Regulation of the human apoptotic DNase/RNase endonuclease G: involvement of Hsp70 and ATP

Endonuclease G (EndoG) is a mitochondrial enzyme that becomes an apoptotic nuclease when released from the mitochondrial intermembrane space. EndoG will digest either DNA or RNA, but at physiological ionic strength, RNA is a much more favorable substrate as compared to chromatin. This indicates that EndoG's major in vivo function(s) may be: (i) an apoptotic RNase, and/or (ii) an apoptotic DNase in the presence of additional co-activators. In the present study we have searched for factors that modulate the activity of human EndoG on DNA substrates. We demonstrate that EndoG forms complexes with AIF and FEN-1 but not with PCNA. Interestingly, heat shock proteins 70 interact with EndoG and are involved in the regulation of its activity. Purified Hsp70 prevented stimulation of EndoG DNase activity by other nuclear factors in the ATP-dependent manner.

Action of recombinant human apoptotic endonuclease G on naked DNA and chromatin substrates: cooperation with exonuclease and DNase I

Endonuclease G (endoG) is released from mitochondria during apoptosis and is in part responsible for internucleosomal DNA cleavage. Here we report the action of the purified human recombinant form of this endonuclease on naked DNA and chromatin substrates. The addition of the protein to isolated nuclei from non-apoptotic cells first induces higher order chromatin cleavage into DNA fragments > or = 50 kb in length, followed by inter- and intranucleosomal DNA cleavages with products possessing significant internal single-stranded nicks spaced at nucleosomal ( approximately 190 bases) and subnucleosomal ( approximately 10 bases) periodicities. We demonstrate that both exonucleases and DNase I stimulate the ability of endoG to generate double-stranded DNA cleavage products at physiological ionic strengths, suggesting that these activities work in concert with endoG in apoptotic cells to ensure efficient DNA breakdown.

Ionic and cofactor requirements for the activity of the apoptotic endonuclease DFF40/CAD

The endonuclease DFF40/CAD mediates regulated DNA fragmentation and chromatin condensation in cells undergoing apoptosis. Here we report the enzyme's co-factor requirements, and demonstrate that the ionic changes that occur in apoptotic cells maximize DFF40/CAD activity. The nuclease requires Mg2+, exhibits a trace of activity in the presence of Mn2+, is not costimulated by Ca2+, is inhibited by Zn2+ or Cu2+, and has high activity over a rather broad pH range (7.0-8.5). The enzyme is thermally unstable, and is rapidly inactivated at 42 degrees C. Enzyme activity is markedly affected by ionic strength. At the optimal [K+] of 50-125 mM, which is in the range of the cytoplasmic [K+] for cells undergoing apoptosis, the activity of DFF40/CAD for naked DNA cleavage is about 100-fold higher than at 0 or 200 mM [K+]. Although these ranges of ionic strength do not affect DFF40 homo-oligomer formation, at higher ionic strengths the enzyme introduces single-stranded nicks into supercoiled DNA.

Cleavage preferences of the apoptotic endonuclease DFF40 (caspase-activated DNase or nuclease) on naked DNA and chromatin substrates

Here we report the co-factor requirements for DNA fragmentation factor (DFF) endonuclease and characterize its cleavage sites on naked DNA and chromatin substrates. The endonuclease exhibits a pH optimum of 7.5, requires Mg(2+), not Ca(2+), and is inhibited by Zn(2+). The enzyme generates blunt ends or ends with 1-base 5'-overhangs possessing 5'-phosphate and 3'-hydroxyl groups and is specific for double- and not single-stranded DNA or RNA. DFF endonuclease has a moderately greater sequence preference than micrococcal nuclease or DNase I, and the sites attacked possess a dyad axis of symmetry with respect to purine and pyrimidine content. Using HeLa cell nuclei or chromatin reconstituted on a 5 S rRNA gene tandem array, we prove that the enzyme attacks chromatin in the internucleosomal linker, generating oligonucleosomal DNA ladders sharper than those created by micrococcal nuclease. Histone H1, high mobility group-1, and topoisomerase II activate DFF endonuclease activity on naked DNA substrates but much less so on chromatin substrates. We conclude that DFF is a useful reagent for chromatin research.

MARs of antigen receptor and co-receptor genes

MARs are cis-acting DNA sequences that function both negatively and positively in conjunction with transcriptional enhancers to regulate antigen receptor and co-receptor genes. Evidence exists that certain tissue-specific nuclear proteins are involved in this regulation, including SATB1, Bright, and Cux/CDP, possibly by modulating intranuclear gene location, histone acetylation, DNA methylation, and/or nucleosome positioning.

