The murine dihydrofolate reductase (DHFR) gene gives rise to multiple polyadenylated mRNAs displaying heterogeneity in the length of the 3' untranslated region. These species are present in the cytoplasm at levels that vary over 2 orders of magnitude, suggesting that certain poly(A) sites are preferred over others. Previous observations have shown that three out of the four major sites of polyadenylation do not display consensus hexanucleotide (AATAAA, ATTAAA) signals. We have further analyzed the sequences involved in directing multiple polyadenylation events on the DHFR gene by focusing our attention on the 4.1- and 5.6-kilobase mRNAs, the lowest abundance DHFR species observed on RNA blot analysis. Identification and sequence analysis of the poly(A) addition sites corresponding to these species revealed appropriately positioned consensus hexanucleotide signals; additional nearby poly(A) sites were also detected which apparently do not use consensus hexanucleotides to direct poly(A) addition to DHFR mRNAs of relatively lower abundance. We have also identified polyadenylation sites downstream of the 4.1- and 5.6-kilobase sites which display consensus hexanucleotide signals and correspond to messenger species too rare for detection by routine RNA blot analysis. Our data bring to 11 the number of known functional poly(A) addition sites associated with the DHFR gene.

We observed equimolar transcription throughout the 35-kilobase mouse dihydrofolate reductase structural gene. Transcription termination occurred within a discrete region (900 base pairs) located 1 kilobase beyond the last of seven functional polyadenylation sites and near a repetitive DNA sequence element. The results imply that a distinct genetic signal may be associated with the process of transcription termination.

Infection of human cells by adenovirus results in multiple alterations of host gene expression. To examine the effects of viral infection on the expression of a single gene, a line of human cells was developed which is resistant to growth in methotrexate and which contains amplified RNA and protein specific for dihydrofolate reductase (DHFR). Cytogenetic evidence indicated the presence of amplified DNA. Adenovirus infection of these cells caused an induction and subsequent decline in the synthesis of DHFR protein. The maximum DHFR induction occurred 16 to 19 h after infection and reached a level 2.5-fold greater than that observed in uninfected cells. Induction of DHFR protein synthesis was accompanied by concomitant increases in the level of steady-state DHFR-specific cytoplasmic RNA. The relative rate of DHFR mRNA production (i.e., the appearance of DHFR-specific mRNA sequences in the cytoplasm) also increased 2.5-fold during induction. Later in infection, the relative rate of DHFR protein synthesis declined, reaching a level below that observed in uninfected cells. This decline was accompanied by a similar decline in the steady-state levels of DHFR RNA and in the relative rate of synthesis of DHFR mRNA. These data suggest that adenovirus infection controls DHFR gene expression by increasing and subsequently decreasing the relative rate at which DHFR-specific mRNA sequences appear in the cytoplasm and enter the pool of mRNA available for translation.

A previously isolated mouse Cl-1D derived cell line (B-1/25) overproduces adenosine deaminase (EC 3.5.4.4) by 3200-fold. The present studies were undertaken to determine the molecular basis of this phenomenon. Rabbit reticulocyte lysate and Xenopus oocyte translation studies indicated that the B-1/25 cells also overproduced adenosine deaminase mRNA. Total poly(A+) RNA derived from B-1/25 was used to construct a cDNA library. After prehybridization with excess parental Cl-1D RNA to selectively prehybridize nonamplified sequences, 32P-labeled cDNA probe synthesized from B-1/25 total poly(A+) RNA was used to identify recombinant colonies containing amplified mRNA sequences. Positive clones containing adenosine deaminase gene sequences were identified by blot hybridization analysis and hybridization-selected translation in both rabbit reticulocyte lysate and Xenopus oocyte translation systems. Adenosine deaminase cDNA clones hybridized with three poly(A+) RNA species of 1.5, 1.7, and 5.2 kilobases in length, all of which were overproduced in the B-1/25 cell line. Dot blot hybridization analysis using an adenosine deaminase cDNA clone showed that the elevated adenosine deaminase level in the B-1/25 cells was fully accounted for by an increase in adenosine deaminase gene copy number. The adenosine deaminase cDNA probes and the cell lines with amplified adenosine deaminase genes should prove extremely useful in studying the structure and regulation of the adenosine deaminase gene.