Analysis of the mouse dhfr promoter region: existence of a divergently transcribed gene

Murine dihydrofolate reductase transcripts through the cell cycle

The murine dihydrofolate reductase gene codes for mRNAs that differ in the length of their 3' untranslated region as well as in the length of their 5' leader sequence. In addition, the dihydrofolate reductase promoter functions bidirectionally, producing a series of RNAs from the opposite strand than the dihydrofolate reductase mRNAs. We have examined the production of these RNAs and their heterogeneous 5' and 3' termini as mouse 3T6 cells progress through a physiologically continuous cell cycle. We found that all of the transcripts traverse the cell cycle in a similar manner, increasing at the G1/S boundary without significantly changing their ratios relative to one another. We conclude that cell-cycle regulation of dihydrofolate reductase is achieved without recruiting new transcription initiation sites and without a change in polyadenylation sites. It appears that the mechanism responsible for the transcriptional cell-cycle regulation of the dihydrofolate reductase gene is manifested only by transiently increasing the efficiency of transcription at the dihydrofolate reductase promoter.

Bidirectional transcription from a solo long terminal repeat of the retrotransposon TED: symmetrical RNA start sites

A single copy of the retrotransposon TED was found integrated within the DNA genome of the insect baculovirus, Autographa californica nuclear polyhedrosis virus. After excision of the element from the viral genome, a single long terminal repeat (LTR) remained behind. We have examined the effect of this solo TED LTR on the local pattern of viral transcription. Most prominent was the transcription of two sets of abundant RNAs; both originated within the LTR but extended in opposite directions into flanking viral genes. By promoting symmetric transcription of adjacent genes, the solo LTR has the capacity to activate or repress gene expression in two directions. Primer extension analysis demonstrated that the divergent LTR transcripts were initiated near the same point within a 22-base-pair sequence having hyphenated twofold symmetry. Analogous symmetries at the initiation sites of other retrotransposon LTRs, including copia and Ty, suggested that these sequences serve to establish the precise start for transcription.

Fine structure of the human hypoxanthine phosphoribosyltransferase gene

The human hypoxanthine phosphoribosyltransferase (HPRT) gene has been characterized by molecular cloning, mapping, and DNA sequencing techniques. The entire gene, which is about 44 kilobases in length, is composed of nine exon elements. The positions of the introns within the coding sequence are identical to those of the previously-characterized mouse HPRT gene, although there are significant differences between intron sizes for the two genes. HPRT minigenes have been used in a transient expression assay involving microinjection into HPRT- cells to demonstrate functional promoter activity within a 234-base-pair region upstream from the ATG codon. The promoter of this gene resembles those of other recently characterized "housekeeping" genes in that it lacks CAAT- and TATA-like sequences, but contains several copies of the sequence GGGCGG. Both RNase protection and primer extension analysis indicate that human HPRT mRNA is heterogeneous at the 5' terminus, with transcription initiation occurring at sites located congruent to 104 to congruent to 169 base pairs upstream from the ATG codon. Comparison of the mouse and human HPRT 5'-flanking sequences indicates that there are only limited stretches of conserved sequence, although there are other shared features, such as an extremely high density of potential methylation sites, that may have functional significance.

5' Nucleotide sequences influence serum-modulated expression of a human dihydrofolate reductase minigene

