REPORTS

17. L. Wu, J. Fan, J. G. Belasco, Proc. Natl. Acad. Sci. U.S.A. 28. J. E. Braun, E. Huntzinger, M. Fauser, E. Izaurralde, data analysis. A.A.B., M.T.L., and A.J.G. interpreted the data
103, 4034 (2006). Mol. Cell 44, 120 (2011). and wrote the manuscript. Sequencing data are deposited
18. E. Huntzinger, E. Izaurralde, Nat. Rev. Genet. 12, 99 29. A. Cooke, A. Prigge, M. Wickens, J. Biol. Chem. 285, in the Gene Expression Omnibus (GEO) database with
(2011). 28506 (2010). accession number GSE34743.
19. T. Fukaya, Y. Tomari, EMBO J. 30, 4998 (2011). 30. E. Wu et al., Mol. Cell 40, 558 (2010).
20. Y. Mishima et al., Proc. Natl. Acad. Sci. U.S.A. 109, Supporting Online Material
1104 (2012). Acknowledgments: We thank S. Wolin and the Wolin www.sciencemag.org/cgi/content/full/science.1215704/DC1
21. M. R. Fabian et al., Mol. Cell 35, 868 (2009). laboratory for vital intellectual and technical support and Materials and Methods
22. M. R. Fabian, Y. V. Svitkin, N. Sonenberg, Methods Mol. access to equipment and reagents; N. Ignolia, J. Weissman, SOM Text
Biol. 725, 207 (2011). D. Schoenberg, D. Patil, A. Staton, C. Takacs, and Y. Mishima Figs. S1 to S11
23. A. Zdanowicz et al., Mol. Cell 35, 881 (2009). for reagents and discussions; and S. Baserga, D. Cazalla, Tables S1 and S2
24. A. Eulalio et al., RNA 15, 21 (2009). D. Cifuentes, and S. Moxon for discussions. Supported by the Reference (31)
25. T. H. Beilharz et al., PLoS ONE 4, e6783 (2009). Pew Fellows Program in Biomedical Sciences (A.A.B.), the Database S1
26. M. Chekulaeva et al., Nat. Struct. Mol. Biol. 18, 1218 Pew Scholars Program in the Biomedical Sciences, the Yale
(2011). Scholar Program, and NIH grants R01GM081602-05 (A.J.G). 24 October 2011; accepted 5 March 2012
27. M. R. Fabian et al., Nat. Struct. Mol. Biol. 18, 1211 Contributions: A.A.B. and A.J.G. designed the project. A.A.B. Published online 15 March 2012;
(2011). performed the experiments. A.A.B. and M.T.L. performed 10.1126/science.1215704

zation and translational repression (11, 12). Be-

miRNA-Mediated Gene Silencing cause only slightly more translational repression

Downloaded from http://science.sciencemag.org/ on July 22, 2016

is observed than mRNA destabilization, it is pos-
by Translational Repression Followed sible that most of the loss in protein synthesis
could directly result from effects on mRNA sta-
by mRNA Deadenylation and Decay bility. Most of these studies have not, however,
evaluated the kinetics of the miRNA-related cel-
lular processes (5, 10, 13, 14). Exceptions include
Sergej Djuranovic, Ali Nahvi, Rachel Green* several analyses of in vitro systems that con-
cluded that the effects of miRNAs on transla-
microRNAs (miRNAs) regulate gene expression through translational repression and/or messenger tional repression precede effects on mRNA target
RNA (mRNA) deadenylation and decay. Because translation, deadenylation, and decay are deadenylation or decay (1517), but concerns
closely linked processes, it is important to establish their ordering and thus to define the molecular remain that the in vitro reactions may not fully
mechanism of silencing. We have investigated the kinetics of these events in miRNA-mediated recapitulate the in vivo situation.
gene silencing by using a Drosophila S2 cell-based controllable expression system and show that We used an in vivo luciferase-based reporter
mRNAs with both natural and engineered 3 untranslated regions with miRNA target sites are system in Drosophila melanogaster S2 cells un-
first subject to translational inhibition, followed by effects on deadenylation and decay. We next der the control of an inducible metallothionein
used a natural translational elongation stall to show that miRNA-mediated silencing inhibits promoter (Mtn) (18). The reporter constructs con-
translation at an early step, potentially translation initiation. sist of one of the luciferase reporter genes [Firefly
(F-Luc) or Renilla (R-Luc)] fused at its 5 end to

m
expression
icroRNAs (miRNAs) are short en-
dogenous RNAs that regulate protein
1 from targeted genes by pairing to
studies showed that miRNAs principally affect
protein expression of miRNA-targeted genes with-
out obvious effects on mRNA abundance (710).
the Mtn promoter and at its 3 end to synthetic or

