Individual microRNAs (miRNAs) display distinct mRNA targeting "rules"

Abstract

MicroRNAs (miRNAs) guide Argonaute (AGO)-containing microribonucleoprotein (miRNP) complexes to target mRNAs.It has been assumed that miRNAs behave similarly to each other with regard to mRNA target recognition. The usual assumptions, which are based on prior studies, are that miRNAs target preferentially sequences in the 3'UTR of mRNAs,guided by the 5' "seed" portion of the miRNAs. Here we isolated AGO- and miRNA-containing miRNPs from human H4 tumor cells by co-immunoprecipitation (co-IP) with anti-AGO antibody. Cells were transfected with miR-107, miR-124,miR-128, miR-320, or a negative control miRNA. Co-IPed RNAs were subjected to downstream high-density Affymetrix Human Gene 1.0 ST microarray analyses using an assay we validated previously-a "RIP-Chip" experimental design. RIP-Chip data provided a list of mRNAs recruited into the AGO-miRNP in correlation to each miRNA. These experimentally identified miRNA targets were analyzed for complementary six nucleotide "seed" sequences within the transfected miRNAs. We found that miR-124 targets tended to have sequences in the 3'UTR that would be recognized by the 5' seed of miR-124, as described in previous studies. By contrast, miR-107 targets tended to have 'seed' sequences in the mRNA open reading frame, but not the 3' UTR. Further, mRNA targets of miR-128 and miR-320 are less enriched for 6-mer seed sequences in comparison to miR-107 and miR-124. In sum, our data support the importance of the 5' seed in determining binding characteristics for some miRNAs; however, the "binding rules" are complex, and individual miRNAs can have distinct sequence determinants that lead to mRNA targeting.

Figure 1

(A) Experimental design. Following miRNA transfections (performed with three biological replicates), Anti-AGO RIP-Chip was performed. Analyses of Affymetrix Human Gene 1.0 ST microarray data were used to identify mRNA targets enriched following miRNA transfections. (B) Further analyses were performed on putative miRNA targets (PmiTs). The 5′ UTR, open reading frames, and 3′ UTR of PmiTs were evaluated for ‘seed’ sequences corresponding to the complementary 6mers from different portions of transfected miRNAs. For example, “seed 1” for miR-107 corresponds to the 6 nucleotides at the 5′ end of miR-107, whereas “seed 18” corresponds to the 6 nucleotides at the 3′ end of miR-107. Most experiments were performed using the “anti-sense” seeds as shown, but for control experiments, “sense” sequences were evaluated.

Figure 2

As has been shown previously, miR-124 targets tend to have “seed” sequences corresponding to nucleotides #2–7 in miR-124, specifically in the 3′ UTR. This chart shows the −log(p value) of Fisher’s exact test analyses to determine the likelihood that the number of sequences found in the putative miRNA targets would have occurred by chance. Separate analyses were performed for the 5′ UTR, open reading frame (“Coding”), and 3′ UTR. Note that the 5′ seeds (6mer sequences corresponding to #1–6, #2–7 and #3–8) of miR-124 are enriched in miR-124 putative miRNA targets, but only in the 3′ UTR.

Figure 3

As was the case for miR-124, putative miR-128 and miR-320 putative targets tend to have sequences corresponding to nucleotides #2–7 in the 3′UTR. However, the tendency for these two miRNAs is much less strong (p < 10⁻⁵ versus p < 10⁻¹¹). Moreover, unlike miR-124, both miR-128 and miR-320 have a tendency—albeit far weaker than the 5′ seed—for the 3′ sequences of the miRNAs to have possible effects (enrichment of complementary 6mers in putative targets). Since the current study was limited to studying 6mer seeds, these data do not test whether sequence determinants with fewer hybridizing nucleotides (5mers and smaller) may be more important in the case of these other miRNAs.

Figure 4

(A) Unlike for miR-124 and the other miRNAs tested in the current study, putative miR-107 targets have a marked tendency to have “seed” sequences corresponding to miR-107 nucleotides #2–7 in the open reading frame (“Coding” sequence) as opposed to the 3′ UTR. Equally remarkable is the apparent lack of a tendency for miR-107 seed sequences to be enriched in the 3′ UTR of miR-107 targets identified experimentally using anti-AGO RIP-Chip. (B) To evaluate whether miR-107 seed sequences are present in other mRNAs (i.e., nonspecifically), we tested other groups of mRNAs and such was not the case. For example, we sampled those mRNAs that were recruited into the AGO-miRNP according to anti-AGO RIP-Chip after miR-124 transfection. There is no “nonspecific” tendency for miR-107 nucleotides #2–7 (in antisense) to be present in these or other mRNAs that were analyzed.

Figure 5

A cartoon to demonstrate that mRNAs strongly recruited into the AGO-microribonucleoparticle following transfection with miR-107 tended to have “seed” sequences corresponding to miR-107 nucleotides #2–7 in the open reading frame. miR-124 ‘seed’ sequences on the PTBP1 are shown at top because this gene has been shown previously to be a miR-124 target and it was a strong miR-124 target on the RIP-Chip assay. Note that the seed sequences corresponding to miR-124 (in antisense) are present mostly within the 3′ UTR of miR-124 targets. Even those miR-124 seed sequences that would rely on G-U binding are mostly in the 3′UTR. By contrast, the miR-107 putative targets tend to contain 5′ seed sequences of miR-107 in the open reading frame. Again, this tendency is also seen for 6mers that would theoretically cause binding with the 5′ seed of miR-107 through G-U binding.

Figure 6

The putative miR-107 targets identified experimentally by RIP-Chip were subjected to biochemical validation. These proteins are GRN, INSIG1 and PPIB, which are targeted by miR-107 in the open reading frame as shown in Figure 5. Immunoblot analyses of these proteins’ expression following transfections of miR-107 (versus control miRNA) show that the expression of these genes are effectively suppressed by miR-107. Beta-Actin immunoblots following identical transfections provides a negative control.

Figure 7

Control experiments show that there is relatively low background “noise” in the anti-AGO RIP-Chip results. These analyses were performed testing for the presence of 6mer sequences corresponding to portions of miRNAs in the “sense” orientations that are predicted NOT to bind to the mRNA targets. Shown here are the results for miR-320, which are representative. As in Figures 2–4 and 7, the Y axis shows the results of −log(p value) for Fisher’s exact test to assess the likelihood that the sequences occurred by chance.

Figure 8

Dual luciferase reporter assays were used to validate miR-107 target sites on INSIG1 mRNA. A sequence in the INSIG1 mRNA open reading frame (nts 370–518) that contains two TGCTGC iterations (top) was subcloned into pRL-TK reporter plasmid (INSIGMRE). The pRL-TK plasmid containing the mutated sequence of INSIGMRE (INSIGMREmut, which is only subtly different as shown) was constructed in parallel. Plasmid transfection, miRNA transfection, and dual luciferase activity assay were followed our published protocols (See Methods section). The figure showed that INSIGMRE specifically responds to miR-107 transfection but not to miR-320, or to a negative control miRNA. “*”-Student’s one-tailed t-test p < 0.001.