Primers for mitochondrial DNA replication generated by endonuclease G.

Endonuclease G (Endo G) is widely distributed among animals and cleaves DNA at double-stranded (dG)n.(dC)n and at single-stranded (dC)n tracts. Endo G is synthesized as a propeptide with an amino-terminal presequence that targets the nuclease to mitochondria. Endo G can also be detected in extranucleolar chromatin. In addition to deoxyribonuclease activities, Endo G also has ribonuclease (RNase) and RNase H activities and specifically cleaves mouse mitochondrial RNA and DNA-RNA substrates containing the origin of heavy-strand DNA replication (OH). The cleavage sites match those found in vivo, indicating that Endo G is capable of generating the RNA primers required by DNA polymerase gamma to initiate replication of mitochondrial DNA.

Histone H5 in the control of DNA synthesis and cell proliferation.

The linker histones (H1, H5, H1 degrees) are involved in the condensation of chromatin into the 30-nanometer fiber. This supranucleosome organization correlates with the resting state of chromatin, and it is therefore possible that the linker histones play an active role in the control of chromatin activity. The effect of H5 has been directly determined by expression of an inducible transfected H5 gene in rat sarcoma cells, which do not produce H5. Transfection resulted in the reversible inhibition of DNA replication and arrest of cells in G1, at which time H5 concentrations approached that of terminally differentiated avian erythrocytes. The arrest of proliferation was accompanied by specific changes in gene expression probably related to the cell cycle block. The selectivity of these effects suggest that H5 plays an active role in the control of DNA replication and cell proliferation.

Regulation of histone and beta A-globin gene expression during differentiation of chicken erythroid cells.

The expression of the genes for several histones and beta A-globin was examined in the chicken erythroid cells lineage. During the transition from CFU-(E) to the mature erythrocyte, histone H5 gradually increased fourfold in nuclei with little concomitant displacement of the H1 histones. This resulted in a 70% net increase in linker histone (H1 plus H5) content. The differential accumulation of H5 reflected (i) an increase in the transcriptional activity of the H5 gene occurring at the erythroblast stage, (ii) an apparent longer half-life of H5 mRNA, and (iii) a higher stability of the protein. Although the transcriptional activity of the histone genes (except H5) decreased with cell age, it was not tightly coupled to the S phase. On the other hand, the mRNA levels for these histones were tightly regulated during the cell cycle. Use of protein and DNA synthesis inhibitors indicated that the content of H5 mRNA was regulated at the posttranscriptional level by a control mechanism(s) differing from those for the other histones. Although the transcription rates of the H5 and beta A-globin genes were comparable, differential accumulation of beta A-globin mRNA led to a 30- to 170-fold-higher copy number of the beta A-globin mRNA as the cell matured.

Transcription of the histone H5 gene is regulated by three differentiation-specific enhancers.

Histone H5, an early marker of the avian erythroid lineage, is expressed at low levels in early erythroid precursors and at higher levels in more mature cells. We show that the increase in H5 expression is due to transcriptional activation of the H5 gene following differentiation of precursor CFU(E). We have found and characterized two upstream enhancers, E1 (between -2233 and -1878 from the site of transcription initiation, +1) and E3 (between -1321 and -1163), and confirmed the presence of a downstream enhancer (C. D. Trainor, S. J. Stamler, and J. D. Engel, Nature [London] 328:827-830, 1987) E7 (between +846 and +1181) which are responsible for the increase in H5 gene transcription. The enhancers had a weak effect in nondifferentiated CFU(E) but a strong effect when the cells were induced to differentiate. Cooperation among the three enhancers, however, was not required for H5 gene activity in the differentiated cells. The enhancers contain binding sites for several ubiquitous and erythroid cell-specific nuclear proteins, including GATA-1, as demonstrated with GATA-1-specific antibodies. Although the GATA sites were required for enhancer function, the concentration of GATA-1, GATA-2, and GATA-3 decreased during cell differentiation, and overexpression of these factors had little effect on H5 transcription. Hence, the differentiation-specific effect of the enhancers is not mediated by changes in relative levels of the GATA factors. Functional analysis of the H5 promoter indicated that the requirement of several elements, including a GC box necessary for transcription enhancement, did not change during the early stages of CFU(E) differentiation. However, the UPE, a positive element in proliferating CFU(E) recognized by the transcription factor H4TF2, was dispensable in the differentiated cells. These results suggest that as the cells enter the final stages of differentiation, there is a reprogramming of the regulatory factors that control H5 transcription and that the enhancers rescue and increase the activity of the promoter.

