Limits...
A-to-I editing in the miRNA seed region regulates target mRNA selection and silencing efficiency.

Kume H, Hino K, Galipon J, Ui-Tei K - Nucleic Acids Res. (2014)

Bottom Line: Hydrolytic deamination of adenosine to inosine (A-to-I) by adenosine deaminases acting on RNA (ADARs) is a post-transcriptional modification which results in a discrepancy between genomic DNA and the transcribed RNA sequence, thus contributing to the diversity of the transcriptome.The difference in base-pairing stability, deduced by melting temperature measurements, between seed-target duplexes containing either C:G or I:C pairs may account for the observed silencing efficiency.These findings unequivocally show that C:G and I:C pairs are biologically different in terms of gene expression regulation by miRNAs.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Faculty of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.

Show MeSH
Possible thermodynamic control of miRNA-mediated gene-silencing activity and correlations between Tm values of 7-bp seed-target duplexes and differential fold-changes of expression levels of target transcripts determined by microarray experiments or relative luciferase activities determined by reporter assays. (A) The miRNA with low miTm1–5 value promotes miRNA unwinding into a single-stranded RNA in the RISC, and that with high Tm2–8 value promotes stable base-pairing between miRNA seed region and target mRNA. Thus, the efficacies of miRNA-mediated silencing are determined by the combined thermodynamic parameters that might reflect their unwinding properties (miTm1–5) in addition to their base-pairing stabilities in the seed-target duplex (Tm2–8) shown as a formula, miScore = Tm2–8 − 0.5 x miTm1–5. (B) The duplex structures formed between 7-mer seed sequence of miR-376a-3p containing adenosine, inosine, or guanosine in the possible editing site and target mRNA sequence with uridine or cytidine at the opposite site of editing position, and the measured Tm values of these 7-bp duplexes. The Tm value of the duplex formed between AGAUACU and UCCAUGA could not be measured, probably due to fairly unstable base-pairing and was shown as not determined (N.D). (C) The correlations between the 7-bp Tm values and fold-changes in the expression levels of target mRNAs containing seed complementary sequences in their 3′-UTRs (red), or relative luciferase activities at 50 nM of miRNA duplex (blue). The correlation coefficient (R) between Tm values and differential fold-changes was 0.82, and R between Tm values and relative luciferase activities was -0.91.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4150774&req=5

Figure 7: Possible thermodynamic control of miRNA-mediated gene-silencing activity and correlations between Tm values of 7-bp seed-target duplexes and differential fold-changes of expression levels of target transcripts determined by microarray experiments or relative luciferase activities determined by reporter assays. (A) The miRNA with low miTm1–5 value promotes miRNA unwinding into a single-stranded RNA in the RISC, and that with high Tm2–8 value promotes stable base-pairing between miRNA seed region and target mRNA. Thus, the efficacies of miRNA-mediated silencing are determined by the combined thermodynamic parameters that might reflect their unwinding properties (miTm1–5) in addition to their base-pairing stabilities in the seed-target duplex (Tm2–8) shown as a formula, miScore = Tm2–8 − 0.5 x miTm1–5. (B) The duplex structures formed between 7-mer seed sequence of miR-376a-3p containing adenosine, inosine, or guanosine in the possible editing site and target mRNA sequence with uridine or cytidine at the opposite site of editing position, and the measured Tm values of these 7-bp duplexes. The Tm value of the duplex formed between AGAUACU and UCCAUGA could not be measured, probably due to fairly unstable base-pairing and was shown as not determined (N.D). (C) The correlations between the 7-bp Tm values and fold-changes in the expression levels of target mRNAs containing seed complementary sequences in their 3′-UTRs (red), or relative luciferase activities at 50 nM of miRNA duplex (blue). The correlation coefficient (R) between Tm values and differential fold-changes was 0.82, and R between Tm values and relative luciferase activities was -0.91.