High-level rearrangement and transcription of yeast artificial chromosome-based mouse Ig kappa transgenes containing distal regions of the contig

The mouse Ig kappa L chain gene locus has been extensively studied, but to date high-level expression of germline transgenes has not been achieved. Reasoning that each end of the locus may contain regulatory elements because these regions are not deleted upon V kappa-J kappa joining, we used yeast artificial chromosome-based techniques to fuse distal regions of the contig to create transgene miniloci. The largest minilocus (290 kb) possessed all members of the upstream V kappa 2 gene family including their entire 5' and 3' flanking sequences, along with one member of a downstream V kappa 21 gene family. In addition, again using yeast artificial chromosome-based technology, we created Ig kappa miniloci that contained differing lengths of sequences 5' of the most distal V kappa 2 gene family member. In transgenic mice, Ig kappa miniloci exhibited position-independent and copy number-dependent germline transcription. Ig kappa miniloci were rearranged in tissue and developmental stage-specific manners. The levels of rearrangement and transcription of the distal and proximal V kappa gene families were similar to their endogenous counterparts and appeared to be responsive to allelic exclusion, but were differentially sensitive to numerous position effects. The minilocus that contained the longest 5' region exhibited significantly greater recombination of the upstream V kappa 2 genes but not the downstream V kappa 21 gene, providing evidence for a local recombination stimulating element. These results provide evidence that our miniloci contain nearly all regulatory elements required for bona fide Ig kappa gene expression, making them useful substrates for functional analyses of cis-acting sequences in the future.

RNA polymerase-specific nucleosome disruption by transcription in vivo

The nucleosomal chromatin structure within genes is disrupted upon transcription by RNA polymerase II. To determine whether this disruption is caused by transcription per se as opposed to the RNA polymerase source, we engineered the yeast chromosomal HSP82 gene to be exclusively transcribed by bacteriophage T7 RNA polymerase in vivo. Interestingly, we found that a fraction of the T7-generated transcripts were 3' end processed and polyadenylated at or near the 3' ends of the hsp82 and the immediately downstream CIN2 genes. Surprisingly, the nucleosomal structure of the T7-transcribed hsp82 gene remained intact, in marked contrast to the disrupted structure generated by much weaker, basal level transcription of the wild type gene by RNA polymerase II under non-heat shock conditions. Therefore, disruption of chromatin structure by transcription is dependent on the RNA polymerase source. We propose that the observed RNA polymerase dependence for transcription-induced nucleosome disruption may be related either to the differential recruitment of chromatin remodeling complexes, the rates of histone octamer translocation and nucleosome reformation during polymerase traversal, and/or the degree of transient torsional stress generated by the elongating polymerase.

Evidence that the Igkappa gene MAR regulates the probability of premature V-J joining and somatic hypermutation

The Igkappa gene contains an evolutionarily conserved nuclear matrix association region (MAR) adjacent to the intronic enhancer. To test for the function of this MAR, we created mouse lines with a targeted MAR deletion. In MAR knockout animals, the immune system was normal in nearly all respects, including the distributions of various B cell populations and Ab levels. However, in pro-B cells, enhanced rearrangement was noted on the MAR- allele in heterozygotes. In addition, the efficiencies for targeting and generating somatic mutations were reduced on MAR-deleted alleles. These results provide evidence for the MAR negatively regulating the probability of premature rearrangement and positively regulating the probability of somatic hypermutation.

Targeted linearization of DNA in vivo

In the past decade, site-specific chromosomal DNA cleavage mediated by DNA endonucleases has been used to examine diverse aspects of chromosome structure and function in eukaryotes, such as DNA topology, replication, transcription, recombination, and repair. Here we describe a method with which chromosomes can be linearized at any predefined position in vivo. Yeast homothallic switching endonuclease (HO endo), a sequence-specific double-strand nuclease involved in mating-type switching, is employed for targeting DNA cleavage. HO endo contains discrete functional domains: a N-terminal nuclease and a C-terminal DNA-binding domain, thereby allowing construction of a chimeric nuclease with the cutting site distinct from the original HO recognition sequence. The expression of the nuclease is engineered to be controlled by a tightly regulated, inducible promoter. The cut sites recognized by HO endo or its derivatives are introduced specifically at desired positions in the yeast genome by homologous recombination. Here we present experimental procedures and review some applications based on this approach in yeast and other biological systems.

Nucleosomes and regulation of gene expression. Structure of the HIV-1 5'LTR

Packaging of DNA into chromatin adds complexity to the problem of regulation of gene expression. Nucleosomes affect the accessibility of transcription factors to occupy their binding sites in chromatin of eukaryotic cells. The disruption of nucleosome structure within the enhancer/promoter region of the integrated HIV-1 proviral genome is an instructive example of a chromatin remodeling process during transcriptional activation. To investigate the mechanism responsible for generating nuclease hypersensitive sites that exist in vivo in the promoter/enhancer region of the 5'LTR (long terminal repeat) of integrated HIV-1 we have utilized an in vitro chromatin assembly system with Xenopus oocyte extracts. Chromatin assembly in the presence of Sp1 and NFkappaB transcription factors induces DNase I hypersensitive sites on either side of their binding sites and positions the adjacent nucleosomes. This structure can also be formed in a factor-induced, ATP-dependent chromatin remodeling process and closely resembles the in vivo chromatin structure. The DNase I hypersensitive sites that form within the HIV LTR are probably histone-free and remain after removal of transcription factors.

