Characterization of rat ribosomal DNA. The highly repetitive sequences that flank the ribosomal RNA transcription unit are homologous and contain RNA polymerase III transcription initiation sites

Nucleolar DNA: the host and the guests

Nucleoli are formed on the basis of ribosomal genes coding for RNAs of ribosomal particles, but also include a great variety of other DNA regions. In this article, we discuss the characteristics of ribosomal DNA: the structure of the rDNA locus, complex organization and functions of the intergenic spacer, multiplicity of gene copies in one cell, selective silencing of genes and whole gene clusters, relation to components of nucleolar ultrastructure, specific problems associated with replication. We also review current data on the role of non-ribosomal DNA in the organization and function of nucleoli. Finally, we discuss probable causes preventing efficient visualization of DNA in nucleoli.

Peroxynitrite modulates MnSOD gene expression in lung epithelial cells

Peroxynitrite (ONOO-) is a strong oxidant derived from nitric oxide ('NO) and superoxide (O2.-), reactive nitrogen (RNS) and oxygen species (ROS) present in inflamed tissue. Other oxidant stresses, e.g., TNF-alpha and hyperoxia, induce mitochondrial, manganese-containing superoxide dismutase (MnSOD) gene expression. These experiments tested whether ONOO regulated MnSOD gene expression in human lung epithelial (A549) cells. 3-morpholinosydnonimine HCI (SIN-1) (10 or 1000 microM) increased MnSOD mRNA, but did not change hypoxanthine guanine phosphoribosyl transferase (HPRT) mRNA. Authentic peroxynitrite (ONOO ) (100-500 microM) also increased MnSOD mRNA but did not change constitutive HPRT mRNA expression. ONOO stimulated luciferase gene expression driven by a 2.5 kb fragment of the rat MnSOD gene 5' promoter region. MnSOD gene induction due to ONOO- was inhibited effectively by L-cysteine (10 mM) and partially inhibited by N-acetyl cysteine (50 mM) or pyrrole dithiocarbamate (10 mM). .NO from 1-propanamine, 3-(2-hydroxy-2-nitroso-1-propylhydrazine) (PAPA NONOate) (100 or 1000 microM) did not change MnSOD or HPRT mRNA. Neither H202 nor NO2-, breakdown products of SIN-1 and ONOO , had any effect on MnSOD mRNA expression; however, ONOO- and SIN-1 did not increase MnSOD protein content detectable by western blots, nor did they increase MnSOD enzymatic activity. Increased steady state [O2.-] in the presence of .NO yields ONOO , and ONOO has direct, stimulatory effects on MnSOD transcript expression.

Differential activation of myc gene family members in hepatic carcinogenesis by closely related hepatitis B viruses

Woodchucks infected with woodchuck hepatitis virus (WHV) and ground squirrels infected with ground squirrel hepatitis virus (GSHV) both develop hepatocellular carcinoma (HCC), but WHV-associated tumors arise more frequently and much earlier in life. These differences are preserved when the oncogenic potentials of the two viruses are examined in the same host (woodchucks). We examined RNA and genomic DNA from tumors arising from WHV- and GSHV-infected woodchucks to determine whether these viruses use the same oncogenic pathway. N-myc RNA was not expressed in normal liver but was expressed in 10 of 13 WHV-associated HCCs examined. Southern blot analysis showed that 7 of 17 WHV-induced tumors (41%) contained rearrangements at N-myc loci due to viral genomic integration. Six of these seven inserts affected N-myc2, and most of these were at the 5' end of the gene. In contrast, only two of seven GSHV-induced woodchuck HCCs expressed N-myc RNA, and only 1 of the 16 tumors (6%) contained a rearranged N-myc allele. The GSHV-associated HCCs all contained numerous viral insertions, so the low frequency of integration into N-myc loci by GSHV was not due to a general block to integration. Four of sixteen GSHV-induced tumors harbored amplified c-myc alleles, and five of seven GSHV tumors tested contained elevated c-myc RNA levels. By contrast, enhanced c-myc RNA levels were observed in only 2 of 13 WHV-induced HCC. We conclude that N-myc overexpression is a regular feature of WHV- but not GSHV-associated hepatocarcinogenesis in a common host. In contrast, c-myc transcriptional deregulation is rarely encountered in WHV-induced HCC but is frequent in GSHV-induced HCC.

