Analysis of CDS-located miRNA target sites suggests that they can effectively inhibit translation

Abstract

Most of what is presently known about how miRNAs regulate gene expression comes from studies that characterized the regulatory effect of miRNA binding sites located in the 3' untranslated regions (UTR) of mRNAs. In recent years, there has been increasing evidence that miRNAs also bind in the coding region (CDS), but the implication of these interactions remains obscure because they have a smaller impact on mRNA stability compared with miRNA-target interactions that involve 3' UTRs. Here we show that miRNA-complementary sites that are located in both CDS and 3'-UTRs are under selection pressure and share the same sequence and structure properties. Analyzing recently published data of ribosome-protected fragment profiles upon miRNA transfection from the perspective of the location of miRNA-complementary sites, we find that sites located in the CDS are most potent in inhibiting translation, while sites located in the 3' UTR are more efficient at triggering mRNA degradation. Our study suggests that miRNAs may combine targeting of CDS and 3' UTR to flexibly tune the time scale and magnitude of their post-transcriptional regulatory effects.

Figure 1.

(A) CDS- and 3′ UTR-located miRNA binding sites that induce mRNA degradation or are isolated in EIF2C2 CLIP experiments are under stronger selection pressure compared with miRNA-complementary sites that were not functional in these experiments; 2274 CDS- and 3956 3′ UTR-located sites that induce mRNA degradation are compared with 3513 CDS- and 3268 3′ UTR-located sites that do not induce mRNA degradation; 751 CDS- and 786 3′ UTR-located sites isolated in EIF2C2 CLIP experiments are compared with 1059 CDS- and 956 3′ UTR-located sites that do not induce EIF2C2 binding. (B) Scatter plot of the inferred probabilities that CDS- and 3′ UTR-located sites complementary to individual miRNAs are under selection. (Dashed red line) First principal component of the scatter.

Figure 2.

Individual miRNAs differ in their preference for targeting the CDS and the 3′ UTR as well as in their tendency to simultaneously target the CDS and the 3′ UTR of individual genes. (A) Number of complementary motifs, weighted by their respective ElMMo posteriors, of individual miRNAs in CDS vs. 3′ UTRs. (Red dashed line) Scaling between the number of CDS and 3′ UTR sites, defined as the line that goes through the origin and maximizes the projected variance. (B) Scatter plot of the CDS vs. 3′ UTR targeting preference of individual miRNAs against the fold enrichment in transcripts that are targeted in both the CDS and 3′ UTR relative to what would be expected if the sites were independently distributed. (Red dots) miRNAs with statistically significant co-targeting enrichment (P < 0.05 in Fisher's test after Bonferroni correction).

Figure 3.

CDS and 3′ UTR sites share common sequence and structure properties. Sets of functional and nonfunctional binding sites were defined according to four different criteria—selective pressure, efficacy in mRNA degradation, efficacy in reducing protein levels, and binding to EIF2C2—each corresponding to a different panel. We then compared t-values obtained in comparing functional and nonfunctional sites from CDS (x-axis) and 3′ UTR (y-axis) regions. Each property is represented in each plot as a dot. Positive and negative values denote positive and negative predictors of functional miRNA binding sites, respectively.

Figure 4.

mRNA destabilization occurs mainly through sites located in the 3′ UTR. Shown are log₂ fold changes in mRNA levels upon miRNA transfection in the experiments of (A) Linsley et al. (2007) and (B) Grimson et al. (2007). mRNAs with binding sites located in the CDS only, in the 3′ UTR only, and in both CDS and in 3′ UTR were analyzed separately. Fold changes were normalized to the average fold change of mRNAs that did not contain canonical binding sites to the transfected miRNA.

Figure 5.

CDS-located binding sites transiently inhibit translation in miRNA transfection experiments. The figure shows log₂ fold changes in mRNA levels (mRNA-seq), and ribosome-protected fragments (rpf) 12 and 32 h after hsa-miR-155-5p and hsa-miR-1 transfection. Changes in translation were estimated from the difference between changes in rpf and changes in mRNA levels. mRNAs with precisely one seed match to the transfected miRNA in the CDS and no seed match in the 3′ UTR were analyzed separately from mRNAs with precisely one seed match in the 3′ UTR and no seed match in the CDS. Fold changes were determined relative to the average fold change of mRNAs with no seed matches.