The 40-kDa subunit of DNA fragmentation factor induces DNA fragmentation and chromatin condensation during apoptosis

We report here the reconstitution of a pathway that leads to the apoptotic changes in nuclei by using recombinant DNA fragmentation factor (DFF), a heterodimeric protein of 40 and 45 kDa. Coexpression of DFF40 and DFF45 is required to generate recombinant DFF, which becomes activated when DFF45 is cleaved by caspase-3. The cleaved fragments of DFF45 dissociate from the DFF40, the active component of DFF. Purified DFF40 exhibited an intrinsic DNase activity that was markedly stimulated by chromatin-associated proteins histone H1 and high mobility group proteins. DFF40 also triggered chromatin condensation when incubated with nuclei. These data suggest that DFF40 is sufficient to trigger both DNA fragmentation and chromatin condensation during apoptosis.

A targeted kappa immunoglobulin gene containing a deletion of the nuclear matrix association region exhibits spontaneous hyper-recombination in pre-B cells

Previous studies employing ectopic integration of reporter genes have shown that the nuclear matrix association region (MAR) adjacent to the intronic enhancer of the mouse kappa immunoglobulin (Ig) gene is required for high level transcription of rearranged genes, demethylation, reduction of position effects and maximal somatic hypermutation in B cells. To test for the function of this MAR in its natural chromosomal environment, we pursued the 'HIT-and-RUN' procedure with the mouse pre-B cell line 103 to create a targeted MAR deletion. We observed a 'HIT' targeting frequency of 1/684 but 0/2100 'RUN' clones maintained the MAR-deleted germline locus because of an unexpected hyper-recombination for Vkappa-Jkappa joining, specifically to the MAR-deleted allele, and primarily at Jkappa4 and Jkappa5. This hyper-recombination was correlated with undermethylation of the Jkappa-Ckappa region but not with the level of local transcription. These results are consistent with the possibility that the MAR and/or DNA methylation negatively regulate(s) Vkappa-Jkappa joining during the pre-B cell stage of development.

Nucleosomes and regulation of gene expression. Structure of the HIV-1 5'LTR

Packaging of DNA into chromatin adds complexity to the problem of regulation of gene expression. Nucleosomes affect the accessibility of transcription factors to occupy their binding sites in chromatin of eukaryotic cells. The disruption of nucleosome structure within the enhancer/promoter region of the integrated HIV-1 proviral genome is an instructive example of a chromatin remodeling process during transcriptional activation. To investigate the mechanism responsible for generating nuclease hypersensitive sites that exist in vivo in the promoter/enhancer region of the 5'LTR (long terminal repeat) of integrated HIV-1 we have utilized an in vitro chromatin assembly system with Xenopus oocyte extracts. Chromatin assembly in the presence of Sp1 and NFkappaB transcription factors induces DNase I hypersensitive sites on either side of their binding sites and positions the adjacent nucleosomes. This structure can also be formed in a factor-induced, ATP-dependent chromatin remodeling process and closely resembles the in vivo chromatin structure. The DNase I hypersensitive sites that form within the HIV LTR are probably histone-free and remain after removal of transcription factors.

A method for genome comparisons and hybridization studies using known megabase-scale DNA sequences as a reference

We present a method for genome comparisons and high-resolution hybridization analyses using megabase stretches of known DNA sequences as a reference. The method employs two-dimensional gel electrophoresis, separating genomic segments cut with different restriction endonucleases in the first and second dimensions, to generate filters suitable for image analysis and repeated nucleic acid hybridizations. The corresponding two-dimensional pattern is computed from the reference nucleotide sequence and matched to the observed pattern, thereby identifying each fragment on the filter; at the same time the technique uncovers discrepancies from the reference sequence. This permits genome comparisons as well as automated identification and quantification of hybridization patterns with various probes. The technique is illustrated by an analysis of Saccharomyces cerevisiae chromosome IX.

