Identification of (CAG)(n) and (CGG)(n) repeat-binding proteins, CAGERs expressed in mature neurons of the mouse brain.

The trinucleotide repeats (CAG)(n) and (CGG)(n) have been shown to be expanded in responsible genes of several human hereditary neurological disorders. In studies of mice, we previously identified two homologous single-stranded (ss)(CAG) and ss(CGG) repeat-binding proteins, CAGER-1 (44 kDa) and CAGER-2 (40 kDa) (CAG-element-recognizing proteins). The specific binding activities of these proteins were predominantly detected in the mouse brain. We have isolated the cDNAs encoding CAGER-1 and CAGER-2 and found that they were identical to previously reported cDNAs for Puralpha and Purbeta, respectively. Puralpha of 28 kDa was previously identified as a replication-origin-binding protein that is ubiquitously expressed in proliferating cells. We show that the transcripts of CAGERs increase after birth and are detected at high levels in the adult mouse brain but at very low or virtually undetectable levels in other mouse tissues. Biochemical properties and molecular weights are different between CAGERs and Puralpha/beta. Immunostaining with specific antibodies against CAGERs indicates that CAGERs in the mouse brain reside in nonproliferating neurons but not in proliferating glia. We conclude that CAGERs and Puralpha/beta are unrelated proteins, and CAGERs are neuronal single-stranded sequence-binding proteins in the mouse brain. Misassignment of cDNAs is described.

Molecular characterization of a common fragile site (FRA7H) on human chromosome 7 by the cloning of a simian virus 40 integration site.

Common fragile sites are chromosomal loci prone to breakage and rearrangement, hypothesized to provide targets for foreign DNA integration. We cloned a simian virus 40 integration site and showed by fluorescent in situ hybridization analysis that the integration event had occurred within a common aphidicolin-induced fragile site on human chromosome 7, FRA7H. A region of 161 kb spanning FRA7H was defined and sequenced. Several regions with a potential unusual DNA structure, including high-flexibility, low-stability, and non-B-DNA-forming sequences were identified in this region. We performed a similar analysis on the published FRA3B sequence and the putative partial FRA7G, which also revealed an impressive cluster of regions with high flexibility and low stability. Thus, these unusual DNA characteristics are possibly intrinsic properties of common fragile sites that may affect their replication and condensation as well as organization, and may lead to fragility.

A matrix attachment region (MAR)-binding activity due to a p114 kilodalton protein is found only in human breast carcinomas and not in normal and benign breast disease tissues.

A M(r) 114,000 protein (p114) that specifically binds to nuclear matrix attachment DNA (matrix attachment region, MAR) from a breast carcinoma cell line SK-BR-3 was purified to near homogeneity. p114 strongly binds to a wild-type A+T-rich MAR probe with high unwinding propensity with a dissociation constant (Kd) of 10(-9), while it exhibits substantially reduced binding to a mutated A+T-rich non-MAR probe, which lacks unwinding propensity. This binding specificity and affinity is similar to the previously cloned thymocyte-associated MAR-binding protein SATB1. By Southwestern blot analysis, the MAR-binding activity of p114 is detectable in human breast carcinomas but is undetectable in normal breast tissues, benign breast diseases, and immortalized epithelial MCF-10A cells. Thus, the MAR-binding activity of p114 is not merely reflecting cell proliferation, but it strongly associates with breast carcinomas. The p114 MAR-binding activity was found in all 43 human breast carcinoma specimens tested, without exception. Much stronger p114 MAR-binding activity was detected in poorly differentiated than well-differentiated carcinomas. p114 may be a reliable diagnostic and possibly prognostic marker for breast cancer.

Single-stranded DNA binding proteins isolated from mouse brain recognize specific trinucleotide repeat sequences in vitro.

