Antioxidants stimulate transcriptional activation of the c-fos gene by multiple pathways in human fetal lung fibroblasts (WI-38).

We have examined the effects of three structurally distinct antioxidants (N-acetylcysteine [NAC], Trolox C [a water-soluble vitamin E derivative], and nordihydroguaiaretic acid [NGA]) on the expression of the c-fos gene over a 2-hour period. Determination of cellular glutathione concentration (the primary determinant of the cellular redox state) over the same time-course verifies that all the compounds studied cause an increase in cellular reduction potential. The level of c-fos messenger RNA increased rapidly in response to micromolar concentrations of these compounds, reaching a peak in 30-60 minutes. Induction of c-fos expression by these antioxidants is at least partly due to an increase in transcription, as determined by nuclear run-on assay. Down regulation of protein kinase C (PKC) by pretreatment for 24 hours with 500 nm PMA prevents induction by subsequent stimulation with either PMA or NGA. NAC induction of c-fos is unaffected by PMA pretreatment, while Trolox C superinduced c-fos following PMA pretreatment. None of these treatments stimulated translocation of PKC-alpha from the cytosol to the membrane. These results suggest that increasing the intracellular reducing potential induces c-fos expression through multiple pathways.

Abundance of alpha 1(I) and alpha 1(III) procollagen and p21 mRNAs in fibroblasts cultured from fetal and postnatal dermis.

The steady-state abundance of alpha 1(I) and alpha 1(III) procollagen mRNAs, p21Sdi1 mRNA, and beta-actin mRNA was determined in 29 skin fibroblast lines established from fetal, young and old donors. Donor ages ranged from 12 gestational weeks to nonagenarian. Adult donors were members of the Baltimore Longitudinal Study of Aging. The abundance of alpha 1(I) procollagen mRNA was decreased in cell lines from both young and old donors compared with fetal lines. Additionally, one alpha 1(I) transcript observed in the fetal lines was not detected in postnatal lines. The abundance of alpha 1(III) procollagen mRNA was decreased in postnatal lines from old donors compared with fetal lines. The abundance of beta-actin mRNA was lower in postnatal lines from both young and old donors compared to fetal lines. These results suggest that cultures of fetal skin fibroblasts exhibit a greater capacity for synthesis of procollagens and beta-actin than postnatal lines. In contrast, the abundance of p21Sdi1 mRNA was elevated in lines established from postnatal donors. These results are consistent with developmental changes in amounts of procollagen, beta-actin and p21.

Construction of a viable BHV1 mutant lacking most of the short unique region.

The 8.2 kb HindIII K-fragment of bovine herpesvirus 1 (BHV1) lies entirely within the short unique region of the genome and contains all or parts of 7 coding regions. We have probed this fragment for the presence of nonessential genes by a simple, rapid method of insertional mutagenesis. Our analysis indicates that all the genes present in the K-fragment, except the glycoprotein D gene, are nonessential for replication of BHV1 in tissue culture. After inserting a copy of the glycoprotein D gene into the thymidine kinase locus of BHV1 it was possible to delete the entire HindIII K-fragment and the contiguous 0.36 kb O-fragment in one step. The deletion mutant, which contains 7 kb less BHV1 DNA than wt virus and lacks ORF1, US2, the genes for protein kinase, glycoprotein G, glycoprotein I, and most of glycoprotein E was still replication-competent.

The steady-state levels of type I collagen mRNA are reduced in senescent fibroblasts.

The decreased collagen content in aging skin could be a consequence of decreased synthesis or increased degradation. The possibility that decreased synthesis of collagen results from decreased synthesis of mRNAs for Type I collagen, the major collagen in skin, was investigated by assessing the steady-state levels of alpha 1(I) and alpha 2(I) collagen mRNAs in actively proliferating and senescent WI-38 fibroblasts. The levels of both alpha 1(I) mRNA and alpha 2(I) mRNA were significantly lower in senescent fibroblasts, suggesting that one factor contributing to the decreased collagen content of aging skin may be decreased synthesis of these collagen mRNAs by senescent fibroblasts. Both mRNAs were reduced to the same extent, suggesting that coordinate regulation of the two Type I collagen genes is maintained in senescent fibroblasts.

Genes for collagen types I, IV, and V are transcribed in HeLa cells but a postinitiation block prevents the accumulation of type I mRNA.