In vitro chromatin assembly of the HIV-1 promoter. ATP-dependent polar repositioning of nucleosomes by Sp1 and NFkappaB

Nuclease hypersensitive sites exist in vivo in the chromatin of the integrated human immunodeficiency virus (HIV)-1 proviral genome, in the 5'-long terminal repeat (LTR) within the promoter/enhancer region near Sp1 and NFkappaB binding sites. Previous studies from the Kadonaga and Jones laboratories have shown that Sp1 and NFkappaB can establish hypersensitive sites in a truncated form of this LTR when added before in vitro chromatin assembly with Drosophila extracts, thus facilitating subsequent transcriptional activation of a linked reporter gene upon the association of additional factors (Pazin, M. J., Sheridan, P. L., Cannon, K., Cao, Z., Keck, J. G., Kadanaga, J. T., and Jones, K. A. (1996) Genes & Dev. 10, 37-49). Here we assess the role of a full-length LTR and 1 kilobase pair of downstream flanking HIV sequences in chromatin remodeling when these transcription factors are added after chromatin assembly. Using Xenopus laevis oocyte extracts to assemble chromatin in vitro, we have confirmed that Sp1 and NFkappaB can indeed induce sites hypersensitive to DNase I, micrococcal nuclease, or restriction enzymes on either side of factor binding sites in chromatin but not naked DNA. We extend these earlier studies by demonstrating that the process is ATP-dependent when the factors are added after chromatin assembly and that histone H1, AP1, TBP, or Tat had no effect on hypersensitive site formation. Furthermore, we have found that nucleosomes upstream of NFkappaB sites are rotationally positioned prior to factor binding and that their translational frame is registered after binding NFkappaB. On the other hand, binding of Sp1 positions adjacent downstream nucleosome(s). We term this polar repositioning because each factor aligns nucleosomes only on one side of its binding sites. Mutational analysis and oligonucleotide competition each demonstrated that this remodeling required Sp1 and NFkappaB binding sites.

Template topology and transcription: chromatin templates relaxed by localized linearization are transcriptionally active in yeast

To address the role of transient torsional stress in transcription, we have utilized the regulated expression of HO endonuclease in yeast to create double-strand breaks in DNA templates in vivo at preselected sites. Linearization of circular minichromosomes, either 2 kb upstream or immediately downstream of a lacZ reporter gene controlled by the yeast metallothionein gene (CUP1) promoter, did not alter the copper induction profile of lacZ RNA transcripts compared to that of nonlinearized controls. Constructs site-specifically integrated into yeast chromosome II gave similar results. In vivo cross-linking with psoralen as a probe for negative DNA supercoiling demonstrated that template linearization efficiently dissipated DNA supercoiling induced by transcription. Therefore, the efficient transcription of linearized, relaxed templates found here demonstrates that transient torsional tension is not required for transcription of chromatin templates in yeast.

Yeast artificial chromosome contigs reveal that distal variable-region genes reside at least 3 megabases from the joining regions in the murine immunoglobulin kappa locus

The immunoglobulin kappa gene locus encodes 95% of the light chains of murine antibody molecules and is thought to contain up to 300 variable (V kappa)-region genes generally considered to comprise 20 families. To delineate the locus we have isolated 29 yeast artificial chromosome genomic clones that form two contigs, span > 3.5 megabases, and contain two known non-immunoglobulin kappa markers. Using PCR primers specific for 19 V kappa gene families and Southern analysis, we have refined the genetically defined order of these V kappa gene families. Of these, V kappa 2 maps at least 3.0 Mb from the joining (J kappa) region and appears to be the most distal V kappa gene segment.

Yeast calmodulin and a conserved nuclear protein participate in the in vivo binding of a matrix association region

Chromatin becomes reorganized during mitosis each cell cycle. To identify genes potentially involved in these supramolecular events, we have used a colony-color assay to screen temperature-sensitive mutants of Saccharomyces cerevisiae. When a sequence that mediates attachment to the nuclear matrix in vitro was inserted into the GAL1 promoter of a lacZ fusion gene, beta-galactosidase synthesis was inhibited. This observation permitted screening for temperature-sensitive-inducible mutants on 5-bromo-4-chloro-3-indolyl beta-D-galactoside plates. Only 1 of 20 complementation groups of newly isolated mutants exhibited temperature-sensitive inducibility for the matrix association region but not for control CEN3 or STE6 inserts--a cmd1 mutant in which the last 7 amino acids of calmodulin were truncated by an ochre termination codon. Another mutant (smi1) exhibited a rare phenotype at the nonpermissive condition, which included S phase and budding arrest. We cloned and sequenced the SMI1 gene, which encodes a 57-kDa polypeptide with evolutionarily conserved epitope(s) found in mammalian cell nuclei. Thus, we provide evidence for involvement of calmodulin and another conserved protein in the in vivo binding of a matrix association region.

