Interaction of Proteins with Inverted Repeats and Cruciform Structures in Nucleic Acids

Abstract

Cruciforms occur when inverted repeat sequences in double-stranded DNA adopt intra-strand hairpins on opposing strands. Biophysical and molecular studies of these structures confirm their characterization as four-way junctions and have demonstrated that several factors influence their stability, including overall chromatin structure and DNA supercoiling. Here, we review our understanding of processes that influence the formation and stability of cruciforms in genomes, covering the range of sequences shown to have biological significance. It is challenging to accurately sequence repetitive DNA sequences, but recent advances in sequencing methods have deepened understanding about the amounts of inverted repeats in genomes from all forms of life. We highlight that, in the majority of genomes, inverted repeats are present in higher numbers than is expected from a random occurrence. It is, therefore, becoming clear that inverted repeats play important roles in regulating many aspects of DNA metabolism, including replication, gene expression, and recombination. Cruciforms are targets for many architectural and regulatory proteins, including topoisomerases, p53, Rif1, and others. Notably, some of these proteins can induce the formation of cruciform structures when they bind to DNA. Inverted repeat sequences also influence the evolution of genomes, and growing evidence highlights their significance in several human diseases, suggesting that the inverted repeat sequences and/or DNA cruciforms could be useful therapeutic targets in some cases.

Keywords: DNA base sequence; DNA structure; DNA supercoiling; cruciform; epigenetics; genome stability; inverted repeat; replication; transcription.

Figure 1

Inverted repeat sequences can form different types of double-stranded conformations. (A) Transition of inverted repeat in a linear conformation to a double hairpin, cruciform state. For the sequence indicated, the cruciform structure consists of four branchpoints and two 7 bp-long stems, each with 4 nt loops. (B) Decisive factors for the resulting thermodynamic stability and genomic occurrence of cruciform structures are: (1) stem size indicated in blue; (2) loop length indicated in purple; (3) possible mismatches in base pairing indicated in red. The arrows at the top and bottom of part (B) highlight the relative stability and occurrence of the represented cruciforms, with the larger and darker part of the arrows indicating those that are most stable and are most likely to occur in genomes. For all schematic molecules, the arrow indicates the 3′-end of the DNA strand.

Figure 2

High-resolution structure of a cruciform (four-way junction) formed by an inverted repeat DNA sequence. Images show the X-ray crystallographic structure determined at 2.10 Å for DNA with the sequence 5′-CCGGTACCGG-3′ (1DCW) [37]. The DNA alone forms a four-way junction in a stacked-X conformation, in which duplexes coaxially stack, with each pair of stacked duplexes related by +30° to +60° (right-handed) rotation. The continuous (least distorted relative to B-DNA) strands are coloured as green and red, while those of the crossing strands (making a tight U-turn) are coloured blue and cyan. In each panel, the images show the structure visualised via different axes viewpoints as indicated by the coloured squares. (A) The upper image provides a schematic view of the molecule, the distinct strands (in different colours), and their sequences, with arrows indicating the 3′-ends of the DNA strands. The lower image presents the high-resolution structure of 1DCW, illustrating its arrangement of base pairs. (B) The upper image views the structure down the helix axis of one pair of stacked duplexes, while the lower image views it from a rotational shift of approximately 90°. (C) The images zoom in on the central part of the structure (dashed bracketed region in (B)) to highlight the electrostatic interactions, particularly close to the Na⁺ ion at its centre. The lower image views the same face of the dyad axis shown in (B), and the upper image shows the opposite face of the axis, viewed from a rotational shift of approximately 180°.

Figure 3

The occurrence of inverted repeat sequences in gene features as determined by bioinformatic analyses. An idealised gene and its regulatory sequences are shown, with UTR referring to “untranslated regions”. A relative abundance (+) or depletion (−) of inverted repeats in the indicated genomes is highlighted above and below the idealised gene, respectively. For E. coli and S. cerevisiae, inverted repeats with a stem length from 5 bp and a spacer length up to 8 bp were considered [50,51], while for H. sapiens and viruses from the Nidovirales order, inverted repeats with the stem length from 6–30 bp and spacer length up to 10 bp were taken into account [18,42].

Figure 4

Cellular processes influenced by cruciform structures.