Differential activation of myc gene family members in hepatic carcinogenesis by closely related hepatitis B viruses

Woodchucks infected with woodchuck hepatitis virus (WHV) and ground squirrels infected with ground squirrel hepatitis virus (GSHV) both develop hepatocellular carcinoma (HCC), but WHV-associated tumors arise more frequently and much earlier in life. These differences are preserved when the oncogenic potentials of the two viruses are examined in the same host (woodchucks). We examined RNA and genomic DNA from tumors arising from WHV- and GSHV-infected woodchucks to determine whether these viruses use the same oncogenic pathway. N-myc RNA was not expressed in normal liver but was expressed in 10 of 13 WHV-associated HCCs examined. Southern blot analysis showed that 7 of 17 WHV-induced tumors (41%) contained rearrangements at N-myc loci due to viral genomic integration. Six of these seven inserts affected N-myc2, and most of these were at the 5' end of the gene. In contrast, only two of seven GSHV-induced woodchuck HCCs expressed N-myc RNA, and only 1 of the 16 tumors (6%) contained a rearranged N-myc allele. The GSHV-associated HCCs all contained numerous viral insertions, so the low frequency of integration into N-myc loci by GSHV was not due to a general block to integration. Four of sixteen GSHV-induced tumors harbored amplified c-myc alleles, and five of seven GSHV tumors tested contained elevated c-myc RNA levels. By contrast, enhanced c-myc RNA levels were observed in only 2 of 13 WHV-induced HCC. We conclude that N-myc overexpression is a regular feature of WHV- but not GSHV-associated hepatocarcinogenesis in a common host. In contrast, c-myc transcriptional deregulation is rarely encountered in WHV-induced HCC but is frequent in GSHV-induced HCC.

Structural alterations of the ribosomal RNA genes in leukemic cells

Cloned 6.7 kb EcoR1 fragment of mice rDNA was used as a hybridization probe for rDNA structure analysis in mice, rat and calf haemopoietic tumor and normal cells. EcoR1, BglII and Pst1 restriction fragment length polymorphism (RFLP) was found in neoplastic rDNA and was not revealed in normal ones. The rRNA gene rearrangements were observed not only in spacer region but in coding sequences of the genes. Leukemic cells reveal also rDNA amplification. A role of genetic rearrangements of rDNA for mechanisms of carcinogenesis is suggested.

Regulation of murine type 1 plasminogen activator inhibitor gene expression in vivo. Tissue specificity and induction by lipopolysaccharide, tumor necrosis factor-alpha, and transforming growth factor-beta

The regulation of type 1 plasminogen activator inhibitor (PAI-1) gene expression was studied in vivo employing a murine model system. Nuclease protection analysis revealed relatively high concentrations of PAI-1 mRNA in the aorta, adipose tissue, heart, and lungs of untreated CB6 (BalbC X C57B16) mice. Treatment of CB6 mice with LPS, TNF-alpha, or transforming growth factor-beta (TGF-beta) increased the steady-state levels of PAI-1 mRNA within 3 h in all tissues examined. However, the greatest responses to TGF-beta were observed in adipose tissue and the kidney, while LPS and TNF-alpha strongly stimulated PAI-1 gene expression in the liver, kidney, lung, and adrenals. In C3H/HeJ mice, which exhibit defective TNF-alpha release in response to LPS, the response of the PAI-1 gene to LPS was severely attenuated. However, injection of these mice with TNF-alpha increased PAI-1 mRNA in a tissue-specific pattern strikingly similar to that observed in LPS-treated CB6 mice. These results demonstrate that the PAI-1 gene is regulated in a complex and tissue-specific manner in vivo, and suggest a role for TNF-alpha in the response of the PAI-1 gene to sepsis.

Hepatic transcription of the acute-phase alpha 1-inhibitor III gene is controlled by a novel combination of cis-acting regulatory elements

mRNA coding for the abundant broad-range plasma proteinase inhibitor alpha 1-inhibitor III (alpha 1I3) was detected only in rat liver, while mRNA for the related proteins alpha 1-macroglobulin and alpha 2-macroglobulin was also found in a variety of nonhepatic tissues. cis-Acting control elements necessary for the hepatic transcription of alpha 1I3 were mapped by transfection and expression studies of control-region constructs in cultured hepatic and nonhepatic cells. The promoter-proximal 5'-flanking region contained four control elements, I to IV, located between -109 and -196 base pairs upstream of the transcriptional start site relevant for the hepatic transcription of this gene. Elements II and III were essential, and I and IV exerted strong modulatory effects. Elements I to III acted as positive regulators, and IV acted as a negative element. Element II contained the sequence TGGCA and is probably a binding site for a nuclear factor related to the known transcription factor NF1. The other three elements did not resemble consensus binding sites for known transcription factors that are involved in the hepatocyte-specific transcription of other well-characterized plasma protein genes, such as the prototype factor HNF-1. Thus, the alpha 1I3 gene achieves its highly hepatocyte-specific transcription through a novel combination of cis-acting control elements and trans-acting factors.