Immunoglobulin kappa gene enhancers synergistically activate gene expression but independently determine chromatin structure

Previous studies have located transcriptional enhancer elements within both the intron and 3'-region of the mouse kappa immunoglobulin gene. Here we address the role of these two enhancers in specifying gene activity and specific chromatin structures. MOPC41 kappa gene constructs, either intact or containing deletions of one or both enhancers, were introduced into S194 mouse plasmacytoma cells for transient and stable expression studies. Transient expression assays revealed that the basal level expression exhibited by enhancerless constructs was activated 100-200-fold by the two enhancers together in a synergistic fashion. A similar trend was observed when both enhancers were present in stably integrated constructs, although the synergy was less pronounced. Analysis of DNase I hypersensitive sites in the chromatin revealed that stably integrated constructs established hypersensitive sites about the enhancer sequences. These sites demonstrated the same nuclease susceptibility as those associated with the endogenous gene(s), and their establishment was independent of the presence of the other enhancer. Thus, although both enhancers are required for maximal gene expression, the elements act independently in determining specific chromatin structures.

Uncoupling gene activity from chromatin structure: promoter mutations can inactivate transcription of the yeast HSP82 gene without eliminating nucleosome-free regions

DNase I-hypersensitive sites represent "nucleosome-free" regions in chromatin where the underlying DNA sequence is highly accessible to trans-acting proteins. Here we demonstrate that it is possible to uncouple gene activity from hypersensitive site formation. Point or substitution mutations were introduced into the promoter of the yeast chromosomal HSP82 gene, encoding the 83-kDa heat shock protein (HSP), via site-directed integration. Mutating either the TATA box or heat shock element 1 (HSE1) significantly reduced basal and heat-induced transcription while mutating both essentially inactivated expression. Dormant transcription units exhibited arrays of sequence-positioned nucleosomes; nevertheless, the inactivated genes still retained a hypersensitive site within their mutated promoters. In addition, all yeast strains maintained a heat-inducible hypersensitive site at -600 base pairs (bp), while several mutant strains converted a constitutive hypersensitive site at -300 bp into a heat-inducible one. Thus, mutations in cis-acting elements within a promoter can inactivate transcription without eliminating nucleosome-free regions.

Positive DNA supercoiling generates a chromatin conformation characteristic of highly active genes

During transcription, positive DNA supercoils generated ahead of RNA polymerase could theoretically uncoil the negative DNA supercoils associated with nucleosomes and thereby decondense the chromatin fiber in preparation for RNA polymerase passage. Here we examine the effect of positive DNA supercoiling on the structure of yeast 2-microns minichromosomes. We utilized a conditional topoisomerase mutant expressing Escherichia coli topoisomerase I to convert the DNA supercoiling state from negative to positive in vivo. Minichromosomes containing positively supercoiled DNA exhibited a striking increase in DNase I sensitivity. They also displayed additional micrococcal nuclease cleavage sites but yielded nearly typical nucleosomal ladders after extensive digestion. Upon in vitro relaxation with eukaryotic topoisomerase I, the minichromosomes remained DNase I sensitive but were converted to negative DNA supercoiling with a slightly increased linking number compared to typical minichromosomes, thus indicating the presence of bound histones. Therefore, positive DNA supercoiling provides a mechanism for generating, but is not required for maintaining, a conformation in chromatin characteristic of highly transcribed genes.

Transcription-induced nucleosome 'splitting': an underlying structure for DNase I sensitive chromatin

Utilizing yeast strains containing promoter mutations, we demonstrate that transcription of the HSP82 gene causes nucleosomes toward the 3'-end to become DNase I sensitive and 'split' into structures that exhibit a 'half-nucleosomal' cleavage periodicity. Splitting occurs even when only a few RNA polymerase II molecules are engaged in basal level transcription or during the first round of induced transcription. The split nucleosomal structure survives nuclear isolation suggesting that it may be stabilized by post-translational modifications or non-histone proteins, and may require DNA replication for reversal to a whole nucleosomal structure. Split nucleosomes represent a structure for DNase I sensitive chromatin and are probably of common occurrence but difficult to detect experimentally. We suggest that transient positive supercoils downstream of traversing RNA polymerase lead to nucleosome splitting.

Identification within the simian virus 40 genome of a chromosomal loop attachment site that contains topoisomerase II cleavage sites

We demonstrate that the simian virus 40 genome contains a single MAR (matrix association region) that maps within a large T-antigen coding region (nucleotides 4071 to 4377). This region contains topoisomerase II cleavage sites, exhibits sequence similarity with cellular MARs, and recognizes the same evolutionarily conserved, abundant nuclear binding sites seen by cellular MARs.

Immunoglobulin kappa gene expression after stable integration. II. Role of the intronic MAR and enhancer in transgenic mice

MOPC41 immunoglobulin kappa gene constructs have been stably introduced into the mouse germ line to investigate the effects of deleting the conserved intronic sequences on gene expression. Intact kappa genes containing 1.5 kilobase pairs of upstream and 8.5 kilobase pairs of downstream flanking sequences were highly expressed tissue-specifically, raising the total level of kappa mRNA in spleens severalfold in most transgenic animals. This high expression was often accompanied by marked suppression of endogenous kappa gene activity. Transgenes containing a deletion of the matrix association region (MAR) or both the MAR and enhancer were expressed tissue-specifically at mean levels only 2- and 3-fold lower, respectively, than that of intact transgenes. Therefore, while the intronic enhancer and MAR appear to play a quantitative role in gene expression, these sequences are not absolutely essential for transcriptional activation of rearranged kappa genes in a normal developmental environment.