Initiation binding repressor, a factor that binds to the transcription initiation site of the histone h5 gene, is a glycosylated member of a family of cell growth regulators [corrected].

Initiation binding repressor [corrected] (IBR) is a chicken erythrocyte factor (apparent molecular mass, 70 to 73 kDa) that binds to the sequences spanning the transcription initiation site of the histone h5 gene, repressing its transcription. A variety of other cells, including transformed erythroid precursors, do not have IBR but a factor referred to as IBF (68 to 70 kDa) that recognizes the same IBR sites. We have cloned the IBR cDNA and studied the relationship of IBR and IBF. IBR is a 503-amino-acid-long acidic protein which is 99.0% identical to the recently reported human NRF-1/alpha-Pal factor and highly related to the invertebrate transcription factors P3A2 and erected wing gene product (EWG). We present evidence that IBR and IBF are most likely identical proteins, differing in their degree of glycosylation. We have analyzed several molecular aspects of IBR/F and shown that the factor associates as stable homodimers and that the dimer is the relevant DNA-binding species. The evolutionarily conserved N-terminal half of IBR/F harbors the DNA-binding/dimerization domain (outer limits, 127 to 283), one or several casein kinase II sites (37 to 67), and a bipartite nuclear localization signal (89 to 106) which appears to be necessary for nuclear targeting. Binding site selection revealed that the alternating RCGCRYGCGY consensus constitutes high-affinity IBR/F binding sites and that the direct-repeat palindrome TGCGCATGCGCA is the optimal site. A survey of genes potentially regulated by this family of factors primarily revealed genes involved in growth-related metabolism.

Effect of histone composition on the stability of chromatin structure.

The mode of fragmentation of chromatin by micrococcal nuclease has been studied in nuclei from different sources at physiological ionic strength and low temperature. During digestion, the size of chromatin was reduced until an average S value of 95-100 (hen erythrocyte) or 60-65 (rat liver) was attained. The accumulation of these structures correlated with the period of maximum solubility (80%), indicating that the bulk of chromatin behaved in this manner. Further digestion did not result in a corresponding decrease in S value but in a bimodal sedimentation pattern. As opposed to this behavior, chromatin containing actively acetylated core histones showed a continuous variation in size during the digestion. Indirect immunoprecipitation of chromatin by anti-H5 antibody and sheep anti-rabbit antibody revealed that the acetylated chromatin is partially depleted of H5.

Endonuclease G: a mitochondrial protein released in apoptosis and involved in caspase-independent DNA degradation.

A hallmark of apoptosis is the fragmentation of nuclear DNA. Although this activity involves the caspase-3-dependent DNAse CAD (caspase-activated DNAse), evidence exists that DNA fragmentation can occur independently of caspase activity. Here we report on the ability of truncated Bid (tBid) to induce the release of a DNAse activity from mitochondria. This DNAse activity was identified by mass spectrometry as endonuclease G, an abundant 30 kDa protein released from mitochondria under apoptotic conditions. No tBid-induced endonuclease G release could be observed in mitochondria from Bcl-2-transgenic mice. The in vivo occurrence of endonuclease G release from mitochondria during apoptosis was confirmed in the liver from mice injected with agonistic anti-Fas antibody and is completely prevented in Bcl-2 transgenic mice. These data indicate that endonuclease G may be involved in CAD-independent DNA fragmentation during cell death pathways in which truncated Bid is generated.

Fine analysis of the active H5 gene chromatin of chicken erythroid cells at different stages of differentiation.

We have analyzed the chromatin structure of a region that encompasses 14.4 X 10(3) base-pairs of the chicken histone H5 locus in adult erythroid cells at different stages of maturation. Seven of eight major lineage-specific DNase I-hypersensitive sites, some of which show complex substructure, were found in the flanking regions of the gene. The hypersensitivity of some of these sites is modulated during erythrocyte maturation in a way that parallels the transcriptional activity of the gene. DNase I, micrococcal nuclease, and S1 nuclease recognize the same regions, which differ from those cleaved by S1 on supercoiled plasmid DNA. This suggests that hypersensitivity of DNA in chromatin reflects a greater accessibility of the DNA rather than its altered conformation. The DNA sequence of some of the DNase I target sites contains repeated motifs, (T-C-C-C)2, (T-C-C)2, (T-G-G-G-G)2, which are found in the hypersensitive sites of other genes. Detailed analysis across sections of the H5 gene and flanking sequences revealed differences in the DNase I sensitivity of the different regions examined. Notably, the first one-third of the gene is more sensitive than the rest. The sequences downstream from the region where most RNA polymerases terminate transcription were found to be the most resistant.