Hepatic transcription of the acute-phase alpha 1-inhibitor III gene is controlled by a novel combination of cis-acting regulatory elements

mRNA coding for the abundant broad-range plasma proteinase inhibitor alpha 1-inhibitor III (alpha 1I3) was detected only in rat liver, while mRNA for the related proteins alpha 1-macroglobulin and alpha 2-macroglobulin was also found in a variety of nonhepatic tissues. cis-Acting control elements necessary for the hepatic transcription of alpha 1I3 were mapped by transfection and expression studies of control-region constructs in cultured hepatic and nonhepatic cells. The promoter-proximal 5'-flanking region contained four control elements, I to IV, located between -109 and -196 base pairs upstream of the transcriptional start site relevant for the hepatic transcription of this gene. Elements II and III were essential, and I and IV exerted strong modulatory effects. Elements I to III acted as positive regulators, and IV acted as a negative element. Element II contained the sequence TGGCA and is probably a binding site for a nuclear factor related to the known transcription factor NF1. The other three elements did not resemble consensus binding sites for known transcription factors that are involved in the hepatocyte-specific transcription of other well-characterized plasma protein genes, such as the prototype factor HNF-1. Thus, the alpha 1I3 gene achieves its highly hepatocyte-specific transcription through a novel combination of cis-acting control elements and trans-acting factors.

The 5' end of U3 snRNA can be crosslinked in vivo to the external transcribed spacer of rat ribosomal RNA precursors

From previous work it was known that U3 RNA is hydrogen bonded to nucleolar 28 S to 35 S RNA and can be covalently crosslinked to RNA of greater than 28 S by irradiation in vivo with long-wave ultraviolet light in the presence of 4'-aminomethyl-4,5',8-trimethylpsoralen (AMT psoralen). Here we use a novel sandwich blot technique to identify these large nucleolar RNA species as rRNA precursors and to map the site(s) of crosslinking in vivo. The crosslink occurs between one or more residues near the 5' end of U3 RNA and a 380 nucleotide region of the rat rRNA external transcribed spacer (ETS1). We have sequenced this region of the rat ETS and we show that it includes an RNA-processing site analogous to those previously mapped to approximately 3.5 kb upstream from the 5' end of mouse and human 18 S rRNAs.

New tandem repeat region in the non-transcribed spacer of human ribosomal RNA gene

A new repetitive DNA region was identified in the non-transcribed spacer of human rDNA, namely a long (4.6 kb) sequence motif (Xbal element) was present in two copies. The repeating unit composed of two parts. One of them consisted of unique nucleotide sequences, interrupted by some simple sequences. The other, about 3.1 kb long one assembled only from highly repeated simple sequences. The unique sequence region contained two, inverted copies of the human AluI type repetitive DNA family. The authors suggest that the XbaI elements may flank the tandem arrays of human rRNA genes as terminal repeats and they might function both as the origin of rDNA replication and/or site of homologous recombination.

Purification and characterization of a high-mobility-group-like DNA-binding protein that stimulates rRNA synthesis in vitro

A 16,000-dalton, high-mobility-group-like (HMG-like) DNA-binding protein, referred to as p16, has been purified to homogeneity from Novikoff hepatoma ascites cells. p16 binds specifically to a portion of the 5' flanking region of the rat rRNA gene (-620 to -417), which is part of the upstream activator sequence identified previously (B. G. Cassidy, H.-F. Yang-Yen, and L. I. Rothblum, Mol. Cell. Biol. 6:2766-2773, 1986). p16 also binds to a segment of the external transcribed spacer (+352 to +545). In vitro reconstituted transcription experiments demonstrated that the addition of p16 stimulated rRNA synthesis up to ca. fourfold. The stimulation was dose dependent and saturable. The effect of p16 on ribosomal gene transcription was also dependent on the presence of either the upstream or the downstream DNA-binding site, or both. The amino acid composition of p16 is very similar to that of HMG-I, suggesting that p16 may be a member of the HMG-I family of proteins. In this case, our results suggest that HMG proteins may play an important role in the regulation of the rRNA gene expression.

Purification and characterization of a high-mobility-group-like DNA-binding protein that stimulates rRNA synthesis in vitro

A 16,000-dalton, high-mobility-group-like (HMG-like) DNA-binding protein, referred to as p16, has been purified to homogeneity from Novikoff hepatoma ascites cells. p16 binds specifically to a portion of the 5' flanking region of the rat rRNA gene (-620 to -417), which is part of the upstream activator sequence identified previously (B. G. Cassidy, H.-F. Yang-Yen, and L. I. Rothblum, Mol. Cell. Biol. 6:2766-2773, 1986). p16 also binds to a segment of the external transcribed spacer (+352 to +545). In vitro reconstituted transcription experiments demonstrated that the addition of p16 stimulated rRNA synthesis up to ca. fourfold. The stimulation was dose dependent and saturable. The effect of p16 on ribosomal gene transcription was also dependent on the presence of either the upstream or the downstream DNA-binding site, or both. The amino acid composition of p16 is very similar to that of HMG-I, suggesting that p16 may be a member of the HMG-I family of proteins. In this case, our results suggest that HMG proteins may play an important role in the regulation of the rRNA gene expression.