Mentions: The strand with relatively lower internal stability at the 5′-terminus of the miRNA duplex is preferentially loaded onto the RISC (Figure 7A, 16–18). Furthermore, miRNA canonically recognizes target transcripts using seed-complementary sequences to direct post-transcriptional repression (Figure 7A, 13,14). Consistent with this knowledge, we have published a mathematical model of miRNA base-pairing stability using known thermodynamic parameters of Watson–Crick base-pairing (35,36), and have demonstrated that the silencing efficiency of miRNA is strongly and positively correlated, with the correlation score ‘miScore’ calculated as follows:\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy}\usepackage{upgreek}\usepackage{mathrsfs}\setlength{\oddsidemargin}{-69pt}\begin{document}}{}\begin{equation*} {\rm miScore} = {\rm Tm}_{2 - 8} - 0.5 \times {\rm miTm}_{1 - 5} \end{equation*}\end{document}where Tm2–8 means the melting temperature (Tm) value of the seed-target duplex (positions 2–8) and miTm1–5 means Tm value of 5′-terminal miRNA duplex (positions 1–5) (Figure 7A). Thus, the seed-target (Tm2–8) and 5′-terminal 5-bp duplex (miTm1–5) stabilities have opposite effects on gene silencing: miRNAs comprising of an unstable 5′-terminal 5-bp duplex and a stable 7-bp seed-target duplex exhibit strong silencing activity, although the contribution of miTm1–5 might be only about half that of Tm2–8 because the multiplier coefficient was about 0.5 (12). Therefore, both the difference in thermodynamic stability between the 5′-terminal regions of G-type and I-type miRNA duplexes, and the difference in seed-target binding strength, are likely to contribute to the observed difference in silencing efficiency. Thermodynamic stability in the RNA duplex can be calculated using known parameters (35–37). However, since the thermodynamic parameters of base-pairing between inosine and cytidine have not yet been determined, we annealed both strands of chemically synthesized 5-mer and 7-mer RNA oligonucleotides to form the equivalent of either the 5′-terminal 5-bp duplex and the seed-duplex, respectively, and measured their Tm values using a UV spectrophotometer. In our sequence contexts, sufficient melting curves were obtained only with 7-bp RNA duplexes but not with 5-bp duplexes in 1M NaCl. We selected miR-376a-2 to examine the relationship between silencing efficiency and miScore. Since the editing site is positioned at nucleotide +6 counting from the 5′ end in miR-376a-2, the miTm1–5 value of miR-376a-2 is the same for A-type, I-type and G-type miR-376a-2 duplexes, meaning that we can use the Tm2–8 value alone instead of miScore. As shown in Figure 7, the measured Tm2–8 value for the A-type seed-duplex (formed by the wild-type 7-mer RNA oligonucleotide seed sequence of miR-376a-3p-A and the wild type U-target 7-mer RNA) was 26.6°C, whereas that of the G-type seed-duplex (formed by the C-target 7-mer and the miR-376a-3p-G seed sequence) was the highest at 39.0°C. This is conceivable, as G:C pairs are widely known to be more stable than A:U pairs. Strikingly, the Tm2–8 value of the I-type seed-duplex (formed by the C-target 7-mer and the miR-376a-3p-I seed sequence) was 23.3°C, which is low compared to that of the G-type duplex and similar to A:U base-pairing. Furthermore, G:U wobble base-pairing showed even weaker stability at 19.3°C, and I:U pairing showed the lowest Tm2–8 value at 9.3°C. The Tm2–8 value of the duplex formed between AGAUACU and UCCAUGA (A:C pair) could not be measured, probably due to fairly unstable base-pairing. The Tm2–8 values and their silencing activities determined by microarray experiments showed a strong positive correlation (R = 0.82). Consistently, the Tm2–8 value also showed a strong negative correlation with relative luciferase activity measured at 50 nM of miRNA duplex by reporter assays (R = −0.91), suggesting that at least the Tm2–8 value behaves like a major determinant for miRNA-mediated gene silencing efficiency for a fixed miTm1–5 value. As for miR-22 and miR-191, the editing sites are positioned at +2 and +3 from the 5′, respectively, indicating that the accurate miScores could not be determined, since the miTm1–5 values were not successfully measured. In fact, in the case of the miR-22 duplex, the Tm2–8 values alone were correlated neither with their silencing activities determined by microarray experiments (R = 0.36), nor with luciferase activity (R = −0.49) (Supplemental Figure S4). This result suggests that the 5′-terminal 5-bp duplex might also be responsible for miRNA silencing efficacy in the case of inosine-mediated base-pairing. Thus, we propose that inosine in the miRNA 5′-terminal and seed region has the potential to regulate miRNA silencing efficiency by regulating base-pairing stability within RNA duplexes.


