more than one mRNA species by differential splicing. The heterogeneous 5' ends of the SURF transcripts are separated by only 50 to 75 base pairs, and this sequence possesses a high G + C content (65%) and contains neither the TATA and CAAT box motifs normally associated with many highly expressed genes nor the GC box motif (Sp1-binding site) associated with a number of housekeeping genes. Although MES-1 appears to have enhancerlike properties when linked to heterologous genes, its normal genomic location suggests that it functions as a bidirectional promoter. Thus, MES-1 may represent a new class of enhancer-promoter element.

The MES-1 murine enhancer element is closely associated with the heterogeneous 5' ends of two divergent transcription units

The location in the mouse genome of the 149-base pair MES-1 element, previously isolated by its ability to restore expression to an enhancerless selectable gene, was analyzed. The active moiety of the single-copy MES-1 element is located between the 5' ends of two divergent transcription units, SURF-1 and SURF-2, both of which specify more than one mRNA species by differential splicing. The heterogeneous 5' ends of the SURF transcripts are separated by only 50 to 75 base pairs, and this sequence possesses a high G + C content (65%) and contains neither the TATA and CAAT box motifs normally associated with many highly expressed genes nor the GC box motif (Sp1-binding site) associated with a number of housekeeping genes. Although MES-1 appears to have enhancerlike properties when linked to heterologous genes, its normal genomic location suggests that it functions as a bidirectional promoter. Thus, MES-1 may represent a new class of enhancer-promoter element.

Activation of an enhancerless gene by chromosomal integration

Expression of enhancerless (E-) and enhancer-containing (E+) genes that are chromosomally integrated was examined. An E- plasmid (pE-cat) containing a chloramphenicol acetyltransferase (cat) gene linked to the simian virus 40 (SV40) early promoter or its E+ counterpart plasmid (pE+-cat) containing the SV40 enhancer was cotransfected into thymidine kinase (TK)-deficient L cells with a cloned tk gene. A number of TK+ transformants were isolated, and expression of the cointegrated cat gene in these cell lines was quantitatively determined by the assay of CAT activity. The results indicated unexpectedly that the E- cat gene was as actively expressed as the E+ cat gene. Analysis of CAT mRNA by primer extension indicated that the E- cat gene, as well as the E+ cat gene, was transcribed from the "native" initiation site contained in the SV40 early promoter region. The active expression of the E- cat gene was maintained in secondary TK+ transformants that arose by transfection with genomic DNA from the primary transformant. These results suggest that expression of the integrated E- cat gene is activated by endogenous enhancer elements.

Random isolation of gene activator elements from the human genome

Long-range-acting gene activator elements were randomly isolated from the human genome by functional selection. HeLa cells were transfected with an enhancer trap, a plasmid containing an enhancerless xanthine-guanosine phosphoribosyltransferase (gpt) gene transcribed from the simian virus 40 early promoter, and stably transformed GPT+ cells were selected. From several transformants, human DNA sequences flanking the enhancer trap were cloned. Two gene activators (GA1 and GA2) were found in the cloned human DNAs. GA1 and GA2 showed strong enhancer activity both in a stable transformation assay and in a transient expression assay. They had functional properties similar to those of other known enhancers: GA1 and GA2 activated the expression of a linked gene over distances of at least 5 kilobases both upstream and downstream in an orientation-independent fashion. GA1 may be required for the initial establishment of gene activation but was not essential for the maintenance of active expression. GA1 and GA2 were active not only in HeLa cells but also in other types of human cells, such as neuroblastoma cells. This indicates a limited but relatively broad cell type specificity. The HeLa genome contains multiple copies of GA1, while GA2 exists once in the genome.

Isolation of tobacco DNA segments with plant promoter activity

We constructed a promoter probe vector, pGVL120, to isolate plant DNA segments with promoter activity in tobacco. Plant nuclear DNA Sau3A fragments were inserted in front of the npt-II sequence, and a mixture of recombinant plasmids was mobilized to Agrobacterium sp. and used to transform tobacco protoplasts. By kanamycin selection, transformed plant cell lines containing NPT-II T-DNAs were isolated. Eight of these cell lines were regenerated and analyzed for the levels of NPT-II activity in stem, root, midrib, and leaf. These levels demonstrated novel regulation patterns in each isolate. One cell line, T20, was analyzed in detail and found to contain four different T-DNAs. One of the recloned T-DNAs, T20-2, contains an insert of 401 base pairs in front of the NPT-II sequence, and by reintroducing this T-DNA into plant cells we could demonstrate that this insert provides a promoter sequence. The NPT-II enzyme activity under the control of the P20 promoter is especially high in stem and root, but low in leaf and callus, both in the originally isolated T20 plant and in independently isolated transformants with the T20-2 T-DNA.