Immunoglobulin kappa gene expression after stable integration. I. Role of the intronic MAR and enhancer in plasmacytoma cells

Rearranged MOPC41 immunoglobulin kappa gene constructs have been stably introduced into cultured S194 mouse plasmacytoma cells to investigate the effects of deleting the intronic enhancer and/or matrix association region (MAR) on gene expression. Intact single-copy kappa genes containing 1.5 kilobase pairs of upstream and 8.5 kilobase pairs of downstream flanking sequences exhibited sensitivity to chromosome position effects and were expressed at a mean level of 27% relative to the endogenous kappa gene expression or only 6% with respect to the MOPC41 kappa mRNA levels in the tumor. Deletion of the intronic MAR led to a 4-fold decrease in expression, while deletion of both the MAR and enhancer led to an 11-fold decline. These effects were dampened by preselecting for integration into a transcriptionally poised chromatin location as demonstrated by linkage to a selectable marker which lacked both a MAR and an enhancer. Significantly, we found that sequences downstream of the poly(A) addition site compensated 150-fold for deletion of the intronic enhancer.

Basal-level expression of the yeast HSP82 gene requires a heat shock regulatory element

Previous studies have shown that heat shock factor is constitutively bound to heat shock elements in Saccharomyces cerevisiae. We demonstrate that mutation of the heat shock element closest to the TATA box of the yeast HSP82 promoter abolishes basal-level transcription without markedly affecting inducibility. The mutated heat shock element no longer bound putative heat shock factor, either in vitro or in vivo, but still resided within a nuclease-hypersensitive site in the chromatin. Thus, constitutive binding of heat shock factor to heat shock elements in S. cerevisiae appears to functionally direct basal-level transcription.

Dysfunction of chromosomal loop attachment sites: illegitimate recombination linked to matrix association regions and topoisomerase II

A family of A + T-rich sequences termed MARs ("matrix association regions") mediate chromosomal loop attachment. Here we demonstrate that several MARs both specifically bind and contain multiple sites of cleavage by topoisomerase II, a major protein of the mitotic chromosomal scaffold. Interestingly, "hotspots" of enzyme cutting occur within the MAR of the mouse immunoglobulin kappa-chain gene at the breakpoint of a previously described chromosomal translocation. Since topoisomerase II can mediate illegitimate recombination in prokaryotes, we explored further the possibility that MARs might be targets for this process in eukaryotes. We found that a MAR had been deleted from one of the two rabbit immunoglobulin kappa-chain genes and that MARs reside next to a long interspersed repetitive element within the recombination junction of a human ring chromosome 21. These results, taken together with other accounts of nonhomologous recombination, lead to the proposal that a dysfunction of MARs is illegitimate recombination.

Protein:DNA interactions at chromosomal loop attachment sites

We have recently identified an evolutionarily conserved class of sequences that organize chromosomal loops in the interphase nucleus, which we have termed "matrix association regions" (MARs). MARs are about 200 bp long, AT-rich, contain topoisomerase II consensus sequences and other AT-rich sequence motifs, often reside near cis-acting regulatory sequences, and their binding sites are abundant (greater than 10,000 per mammalian nucleus). Here we demonstrate that the interactions between the mouse kappa immunoglobulin gene MAR and topoisomerase II or the "nuclear matrix" occur between multiple and sometimes overlapping binding sites. Interestingly, the sites most susceptible to topoisomerase II cleavage are localized near the breakpoints of a previously described illegitimate recombination event. The presence of multiple binding sites within single MARs may allow DNA and RNA polymerase passage without disrupting primary loop organization.

Vacuum blotting: a simple method for transferring DNA from sequencing gels to nylon membranes

We describe a vacuum blotting procedure for transferring DNA fragments from conventional polyacrylamide sequencing gels to nylon membranes. The method employs a combination of vacuum-assisted diffusion (effected by a standard gel drier) and an osmotic gradient (effected by over- and underlying filters presoaked in ammonium acetate). Fragments up to 310 nucleotides in length transfer at 40-60% efficiency within 90 min. When combined with indirect end-labelling, the method allows genomic sequencing of a single-copy gene of Saccharomyces cerevisiae employing as little as 5 micrograms DNA per lane.

