Replication protein A binds RNA and promotes R-loop formation

Olga M. Mazina, Srinivas Somarowthu, Lyudmila Y. Kadyrova, Andrey G. Baranovskiy, Tahir H. Tahirov, Farid A. Kadyrov, Alexander V. Mazin

Preprint posted on 12 February 2020

Article now published in Journal of Biological Chemistry at

RPA can handle more than just binding ssDNA: a novel function for RPA in mediating R-loop formation and replication restart.

Selected by Giuseppina D'Alessandro

Categories: molecular biology


Figure 1: ssDNA or ssRNA invasion into the DNA duplex generates D-loop or R-loop, respectively. The hybridization of DNA and RNA generates a DNA:RNA hybrid.

Invasion of single-stranded DNA (ssDNA) or RNA (ssRNA) within the double-stranded DNA helix generates a D-loop or an R-loop, respectively (Figure 1). R-loops, which are composed of a DNA:RNA hybrid and a displaced ssDNA, form physiologically during transcription, when the nascent RNA anneals back to its template DNA. Recently, the focus on DNA:RNA hybrids changed and, from undesirable by-products of transcription and threats to genome stability, they became key players in several physiological processes, from the DNA damage response to the control of gene expression. The ssDNA binding protein RPA recognizes the displaced ssDNA of the R-loop (Nguyen, 2017), as well as many other ssDNA substrates generated during various physiological processes. Until now, while the ability of RPA to bind ssDNA was clear, the affinity for ssRNA or DNA:RNA hybrids remained obscure.

Mazina and colleagues reported a novel and unexpected function for RPA. They propose that RPA binds the ssRNA, promotes R-loop formation and that the human DNA polymerases use these R-loops to initiate DNA replication, most likely after replication blockage (Figure 2). With the exception of the RNA immunoprecipitation to prove RPA binding to RNA in cells, the study is mainly conducted in vitro. The authors monitor RPA affinity for ssRNA by electrophoretic mobility shift assays with synthetic radiolabeled oligonucleotides. They also evaluate the ability of radiolabeled ssRNA oligonucleotides to invade homologous supercoiled plasmids and form R-loops, in an adaptation of the classical D-loop assay. Of note, neither RPA promoted D-loop formation nor the canonical recombinase RAD51 mediated R-loop formation. Finally, by adding DNA polymerases and, eventually, accessory factors to the plasmid containing the RPA-generated R-loop, they observe that DNA polymerases can use the RNA to initiate DNA synthesis.

Figure 2: Invasion of the RPA-coated ssRNA into the duplex DNA generates an R-loop. DNA polymerases uses the RNA to restart replication after a blockage (adapted from Mazina et al).

Future directions and questions to the author:

This paper proposes an out-of-the-box model arising from the observations that RPA binds ssRNA and favours the formation of R-loops, which are extended by the DNA polymerases in vitro. These findings are supported by RNA-immunoprecipitation experiments in mouse cells, which also identified RPA as an RNA binding protein (He, 2016). Recently, Nguyen and colleagues also linked RPA to R-loops and proposed a model in which RPA binds the displaced ssDNA of the R-loops and promotes their degradation by RNase H1, thus safeguarding genome stability. How can the same protein promote DNA:RNA hybrids formation and preserve genome integrity? An intriguing question that already arose when, in 2013, the Koshland group reported that the yeast Rad51, one of the key factors in repair of DNA damage, promotes DNA:RNA hybrid formation.

The specificity of the model described arises from the observation that only RPA, but not the recombinases RAD51 and RAD52, promotes R-loops formation. Since the same author demonstrated that RAD52 binds RNA, I would be curious to see whether the observation also extends to RAD51.

In the future it will be interesting to validate these studies in cells. According to the proposed model, RPA favours the annealing of the presumably newly-transcribed RNA to its template DNA and forms an R-loop. As a consequence, replication restart could be favoured in transcribed regions rather than in non-transcribed ones. However, we cannot exclude that RPA-coated ssRNA also anneals to distant homologous regions within the genome, similarly to the yeast Rad51 (Wahba, 2013). Moreover, being the RPA affinity for RNA influenced by the sequence, different regions could restart replication with different efficiency. An additional future direction could involve testing the presence of consensus sequences or common features in the RNAs retrieved in the RNA-immunoprecipitation experiments.

Further reading:

  1. He C, Sidoli S, Warneford-Thomson R, Tatomer DC, Wilusz JE, Garcia BA, Bonasio R. 2016. High-Resolution Mapping of RNA-Binding Regions in the Nuclear Proteome of Embryonic Stem Cells. Mol Cell 64, 416-430.

This study combines protein-RNA photocrosslinking and mass spectrometry to identify the RNA-binding regions of nuclear proteins in embryonic stem cells.

  1. Nguyen HD, Yadav T, Giri S, Saez B, Graubert TA, Zou L. 2017. Functions of Replication Protein A as a Sensor of R-loops and a Regulator of RNaseH1. Mol Cell 65(5):832–847.e4.

The paper reporting that RPA interacts with RNase H1, stimulates its activity, and suppresses R-loops and associated genomic instability in cells.

  1. Wahba L, Gore SK, Koshland D. 2013. The homologous recombination machinery modulates the formation of RNA–DNA hybrids and associated chromosome instability. eLife 2: e00505.

The paper proving that the yeast Rad51 mediates DNA:RNA hybrids formation both at the site of transcription and at distant genomic locations.


Tags: dna:rna hybrids, r loops, rpa

Posted on: 22 June 2020


Read preprint (No Ratings Yet)

Have your say

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here