Human dihydrofolate reductase (DHFR) gene sequences were isolated from DHFR gene-amplified breast cancer cell line MCF-7. These genomic sequences plus human DHFR cDNA sequences were used to construct a DHFR minigene. Calcium phosphate-mediated transfer of minigene DNA into DHFR gene-deleted Chinese hamster ovary cells converted these cells to a DHFR+ phenotype at a frequency of 0.12%. Minigene-transfected cells contained 20 to 30 minigene copies per cell and had DHFR enzyme levels similar to those of wild-type MCF-7 human cells (1.4 pmol/mg of protein). In contrast to gene-amplified MCF-7 cells, which contained multiple DHFR mRNA species (1.1, 1.6, 3.8, and 5.3 kilobases), only a single 3.8-kilobase DHFR mRNA was found in minigene-transfected cells. Previous studies on normal cells demonstrated modulation of DHFR levels by a variety of conditions which altered cell growth. When cell growth was induced in minigene-transfected cells by release from serum deprivation and DHFR levels were assayed at the time of maximum DNA synthesis, these levels were increased 2.4 to 3.7-fold. In contrast, the DHFR levels in cells transfected with a construct made from DHFR cDNA and viral promoter, intron, and termination sequences were unchanged. Minigene deletions were made and analyzed to determine the DHFR gene sequences responsible for regulation. Deletion of sequences upstream from 322 base pairs 5' to the start of transcription or 90 base pairs downstream from the termination of translation (which removed most of the 3' nontranslated region of the gene) did not alter the responsiveness of minigene-transfected cells to serum deprivation. However, when sequences between 322 and 113 base pairs 5' to the start of transcription were deleted, serum-dependent expression in minigene-transfected cells was affected.

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.

Preferential DNA repair of an active gene in human cells

Removal of pyrimidine dimers was measured in defined sequences in human cells amplified for the dihydrofolate reductase (DHFR) gene. We quantitated repair in specific restriction fragments by using the dimer-specific bacteriophage T4 endonuclease V and analysis by Southern blotting. Within 4 hr after 5- or 10-J/m2 UV irradiation, more than 60% of the dimers had been removed from a 20-kilobase fragment that lies entirely within the transcription unit of the DHFR gene and from a 25-kilobase fragment located in the 5' flanking region of the gene. Repair in the overall genome was measured by analyzing cellular DNA treated with T4 endonuclease V in alkaline sucrose gradients. Sixty-nine percent of the dimers were removed from the genome overall within 24 hr after irradiation, but only 25% were removed within 4 hr and 38% were removed within 8 hr. These results demonstrate a strong preferential rate of removal of dimers from the 50-kilobase region that includes the transcriptionally active DHFR gene compared to that in total cellular DNA. We confirmed that DHFR-containing DNA is repaired more rapidly than bulk DNA by using an approach that provides a direct comparison between repair in specific sequences and repair in total cellular DNA. We also show that the DHFR-containing sequences are repaired more rapidly than the nontranscribed repetitive alpha DNA sequences. Our finding of preferential early repair in a transcriptionally active region in overall repair-proficient cells suggests that selective dimer removal from active sequences may be a general characteristic of mammalian DNA repair.

Novel promoter upstream of the human c-myc gene and regulation of c-myc expression in B-cell lymphomas

A new promoter of the human c-myc gene called P0, with multiple RNA start sites, was mapped over 500 bases upstream of the two previously identified promoters, P1 and P2. Sequencing full-length cDNA clones of P0 RNAs revealed two open reading frames upstream of that for the P64c-myc protein. P0 RNA is located on polyribosomes and released by puromycin, indicating that it functions as an mRNA. In vitro translation of RNA synthesized from the cloned cDNAs predicts that P0 transcripts are translated into a novel 12.5-kilodalton protein corresponding to the first open reading frame. The regulation of P0 RNA was studied in the B-cell lymphoma cell line Manca, in which only the translocated c-myc allele lacking exon 1 was thought to be active. However, we found that P0 transcription and the DNase I-hypersensitive site associated with this promoter persist on the untranslocated allele, even though P1/P2 transcription as measured by a nuclear runoff assay was repressed. These results suggest that allelic exclusion of c-myc expression in this B-cell lymphoma is caused by a repression of transcription which is specific to the P1/P2 promoters. We previously reported a block to elongation of transcription near the 3' end of exon 1 in the wild-type c-myc gene, which results in an excess of exon 1 over exon 2 transcription (5a). In contrast, we found that in the Daudi B-cell lymphoma, which retains exon 1 in the active allele, equimolar transcription of exons 1 and 2 occurs. This result suggests a model for the activation of c-myc in B-cell lymphomas.