The enhancer of the immunoglobulin heavy chain locus is flanked by presumptive chromosomal loop anchorage elements

We have located presumptive chromosomal loop anchorage elements within the mouse heavy chain immunoglobulin locus. Analysis of 31 kilobases spanning diversity, joining, enhancer, switch, and the mu and delta constant regions reveals that only a single 1-kilobase segment exhibits specific binding to nuclear matrices. It is of particular significance that the transcriptional enhancer element resides within this matrix association region (MAR). Fine structure mapping indicates that binding is mediated by A+T-rich approximately 350-base pair segments that reside on either side of the enhancer. The MAR sequences residing 5' of the enhancer contain topoisomerase II consensus sequences like the MAR located upstream of the kappa light chain gene enhancer. The heavy chain gene MARs, however, exhibit a lower affinity for matrix association compared to the kappa gene MAR. Significantly, the juxtaposition of enhancer elements with MARs appears to be evolutionarily conserved within the immunoglobulin genes, suggesting that MARs may act as positive and/or negative regulators of enhancer function.

Sharp boundaries demarcate the chromatin structure of a yeast heat-shock gene

In both induced and basally transcribed states, the chromatin structure of the yeast HSP82 heat-shock locus exhibits a remarkable degree of organization with respect to DNA sequence. The promoter region contains a constitutive DNase I hypersensitive site. The transcription unit is markedly sensitive to DNase I, and exhibits a sharp transition from a phased half- to a whole nucleosomal cleavage periodicity at the 3' end. Distant upstream and downstream regions are also organized into distinct arrays of phased nucleosomes. Each array is demarcated by DNase I hypersensitive sites that display internal protected regions, suggesting the presence of DNA binding proteins. In addition, since these sites are of mononucleosomal DNA length, they may acquire a nucleosomal structure under certain environmental conditions without disrupting flanking nucleosomal phasing frames. Thus, the HSP82 locus is organized into specific, phased, chromatin structures that appear to function in transcriptional initiation, RNA polymerase passage, transcriptional termination, and the establishment of chromatin-domain microenvironments.

Chromosomal loop anchorage sites appear to be evolutionarily conserved

We have previously identified a class of DNA sequence elements, termed matrix association regions (MARs), which specifically bind to nuclear matrices in vitro and are believed to be at the bases of chromosomal loops in vivo. Here we demonstrate that nuclear matrices prepared from the yeast Saccharomyces cerevisiae will specifically bind an MAR sequence derived from the mouse kappa light chain immunoglobulin gene. This suggests that both MAR sequences and their binding sites have been strongly evolutionarily conserved.

The ubiquitous potential Z-forming sequence of eucaryotes, (dT-dG)n . (dC-dA)n, is not detectable in the genomes of eubacteria, archaebacteria, or mitochondria

The potential Z-forming sequence (dT-dG)n . (dC-dA)n is an abundant, interspersed repeat element that is ubiquitous in eucaryotic nuclear genomes. We report that in contrast to eucaryotic nuclear DNA, the genomes of eubacteria, archaebacteria, and mitochondria lack this sequence, since even a single tract of greater than or equal to 14 base pairs in length is not detectable through either hybridization or sequence analysis. Interestingly, the phylogenetic distribution of the (dT-dG)n . (dC-dA)n repeat exhibits a striking parallel to that of (dT-dC)n . (dG-dA)n, but not to other homocopolymeric sequences such as (dC-dG)n . (dC-dG)n or (dT-dA)n . (dT-dA)n.

The active immunoglobulin kappa chain gene is packaged by non-ubiquitin-conjugated nucleosomes

To elucidate the molecular features of active chromatin, we have mapped, by two-dimensional electrophoresis, the protein composition of nucleosomes that package the immunoglobulin kappa chain gene of mouse plasmacytoma cells. Nucleoprotein particles that possess the active kappa chain gene comigrate with bulk mononucleosomes that contain high mobility group proteins HMG-14 or -17 but lack histone H1. High electrophoretic resolution of the underlying core particles, after removal of ubiquitin by isopeptidase treatment, reveals that these nucleosomes are nonubiquitinated, even though they coincidently migrate with bulk ubiquitinated particles. This distinctive electrophoretic behavior may be correlated with the presence of histone H2A.X. Nucleosomes exhibiting these unusual properties appear to span at least 10 kilobases, in both transcribed and nontranscribed regions, suggesting that mechanisms independent of transcription exist to initiate, maintain, and propagate a common chromatin phenotype over long distances along the kappa chain locus.

Temperature-dependent cleavage of chromatin by micrococcal nuclease near the nucleosome center

Digestion of nuclei at 4 degrees C with micrococcal nuclease results in significant intranucleosomal cleavage compared to digestion conducted at 37 degrees C. Employing nucleoprotein gel electrophoresis in one dimension followed by DNA electrophoresis in a second dimension, we demonstrate that such temperature-sensitive, internal cleavage predominantly occurs about 20 bp from the nucleosome center. We suggest that lower temperatures reduce the stability of hydrophobic interactions within the histone octamer and lead to a conformational alteration in nucleosomes that is detected by micrococcal nuclease.