Gene activation properties of a mouse DNA sequence isolated by expression selection

The MES-1 element was previously isolated from restricted total mouse cellular DNA by "expression selection"--the ability to reactivate expression of a test gene devoid of its 5' enhancer sequences. Mes-1 has been tested in long-term transformation and short-term CAT expression assays. In both assays MES-1 is active independent of orientation and at a distance when placed 5' to the test gene. The element is active with heterologous promoters and functions efficiently in both rat and mouse cells. MES-1 activates expression by increasing transcription from the test gene's own start (cap) site. Thus the expression selection technique can be used for the isolation of DNA sequences with enhancer-like properties from total cellular DNA.

A mouse locus at which transcription from both DNA strands produces mRNAs complementary at their 3' ends

The organization and large size of the mammalian cell genome allows spatial separation of different transcription units. In those cases where more than one species of messenger are synthesized from the same cellular DNA sequence, they have been found to be generated from transcription proceeding in the same direction. These mRNAs always share regions of homology and can differ from one another as a result of differential processing (splicing and/or polyadenylation) or alternative initiation. In contrast, complementary mRNAs transcribed from opposite strands of the same cellular DNA sequence have not previously been observed. Here we have identified a region of mouse DNA at which processed mRNAs from two adjacent convergent transcription units overlap by 133 base pairs (bp) at their 3'-untranslated ends. One of the transcription units appears to encode a second mRNA which does not contain this overlapping region. This represents the first description of the natural occurrence of processed mammalian cell mRNAs transcribed from opposite strands of the same DNA sequence. The implications of these complementary regions in normal gene regulation are discussed in the context of the finding that the artificial introduction into cells of DNA constructs synthesizing anti-sense RNAs complementary to regions of mRNA transcribed from a chromosomal gene, can inhibit the gene's activity, presumably by the formation of double-stranded RNA.

Control of eukaryotic messenger RNA synthesis by sequence-specific DNA-binding proteins

The enzymatic machinery that carries out RNA synthesis provides the cell with the means to adjust the patterns of transcription in response to environmental and developmental signals. In eukaryotes, this regulation is mediated in part by promoter-specific transcription factors, which are DNA-binding proteins with the ability to discriminate between distinctive DNA sequence elements found in the promoter regions of different genes. The presence of these factors bound to DNA enables other components of the transcriptional machinery, including the RNA polymerase, to initiate transcription with selectivity and accuracy.

Isolation of a gene enhancer within an amplified inverted duplication after "expression selection"

We have attempted to isolate and identify cellular expression sequences from F9 teratocarcinoma DNA by utilizing their ability to reactivate a selectable gene devoid of its own expression sequences (expression selection). Restriction nuclease-digested F9 cellular DNA was ligated to a polyoma virus (Py) DNA fragment which contains an intact transforming region but is incapable of inducing transformation because it lacks the viral 5' enhancer sequence. The ligation mixture was used to transfect Rat-1 cells and a transformed cell line, 3B, was isolated. The 3B cell line contained a single type of Py DNA insert, which was molecularly cloned as an 18-kilobase BglII fragment. A weak cellular enhancer was identified in a 4.7-kilobase BamHI fragment upstream from the Py sequences. Both the Py DNA and the enhancer sequences were found to be present in an inverted duplication in the 3B clone. The presence of this structure in 3B genomic DNA was confirmed by the analysis of selectively isolated inverted duplicated sequences, and the structure was found to be at least 22 kilobases long. In the 3B cell line, the inverted duplicated sequences containing the Py and enhancer sequences are quite stable and are amplified 20- to 40-fold. The strongly transformed phenotype of the 3B cells may be a result of this amplification. The formation of inverted duplications as a part of the amplification mechanism as well as a general strategy for the cloning of inverted duplicated (amplified) sequences is discussed.