Expansion of trinucleotide repeats (CAG)n and (CGG)n is found in genes responsible for certain human hereditary neurodegenerative diseases. By gel-mobility shift assay, we detected a single-stranded (AGC)n repeat-binding activity primarily in mouse brain extracts and very low or undetectable activity in other tissue extracts. Two (AGC)n-repeat binding proteins, with apparent molecular weights of 44 and 40 kDa, have been purified from mouse adult brain by a DNA affinity column and fast protein liquid chromatography. UV-cross linking of radiolabeled (AGC)n repeats with crude brain extracts and with purified two proteins of 44 and 40 kDa produced identical doublet bands, indicating that these proteins are in fact responsible for the (AGC)n-binding activity in brain extracts. We designated these two proteins TRIP-1 for the 44 kDa protein and TRIP-2 for the 40 kDa protein, where TRIP represents trinucleotide repeat-binding protein. TRIP-1 and TRIP-2 bind to a specific subset of trinucleotide repeat sequences including (AGC)n, (AGT)n, (GGC)n, and (GGT)n repeats but not to various other trinucleotide repeats. A minimum of eight (AGC) trinucleotide repeating units is required for TRIP-1 and -2 recognition and binding. The (AGC)n repeat-binding activity increases in the brain after birth and reaches a plateau within 3 weeks. In the brain, TRIP-1 and TRIP-2 may alter the function of the genes containing the expanded-trinucleotide repeats.

A novel DNA-binding motif in the nuclear matrix attachment DNA-binding protein SATB1.

The nuclear matrix attachment DNA (MAR) binding protein SATB1 is a sequence context-specific binding protein that binds in the minor groove, making virtually no contact with the DNA bases. The SATB1 binding sites consist of a special AT-rich sequence context in which one strand is well-mixed A's, T's, and C's, excluding G's (ATC sequences), which is typically found in clusters within different MARs. To determine the extent of conservation of the SATB1 gene among different species, we cloned a mouse homolog of the human STAB1 cDNA from a cDNA expression library of the mouse thymus, the tissue in which this protein is predominantly expressed. This mouse cDNA encodes a 764-amino-acid protein with a 98% homology in amino acid sequence to the human SATB1 originally cloned from testis. To characterize the DNA binding domain of this novel class of protein, we used the mouse SATB1 cDNA and delineated a 150-amino-acid polypeptide as the binding domain. This region confers full DNA binding activity, recognizes the specific sequence context, and makes direct contact with DNA at the same nucleotides as the whole protein. This DNA binding domain contains a novel DNA binding motif: when no more than 21 amino acids at either the N- or C-terminal end of the binding domain are deleted, the majority of the DNA binding activity is lost. The concomitant presence of both terminal sequences is mandatory for binding. These two terminal regions consist of hydrophilic amino acids and share homologous sequences that are different from those of any known DNA binding motifs. We propose that the DNA binding region of SATB1 extends its two terminal regions toward DNA to make direct contact with DNA.

The non-B-DNA structure of d(CA/TG)n differs from that of Z-DNA.

Chemical probing of two predominantly alternating purine-pyrimidine d(CA/TG)n repeats led us to propose previously that in supercoiled plasmids these elements adopt a non-B-DNA structure distinct from that of Z-DNA formed by d(CG)n sequences. Here, we present further evidence supporting this contention. Reactivity with the conformation-sensitive reagent chloroacetaldehyde, which reacts with unpaired adenines and cytosines, was confined strictly to adenines in the d(CA/TG)n repeat. In contrast, only bases outside the d(CG)n repeat exhibited chloroacetaldehyde reactivity. Two-dimensional gel analysis of topoisomers containing d(CA/TG)n tracts with bases out of strict purine-pyrimidine alteration revealed multiple superhelical-dependent transitions to an alternative left-handed structure. Within individual plasmid molecules, these multiple transitions resulted from the stepwise conversion of contiguous segments of alternating purine-pyrimidine sequence, which are delimited by bases out of alternation, to the full-length alternative conformation. When the left-handed helices increased in length to include more bases out of alternation, the average helical pitch changed substantially to produce a less tightly wound left-handed helix. Overall, these data indicate that d(CA/TG)n tracts adopt a left-handed conformation significantly different from that of the canonical Z-DNA structure of d(CG)n sequences.