Ribosomal DNA sequences attached to the nuclear matrix

The organization of rat liver ribosomal DNA (rDNA) as matrix-attached DNA loops was examined using a protocol which fractionates chromatin from discrete regions of DNA loops. Southern blot analysis of matrix-attached and solubilized chromatin DNA fragments demonstrated that rDNA is associated with the matrix via its 5' and 3' nontranscribed spacer sequences (NTS). Although the 45 S rRNA coding sequences were approximately threefold enriched in matrix preparations, the recovery of this DNA (unlike the NTS) was dependent on the extent of nuclease digest and proportional to the length of the matrix-attached DNA fragments. The data suggest that rDNA is organized as matrix-attached DNA loops and only the NTS are directly involved in matrix binding. Further, we demonstrated that while the kinetics and extent of nuclease digestion were similar in all regions of the DNA loops, the nuclease digestion pattern of bulk nuclear and matrix DNA showed a typical nucleosome organization, but the rDNA fragments retained with the nuclear matrix did not.

Transcription of spacer sequences flanking the rat 45S ribosomal DNA gene

The transcriptional activity of spacer sequences flanking the rat 45S ribosomal DNA (rDNA) gene were studied. Nascent RNA labeled in in vitro nuclear run-on reactions hybridized with both 5' and 3' spacer regions. The highest level of hybridization was seen with an rDNA fragment containing tandem repeats of a 130-base-pair sequence upstream of the 45S rRNA initiation site. Synthesis of RNA transcripts homologous to this internally repetitious spacer region was insensitive to high levels of alpha-amanitin, suggesting that it is mediated by RNA polymerase I. Analysis of steady-state RNA showed that these transcripts were present at extremely low levels in vivo relative to precursor rRNA transcripts. In contrast, precursor and spacer run-on RNAs were synthesized at similar levels. This suggests that spacer transcripts are highly unstable in vivo; therefore, it may be the process of transcription rather than the presence of spacer transcripts that is functionally important. Transcription in this upstream rDNA region may be involved in regulation of 45S rRNA synthesis in rodents, as has been suggested previously for frog rRNA. In addition, the presence of transcriptional activity in other regions of the spacer suggests that some polymerase I molecules may transcribe through the spacer from one 45S gene to the next on rodent rDNA.

The evolutionarily conserved repetitive sequence d(TG.AC)n promotes reciprocal exchange and generates unusual recombinant tetrads during yeast meiosis

We have studied the genetic behavior of the alternating copolymer d(TG.AC)n inserted into a defined position in the genome of the yeast Saccharomyces cerevisiae. When d(TG.AC)n sequences were present at the HIS3 locus on homologous chromosomes, diploid cells undergoing meiosis generated an excess of tetrads containing reciprocally recombined products with crossover points close to the repetitive DNA insert. Most of these tetrads exhibited gene conversion of a d(TG.AC)n insert. However, the insertion of d(TG.AC)n sequences had no effect on the frequency of gene conversion of closely linked marker genes. Surprisingly, when d(TG.AC)n sequences were present on only one homolog at the HIS3 locus, one-half of the tetrads exhibiting nonparental segregation for marker genes that flanked the repetitive DNA insert were very unusual and appeared to have arisen by multiple recombination events in the vicinity of the d(TG.AC)n insert. Similar multiply recombinant tetrads were seen in crosses in which d(TG.AC)n sequences were present on both homologs. Combined, the data strongly suggest that d(TG.AC)n sequences significantly enhance reciprocal meiotic recombination and may be important in causing multiple recombination events to occur within a relatively small region of the yeast chromosome. Molecular evidence is presented that clearly documents the postmeiotic segregation of an 80-base stretch of d(TG.AC)n.

Structure of the Bombyx mori rDNA: initiation site for its transcription

The initiation site of rDNA transcription in Bombyx mori was determined to be located at 909bp upstream from the 5'- end of mature 18s rRNA by S1-nuclease mapping and primer extension experiment. An in vitro transcription system, which was constructed using posterior silk glands of Bombyx larvae, initiated the transcription of cloned rDNA at exactly the same site as determined for the in vivo transcription above. The primary transcript seemed to be processed at about 200b downstream of the initiation site both in vivo and in vitro. Sequence analyses of the flanking region of the initiation site revealed that short repetitive sequences are widely distributed throughout the NTS region, and that a highly AT-rich region resides immediately upstream of the initiation region.

