Acta Biochimica et Biophysica Sinica
Characterization of the association and sequestration of RNA-binding proteins by single-stranded DNA chimera
The biomolecular assemblies (condensates or aggregates) formed by mutant proteins are a pathological hallmark of neurodegenerative diseases. Some RNA-binding proteins (RBPs) are typically prone to aggregation that is closely associated with disease pathologies. These RBPs include numerous well-recognized pathogenic proteins, such as TAR DNA binding protein of 43 kDa (TDP-43), fused in sarcoma (FUS), ataxin-2 (Atx2), and poly(A)-binding protein nuclear 1 (PABPN1). Recent studies have revealed that liquid-liquid phase separation (LLPS), as a mechanism, underlies the highly dynamic and reversible granule formation of RBPs, and highlighted that multivalent RNA molecules play crucial roles in this process. These granules are necessary for diverse physiological functions, such as RNA splicing, trafficking, and even RNA storage, during stress. However, the aberrant phase transition of these mutant RBPs usually results in the formation of solid-like aggregates or inclusions within both the cytoplasm and nucleus. More importantly, aggregates formed by RBPs can sequester specific proteins, RNAs or other interacting partners, consequently contributing to RBP-related pathologies. For example, wild-type PABPN1 forms dynamic nuclear speckles with the assistance of poly(A) RNAs, whereas Ala expansion of PABPN1 results in the formation of aggregates, which are involved in the disease progression of oculopharyngeal muscular dystrophy (OPMD). Although the biological importance of various RBP granules is realized in either the cytoplasm or nucleus, how RNA regulates the formation of granules and the transition to aberrant RBP aggregates remains largely unknown. The interaction of a protein with other biomolecules (proteins, nucleic acids, etc.) is the prerequisite for the protein executing its normal biological function in cells. Identifying protein-protein and protein-RNA interactions is fundamental for the biochemical investigation of an individual protein and for attempts to understand the functional role of the protein. To date, many methods for studying protein-protein interactions have been developed on the basis of various principles, but it is still difficult to clarify whether the interactions between proteins, especially the RBPs involved, are direct or indirect, since RBPs generally bind to diverse RNAs closely and are incorporated into macromolecular ribonucleoprotein (RNP) complexes. We have taken several pairs of RBPs as examples, including TDP-35 (C-terminal 35-kDa fragment of TDP-43) with TDP-43 or TIA1, PABPN1 with a 25-kDa component of the mammalian cleavage factor I complex (CFIm25) and Atx2 with DEAD-box RNA helicase 6 (DDX6), and applied modified co-immunoprecipitation (Co-IP) and supernatant/pellet (S/P) fractionation experiments to characterize the association and sequestration of RBPs by using single-stranded DNA (ssDNA) chimera under ribonuclease (RNase) treatment. We designed several pieces of ssDNA oligonucleotides to mimic particular RNAs in cells that may mediate the association and sequestration of RBPs. The association of RBP proteins generally requires binding with multivalent RNA chains, since the bound RNAs tend to incorporate into a large protein-RNA complex with the help of RNA molecules. In Co-IP assay, especially for RBPs, RNase is often utilized to digest RNA in cell lysates to characterize whether the association of different RBPs is direct or indirect. It is important for us to demonstrate the active role of particular RNAs in the association or interaction of RBPs. Therefore, we designed and synthesized ssDNA chimeras to mimic the corresponding RNA that specifically bind to both RBPs simultaneously. In this case, ssDNA is used for rescuing the association of RBPs under the condition of RNase treatment, since the ssDNA oligonucleotide is resistant to nuclease activity. To design ssDNA chimeras for the RBPs of interest, first, the RNA sequences that bind to the two RPBs should be defined. The ssDNA should contain at least two portions (motifs) that specifically bind to each RBP, and each ssDNA portion may include 2–3 repeats of the binding sequence, so that the ssDNA can be recognized and bound efficiently by each RBP. Notably, the T base in ssDNA may sometimes be replaced with the U base (dU) for some more specific-binding RBPs, such as PABPN1. In the case of TDP-43 with Atx2, the binding specificities of the RNA sequences for TDP-43 and Atx2 are UG-rich and AUUUUU (AU5), respectively; then, the TG repeat portion is designed to bind to TDP-43, and the AT5 repeat is to bind to Atx2. Thus, an integrated method of co-IP and S/P fractionation was applied to characterize the association and sequestration of RBPs by combining ribonuclease (RNase) and ssDNA treatments.
Graphical AbstractVerified