Genomic organization of the genes coding for the six main histones of the chicken: complete sequence of the H5 gene.

The organization of the genes coding for histones in the chicken has been examined, with special reference to that coding for the tissue-specific, developmentally regulated histone H5. Two recombinant phages containing sequences complementary to cloned H5 cDNA have been isolated from a genomic chicken library. The clones have been characterized by heteroduplex formation, restriction nuclease analysis, hybridization to cloned homologous histone gene probes, and DNA sequencing. Hybridization to genomic DNA has shown that there is only one copy of the H5 gene per haploid genome, whereas there are six to 11 copies of the genes for the other histones. Examination of 29 X 10(3) base-pairs of DNA sequences flanking the H5 gene has revealed the absence of any other histone genes which, although not tandemly reiterated, for the most part appear to reside in loosely organized clusters. The complete DNA sequence of the H5 gene and flanking regions, as well as the mapping of the 5'-end of its messenger RNA by primer extension with AMV reverse transcriptase, has shown that the gene has no introns and little homology to other histone genes, including those for H1.

Isolation of a 167 basepair chromatosome containing a partially digested histone H5.

A test has been made of the proposal that protection of the 167 basepair DNA length in the 'chromatosome' is due only to the central globular domain of the lysine-rich histones. Chicken erythrocyte chromatin was treated with trypsin to leave only the limit peptide from histones H1 and H5. Nucleosome monomers were then isolated on sucrose gradients following micrococcal nuclease digestion and were found to contain the 167 basepair DNA band as in intact chromatin. The presence of the limit peptide from H5 on the monomers was confirmed using an antibody to H5. It is concluded that the trypsin-susceptible domains of the lysine-rich histones are not involved in the protection of the 2-turn 167 basepair length of DNA in the nucleosome.

Elements regulating differential activity of chicken histone H1 gene promoters.

The chicken genome contains six closely related histone H1 genes, each of which encodes a different H1 protein. The four common regulatory elements previously identified in H1 histone promoters are very similar in sequence and location in all chicken H1 genes, which gives rise to the question of how the six H1 variants are expressed at significantly different levels. Transient transfections of reporter gene transcriptional fusions indicate that approximately 200 bp of each promoter is sufficient to generate the observed spectrum of H1 promoter activity. The differences in H1 promoter-driven expression are shown to be explained by the relative activity of the previously characterized G box region and that of a novel element found between CCAAT and TATA that we have termed differential upstream sequence (Dus). Gel shift analysis indicated that the primary nuclear binding protein to the G box is one or more avian homologues of the Sp1 transcription factor. The Dus region binds multiple nuclear proteins, one of which is the recently described IBR/IBF factor. The differential affinities of the G box and Dus sequences of the H1 promoters for their respective nuclear binding factors correlate well with their relative promoter activities.

Characterization and expression of the mouse endonuclease G gene.

Endonuclease G (Endo G) is a nuclease of prokaryotic lineage found in the mitochondria of vertebrates that has been suggested to play a role in mitochondrial DNA (mtDNA) replication. We have isolated and sequenced the entire mouse endo G gene, determined the limits of the mRNA, and mapped the promoter region. The coding sequence of the single copy gene is interrupted by two introns and analysis of the transcripts does not support a model by which more than one Endo G isoform could be produced by alternative splicing. We have also characterized a full-length human Endo G cDNA and comparison at the protein level of the human, bovine, and murine nucleases indicates a high degree of conservation except in the respective mitochondrial targeting signals. Endo G is ubiquitously expressed and the steady-state levels of its mRNA vary by a factor greater than seven between different tissues. The relationship between the mtDNA copy number and Endo G mRNA levels is not strictly proportional but tissues richer in mtDNA have higher amounts of the mRNA and vice versa.

The histone H5 gene is flanked by S1 hypersensitive structures.

The potential of the cloned histone H5 gene to form altered DNA structures has been examined by S1 nuclease digestion of supercoiled recombinant plasmids containing up to 8.8 kbp of chicken DNa. The three main nicking sites map at the upstream and downstream sequences flanking the structural gene. The cleavage sites share sequence homology, strand specificity, and do not seem to be single-stranded. The sequence of the S1-sensitive sites does not suggest that the fragments can adopt any of the known DNA secondary structures.

Endonuclease G: a (dG)n X (dC)n-specific DNase from higher eukaryotes.