Activation of major histocompatibility complex class I mRNA containing an Alu-like repeat in polyoma virus-transformed rat cells

Class I genes of the major histocompatibility complex (MHC) appear to be activated in mouse cells transformed by the DNA tumour virus simian virus 40 (SV40). Conversely, suppression of MHC class I genes has been reported in adenovirus-12-transformed baby kidney rat cells. We have now investigated the expression of genes encoded by the rat MHC locus in rat fibroblast cells transformed by polyoma virus (Py). Using a mouse genomic H-2 clone as a probe in Northern transfer hybridization analysis, we have observed a high level of expression of rat MHC class I messenger RNA in all the transformed rat cell lines analysed. The class I 1.6-kilobase (kb) mRNA activated in Py-transformed rat cells appears to contain an Alu-like type II repeat element, as the same 1.6-kb mRNA is detected using either the H-2 class I sequence or a repetitive Alu-like type II element as a probe. High levels of heterogeneous poly(A)+ transcripts of 0.5-0.8 kb are also observed in Py-transformed rat cells using probes containing an Alu-like type II repetitive element.

Control of gene expression by glucocorticoid hormones

Glucocorticoids control the expression of a small number of transcriptionally active genes by increasing or decreasing mRNA concentration. Either effect can result from a transcriptional or a post-transcriptional mechanism. Induction of mouse mammary tumour virus RNA results from a stimulation of transcription initiation and depends on the presence of defined regions in proviral DNA. These regions bind the glucocorticoid receptor and behave functionally as proto-enhancers. Glucocorticoid-inducible genes can retain their sensitivity to the hormone after transfer to a heterologous cell by transfection techniques. Non-inducible genes can become inducible when linked to the promoter region of an inducible gene. The mechanisms by which the receptor-steroid complex stimulates or inhibits transcription or influences mRNA stability are unknown. Receptor binding to nucleic acids appears to be a necessary but not sufficient condition. It is likely that the receptor also interacts with chromatin proteins. This might lead to a catalytic modification of these proteins, resulting in a modulation of gene expression. Development of glucocorticoid-sensitive, biochemically defined, cell-free transcription systems should provide a tool to delineate the molecular determinants of this essential regulatory mechanism.

A plasmid vehicle suitable for the molecular cloning and characterization of mammalian promoters

We have constructed a plasmid, pPSG4, that carries only the coding sequences for polyomavirus (Py) small, middle and truncated large T-antigens. A unique HindIII site allows the introduction of foreign promoters directly in front of viral coding sequences. The simian virus 40 (SV40) early and late, adenovirus-2 (Ad-2) major late and herpes simplex virus-1 (HSV-1) thymidine kinase (TK) promoters all confer on pPSG4 the ability to transform rat embryonic fibroblasts with high efficiency. Sequential deletion of the 72 bp repeats, the 21 bp repeats and the TATA box from the SV40 early region in pPSG4 produced a 50, then 30 and then a further 5 to 10-fold decrease in transformation efficiency, respectively. Thus pPSG4 is a convenient vector for the cloning and characterization of mammalian promoters.

Small fragments of herpesvirus DNA with transforming activity contain insertion sequence-like structures

A 737-base-pair fragment of herpes simplex virus type 2 DNA with morphological-transforming ability was identified by transfecting into rodent cells deleted fragments of the left-hand end of the Bgl II N fragment region (map position 0.58-0.625), which were constructed in vitro. The transforming sequences lie within the coding region for a Mr 61,000 protein, but the fragment itself does not appear to specify a viral polypeptide. Contained within the transforming fragment are sequences that can be drawn as a stem-loop structure flanked by direct repeats, similar to an insertion sequence-like element. An insertion sequence-like structure was also found in a small fragment of human cytomegalovirus DNA that has transforming activity. Possible mechanisms of herpesvirus transformation are discussed.

BK viral enhancer element and a human cellular homolog

Comparison of two closely related primate papovaviruses, simian virus 40 (SV40) and human BK virus (BKV), reveals that the only region of extensive divergence, the tandem sequences adjacent to the origins of DNA replication, is responsible in SV40 for enhancing early gene expression. This study demonstrates a similar enhancer function for the analogous repeated region in BKV. The dissimilarity in sequence of the BKV and SV40 enhancer elements suggests that they may have been acquired since SV40 and BKV diverged. A locus cloned from the human genome homologous to the BKV tandem repeats has been shown to function as low level enhancer element in mammalian cells. These data support the hypothesis that viral enhancer sequences may be evolutionarily related to host cell sequences.