Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 8 (1), 1127

Identification of Functional Tetramolecular RNA G-quadruplexes Derived From Transfer RNAs

Affiliations

Identification of Functional Tetramolecular RNA G-quadruplexes Derived From Transfer RNAs

Shawn M Lyons et al. Nat Commun.

Erratum in

Abstract

RNA G-quadruplex (RG4) structures are involved in multiple biological processes. Recent genome-wide analyses of human mRNA transcriptome identified thousands of putative intramolecular RG4s that readily assemble in vitro but shown to be unfolded in vivo. Previously, we have shown that mature cytoplasmic tRNAs are cleaved during stress response to produce tRNA fragments that function to repress translation in vivo. Here we report that these bioactive tRNA fragments assemble into intermolecular RG4s. We provide evidence for the formation of uniquely stable tetramolecular RG4 structures consisting of five tetrad layers formed by 5'-terminal oligoguanine motifs of an individual tRNA fragment. RG4 is required for functions of tRNA fragments in the regulation of mRNA translation, a critical component of cellular stress response. RG4 disruption abrogates tRNA fragments ability to trigger the formation of Stress Granules in vivo. Collectively, our data rationalize the existence of naturally occurring RG4-assembling tRNA fragments and emphasize their regulatory roles.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Fig. 1
Fig. 1
5ʹTOG containing tiRNAs exist in two distinct and interconvertible forms. a G-quartet formation by guanines. b Predicted secondary structure of 5ʹtiRNAAla with a predominant hairpin structure. 5ʹTOG motif is indicted in green. c Indicated RNAs were ran on a 15% acrylamide denaturing gel (left panels) or native gel (right panels). Gels were post-stained with either SYBR gold to detect total RNA or NMM to detect RG4 structures. 5ʹtiRNAAla ran at the expected 31 nts, but also at ~120 nts. Only the more slowly running band stained with NMM, indicating it contained RG4s. This more slowly migrating band persisted under denaturing conditions. d Indicated 5ʹtiRNAs were ran on a 15% denaturing acrylamide gel and either upper or lower species were gel purified and reanalysed on a denaturing acrylamide gel and post-stained with SYBR gold. A portion of the lower band re-equilibrated to the upper band and vice-a-versa. e 5ʹtiRNAAla was equilibrated overnight in 100 mM of indicated salts. RNAs were analyzed on denaturing (left panels) or native gels (right panels) and post-stained with SYBR gold or NMM. Results demonstrate that NMM staining, more slowly migrating band disassembles in the presence of Li+ ions
Fig. 2
Fig. 2
NMR spectra confirm G-quadruplex formation. a Imino region of the1H NMR spectra of 5ʹtiRNAAla. Spectra were recorded at 25 °C in 90% H2O/10% D2O(v/v) in the presence of 50 mM KCl, 10 mM potassium phosphate, 0.1 mM EDTA, pH 6.8 (top panel), 150 mM NaCl, 10 mM sodium phosphate, 0.1 mM EDTA, pH 6.8 (middle panel), and 150 mM LiCl, 10 mM Tris-HCl, 0.1 mM EDTA, pH 6.8 (bottom panel). b Temperature dependence of the imino region of the 1H NMR spectra of 5ʹtiRNAAla. Spectra were recorded in 90% H2O/10% D2O(v/v) in the presence of 50 mM KCl, 10 mM potassium phosphate, 0.1 mM EDTA, pH 6.8 (left panels), 150 mM NaCl, 10 mM sodium phosphate, 0.1 mM EDTA, pH 6.8 (middle panels), and 150 mM LiCl, 10 mM Tris-HCl, 0.1 mM EDTA, pH 6.8 (right panels). c Imino region of 5ʹtiRNAAla in 90% H2O/10% D2O (lower panels) and after 12 h in D2O (upper panels) at 45 °C in the presence of 150 mM NaCl, 10 mM sodium phosphate, 0.1 mM EDTA, pH 6.8 (left panels) and 50 mM KCl, 10 mM potassium phosphate and 0.1 mM EDTA, pH 6.8 (right panels). d CD spectra of 5ʹtiRNAAla at 25 °C. The spectra were recorded in the presence of 50 mM KCl, 10 mM potassium phosphate and 0.1 mM EDTA, pH 6.8 (black), 150 mM NaCl, 10 mM sodium phosphate, 0.1 mM EDTA, pH 6.8 (red) and 150 mM LiCl, 10 mM Tris-HCl, 0.1 mM EDTA, pH 6.8 (blue)
Fig. 3
Fig. 3
Deletion of RG4 structures in 5ʹtiRNAAla through ionic equilibration abolishes activity. a Model of tetrameric 5ʹtiRNAAlacoordinated by a central G-quadruplex core containing 5 stacked G-quartets. b The 5ʹTOG motif of 5ʹtiRNAAla was mutated to varying extents. Indicated 5ʹtiRNAAla mutants were run on a 15% denaturing acrylamide gel and post-stained with SYBR gold or NMM to detect total RNA or RG4, respectively. The propensity to form RG4 structures correlated with the bioactivity of each tiRNA (Supplementary Table 2). c 5ʹtiRNAAla was equilibrated overnight in indicated salt or left untreated (NT). In vitro translation assays were carried out in rabbit reticulocyte lysate using polyadenylated NanoLuc RNA as a reporter. To each, 1 μl of 100 pmol salt equilibrated 5ʹtiRNAAla or 1 μl of indicated salt was added prior to initiating translation reaction. Depletion of RG4 structure of 5ʹtiRNAAla through equilibration in Li+ ions abolished the ability of 5ʹtiRNAAla to repress translation. d eIF4F was assembled onto m7GTP agarose and challenged with indicated RNAs equilibrated in NaCl or LiCl. Depletion of RG4 structures through equilibration in Li+ ions abrogated the ability of the tiRNA to displace eIF4G or eIF4E from m7GTP cap. e RNPs were purified from U2OS lysates using biotinylated RNAs equilibrated in indicated salt solutions. Depletion of RG4 structures reduced 5ʹtiRNAAla’s association with YB-1, a key regulator of 5ʹtiRNAAla induced stress granules, to background levels. The interaction of 5ʹtiRNAAla with FUS/TLS or TIAR was unaltered through ionic equilibration
Fig. 4
Fig. 4
Substitution of guanines in 5ʹTOG motif with 7-deazaguanine prevents RG4 formation. a 7-deazaguanine derivative replaces the azide group at position 7 with a methine group (highlighted). Interaction with the hydrogen of the methine group is essential for Hoogsteen basepairing in a G-quartet. b Guanines at position 2 and 4 in the 5ʹTOG motif of 5ʹtiRNAAla were substituted with 7-deazaguanine (c, d) 5ʹtiRNAAla(WT) or 5ʹtiRNAAla(7daG) were equilibrated overnight in indicated salt solution and analyzed on denaturing c or native d gel and post-stained with SYBR gold (left panels) to detect total RNA or NMM (right panels) to detect RG4 structures. Regardless of ionic condition, 5ʹtiRNAAla(7daG) failed to form RG4 structure highlighting the importance of Hoogsteen base pairing for RG4 formation
Fig. 5
Fig. 5
7-deazaguanine substitutions in 5ʹtiRNAAla prevent bioactivity. a In vitro translation assay in rabbit reticulocyte lysate using NanoLuc reporter to monitor translation efficiency. Substitution of two guanines in 5ʹtiRNAAla reduces its ability to repress translation. b Indicated RNAs were transfected into U2OS cells and stress granule formation was monitored through immunofluorescence. 5ʹtiRNAAla (WT) readily induces the formation of SGs, while 5ʹtiRNAAla(7daG) fails to. c RNA affinity purifications using indicated RNAs from U2OS lysates. Loss of RG4 forming ability in 5ʹtiRNAAla (7daG) correlates with loss of binding to YB-1, a protein required for tiRNA mediated SG formation. 5ʹtiRNAAla(7daG) also loses the ability to bind to Fxr1 and DHX36, previously identified RG4 binding proteins. In contrast, Vigilin, GRSF1, and TDP43 bind 5ʹtiRNAAla regardless of its ability to form RG4. d Inhibition of RG4 formation in 5ʹtiRNAAla through incorporation of 7-deazaguanine abrogates its ability to displace eIF4F from m7GTP agarose
Fig. 6
Fig. 6
tRNA fragments from Arabadopsis thaliana form RG4 which are required for their activity. a Secondary structures of three 5ʹtiRNA species previously identified in salt stressed A. thaliana harboring putative 5ʹTOG motifs. b Analysis of A. thaliana tiRNAs on 15% denaturing acrylamide gel post-stained with either SYBR Gold or NMM to detect total RNA or RG4 structures, respectively. AtRNA90 forms a strong RG4 positive band which runs in a similar manner to 5ʹtiRNAAla. Upon equilibration in LiCl, the more slowly migrating RG4 positive band disappears. c In vitro translation in rabbit reticulocyte lysates using salt equilibrated or untreated (NT) A. thaliana RNAs. Depletion of RG4 structures in AtRNA90 inhibits its ability to repress translation. However, other active A. thaliana tiRNAs repress translations despite ionic equilibration suggesting they inhibit translation via an alternative mechanism