A-to-I editing in the miRNA seed region regulates target mRNA selection and silencing efficiency.

Kume H, Hino K, Galipon J, Ui-Tei K - Nucleic Acids Res. (2014)

Possible thermodynamic control of miRNA-mediated gene-silencing activity and correlations between Tm values of 7-bp seed-target duplexes and differential fold-changes of expression levels of target transcripts determined by microarray experiments or relative luciferase activities determined by reporter assays. (A) The miRNA with low miTm1–5 value promotes miRNA unwinding into a single-stranded RNA in the RISC, and that with high Tm2–8 value promotes stable base-pairing between miRNA seed region and target mRNA. Thus, the efficacies of miRNA-mediated silencing are determined by the combined thermodynamic parameters that might reflect their unwinding properties (miTm1–5) in addition to their base-pairing stabilities in the seed-target duplex (Tm2–8) shown as a formula, miScore = Tm2–8 − 0.5 x miTm1–5. (B) The duplex structures formed between 7-mer seed sequence of miR-376a-3p containing adenosine, inosine, or guanosine in the possible editing site and target mRNA sequence with uridine or cytidine at the opposite site of editing position, and the measured Tm values of these 7-bp duplexes. The Tm value of the duplex formed between AGAUACU and UCCAUGA could not be measured, probably due to fairly unstable base-pairing and was shown as not determined (N.D). (C) The correlations between the 7-bp Tm values and fold-changes in the expression levels of target mRNAs containing seed complementary sequences in their 3′-UTRs (red), or relative luciferase activities at 50 nM of miRNA duplex (blue). The correlation coefficient (R) between Tm values and differential fold-changes was 0.82, and R between Tm values and relative luciferase activities was -0.91.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4150774&req=5

