Chromatin remodeling facilitates DNA incision in UV-damaged nucleosomes.

The DNA repair machinery must locate and repair DNA damage all over the genome. As nucleosomes inhibit DNA repair in vitro, it has been suggested that chromatin remodeling might be required for efficient repair in vivo. To investigate a possible contribution of nucleosome dynamics and chromatin remodeling to the repair of UV-photoproducts in nucleosomes, we examined the effect of a chromatin remodeling complex on the repair of UV-lesions by Micrococcus luteus UV endonuclease (ML-UV endo) and T4-endonuclease V (T4-endoV) in reconstituted mononucleosomes positioned at one end of a 175-bp long DNA fragment. Repair by ML-UV endo and T4-endoV was inefficient in mononucleosomes compared with naked DNA. However, the human nucleosome remodeling complex, hSWI/SNF, promoted more homogeneous repair by ML-UV endo and T4-endo V in reconstituted nucleosomes. This result suggests that recognition of DNA damage could be facilitated by a fluid state of the chromatin resulting from chromatin remodeling activities.

Ultraviolet damage and nucleosome folding of the 5S ribosomal RNA gene.

The Xenopus borealis somatic 5S ribosomal RNA gene was used as a model system to determine the mutual effects of nucleosome folding and formation of ultraviolet (UV) photoproducts (primarily cis-syn cyclobutane pyrimidine dimers, or CPDs) in chromatin. We analyzed the preferred rotational and translational settings of 5S rDNA on the histone octamer surface after induction of up to 0.8 CPD/nucleosome core (2.5 kJ/m(2) UV dose). DNase I and hydroxyl radical footprints indicate that UV damage at these levels does not affect the average rotational setting of the 5S rDNA molecules. Moreover, a combination of nuclease trimming and restriction enzyme digestion indicates the preferred translational positions of the histone octamer are not affected by this level of UV damage. We also did not observe differences in the UV damage patterns of irradiated 5S rDNA before or after nucleosome formation, indicating there is little difference in the inhibition of nucleosome folding by specific CPD sites in the 5S rRNA gene. Conversely, nucleosome folding significantly restricts CPD formation at all sites in the three helical turns of the nontranscribed strand located in the dyad axis region of the nucleosome, where DNA is bound exclusively by the histone H3-H4 tetramer. Finally, modulation of the CPD distribution in a 14 nt long pyrimidine tract correlates with its rotational setting on the histone surface, when the strong sequence bias for CPD formation in this tract is minimized by normalization. These results help establish the mutual roles of histone binding and UV photoproducts on their formation in chromatin.

Strand-specific modulation of UV photoproducts in 5S rDNA by TFIIIA binding and their effect on TFIIIA complex formation.

The relationship between UV-induced photoproduct formation and transcription factor binding was studied in a 214 bp fragment containing the entire Xenopus borealis 5S rRNA gene. DNA mobility shift and DNase I footprinting show a strong inhibition of TFIIIA binding to UV-damaged 5S rDNA. An average of approximately 2 cyclobutane pyrimidine dimers (CPDs) per 214 bp fragment, and a lesser amount of pyrimidine-pyrimidone (6-4) dimers, reduced the fraction of TFIIIA bound by approximately 70%. Furthermore, irradiation of the TFIIIA/5S rDNA complex displaces TFIIIA at doses of 0.8-2 CPDs/fragment, indicating the complex is unable to accommodate UV photoproducts. UV photofootprinting of the 50 bp TFIIIA binding region of 5S rDNA (or ICR) shows that TFIIIA binding modulates photoproduct formation primarily in the template strand. Formation of CPDs at six different sites is strongly inhibited, while another CPD site is strongly enhanced, by TFIIIA binding. Most of these sites are located in one of three boxes (A, IE, or C) designated as TFIIIA contact sites in the ICR, while one site is between these boxes. Formation of (6-4) dimers is also inhibited at several sites in the template strand by TFIIIA binding. However, formation of photoproducts in the nontemplate strand is much less affected by TFIIIA binding, where only one CPD site is inhibited in the complex. These data indicate that formation of UV photoproducts in 5S rDNA can be markedly affected by TFIIIA binding, and complex formation is inhibited by UV photoproducts.

The conformation of loop E of eukaryotic 5S ribosomal RNA.

The solution structure of a 27-nucleotide duplex, including the internal loop E from Xenopus laevis 5S ribosomal RNA, has been studied by two-dimensional NMR spectroscopy, followed by restrained molecular dynamics. The highly conserved internal loop closes to form a G.A base pair and a reverse-Hoogsteen A.U base pair. Extensive interstrand stacking between these uncommon base pairs provides a structural explanation for an interstrand ultraviolet-induced cross-link. A guanosine residue is bulged into the major groove and may form a base-triple with the adjacent reverse-Hoogsteen A.U pair. The structure of the less highly-conserved portion of the loop is less well-defined by the NMR data. A single-nucleotide deletion mutant has a very different, open conformation without mismatched base pairs [Varani, G., Wimberly, B., & Tinoco, I. Jr. (1989) Biochemistry 28, 7760-7772]. The implications of the structure for binding of the transcription factor TFIIIA and the cytotoxin alpha-sarcin are discussed.

B----A transitions within a 5 S ribosomal RNA gene are highly sequence-specific.

The UV footprinting technique has been used to detect and map, at single nucleotide resolution, the formation of A conformations within a sea urchin 5S ribosomal RNA gene. Increasing amounts of the dehydrating agent, trifluorethanol, were used to induce the B----A transition. Our measurements argue that the B----A transition is highly sequence-specific. Fourteen different sequences within a fragment of DNA bearing the 5 S gene were found to undergo the B----A transition independently of one another. There is a striking relationship between the midpoint of the B----A transition for each stretch of DNA and its (G+C) content. DNA sequences at the boundary between A and B conformations do not appear to be significantly distorted. A (dAdT)8 tract at the 3' end of the 5 S gene undergoes the B----A transition in two cooperative steps suggesting that for some sequences the B----A transition may actually proceed through the formation of a previously unidentified intermediate. Although the sequence specificity of the B----A transition may be exploited by regulatory proteins when they bind DNA, our measurements argue that binding of the Xenopus laevis transcription factor 111A to 5 S genes does not.