Similar articles

See all similar articles

Cited by 21 PubMed Central articles

See all "Cited by" articles

References

    1. Parkinson GN, Lee MP, Neidle S. Crystal structure of parallel quadruplexes from human telomeric DNA. Nature. 2002;417:876–880. doi: 10.1038/nature755. - DOI - PubMed
    1. Pinnavaia TJ, et al. Alkali metal ion specificity in the solution ordering of a nucleotide, 5’-guanosine monophosphate. J. Am. Chem. Soc. 1978;100:3625–3627. doi: 10.1021/ja00479a070. - DOI
    1. Pinnavaia TJ, Miles HT, Becker ED. Self-assembled 5’-guanosine monophosphate, nuclear magnetic resonance evidence for a regular, ordered structure and slow chemical exchange. J. Am. Chem. Soc. 1975;97:7198–7200. doi: 10.1021/ja00857a059. - DOI - PubMed
    1. Bugaut A, Balasubramanian S. 5’-UTR RNA G-quadruplexes: translation regulation and targeting. Nucleic Acids Res. 2012;40:4727–4741. doi: 10.1093/nar/gks068. - DOI - PMC - PubMed
    1. Balasubramanian S, Hurley LH, Neidle S. Targeting G-quadruplexes in gene promoters: a novel anticancer strategy? Nat. Rev. Drug Discov. 2011;10:261–275. doi: 10.1038/nrd3428. - DOI - PMC - PubMed

Publication types

Feedback