The Emerging Roles of Multimolecular G-Quadruplexes in Transcriptional Regulation and Chromatin Organization

Abstract

The ability of genomic DNA to adopt non-canonical secondary structures known as G-quadruplexes (G4s) under physiological conditions has been recognized for its potential regulatory function of various biological processes. Among those, transcription has recently emerged as a key process that can be heavily affected by G4 formation, particularly when these structures form at gene promoters. While the presence of G4s within gene promoters has been traditionally associated with transcriptional inhibition, in a model whereby G4s act as roadblocks to polymerase elongation, recent genomics experiments have revealed that the regulatory role of G4s in transcription is more complex than initially anticipated. Indeed, earlier studies linking G4-formation and transcription mainly relied on small-molecule ligands to stabilize and promote G4s, which might lead to disruption of protein-DNA interactions and local environments and, therefore, does not necessarily reflect the endogenous function of G4s at gene promoters. There is now strong evidence pointing toward G4s being associated with transcriptional enhancement, rather than repression, through multifaceted mechanisms such as recruitment of key transcriptional proteins, molding of chromatin architecture, and mode of phase separation. In this Account, we explore pivotal findings from our research on a particular subset of G4s, namely, those formed through interactions between distant genomic locations or independent nucleic acid strands, referred to as multimolecular G4s (mG4s), and discuss their active role in transcriptional regulation. We present our recent studies suggesting that the formation of mG4s may positively regulate transcription by inducing phase-separation and selectively recruiting chromatin-remodeling proteins. Our work highlighted how mG4-forming DNA and RNA sequences can lead to liquid-liquid phase separation (LLPS) in the absence of any protein. This discovery provided new insights into a potential mechanism by which mG4 can positively regulate active gene expression, namely, by establishing DNA networks based on distal guanine-guanine base pairing that creates liquid droplets at the interface of DNA loops. This is particularly relevant in light of the increasing evidence suggesting that G4 structures formed at enhancers can drive elevated expression of the associated genes. Given the complex three-dimensional nature of enhancers, our findings underscore how mG4 formation at enhancers would be particularly beneficial for promoting transcription. Moreover, we will elaborate on our recent discovery of a DNA repair and chromatin remodeling protein named Cockayne Syndrome B (CSB) that displays astonishing binding selectivity to mG4s over the more canonical unimolecular counterparts, suggesting another role of mG4s for molding chromatin architecture at DNA loops sites. Altogether, the studies presented in this Account suggest that mG4 formation in a chromatin context could be a crucial yet underexplored structural feature for transcriptional regulation. Whether mG4s actively regulate transcription or are formed as a mere consequence of chromatin plasticity remains to be elucidated. Still, given the novel insights offered by our research and the potential for mG4s to be selectively targeted by chemical and biological probes, we anticipate that further studies into the fundamental biology regulated by these structures can provide unprecedented opportunities for the development of therapeutic agents aimed at targeting nucleic acids from a fresh perspective.

Figure 1

General structural features of G-quadruplexes and their biological relevance. (a) Structure of a G-tetrad composed of four guanines interacting by Hoogsteen hydrogen bonding. M⁺ refers to a monovalent cation, with the order of stability K⁺ > Na⁺ > Li⁺. (b) Schematic representation of different G4 topologies and molecularities. Bimolecular and tetramolecular G4s are shown on the right, whereas different topologies of monomolecular G4s are displayed on the left. (c) Schematic illustration of the various biological processes by which G4 structures have been postulated to play a role in transcriptional regulation.

Figure 2

Proposed formation mechanism of a “super-G4” cluster by nonpromoter G4s. Super-G4 leads to epigenetic rewiring correlated to the increasing expression of genes important for conferring drug resistance in ovarian cancer. Figures were reproduced from ref (3). Available under a CC-BY ND license. Copyright 2023 Robinson et al.

Figure 3

Multimolecular G4s cluster and form a phase-separated entity in the (GGGGCC)_n hexanucleotide repeat. (a) Proposed mechanism of G4-mediated aggregation: (GGGGCC)_n, by virtue of being G-rich, forms multimolecular G4s (mG4), which further cluster into microscopic aggregates. (b) Fluorescence (left) and NMM (right) gels demonstrate the formation of mG4 species, indicated by the appearance of a higher molecular weight species with increasing concentration of K⁺ (G4 stabilizing cation). (c) NMM gel shows that mG4 formation depends on the number of (GGGGCC)_n repeats; a higher number of repeats allows better cross-linking between strands, which more readily form aggregates. (d) Bright field imaging of (GGGGCC)_n (n = 2–12) annealed under mG4-forming conditions at 250 μM oligonucleotide concentration. 100 μm scale bar. Figures were reproduced from ref (2). Available under a Creative Commons CC BY license. Copyright 2023 Raguseo et al.

Figure 4

CSB binds multimolecular G4s (mG4s) with high affinity and specificity. (a) rDNA sequences utilized in the EMSA experiment. The red star symbol refers to the Cy5 dye. (b) EMSA gels on rDNA1 (annealed in KCl and LiCl), rDNA2, and rDNA3 G4s under 0–5 nM of CSB-HD. (c) Binding isotherm showing the percentage of CSB-HD bound intermolecular G4 under increasing protein concentration. The gel images were analyzed using ImageJ, and the binding affinity (K_D) was calculated using Prism, fitting the binding curve to the “one site-specific binding” equation. All of the experiments were performed in biological duplicates. (d) Nucleoli localization of CSB protein in CSB-EGFP-expressing CS1AN cells. As the nuclei were occupied by CSB, probing with a G4-specific antibody, BG4, is inefficient, resulting in black spots in the nuclei locale. (e) Nucleoli of nontransfected CS1AN (top) and HeLa (bottom) cells, both of which are CSB-null, are efficiently stained by BG4. (f) Quantification of the number of cells without BG4 signal in nontransfected CS1AN cells, CSB-reinstated CS1AN cells, and HeLa cells. Figures were reproduced from ref (1). Copyright 2021 American Chemical Society.