A single trinucleotide, 5'AGC3'/5'GCT3', of the triplet-repeat disease genes confers metal ion-induced non-B DNA structure.

Expansion of (AGC)n repeats has been associated with genetic disorders called triplet-repeat diseases such as Huntington's disease (HD), myotonic muscular dystrophy (DM) and Kennedy's disease. To gain insight into the abnormal behavior of these repeats, we studied their structural properties in supercoiled DNA. Chemical probing revealed that, under physiological salt and pH conditions, Zn2+ or Co2+ ions induce (AGC)n repeats to adopt a novel non-B DNA structure in which all cytosine but none of adenine residues in either strand become unpaired. The minimum size of (AGC)n repeat that could form this structure independently of neighboring sequences is a single unit of double-stranded trinucleotide, 5'AGC3'/5'GCT3'. Other trinucleotide units of the same nucleotide composition, 5'CAG3'/5'CTG3' or 5'GCA3'/5'TGC3', do not form non-B DNA structures. This unusual DNA structural properly adopted by a single 5'AGC3'/5'GCT3' trinucleotide may contribute to expansion of (AGC)n sequences in triplet-repeat diseases.

Transcription-dependent recombination induced by triple-helix formation.

The homologous recombination between direct repeat sequences separated by either 200 or 1000 bp was induced by active transcription of the downstream gene when poly(dG)-poly(dC) sequences exist between the two direct repeats. This dG tract-mediated and transcription-induced recombination was RecA independent, and the frequency of recombination was dependent on both the length and the orientation of the poly(dG)-poly(dC) sequences relative to the gene. An intramolecular dG.dG.dC triplex formation was detected in Escherichia coli cells in a length-dependent manner when the transcription of the downstream gene was activated. We suggest that the negative superhelical strain generated by active transcription of the downstream gene induces poly(dG)-poly(dC) sequences to adopt a triple-helix structure in vivo and that this structure brings two remote sequences together to stimulate homologous recombination.

Structural polymorphism of homopurine-homopyrimidine sequences at neutral pH.

Under superhelical strain and mildly acidic pH, polyd(GA).polyd(CT) sequences are known to adopt an H-y3 conformer of H-DNA. We studied the effects of the sequence length, metal ions, and pH on the structures formed by polyd(GA).polyd(CT) sequences in supercoiled plasmid DNA at bacterial superhelicity. The results from experiments that distinguish multiple structures forming in a given plasmid DNA population strongly suggest that polyd(GA).polyd(CT) sequences of 33 base-pairs or more can adopt both H-y3 and H-y5 isomeric forms of H-DNA at neutral pH either with or without magnesium ions. At pH 5 in the presence of zinc ions, H-DNA is formed. However, at pH 7 in the presence of zinc ions, polyd(GA).polyd(CT) sequences form an intramolecular d(GA).d(GA).d(CT) triplex similar to a dG.dG.dC triplex, which is stabilized with magnesium ions for poly(dG).poly(dC) sequences. Zinc ions also stabilize the dG.dG.dC triplex for relatively long poly(dG).poly(dC) sequences (more than 24 base-pairs at pH 7).

A tissue-specific MAR/SAR DNA-binding protein with unusual binding site recognition.