Structural features of the murine dihydrofolate reductase transcription termination region: identification of a conserved DNA sequence element

Structural features of the transcription termination region for the mouse dihydrofolate reductase gene have been determined and compared with those of several other known termination regions for protein coding genes. A common feature identified among these termination regions was the presence of a 20 bp consensus DNA sequence element (ATCAGAATATAGGAAAGTAGCAAT). The results imply that the 20 bp consensus DNA sequence element is important for signaling RNA polymerase II transcription termination at least in the several vertebrate species investigated. Furthermore, the results suggest that for the dhfr gene and possibly for other genes in mice as well, the potential termination consensus sequence can exist as part of a long interspersed repetitive DNA element.

Bidirectional transcription from a solo long terminal repeat of the retrotransposon TED: symmetrical RNA start sites

A single copy of the retrotransposon TED was found integrated within the DNA genome of the insect baculovirus, Autographa californica nuclear polyhedrosis virus. After excision of the element from the viral genome, a single long terminal repeat (LTR) remained behind. We have examined the effect of this solo TED LTR on the local pattern of viral transcription. Most prominent was the transcription of two sets of abundant RNAs; both originated within the LTR but extended in opposite directions into flanking viral genes. By promoting symmetric transcription of adjacent genes, the solo LTR has the capacity to activate or repress gene expression in two directions. Primer extension analysis demonstrated that the divergent LTR transcripts were initiated near the same point within a 22-base-pair sequence having hyphenated twofold symmetry. Analogous symmetries at the initiation sites of other retrotransposon LTRs, including copia and Ty, suggested that these sequences serve to establish the precise start for transcription.

5' Nucleotide sequences influence serum-modulated expression of a human dihydrofolate reductase minigene

Human dihydrofolate reductase (DHFR) gene sequences were isolated from DHFR gene-amplified breast cancer cell line MCF-7. These genomic sequences plus human DHFR cDNA sequences were used to construct a DHFR minigene. Calcium phosphate-mediated transfer of minigene DNA into DHFR gene-deleted Chinese hamster ovary cells converted these cells to a DHFR+ phenotype at a frequency of 0.12%. Minigene-transfected cells contained 20 to 30 minigene copies per cell and had DHFR enzyme levels similar to those of wild-type MCF-7 human cells (1.4 pmol/mg of protein). In contrast to gene-amplified MCF-7 cells, which contained multiple DHFR mRNA species (1.1, 1.6, 3.8, and 5.3 kilobases), only a single 3.8-kilobase DHFR mRNA was found in minigene-transfected cells. Previous studies on normal cells demonstrated modulation of DHFR levels by a variety of conditions which altered cell growth. When cell growth was induced in minigene-transfected cells by release from serum deprivation and DHFR levels were assayed at the time of maximum DNA synthesis, these levels were increased 2.4 to 3.7-fold. In contrast, the DHFR levels in cells transfected with a construct made from DHFR cDNA and viral promoter, intron, and termination sequences were unchanged. Minigene deletions were made and analyzed to determine the DHFR gene sequences responsible for regulation. Deletion of sequences upstream from 322 base pairs 5' to the start of transcription or 90 base pairs downstream from the termination of translation (which removed most of the 3' nontranslated region of the gene) did not alter the responsiveness of minigene-transfected cells to serum deprivation. However, when sequences between 322 and 113 base pairs 5' to the start of transcription were deleted, serum-dependent expression in minigene-transfected cells was affected.

