miRNAs regulate the posttranscriptional fates of most mammalian mRNAs, yet for endogenous mRNAs, the effects of miRNAs on translation, the steps in translation that are regulated by miRNAs, and the relationship between regulation of translation and mRNA decay by miRNAs have not been systematically explored. To address these effects and relationships, we determined the effect of a human miRNA, miR-124, on translation and abundance of hundreds of endogenous mRNAs that were recruited to Argonaute proteins in response to ectopic expression of miR-124 in HEK293T cells.
We developed a simple and economical method to quantitatively measure two key parameters of translation, ribosome occupancy and average ribosome density, on a genome-wide scale with single DNA microarray hybridizations for each (). This method allowed us to address the effects of miR-124 on translation of endogenous mRNAs; it is also more broadly applicable to the study of translational regulation. In this initial application, we found many parallels between the translation programs in proliferating human embryonic kidney cells and S. cerevisiae
(), suggesting common features of translational programs in eukaryotes 
Direct identification of the mRNAs specifically recruited by miR-124 to Ago proteins, core components of miRNA-effector complexes, defined functional targets of this miRNA in this model system, providing a starting point for dissecting miRNA regulation 
. mRNA expression profiling then allowed us to recognize the specific effects of miR-124 on the abundance of these targets.
Three major conclusions emerged from our studies: (i) miR-124 reduces translation and abundance of its mRNA targets over a broad range; changes in mRNA abundance accounted for ~75% of the estimated effect on protein production; (ii) miR-124 predominantly targets translation at the initiation stage or stimulates ribosome drop-off preferentially near the translation start site; and (iii) miR-124–mediated regulation of translation and mRNA decay are correlated, indicating that most mRNAs are not differentially targeted for translational repression versus mRNA decay.
Transfection of miR-124 consistently reduced the translation and abundance of most of its several hundred high-confidence targets; the resulting decrease in translation averaged 12% and the decrease in target mRNA abundance averaged 35% (). The observation that there were several mRNAs (CD164, VAMP3, and DNAJC1) that had about 10-fold reductions in mRNA levels (Figure S7
), and the fact that 90% of control-transfected cells expressed the transfected GFP marker, suggests that more than 90% cells were transfected with functionally significant quantities of miR-124; thus the small magnitude of the effects on translation and abundance of most of the mRNA targets of miR-124 identified by Ago IP was not likely a result of poor transfection efficiency. The correlation between predicted changes in protein synthesis and observed changes in protein levels for 11 of 12 proteins following miR-124 transfection (), suggests that our assays capture most (or all) of the effects of miR-124 on protein synthesis.
Although we need to be cautious in generalizing from these model systems, in these cells under the condition examined, miRNAs appears to modulate production for hundreds of proteins through joint regulation of target mRNA translation and stability over a strikingly large dynamic range. While the repressive effects on most targets were modest (1–3-fold), there were eight targets (DNAJC1, VAMP3, CD164, SYPL1, MAGT1, HADHB, ATP6V0E1, and SGMS2) that were substantially down-regulated with decreases in protein synthesis of 10-fold or greater. In addition, 45 targets were estimated to have greater than 4-fold changes in protein synthesis. Regardless of the magnitude of regulation, mRNA destabilization accounted for ~75% of the change in estimated protein synthesis. This range of regulation is in good accord with previous studies with genetically characterized endogenous miRNAs as well as with studies introducing exogenous miRNAs introduced into human tissue culture 
. However, our observation that miR-124 had only modest effects on the translation of hundreds of targets contrasts dramatically with several previous studies in which miRNAs reduced protein expression by 5–25-fold while only modestly decreasing mRNA levels (1.1–2-fold), suggesting substantial inhibitory effects on translation 
. The previous studies, however, measured the effect of a specific miRNA on reporter constructs in which the 3′-UTRs of the encoded mRNAs were not derived from mammalian mRNAs, but were either short (~250 nts) modified viral sequences or artificial. In contrast, mammalian mRNA 3′-UTRs tend to be much longer (on average ~1,000 nts) and include regulatory sites for RNA-binding proteins and regulatory RNAs that influence mRNA localization, translation, and decay. The basis for the discrepancy in the results from these two experimental designs remains to be determined, and the answer is likely to provide useful mechanistic insights. One possibility is that mRNAs containing exogenous 3′-UTRs might have anomalously long mRNA half-lives that obscure the normal contribution of mRNA degradation to the miRNA-directed inhibition of protein expression. The large magnitude of effects observed in reporter-based assays, compared to what we and others have observed with endogenous mRNAs, is likely to be partially due to the multiple (four to eight) engineered miRNA binding sites in the reporter constructs used in those studies 
. Further, these sites were in close proximity, and adjacent miRNA binding sites have been reported to act cooperatively 
. Indeed, two studies that measured the effects of specific miRNAs on protein and mRNA levels of reporters with endogenous mammalian 3′-UTRs found more modest effects on translation, less than 2-fold on average 
. Moreover, the magnitude of the effects we observed on translation of the mRNAs targeted by miR-124 were in agreement with two recent studies that inferred the repressive effect of miRNAs on translation by measuring miRNA-mediated effects on mRNA and protein abundance 
. Those reports, based on directly measured changes in protein levels by quantitative mass spectrometry, concluded that the effects of miRNAs on translation were small (less than 2-fold for hundreds of target mRNAs).