Collagen mRNA synthesis in HeLa cells was evaluated by in vitro transcription of type I collagen DNA, nuclear run-on studies, and steady-state mRNA analysis. Type I collagen mRNA was accurately initiated by HeLa cell RNA polymerase II in nuclear extracts, and run-on analysis indicted that mRNAs for collagen types alpha 1(I), alpha 2(I), alpha 1(III), alpha 1(IV), and alpha 2(V) were synthesized in HeLa cells. However, on assessing the steady-state levels of mRNAs of collagen types alpha 1(I), alpha 2(I), alpha 1(IV), and alpha 2(V), no type I mRNA was found in HeLa cells while types alpha 1(IV) and alpha 2(V) collagen mRNAs were observed. These results suggest that a postinitiation process prevents the accumulation of type I collagen mRNAs in HeLa cells. Persistence of types IV and V collagen mRNAs is consistent with the involvement of types IV and V collagen in adhesion of HeLa cells to glass or plastic.

Transcription of genomic bovine and Xenopus laevis DNA species by RNA polymerase III in HeLa-cell cytosol extracts.

RNA transcribed from genomic Xenopus laevis DNA by RNA polymerase III in HeLa-cell extract was found in discrete size classes and was transcribed from at least two different Xenopus repeat DNA species. Very little 5S ribosomal RNA was transcribed, contrasting with results obtained on transcription of genomic Xenopus DNA by Xenopus extract [Bogenhagen et al. (1982) Cell 28, 413-421]. The low transcription was not due to an inability to use 5S rDNA templates, since the cloned Xenopus 5S ribosomal gene and pseudogene were effective templates for RNA polymerase III in HeLa extract. RNA transcribed from genomic bovine DNA by RNA polymerase III in HeLa-cell cytosol extract consisted of 120-nucleotide RNA and a larger amount of heterogeneously sized RNA (180-650 nucleotides). Only a small portion of the 120-nucleotide RNA was 5S rRNA. Most of the 120-nucleotide RNA and the larger RNA species were transcribed from one bovine repeat DNA. Genes for 5S rRNA and bovine repeat DNA were transcribed roughly in proportion to their frequency in Bos, contrasting with results in a homologous system in which transcription of repeat genes is repressed [Furth (1985) Biochem. Biophys. Res. Commun. 131, 551-556]. Bovine 5S rRNA genes appear to be concentrated on one DNA fragment obtained by restriction-enzyme-HindIII digestion of genomic bovine DNA.

Transcription of repeat DNA in vitro by RNA polymerase III in cytosol extracts is inhibited when the extract is of the same species as the DNA.

Cytosol extracts containing RNA Polymerase III were obtained from cultured human (HeLa) and bovine (MDBK) cells. Human extract efficiently transcribed genomic bovine DNA and cloned bovine alu-type repeat DNA while genomic human DNA and cloned human alu repeat DNA were inefficient templates. Conversely, genomic bovine DNA and cloned bovine alu-type repeat DNA were less effective templates for bovine extract than were genomic human DNA and alu repeat DNA. These results suggest that species specific factor(s) present in mammalian cells are responsible for the observed in vivo inhibition of transcription of most alu-type repeat DNA.

Transcription of 5S RNA by RNA polymerase III from genomic bovine DNA templates.

RNA polymerase III in an extract of HeLa cells transcribes RNA 5S in size from genomic bovine DNA template. This RNA represents a major fraction of the RNA synthesized. 5S RNA is not transcribed from bovine chromatin at equivalent concentrations and RNA the size of tRNA or tRNA precursor is not detected. At low UTP concentrations RNA slightly smaller in size than 5S is synthesized in addition to 5S RNA.

Transcription of bovine satellite DNA by purified RNA polymerase III.

Purified calf thymus RNA polymerase III synthesizes, from calf thymus DNA template, RNA which hybridizes to the major repeated sequence of Eco R1-digested calf thymus DNA. Similar results are obtained with RNA transcribed from calf thymus chromatin. It is suggested that this DNA sequence, which is derived from bovine satellite DNA, may be genetically active.

Enzymic polyadenylation of 5S ribosomal ribonucleic acid and synthesis of a complementary deoxyribonucleic acid.