Multiple transcription start sites, DNase I-hypersensitive sites, and an opposite-strand exon in the 5' region of the CHO dhfr gene

Transcription of the 26-kilobase (kb) dihydrofolate reductase (dhfr) gene in CHO cells is initiated at two sites: a major site (approximately 85% of the dhfr mRNA) at -63 relative to the translation start and a minor site (approximately 15%) at -107. Transcription also occurs from the opposite DNA strand in the dhfr 5' region, with a probable initiation site at approximately -195 relative to the dhfr translation start. A 4-kb polyadenylated RNA that is derived from the opposite-strand transcription increases threefold in abundance after serum starvation of CHO cells for 24 h. dhfr mRNA levels do not change during this time. The first dhfr exon lies within a 1-kb genomic region marked by exceptionally high G + C content and lack of DNA methylation. This region also includes a 214-base-pair (bp) exon for the opposite-strand transcript and five of the six DNase I-hypersensitive sites identified at the dhfr locus. Analysis of the DNA sequences of hamster, human (M. Chen, T. Shimada, A. D. Moulton, A. Cline, R. K. Humphries, J. Maizel, and A. W. Nienhuis, J. Biol. Chem. 259:3933-3943, 1984), and mouse (M. McGrogan, C. C. Simonsen, D. T. Smouse, P. J. Farnham, and R. T. Schimke, J. Biol. Chem. 260:2307-2314, 1985) dhfr genes reveals the presence of a 29-bp unit that is conserved 45 to 49 bp upstream of major and minor dhfr transcription start sites. This unit follows the consensus: GRGGCGGTGGCCTNNNNTGTCRCAARTRGGTR. The 5' part of the 29-bp unit contains a GC box that agrees with the GGGCGG consensus-binding site for the RNA polymerase II transcription factor Sp1 (D. Gidoni, W. A. Dynan, and R. Tjian, Nature (London) 312:409-413, 1984). Each of the three mammalian dhfr genes has several G-rich GC boxes proximal to the major dhfr transcription start site and several GC boxes of the opposite orientation (C rich) in a distal region about 500 bp upstream.

Murine dihydrofolate reductase transcripts through the cell cycle

The murine dihydrofolate reductase gene codes for mRNAs that differ in the length of their 3' untranslated region as well as in the length of their 5' leader sequence. In addition, the dihydrofolate reductase promoter functions bidirectionally, producing a series of RNAs from the opposite strand than the dihydrofolate reductase mRNAs. We have examined the production of these RNAs and their heterogeneous 5' and 3' termini as mouse 3T6 cells progress through a physiologically continuous cell cycle. We found that all of the transcripts traverse the cell cycle in a similar manner, increasing at the G1/S boundary without significantly changing their ratios relative to one another. We conclude that cell-cycle regulation of dihydrofolate reductase is achieved without recruiting new transcription initiation sites and without a change in polyadenylation sites. It appears that the mechanism responsible for the transcriptional cell-cycle regulation of the dihydrofolate reductase gene is manifested only by transiently increasing the efficiency of transcription at the dihydrofolate reductase promoter.

Fine structure of the human hypoxanthine phosphoribosyltransferase gene

The human hypoxanthine phosphoribosyltransferase (HPRT) gene has been characterized by molecular cloning, mapping, and DNA sequencing techniques. The entire gene, which is about 44 kilobases in length, is composed of nine exon elements. The positions of the introns within the coding sequence are identical to those of the previously-characterized mouse HPRT gene, although there are significant differences between intron sizes for the two genes. HPRT minigenes have been used in a transient expression assay involving microinjection into HPRT- cells to demonstrate functional promoter activity within a 234-base-pair region upstream from the ATG codon. The promoter of this gene resembles those of other recently characterized "housekeeping" genes in that it lacks CAAT- and TATA-like sequences, but contains several copies of the sequence GGGCGG. Both RNase protection and primer extension analysis indicate that human HPRT mRNA is heterogeneous at the 5' terminus, with transcription initiation occurring at sites located congruent to 104 to congruent to 169 base pairs upstream from the ATG codon. Comparison of the mouse and human HPRT 5'-flanking sequences indicates that there are only limited stretches of conserved sequence, although there are other shared features, such as an extremely high density of potential methylation sites, that may have functional significance.