A human cDNA was cloned that encodes a DNA-binding protein (SATB1) that is expressed predominantly in thymus and binds selectively to the nuclear matrix/scaffold-associating DNAs (MARs/SARs). Missing nucleoside experiments showed that SATB1 selectively binds in a special AT-rich sequence context where one strand consists of mixed A's, T's, and C's, excluding G's (ATC sequences). When this feature is destroyed by mutation, SATB1 binding is greatly reduced even if the direct contact sequence remains intact. Conjunctional SATB1-binding sequences become stably unpaired in supercoiled DNA. Specific mutations that diminish the unwinding potential greatly reduce SATB1 binding. However, SATB1 does not bind single-stranded DNA. Chemical interference assays show that SATB1 binds along the minor groove with very little contact with the bases. This suggests that SATB1 recognizes the ATC sequence indirectly through the altered sugar-phosphate backbone structure present in the double-stranded DNA.

Intramolecular dG.dG.dC triplex detected in Escherichia coli cells.

The formation of an intramolecular dG.dG.dC triplex in Escherichia coli cells is demonstrated at single-base resolution. The intramolecular dG.dG.dC triplex structure was probed in situ for E. coli cells containing plasmid DNAs with varying lengths of poly(dG).poly(dC) tracts employing chloroacetaldehyde. This chemical probe reacts specifically with unpaired DNA bases. The triplex structure formed with the poly(dG).poly(dC) tracts of 35 and 44 base-pairs, but not with 25 base-pairs. The triplex was detected only one to two hours after the chloramphenicol treatment: the period at which the extracted plasmid DNA revealed the maximal superhelical density.

Biological significance of unwinding capability of nuclear matrix-associating DNAs.

Matrix attachment regions (MARs) are thought to separate chromatin into topologically constrained loop domains. A MAR located 5' of the human beta-interferon gene becomes stably base-unpaired under superhelical strain, as do the MARs flanking the immunoglobulin heavy chain gene enhancer; in both cases a nucleation site exists for DNA unwinding. Concatemerized oligonucleotides containing the unwinding nucleation site exhibited a strong affinity for the nuclear scaffold and augmented SV40 promoter activity in stable transformants. Mutated concatemerized oligonucleotides resisted unwinding, showed weak affinity for the nuclear scaffold, and did not enhance promoter activity. These results suggest that the DNA feature capable of relieving superhelical strain is important for MAR functions.

Altered gene expression correlates with DNA structure.

We examined the participation of triplex DNA structure in gene regulation using a poly(dG)-poly(dC) sequence as a model. We show that a poly(dG)-poly(dC) sequence, which can adopt an intramolecular dG.dG.dC triplex under superhelical strain, strongly augments gene expression when placed 5' to a promoter. The activity of this sequence exhibits a striking length dependency: dG tracts of 27-30 bp augment the expression of a reporter gene to a level comparable to that observed with the polyoma enhancer in mouse LTK- cells, whereas tracts of 35 bp and longer have virtually no effect. A supercoiled plasmid containing a dG tract of 30 bp competes in vivo for a trans-acting factor as revealed by reduction in the reporter gene transcription driven by the (dG)29/promoter of the test plasmid, while dGs of 35 bp and longer in the competition plasmid failed to compete. In purified supercoiled plasmid DNA at a superhelical density of -0.05, dG tracts of 32 bp and longer form a triplex, whereas those of 30 bp and shorter remain double-stranded under a PBS solution. These results suggest that a localized superhelical strain can exist, at least transiently, in mouse LTK- cells, and before being relaxed by topoisomerases this rapidly induces dG tracts of 35 bp and longer to adopt a triplex preventing the factor from binding. Thus, these data suggest that a poly(dG)-poly(dC) sequence can function as a negative regulator by adopting an intramolecular triple helix structure in vivo.

Detection of triple-helix related structures adopted by poly(dG)-poly(dC) sequences in supercoiled plasmid DNA.