Transcriptional role for the nontranscribed spacer of rat ribosomal DNA

In vitro transcription of the rat rRNA gene led to the identification of a region within a 3.4-kilobase fragment of the nontranscribed spacer (NTS) which significantly increased the transcription of rat ribosomal DNA. Promoter constructs containing this region were transcribed up to 17-fold more efficiently in vitro than templates with only 167 or 286 base pairs of NTS. This effect was also observed when the 3.4-kb fragment of the NTS was subcloned in the opposite orientation and 4 kb upstream of the promoter. The region responsible for the enhanced level of transcription was found between -286 and -1018. The results of order-of-addition experiments suggested that the enhanced level of transcription was the result of the formation of a stable complex between a trans-acting factor and the nontranscribed spacer. DNA-protein binding assays demonstrated that the same region of the NTS determined to have enhancer activity also specifically bound a proteinase K-sensitive factor present in nuclear extracts. The sequence of this region was not found to have any significant homology with the promoter of the rat rRNA gene. This is the first report to assign a transcriptional role to the NTS of a mammalian rRNA gene.

Rat and mouse identifier sequences are preferentially but not exclusively located in cortical neuronal genes expressed postnatally

A genetic element called the identifier (ID) sequence, highly repeated in the rat genome, has previously been reported to be located in the introns of some genes transcribed in the adult rat brain by RNA polymerase II (Pol II). We show that nuclear RNA isolated from neurons of cerebral hemispheres (cortex) of 14-day old rats is enriched more than 10-fold in ID sequences compared to nuclear RNA from liver, kidney, cerebellum, or cortical glia. The developmental onset of the difference is during the first 2 weeks after birth. Mouse cortical neuronal nuclear RNA is similarly enriched in an element related but not identical to the rat ID element, and the enrichment also has postnatal onset. The enriched appearance of ID sequences in transcripts whose expression is increased postnatally in cortical neurons correlates developmentally and spatially with the transcription of ID elements by RNA polymerase III (Pol III) and with a change in chromatin structure.

Transcription of spacer sequences flanking the rat 45S ribosomal DNA gene

The transcriptional activity of spacer sequences flanking the rat 45S ribosomal DNA (rDNA) gene were studied. Nascent RNA labeled in in vitro nuclear run-on reactions hybridized with both 5' and 3' spacer regions. The highest level of hybridization was seen with an rDNA fragment containing tandem repeats of a 130-base-pair sequence upstream of the 45S rRNA initiation site. Synthesis of RNA transcripts homologous to this internally repetitious spacer region was insensitive to high levels of alpha-amanitin, suggesting that it is mediated by RNA polymerase I. Analysis of steady-state RNA showed that these transcripts were present at extremely low levels in vivo relative to precursor rRNA transcripts. In contrast, precursor and spacer run-on RNAs were synthesized at similar levels. This suggests that spacer transcripts are highly unstable in vivo; therefore, it may be the process of transcription rather than the presence of spacer transcripts that is functionally important. Transcription in this upstream rDNA region may be involved in regulation of 45S rRNA synthesis in rodents, as has been suggested previously for frog rRNA. In addition, the presence of transcriptional activity in other regions of the spacer suggests that some polymerase I molecules may transcribe through the spacer from one 45S gene to the next on rodent rDNA.

The evolutionarily conserved repetitive sequence d(TG.AC)n promotes reciprocal exchange and generates unusual recombinant tetrads during yeast meiosis

We have studied the genetic behavior of the alternating copolymer d(TG.AC)n inserted into a defined position in the genome of the yeast Saccharomyces cerevisiae. When d(TG.AC)n sequences were present at the HIS3 locus on homologous chromosomes, diploid cells undergoing meiosis generated an excess of tetrads containing reciprocally recombined products with crossover points close to the repetitive DNA insert. Most of these tetrads exhibited gene conversion of a d(TG.AC)n insert. However, the insertion of d(TG.AC)n sequences had no effect on the frequency of gene conversion of closely linked marker genes. Surprisingly, when d(TG.AC)n sequences were present on only one homolog at the HIS3 locus, one-half of the tetrads exhibiting nonparental segregation for marker genes that flanked the repetitive DNA insert were very unusual and appeared to have arisen by multiple recombination events in the vicinity of the d(TG.AC)n insert. Similar multiply recombinant tetrads were seen in crosses in which d(TG.AC)n sequences were present on both homologs. Combined, the data strongly suggest that d(TG.AC)n sequences significantly enhance reciprocal meiotic recombination and may be important in causing multiple recombination events to occur within a relatively small region of the yeast chromosome. Molecular evidence is presented that clearly documents the postmeiotic segregation of an 80-base stretch of d(TG.AC)n.