The 5S ribosomal RNA has been isolated, pure and intact, from rat liver (5 mg of 5S RNA from 150g of liver). The 5S RNA serves as a primer for calf thymus poly(A) polymerase with 20% of the efficiency of (Ap)3A. Bacterial 5S RNA and transfer RNA also serve as primers; rat liver 18S and 28S ribosomal RNAs support poly(A) synthesis poorly. Neither the 5S RNA primer nor the appended poly(A) tract is nicked or degraded by poly(A) polymerase, and initiation of poly(A) tracts on 5S RNA primers continues throughout the reaction period. The rate of initiation is dependent on the enzyme concentration; the ATP concentration affects the rate of elongation. The polyadenylated material increases in size over time, with the largest material reaching a size of 6.8 S in 5 h, corresponding to an appended poly(A) tract of 140 nucleotides. Using polyadenylated 5S RNA, oliog(dTY as primer, and avian myeloblastosis virus reverse transcriptase, we synthesized DNA complementary to 5S RNA. The complementary DNA has an apparent molecular weight (in alkaline sucrose gradients) of 4.3 X 10(4). Base composition analysis and nearest-neighbor analysis of the DNA are as expected for a complement of 5S RNA, indicating that the entire 5S sequence is copied. The complementary DNA hybridizes to 5S RNA with a R0t1/2 of 8.9 X 10(-4) mol.s.L-1. No hybrid is formed with Escherichia coli 16S and 23S ribosomal RNA, E. coli 5S ribosomal RNA, yeast transfer RNA, rat liver transfer RNA, or rat liver 18S and 28S RIBOSOMAL RNA. The Tm of the 5S RNA:5S DNA hybrid in 15 mM NaCl containing 1.5 mM sodium citrate is 74 degrees C, 2.5 degrees C below the theoretical melting temperature of a DNA duplex of 60% G + C. Analysis of the hybrid in buoyant density gradients also indicates that hybridization is both specific and precise. The complementary DNA anneals to calf thymus, rat liver, and salmon sperm DNAs but not to E. coli DNA. Annealing of 5S cDNA to calf thymus DNA with a C0t1/2 of 2.1 suggests that there are several thousand 5S RNA genes in the calf thymus genome (haploid). At least that number of 5S RNA genes is present in the salmon sperm genome.

Purification and characterization of a DNA single strand specific endonculease from human cells.

An endonuclease with DNA single-strand specificity has been purified from KB cells. The enzyme has a pH optimum at 9.2, requires Mg2+ for activity, and is inhibited by mono- or divalent cations. Its sedimentation coefficient of 4.6 S is based on sucrose gradient sedimentation, and it has a molecular weight of 54 000 as determined by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. The enzyme specifically catalyzes the endonucleolytic cleavage of denatured DNA, yielding acid-soluble oligonucleotides which contain 5'-phosphoryl termini. The rate of hydrolysis of poly(dT) is approximately eightfold greater than that observed with denatured DNA, although the Km for both substrates is 1.74 X 10(-5) M. The relative rates of hydrolysis of homopolymers by the endonuclease are: poly(dG) greater than poly(dT) greater than poly(dA) greater than poly (dC). Purified enzyme preparations also hydrolyze poly(U), releasing acid-soluble products. This activity cosediments in sucrose gradients with the DNA endonuclease activity, suggesting that both activities are contained in the same enzyme molecule.

Composition and template activity of chromatin fractionated by isoelectric focusing.

HeLa cell interphase chromatin has been sheared and fractionated by isoelectric focusing. Chromatin fractions are obtained with a wide range of isoelectric points. No free DNA is observed. While protein/DNA rations are similar in the various fractions, they appear to contain different nonhistone chromosomal proteins. A minor chromatin fraction with isoelectric point congruent to 7.0 does not contain histone H1. This fraction is considerably more active as template with different RNA polymerases than the other fractions. Kinetic studies, in which RNA polymerase activity is assayed at various concentrations of chromatin, indicate that the greater activity of Escherichia coli RNA polymerase is due to an increased rate of transcription at saturating concentrations of template (Vmax) and is not due to a lower concentration required for half-maximal rate of transciption (Km). In contrast, the increased rate of transcription by calf-thymus RNA polymerases II and III is due to a decrease in chromatin concentration required for half-maximal rate of transcription rather than an increased rate of transcription at saturating concentrations of template. These results suggest that chromatin with isoelectric point congruent to 7 offers a greater frequency of binding sites for mammalian RNA polymerases, as would be expected for a "transcriptionally active" fraction.

RNA synthesized in vitro by calf thymus RNA polymerase III (C), as well as by E. coli RNA polymerase, is restricted to a subset of calf thymus DNA.