Transcription termination and chromatin structure of the active immunoglobulin kappa gene locus

To investigate the chromatin surrounding an active gene, we have determined the distribution of RNA polymerase molecules, the intactness of nucleosomal structure, and the subnuclear compartmentalization along 15 kilobase pairs (kb) of the mouse kappa immunoglobulin locus of MPC-11 plasmacytoma cells. Hybridization of in vitro nuclear transcripts to probes specific for the template strand reveals that transcription terminates within the region between 1.1 and 2.3 kb downstream from the poly(A) addition site. Ten different short sequences (8-13 base pairs) reside within 460 base pairs of this termination region that exhibit homology with sequences found in the termination regions of mouse beta-globin and chicken ovalbumin genes. Transcription of the nontemplate strand occurs on either side of this termination region. We find that both within the transcription unit and 6.5 kb downstream of the termination region of the kappa gene, the canonical nucleosomal structure is perturbed, the chromatin exhibits pronounced insolubility, and the nucleosomes liberated by micrococcal nuclease appear to lack histone H1. The insolubility is characterized by interactions that are disrupted by 0.3 to 0.6 M NaCl treatment. We conclude that the active chromatin phenotype spreads a considerable distance along the kappa locus, well beyond the region of transcription termination.

Chromosomal loop anchorage of the kappa immunoglobulin gene occurs next to the enhancer in a region containing topoisomerase II sites

Introduction of torsional stress into active chromatin domains requires that linear DNA molecules be anchored in vivo to impede free rotation. While searching for these anchorage elements, we have localized a nuclear matrix association region (MAR) within the mouse immunoglobulin kappa gene that contains two topoisomerase II sites and is adjacent to the tissue-specific enhancer. The same matrix contact occurs when the kappa locus is in germ-line (inactive) or rear-ranged (transcribed) configurations. This constitutive anchorage site partitions the gene into V-J and C region chromatin domains. We demonstrate that at least 10,000 similar and evolutionarily conserved MAR binding sites exist in the nucleus. We propose that these sites, in association with topoisomerase II and possibly in conjunction with enhancers, play fundamental roles in the functional organization of chromatin loop domains.

Chromatin structure of the potential Z-forming sequence (dT-dG)n X (dC-dA)n. Evidence for an "alternating-B" conformation

The sequence (dT-dG)n X (dC-dA)n is the most abundant purine-pyrimidine dinucleotide repeat in eukaryotic genomes. This sequence and certain others that contain alternating purine-pyrimidine residues have been shown to adopt the left-handed, Z-DNA conformation in vitro when subjected to negative torsional stress or elevated ionic strengths. We have asked whether (dT-dG)n X (dC-dA)n tracts exist in topologically constrained Z-form structures in vivo by examining the chromatin organization of these sequences in cultured mouse cell nuclei. We find that these elements are quantitatively packaged into typical core particles which are embedded in canonical polynucleosomal arrays. In addition, these sequences neither flank nor reside within regions of chromatin that are preferentially sensitive to S1 nuclease. These characteristics suggest that these tracts do not exist predominantly in the Z-form in vivo. Furthermore, employing techniques that permit prominent hybridization to DNA fragments as short as 18 bases, we provide evidence that in vivo, most (dT-dG)n X (dC-dA)n elements instead adopt an "alternating-B" conformation on the nucleosomal surface.

Differentiation-dependent chromatin alterations precede and accompany transcription of immunoglobulin light chain genes

We have studied the nature of chromatin alterations along immunoglobulin light chain (IgL) genes during B cell development using cultured murine cell lines. Employing a chromatin fractionation procedure on micrococcal nuclease-treated nuclei, we demonstrate that transcriptionally active k IgL chromatin lacks a canonical nucleosomal repeat and exhibits a pronounced association with insoluble nuclear material but is processed by nuclease to a soluble nucleosomal component that apparently lacks histone H1 and is enriched in high mobility group proteins. Of particular significance, utilizing a variant plasmacytoma cell line that has transcriptionally inactivated one k allele via a promoter deletion, we demonstrate that transcription per se is not responsible for these novel alterations. Furthermore, we show that the chromatin encompassing germline (unrearranged) and transcriptionally silent lambda IgL alleles in k-producing plasmacytomas exhibit some of the same unusual properties that are displayed by k alleles. Finally, we demonstrate that these alterations only occur in cell lines of the lymphocyte lineage that have progressed past the early pre-B cell stage; when inactive, both k and lambda IgL genes possess typical nucleosomal packaging and co-fractionate with histone H1-containing chromatin. These findings lead us to propose a model that predicts B cell stage-specific alterations in IgL chromatin prior to gene rearrangement and transcription.