Figure 7: Possible thermodynamic control of miRNA-mediated gene-silencing activity and correlations between Tm values of 7-bp seed-target duplexes and differential fold-changes of expression levels of target transcripts determined by microarray experiments or relative luciferase activities determined by reporter assays. (A) The miRNA with low miTm1–5 value promotes miRNA unwinding into a single-stranded RNA in the RISC, and that with high Tm2–8 value promotes stable base-pairing between miRNA seed region and target mRNA. Thus, the efficacies of miRNA-mediated silencing are determined by the combined thermodynamic parameters that might reflect their unwinding properties (miTm1–5) in addition to their base-pairing stabilities in the seed-target duplex (Tm2–8) shown as a formula, miScore = Tm2–8 − 0.5 x miTm1–5. (B) The duplex structures formed between 7-mer seed sequence of miR-376a-3p containing adenosine, inosine, or guanosine in the possible editing site and target mRNA sequence with uridine or cytidine at the opposite site of editing position, and the measured Tm values of these 7-bp duplexes. The Tm value of the duplex formed between AGAUACU and UCCAUGA could not be measured, probably due to fairly unstable base-pairing and was shown as not determined (N.D). (C) The correlations between the 7-bp Tm values and fold-changes in the expression levels of target mRNAs containing seed complementary sequences in their 3′-UTRs (red), or relative luciferase activities at 50 nM of miRNA duplex (blue). The correlation coefficient (R) between Tm values and differential fold-changes was 0.82, and R between Tm values and relative luciferase activities was -0.91.
Mentions: The strand with relatively lower internal stability at the 5′-terminus of the miRNA duplex is preferentially loaded onto the RISC (Figure 7A, 16–18). Furthermore, miRNA canonically recognizes target transcripts using seed-complementary sequences to direct post-transcriptional repression (Figure 7A, 13,14). Consistent with this knowledge, we have published a mathematical model of miRNA base-pairing stability using known thermodynamic parameters of Watson–Crick base-pairing (35,36), and have demonstrated that the silencing efficiency of miRNA is strongly and positively correlated, with the correlation score ‘miScore’ calculated as follows:\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy}\usepackage{upgreek}\usepackage{mathrsfs}\setlength{\oddsidemargin}{-69pt}\begin{document}}{}\begin{equation*} {\rm miScore} = {\rm Tm}_{2 - 8} - 0.5 \times {\rm miTm}_{1 - 5} \end{equation*}\end{document}where Tm2–8 means the melting temperature (Tm) value of the seed-target duplex (positions 2–8) and miTm1–5 means Tm value of 5′-terminal miRNA duplex (positions 1–5) (Figure 7A). Thus, the seed-target (Tm2–8) and 5′-terminal 5-bp duplex (miTm1–5) stabilities have opposite effects on gene silencing: miRNAs comprising of an unstable 5′-terminal 5-bp duplex and a stable 7-bp seed-target duplex exhibit strong silencing activity, although the contribution of miTm1–5 might be only about half that of Tm2–8 because the multiplier coefficient was about 0.5 (12). Therefore, both the difference in thermodynamic stability between the 5′-terminal regions of G-type and I-type miRNA duplexes, and the difference in seed-target binding strength, are likely to contribute to the observed difference in silencing efficiency. Thermodynamic stability in the RNA duplex can be calculated using known parameters (35–37). However, since the thermodynamic parameters of base-pairing between inosine and cytidine have not yet been determined, we annealed both strands of chemically synthesized 5-mer and 7-mer RNA oligonucleotides to form the equivalent of either the 5′-terminal 5-bp duplex and the seed-duplex, respectively, and measured their Tm values using a UV spectrophotometer. In our sequence contexts, sufficient melting curves were obtained only with 7-bp RNA duplexes but not with 5-bp duplexes in 1M NaCl. We selected miR-376a-2 to examine the relationship between silencing efficiency and miScore. Since the editing site is positioned at nucleotide +6 counting from the 5′ end in miR-376a-2, the miTm1–5 value of miR-376a-2 is the same for A-type, I-type and G-type miR-376a-2 duplexes, meaning that we can use the Tm2–8 value alone instead of miScore. As shown in Figure 7, the measured Tm2–8 value for the A-type seed-duplex (formed by the wild-type 7-mer RNA oligonucleotide seed sequence of miR-376a-3p-A and the wild type U-target 7-mer RNA) was 26.6°C, whereas that of the G-type seed-duplex (formed by the C-target 7-mer and the miR-376a-3p-G seed sequence) was the highest at 39.0°C. This is conceivable, as G:C pairs are widely known to be more stable than A:U pairs. Strikingly, the Tm2–8 value of the I-type seed-duplex (formed by the C-target 7-mer and the miR-376a-3p-I seed sequence) was 23.3°C, which is low compared to that of the G-type duplex and similar to A:U base-pairing. Furthermore, G:U wobble base-pairing showed even weaker stability at 19.3°C, and I:U pairing showed the lowest Tm2–8 value at 9.3°C. The Tm2–8 value of the duplex formed between AGAUACU and UCCAUGA (A:C pair) could not be measured, probably due to fairly unstable base-pairing. The Tm2–8 values and their silencing activities determined by microarray experiments showed a strong positive correlation (R = 0.82). Consistently, the Tm2–8 value also showed a strong negative correlation with relative luciferase activity measured at 50 nM of miRNA duplex by reporter assays (R = −0.91), suggesting that at least the Tm2–8 value behaves like a major determinant for miRNA-mediated gene silencing efficiency for a fixed miTm1–5 value. As for miR-22 and miR-191, the editing sites are positioned at +2 and +3 from the 5′, respectively, indicating that the accurate miScores could not be determined, since the miTm1–5 values were not successfully measured. In fact, in the case of the miR-22 duplex, the Tm2–8 values alone were correlated neither with their silencing activities determined by microarray experiments (R = 0.36), nor with luciferase activity (R = −0.49) (Supplemental Figure S4). This result suggests that the 5′-terminal 5-bp duplex might also be responsible for miRNA silencing efficacy in the case of inosine-mediated base-pairing. Thus, we propose that inosine in the miRNA 5′-terminal and seed region has the potential to regulate miRNA silencing efficiency by regulating base-pairing stability within RNA duplexes.

Bottom Line: Hydrolytic deamination of adenosine to inosine (A-to-I) by adenosine deaminases acting on RNA (ADARs) is a post-transcriptional modification which results in a discrepancy between genomic DNA and the transcribed RNA sequence, thus contributing to the diversity of the transcriptome.The difference in base-pairing stability, deduced by melting temperature measurements, between seed-target duplexes containing either C:G or I:C pairs may account for the observed silencing efficiency.These findings unequivocally show that C:G and I:C pairs are biologically different in terms of gene expression regulation by miRNAs.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Faculty of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.

Show MeSH