RNA synthesized in vitro from chromatin and DNA by calf thymus RNA polymerase III was evaluated by hybridization in vast DNA excess. The RNA contains RNA complementary to both moderately repeated and unique DNA sequences. Very highly repeated DNA is not transcribed. A greater portion of RNA transcribed from DNA by RNA polymerase III hybridizes to moderately repeated DNA than RNA transcribed by Escherichia coli RNA polymerase. In studies utilizing DNA absorbed to filters, RNA transcribed from chromatin in short incubations hybridized to a greater extent than RNA transcribed for longer times. Similar results were obtained with RNA transcribed from DNA by E. coli RNA polymerase. These results suggest: 1) RNA polymerase III may be responsible for the synthesis of RNA species in addition to tRNA and 5 S ribosomal RNA and a portion of this RNA is transcribed from unique DNA; and 2) in vitro there may be selectivity in the initiation of transcription by both E. coli RNA polymerase and calf thymus RNA polymerase III.

Mammalian endonuclease, DNase V. Purification and properties of enzyme of calf thymus.

An endonuclease present in partially purified preparations of calf thymus DNA polymerase has been purified to homogeneity. It has a molecular weight of 53,000 +/- 2,500 as determined by sucrose gradient sedimentation. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate indicates the protein is composed of four subunits, each polypeptide possessing a molecular weight of 13,000. Its isoelectric point is 10.3 +/- 0.2. The endonuclease has a pH optimum at 6.6, requires Mg2+ or Mn2+ for activity, and does not attack RNA. The enzyme appears to be present in tissues other than calf thymus. The enzyme catalyzes the endonucleolytic cleavage of both denatured and native eukaryotic DNA. The enzyme introduces a limited number of single strand nicks into native DNA; hydrolysis of denatured DNA produces acid-soluble oligonucleotides. The average size of the limit product, sedimented in an alkaline sucrose gradient, is 1200 nucleotides for native DNA. The product contains 5'-phosphoryl and 3'-hydroxyl termini. While all four deoxynucleotides are found at the 5' termini, pyrimidine residues predominate. Calf thymus DNase V degrades closed circular duplex SV40 DNA and glucosylated T4DNA but not poly(dA-dT). The rate of hydrolysis of homopolymers is: poly(dT) greater than poly(dA) greater than poly(dC) greater than poly(dG) in the presence of Mg2+, and poly(dT) greater than poly(dC) greater than poly(dA) = poly(dG) in the presence of Mn2+.

Inhibition of mammalian polyadenylate polymerase by 2-aza-1,N6-etheno-adenosine triphosphate.

2-Aza-1,N6-etheno-adenosine triphosphate (aza-epsilonATP), a fluorescent analog of adenosine triphosphate, significantly inhibits polyadenylate [poly(A)] polymerase of bovine lymphosarcoma and calf thymus, with 50% inhibition at 200 muM (in the presence of an equal concentration of adenosine triphosphate). Calf thymus RNA polymerases II and III are inhibited 32 and 20%, respectively, by a 3.8-fold excess of aza-epsilonATP; DNA polymerase alpha is not inhibited. The inhibition of poly(A) polymerase by aza-epsilonATP appears to be competitive with adenosine triphosphate; incorporation of aza-epsilonATP is not observed. Polymers of 2-aza 1,N6-etheno-adenosine monophosphate are used as primers, but pootly. 1,N-Etheno-adenosine triphosphate and 9-beta-D-arabinofuranosyladenine triphosphate are poor inhibitors of poly(A) polymerase; adenosine diphosphate is ineffective. Deoxyadenosine triphosphate inhibits to the same extent as aza-epsilonATP, while other naturally occurring nucleotides inhibit poly(A) polymerase to varying degrees, with deoxynucleoside triphosphates more potent than ribonucleoside triphosphates. Inhibition of poly(A) polymerase by naturally occurring nucleoside triphosphates suggests that nucleotides may regulate the enzyme in vivo; inhibition by the fluorescent analog aza-epsilonATP suggests that this compound may be useful in elucidating poly(A) metabolism in both normal and neoplastic cells.

Duplication of single stranded DNA catalyzed by calf thymus DNA polymerase alpha.

Purified calf thymus DNA polymerase alpha is inactive with native DNA as template and shows little activity with denatured DNA. DNA synthesis with denatured DNA as template is greatly stimulated by the addition of a nuclease which initially copurifies with DNA polymerase but is separated from the polymerase on DEAE-cellulose chromatography. A limit digest of nuclease treated native DNA which is then denatured is replicated 80-95%; extensive replication is also obtained with native DNA partially degraded by pancreatic DNase and then denatured. The product of the reaction with calf thymus nuclease-treated DNA as template is double-stranded DNA with a hairpin (looped back) structure.