Transcriptional role for the nontranscribed spacer of rat ribosomal DNA

In vitro transcription of the rat rRNA gene led to the identification of a region within a 3.4-kilobase fragment of the nontranscribed spacer (NTS) which significantly increased the transcription of rat ribosomal DNA. Promoter constructs containing this region were transcribed up to 17-fold more efficiently in vitro than templates with only 167 or 286 base pairs of NTS. This effect was also observed when the 3.4-kb fragment of the NTS was subcloned in the opposite orientation and 4 kb upstream of the promoter. The region responsible for the enhanced level of transcription was found between -286 and -1018. The results of order-of-addition experiments suggested that the enhanced level of transcription was the result of the formation of a stable complex between a trans-acting factor and the nontranscribed spacer. DNA-protein binding assays demonstrated that the same region of the NTS determined to have enhancer activity also specifically bound a proteinase K-sensitive factor present in nuclear extracts. The sequence of this region was not found to have any significant homology with the promoter of the rat rRNA gene. This is the first report to assign a transcriptional role to the NTS of a mammalian rRNA gene.

Regular distribution of length heterogeneities within non-transcribed spacer regions of cloned and genomic rDNA of Saccharomyces cerevisiae

A length difference of about 50 bp in the EcoRI fragment B of the rDNA from two different strains of Saccharomyces cerevisiae has been mapped in detail by sequencing of cloned fragments. This 2.4 kb EcoRI fragment contains the start of the 35S rRNA gene at one end and the 5S rRNA gene in the middle flanked by non-transcribed spacers, NTS1 and NTS2. The difference appeared as short deletions or insertions in five regularly spaced regions within the 1 kb NTS1, 3' to the 5S rRNA gene. The same regions of heterogeneities were displayed when all available sequence data of the NTS1 were compared. Four of the variable regions are located 160-170 bp apart, indicating that they might represent linker sequences between phased nucleosomes. Two variant clones, differing in the length of one subfragment of NTS1, were isolated for each strain. In both cases these represented the major variants among chromosomal NTS1 as revealed by sequencing of genomic fragments.

Isolation of an MSP1-derived transcriptionally active nucleolar particle

A discrete macromolecular structure with inherent transcription activity has been isolated from Novikoff nucleoli. The particle (350 A diameter) was released from the intact nucleoli by digestion with restriction endonuclease (Msp1) and is enriched in repetitive DNA sequences both 5' and 3' to the ribosomal genes. In vitro transcription analyses suggest that the particle contains the rDNA polymerase template, and that its 12-15 major proteins are responsible for transcriptional activity without added soluble protein factors.

Nucleotide sequence analysis of the spacer regions flanking the rat rRNA transcription unit and identification of repetitive elements

We investigated the organization of the rat rDNA non-transcribed spacer (NTS) by determining the sequence of large NTS segments located upstream (2501 bp) and downstream (4025 bp) from the rRNA transcription unit. We identified four B2-like and two ID mobile elements. They may be grouped in three pairs with the members of each pair located in the upstream and downstream NTS. The ID sequences are identical to the consensus sequence, while the pairs of B2-like elements show 85% and 50/65% homology to the consensus B2 sequence. The proximal part of the downstream NTS contains a region of widely diverged SalI tandem repeats. A considerable part of the analyzed upstream and downstream NTS sequences is constituted by different types of simple sequences and long poly(purine) X poly(pyrimidine) tracts. These data show that the rat rDNA NTS regions flanking the rRNA transcription unit are characterized by a combination of short interspersed (B2-superfamily) and various simple sequences.

Characterization of the region around the start point of transcription of ribosomal RNA in the Chinese hamster

The initiation site for ribosomal RNA transcription in the Chinese hamster was identified and the sequence around and upstream determined. The start point region shows considerable homology with comparable regions in the mouse and rat. In the Chinese hamster, the region between bp -700 and -200 consists of imperfect repeats approximately 120-130 bp in length which are flanked by T-rich regions. The region within each repeat which is homologous with an adjacent repeat decreases in length as the start point is approached. The final promoter-proximal repeat preserves only an 11-bp region of the promoter-distal repeats. This short sequence, termed the repeat core, appears with a periodicity of about 120-130 bp in the Chinese hamster, and is conserved in both mouse and rat. In humans, a short repeated sequence occupies similar positions, suggesting that while complete 120-130-bp repeats are not a feature of all mammalian RNA polymerase I promoter-proximal r X DNA spacers, a short sequence repeating with approximate 120-130-bp periodicity may be such a feature.