An endonuclease activity (termed endonuclease G) that selectively cleaves DNA at (dG)n X (dC)n tracts has been partially purified from immature chicken erythrocyte nuclei. Sites where n greater than or equal to 9 are cleaved in a manner that resembles types II and III restriction nucleases. The nicking rate of the G-strand is 4- to 10-fold higher than that of the C-strand depending on the length of the (dG)n X (dC)n tract and/or nucleotide composition of the flanking sequences. Endonuclease G hydrolyzes (dG)24 X (dC)24 of supercoiled DNA in a bimodal way every 9-11 nucleotides, the maxima in one strand corresponding to minima in the opposite, suggesting that it binds preferentially to one side of the double helix. The nuclease produces 5' phosphomonoester ends and its activity is dependent on Mg2+ or Mn2+. The wide distribution and high relative activity of endonuclease G in a variety of tissues and species argues for a general role of the enzyme. The striking correlation between genetic instability and poly(dG) X poly(dC) tracts in DNA suggests that these sequences and endonuclease G are involved in recombination processes.

H3.H4 tetramer directs DNA and core histone octamer assembly in the nucleosome core particle.

The way in which histones interact with DNA during in vitro assembly of nucleohistone has been examined. Chicken erythrocyte core histones H2A, H2B, H3, and H4 and lambdaDNA in 2 M NaCl were allowed to interact by stepwise decrease in the salt concentration. Binding, although weak, was first observed at 1.4 M NaCl and was essentially completed at 0.6 M NaCl. Analysis of the DNA-bound histones revealed that each of the histones in the pairs H2A,H2B and H3,H4 was always present in equimolar amounts and that the relative proportion of each pair was constant between 1.4 and 0.8 M NaCl. Evidence is presented suggesting that binding occurred via complexes of the four histones, the nature of which is likely to reflect the equilibrium among the octamer and its products of dissociation (Ruiz-Carrillo, A., & Jorcano, J.L. (1979) Biochemistry (preceding paper in this issue)). The presence of complexes of the four core histones is, however not required for the correct assembly of the nucleosome core particle. Nucleohistones obtained by adding at progressively lower ionic strengths the dimer H2A.H2B to the H3.H4-DNA complex (split reconstitutions) had the same characteristics as those assembled with the core histone complexes.

An octamer of core histones in solution: central role of the H3-H4 tetramer in the self-assembly.

The association of histones H2A, H2B, H3, and H4 in solution has been studied. In 2 M NaCl and at neutral pH they can assemble in a complex in which each histone is present in equimolar amounts. The complex has a weight average molecular weight of 98,000 (+/- 3700) and a sedimentation coefficient (so20,w) of 4.8. The value of the weight average molecular weight and the histone stoichiometry indicate that the complex is an octamer. The pairs of histones H2A,H2B and H3,H4 studied separately under identical conditions only associated as equimolar complexes consistent with dimeric and tetrameric structures, respectively. The stability of the core histone octamer is a function of the ionic strength, pH, and concentration of protein. The octamer dissociates by losing dimers of H2A,H2B until the main complexes existing in solution are the H3.H4 tetramer and the H2A.H2B dimer. This process is reversible upon reestablishing the original conditions.

Increased histone H1(0) expression in differentiating mouse erythroleukemia cells is related to decreased cell proliferation.

The amount of histone H1(0) increases relative to other H1 subtypes in terminally differentiated cells, and its expression has been associated with the onset of differentiation. We have studied the kinetics of H1(0) accumulation in mouse erythroleukemia (MEL) cells and found that the levels of H1(0) reflect the rate of cell proliferation rather than the state of differentiation. This suggests that changes in the relative amount of H1(0) during MEL cell differentiation are primarily a consequence of cell cycle arrest.

The globular region of histone H5 is equally accessible to antibodies in relaxed and condensed chromatin.

The accessibility of histone H5 in chromatin was examined with monoclonal antibodies recognizing several epitopes of the globular region (GH5) of the histone (Rózalski, M., Lafleur, L., and Ruiz-Carrillo, A. (1985) J. Biol. Chem. 260, 14379-14385). The stoichiometry of the chromatin-antibody complexes indicated that while 0-86% of the H5 molecules were able to react, depending on the particular epitope, the extent of antibody binding to relaxed chromatin (in 5 mM KCl) and condensed chromatin (in 100 mM KCl or 0.35 mM MgCl2) was virtually identical. This indicates that the topography of H5 does not change during the conformational transition of chromatin. The data suggest that H5 is not completely internalized in the 30-nm fiber or that the fiber is flexible enough to allow full exposure of the GH5 epitopes. Several control experiments, including monoclonal antibody binding, sedimentation analysis, DNase II digestion, and glutaraldehyde cross-linking, showed that epitope accessibility is not due to H5 exchange or to perturbation of the chromatin fiber. The accessibility of GH5 suggests ways in which inactive chromatin may be unfolded in vivo.