Although we believe that our experimental design provided a good model of miRNA regulation as it normally operates in vivo, our results do represent the full range of possible regulatory consequences of miRNA–mRNA interactions. Our results suggest that miRNAs have a large dynamic range of effects on endogenous protein expression, achieved via regulation of both translation and mRNA abundance; this pattern is generally quite consistent with previous results from cells grown in culture and limited in vivo observations. However, in specific developmental or physiological programs, or for specific mRNAs, the effects on abundance and translation, as well as the apparent mode of translation regulation may differ from what we observed in this study 
. Thus, the effects we observed for miR-124 targets after ectopically expressing the microRNA in Hek293T cells may not capture the full scope of regulation by miRNAs in their endogenous context; miR-124 is endogenously expressed in neuronal cells, and the regulatory effects of miR-124 interactions may be modulated by the physiological demands of the cell and the specific suite of specific RNA-binding proteins and regulatory RNAs that also associate with miR-124 target mRNAs.
miR-124 negatively affected both the ribosome occupancy and ribosome density of hundreds of its targets (). These parallel effects, combined with the close match between changes in protein synthesis predicted from miRNA-induced effects on mRNA abundance and translation and changes in protein levels for 11 of 12 proteins, suggest that the step in translation principally targeted by miR-124 and presumably other miRNAs is initiation or elongation processivity near the translation start site. We favor the initiation model because it is in accord with several previous studies that focused on one or a few mRNAs 
, and there is a paucity of empirical evidence supporting ribosome drop-off, which predicts that ribosome density of miRNA-regulated mRNAs declines between the 5′- and 3′-ends of the coding sequence and that there should be an overrepresentation of incompletely synthesized N-terminal nascent polypeptides 
The small apparent magnitude of the effects on translation initiation, combined with the strong correlation between changes in translation and mRNA abundance, can be explained by a model in which repression of translational by miR-124 rapidly leads to mRNA decay. Such a model would explain why observable effects on translation appear to be smaller than the changes in mRNA abundance: if mRNAs whose translation is inhibited are quickly destroyed, their diminished translation would not be detected in our translation assay. There is already compelling evidence that translational repression and mRNA decay are linked 
. Our observation that an overwhelming majority of polyadenylated mRNAs are associated with ribosomes in HEK293T cells may be a manifestation of this relationship (). Thus, miRNA-mediated inhibition of translation may be linked to a general system for removal of the mRNA from the translational pool, involving recruitment to P-bodies and subsequent destruction 
. Regulated decoupling of miRNA-mediated translation repression and mRNA decay would then allow organisms to tilt the balance of effects in favor of translational repression during physiological and developmental conditions where mRNA destruction is a disadvantage 
. Our results are also consistent with a model in which miRNA-mediated regulation of translation and mRNA decay are functionally independent, but are similarly controlled by the same cis
-elements. Determining whether the concordant regulation of translation and mRNA abundance represents a mechanistic coupling of miRNA-mediated regulation of translation and mRNA decay, and understanding the molecular links between these two regulatory consequences of miRNA–mRNA interactions are important goals for future investigation.