Characterization of two types of rRNA gene repeat units from the crustacean Artemia

We have previously described that Artemia rRNA genes are organized with a basic repeat unit of 16.5 kb [Cruces et al., Biochem. Biophys. Res. Commun. 98 (1981) 404-409]. Here we describe the organization of the DNA coding for rRNA of a different population of this crustacean that has a repeat unit of 12.2 kb. Both types of repeat units have been cloned and the organization of the external spacers studied by restriction analysis. Both external spacers contain repeated sequences, but they are not homologous to each other. Sequences from the external spacer of the 16.5 kb repeat are also found elsewhere in the genome, within sequences not related to rRNA genes.

The ribosomal RNA genes of three salmonid species

The ribosomal RNA cistrons of three species of trout: Salvelinus namaycush (lake trout), Salvelinus fontinalis (brook trout), and Salmo gairdneri (rainbow trout) were examined by Southern blot analysis of genomic DNA. The repeat length of the cistron of S. namaycush is 26 kb. A repeat-length polymorphism was observed in some of the individual fish examined. These individuals showed 24-kb repeats. In some individuals both forms were present. The restriction maps of the transcribed regions of all three species were similar and showed a site homology with other vertebrate ribosomal RNA genes. Interspecific comparison showed restriction-site differences within the spacer regions examined. A restriction-site polymorphism within the 28 S gene was observed in S. fontinalis. The rDNA of S. namaycush liver showed a high degree of methylation as determined by digestion with the restriction endonucleases MspI and HpaII.

Upstream activation of ribosomal RNA biosynthesis in Saccharomyces cerevisiae

Yeast was transformed with eight recombinants that contained an rRNA minigene and upstream elements of rDNA in different orientations in the multi-copy yeast-Escherichia coli shuttle vector, pJDB207. The effect of these elements of upstream rDNA on the initiation of transcription of the minigene at the site for rRNA biosynthesis was determined by using an S1 nuclease mapping procedure to measure the abundance of the minigene transcript in RNA from the yeast transformants. Transcription of the minigene was enhanced 3-fold by DNA within a 2.2 kb element more than 1.5 kb upstream from the initiation site. Inversion of the 2.2 kb element decreased expression of the minigene by 40%. This 2.2 kb element contained approx. 500 bp from the 25S rRNA coding region at the 3' end of the preceding rRNA gene and 1 kb of adjacent nontranscribed spacer rDNA. The enhancing activity was independent of interference from readthrough that might have contributed to the 7-fold decrease in minigene expression caused by removing all rDNA upstream from -209 bp.

DNase I-resistant nontranscribed spacer segments of mouse ribosomal DNA contain poly(dG-dT).poly(dA-dC)

The nontranscribed spacer regions (NTS) that adjoin the coding portion of mouse ribosomal DNA are protected in nucleoli against exhaustive DNase I digestion. Since these sequences are degraded by the enzyme after they are extracted by phenol, the protection is suggested to result from the binding of specific proteins. The nucleolar structure would thus be organized to protect NTS sequences and expose the coding sequences for transcription. We show here that these protected sequences include tracts of poly(dG-dT).poly(dA-dC). We also report that these sequences are localized in regions flanking the rRNA transcription unit. These sequences can potentially form Z-DNA. The organized DNase I-resistant NTS structure in which they participate could be involved in structuring the nucleolus or in regulating transcription because poly(dG-dT).poly(dA-dC) sequences and portions of spacer rDNA can serve as transcriptional enhancer elements.

Localization of specific rDNA spacer sequences to the mouse L-cell nucleolar matrix

Mouse L-cell nucleoli were isolated from sonicated nuclei by centrifugation and extensively treated with pancreatic DNase or micrococcal nuclease to obtain "core nucleoli." Core nucleoli still contained the precursors to rRNA and about 1% of the total nuclear DNA, which remained tightly bound even after the removal of some chromatin proteins with 2 M NaCl. The core nucleolar DNA electrophoresed in a series of discrete bands, 20 to about 200 base pairs in length. Hybridization tests with specific DNA probes showed that the DNA was devoid of sequences complementary to mouse satellite, mouse Alu-like, and 5S RNA sequences. It also lacked sequences coding for cytoplasmic rRNA species, since it did not hybridize to the 18S to 28S portion of rDNA in Northern blot analyses and none of it was protected by hybridization to a 100-fold excess of total cytoplasmic RNA in S1 nuclease assays. However, the core nucleolar DNA did hybridize to nontranscribed and external transcribed spacer rDNA sequences. We infer that specific portions of rDNA are protected from DNase action by a tight association with nucleolar structural proteins.