Negative superhelical strain induces the poly(dG)-poly(dC) sequence to adopt two totally different types of triple-helices, either a dG.dG.dC triplex in the presence of Mg(+)+ at both neutral and acidic pHs or a protonated dC+.dG.dC triplex in the absence of Mg(+)+ ions at acidic pH (1). To examine whether there are still other types of non-B DNA structures formed by the same sequence, we constructed supercoiled plasmid DNAs harboring varying lengths of the poly(dG) tract, and the structures adopted by each supercoiled plasmid DNA were studied with a chemical probe, chloroacetaldehyde. The potential of a poly(dG)-poly(dC) sequence to adopt non-B DNA structures depends critically on the length of the tract. Furthermore, in the presence of Mg(+)+ and at a mildly acidic pH, in addition to the expected dG.dG.dC triplex detected for the poly(dG) tracts of 14 to 30 base pairs (bp), new structures were also detected for the tracts longer than 35 bp. The structure formed by a poly(dG) tract of 45 bp revealed chemical reaction patterns consistent with a dG.dG.dC triplex and protonated dC+.dG.dC triple-helices fused together. This structure lacks single-stranded stretches typical of intramolecular triplexes.

Torsional stress stabilizes extended base unpairing in suppressor sites flanking immunoglobulin heavy chain enhancer.

DNA sequences surrounding the immunoglobulin heavy chain (IgH) enhancer contain negative regulatory elements which are important for the tissue specificity of the enhancer. We have shown that sequences located both 5' and 3' of the enhancer, corresponding to the negative regulatory elements, become stably and uniformly unpaired over an extended length when subjected to torsional stress. These DNA sequences are also included within matrix association regions. The ability of the sequences to assume a stably unpaired conformation was shown by reactivity with chloroacetaldehyde which is specific for unpaired DNA bases, as well as two-dimensional gel electrophoresis of topoisomers. The sequences located 3' of the enhancer induce base unpairing in the direction of the enhancer. This unpaired region progressively expands to include as much as 200 base pairs as the ionic concentration decreases or superhelical density increases. When an ATATAT motif within a negative regulatory element located 3' of the enhancer was mutated, the extensive base-unpairing property was abolished. This base-unpairing property of DNA may be important for negative regulation of gene expression and attachment to the nuclear matrix.

Hierarchical binding of DNA fragments derived from scaffold-attached regions: correlation of properties in vitro and function in vivo.

On its upstream side, the human interferon-beta gene is flanked by a 7-kb SAR (scaffold-attached region) DNA element. The core of this element is determined and subjected to in vitro reassociations with isolated scaffolds. Binding properties of SAR fragments with decreasing length are quantified and related to consensus sequences like the topoisomerase II box and an ATATTT motif. Characteristics as the stoichiometry, affinity, and cooperativity of the binding process are shown to depend on the length of SAR DNA and suggest a model involving a multiple-site attachment to protein scaffolds. We propose a rational approach for predicting the SAR mediated transcriptional enhancements in vivo from their binding properties in a standardized in vitro assay. The efficiency of this approach is demonstrated for a marker (huIFN-beta) and a selector gene (neor).

Non-B DNA structure: preferential target for the chemical carcinogen glycidaldehyde.

The effect of DNA conformation on the reaction specificities of the chemical carcinogen glycidaldehyde (GDA) was examined. Supercoiled plasma DNA harboring a poly(dG)-poly(dC) tract, which folds sharply into halves from the center of the tract to form a tetra-stranded structure containing either a dG.dG.dC triplex structure in the presence of Mg2+ or a dC+.dC.dC triplex structure in the absence of Mg2+ was chosen as the reaction substrate. The reactive sites alkylated by GDA were determined at a single base resolution after these sites were specifically cleaved with a combination of either the hydrazine and piperidine or formic acid and piperidine reactions. The results show that at pH 5-7, GDA reacts preferentially with DNA bases that are involved in the altered DNA conformations. Interestingly, in addition to the known reaction of GDA with guanine residues, it was also found to be highly reactive with specific cytosine residues that reside in the altered DNA conformations. These GDA-reactive cytosine residues were unpaired as judged by their reactivity with the unpaired DNA base specific probe, chloroacetaldehyde. Therefore, it appears that DNA conformation plays a major role in determining the reaction specificities of GDA.