Left-handed Z-DNA

Since the Watson-Crick proposal of right-handed B-DNA, numerous studies have been devoted to the conformation of DNA. Both natural DNAs of heterogeneous sequences and synthetic DNAs are capable of adopting more than one conformation. The specific conformation a DNA adopts appears to depend mainly on its base sequence and its environmental conditions. For a given DNA, changes in environmental conditions can induce conformational transitions which occur according to cooperative or non-cooperative processes (for general reviews see Ref. 1a, b). Despite many results, molecular biologists did not put much emphasis on the polymorphism of DNA. The discovery of the intraconversion in helical sense between the right-handed B and left-handed Z conformers of DNA has brought a new interest in the polymorphism of DNA. It is now proposed that this polymorphism has important functions in biological reactions. A recent review, 'The Chemistry and Biology of Left-handed Z-DNA', by Rich et al. has just been published. We here report some of the results published in 1984 on Z-DNA.

Characterization of rat ribosomal DNA II. identification of the highly repetitive DNA in the 3' non-transcribed spacer

The nucleotide sequence of one of the non-transcribed spacer subclones, p1.7, from the region 3' to rat 45 S pre-rRNA has been determined. Within 1612 base-pairs, the fragment contains two distinct regions of highly repetitive DNA, one of which can serve as a site for initiation in vitro by RNA polymerase III. The first is the alternating purine-pyrimidine sequence (A-C)21. The second of these regions has 95% homology to the identifier sequence and served as the template for RNA polymerase III transcription in vitro. The in vitro polymerase III template is aligned in opposite polarity to the direction of transcription of 45 S rRNA. Located near the identifier sequence is a region that is 59% homologous to the type-II Alu sequences. It would seem, therefore, that members of more than one highly repetitive sequence family have accumulated in the non-transcribed spacers. These data also suggest that within the non-transcribed spacers these families have evolved (sequence variation) at different rates, until one of them, the Alu type-II-like element, may represent a new Alu type-II subfamily.

The role of HnRNA in the control of ribosomal gene transcription

The hypothesis is put forward whereby a processed fragment of HnRNA acts as activator of ribosomal gene transcription by base-pair interaction with the non-transcribed region of the ribosomal gene.

Localization of specific rDNA spacer sequences to the mouse L-cell nucleolar matrix

Mouse L-cell nucleoli were isolated from sonicated nuclei by centrifugation and extensively treated with pancreatic DNase or micrococcal nuclease to obtain "core nucleoli." Core nucleoli still contained the precursors to rRNA and about 1% of the total nuclear DNA, which remained tightly bound even after the removal of some chromatin proteins with 2 M NaCl. The core nucleolar DNA electrophoresed in a series of discrete bands, 20 to about 200 base pairs in length. Hybridization tests with specific DNA probes showed that the DNA was devoid of sequences complementary to mouse satellite, mouse Alu-like, and 5S RNA sequences. It also lacked sequences coding for cytoplasmic rRNA species, since it did not hybridize to the 18S to 28S portion of rDNA in Northern blot analyses and none of it was protected by hybridization to a 100-fold excess of total cytoplasmic RNA in S1 nuclease assays. However, the core nucleolar DNA did hybridize to nontranscribed and external transcribed spacer rDNA sequences. We infer that specific portions of rDNA are protected from DNase action by a tight association with nucleolar structural proteins.

Structural organization of rat ribosomal RNA genes: interspersed sequences and their putative role in the alignment of nucleosomes

We have observed four regions containing highly repetitive interspersed sequences in the nontranscribed spacer (NTS) of the rat rRNA genes. Two of them (A and B) are located at a distance of 3-5 kb upstream from the transcription start point and two others (C and D) at a distance of 2-5 kb downstream from the 3' end of the 28S rRNA gene. These repetitive sequences are widely dispersed in the genome and are included both in small-copy regions and in the families of extended reiterated sequences. The sequences of three fragments were determined: one from the C2 region, 1100 bp in length and two from A and C1 regions, 110-120 bp long. These regions are characterized by the presence of 'simple' sequences, such as (AC)n, (ACC)n, (GAG)n, (GGGA)n, (TAAG)n, and also of long blocks, (G)n and (A)n. In the C2 region two palindromes, 16 and 14 nt long, were found, one of them including a XhoI site. Mobile element B2 was observed in regions B and C. All four regions, A, B, C and D, contain sets of simple sequences, among which some common elements have been found. Theoretical prediction of the nucleosomal disposition in the C region indicates that the combination of simple sequences existing in the given area secures fixed positions of the nucleosomes, one of the nucleosomes being formed on the B2 element. Moreover, a striking periodicity, with the repeat length close to that of the rat nucleosomal DNA, has been observed. A hypothesis is put forward that the simple sequences can dictate the location of nucleosomes on the adjoining DNA sequences, thereby regulating the gene activity.