Howard Hughes Medical Institute (HHMI) and Department of

sites in the 3 untranslated region (3UTR) (1). By simultaneously measuring translational effi- Molecular Biology and Genetics, Johns Hopkins University
Although some studies showed a strong corre- ciencies (thus indirectly levels of protein synthe- School of Medicine, Baltimore, MD 21205, USA.
lation between the diminution of protein and sis) and mRNA abundance, global analyses have *To whom correspondence should be addressed. E-mail:
mRNA levels of miRNA-targeted genes (26), other shown evidence of significant mRNA destabili- [email protected]

Fig. 1. Steady state evaluation of miRNA-mediated gene silencing using a copper-inducible in vivo reporter system. (A)
Measured protein amounts (luminescence) from transfected nontargeted (NT), targeted (T), and control (con) constructs
24 hours after induction. Additional expression of bantam miRNA, not Argonaute 1, results in increased repression for
synthetic bantam targeted constructs (fig. S1). (B to E) Ratios of steady-state protein amounts for synthetic and natural miRNA-targeted constructs 48 hours
after induction. In each case, mean values T SD from three independent triplicate experiments are shown as a normalized ratio of protein amounts (NT/T).

www.sciencemag.org SCIENCE VOL 336 13 APRIL 2012 237

REPORTS
natural 3UTRs that contain miRNA binding sites We next used a set of synthetic and natural antagomirs to the cell cultures induced levels of
responsive to either endogenously expressed 3UTRs with miRNA binding sites responding to derepression for the reporter constructs similar to
(bantam and miR-279) or ectopically introduced either bantam or miR-9b and miR-279 and their those observed when comparing reporter protein
miRNAs (miR-9b) (fig. S1); control constructs corresponding controls (fig. S1 and tables S1 and expression from constructs with intact and altered
not subject to miRNA-mediated gene regulation S2). We chose natural (endogenous) 3UTRs miRNA binding sites (fig. S2).
are detailed in fig. S1. We first asked whether from hid and Vha68-1 and a pair of synthetic For regulated induction, transcriptional shut-
miRNA-mediated responsiveness is limited by en- 3UTR targets that contain six natural tandem off of the Mtn promoter was accomplished by
dogenous components of the microRNA-induced sites for either bantam or miR-9b and miR-279 using a specific copper chelator, bathocuproine
silencing complex (miRISC) (Argonaute protein miRNAs (19, 20). After transfection and consti- disulphonate (BCS). Although BCS chelates any
or miRNAs). By using a previously character- tutive induction, luciferase expression levels were residual copper in the S2 cell medium, it does not
ized bantam-responsive synthetic construct and measured after 48 hours and normalized to assess penetrate the cell membrane and thus does not
ectopically expressed additional miRISC compo- the end-point effects of miRNA-mediated repres- affect normal cellular homeostasis (21). A 90-min
nents (bantam and Ago1), we followed repres- sion (Fig. 1, B to E). The synthetic bantam pulse induction by copper (II) sulfate induced ex-
sion levels of the target mRNAs 24 hours after 3UTR exhibited repression levels up to 80-fold pression of the various reporter constructs to lev-
induction (Fig. 1A) (19). The results indicate that when compared with the control (Fig. 1B); the els that could be monitored over the subsequent
Ago1 is not limiting for repression in Drosophila reporter containing the 3UTR of the hid gene 48 hours. By using this optimized pulse-induction
S2 cells, whereas bantam is limiting because greater with as many as eight miRNA binding sites ex- protocol (fig. S3), we determined how our reporter
repression was observed when it was overex- hibited ~ninefold repression (Fig. 1D) (19). Ad- pairs with synthetic and natural 3UTRs respond to
pressed (Fig. 1A). In the remaining experiments, ditionally, both synthetic and natural (20) 3UTRs miRNAs during an extended period of time. Nor-

Downloaded from http://science.sciencemag.org/ on July 22, 2016

we expressed additional amounts of endogenous- responding to ectopically expressed miR-9b and malized levels of luciferase luminescence for both
ly present miRNAs (bantam and miR-279) from miR-279 exhibited strong repression 48 hours miRNA-targeted and nontargeted constructs were
plasmids under the constitutive actin promoter. after induction (Fig. 1, C and E). The addition of used to assess miRNA-mediated gene silencing at

Fig. 2. Time-resolved progression of miRNA-mediated gene silencing es- reporter genes from oligo(dT)25 resin precipitation or from total RNA
tablishes that repression of protein synthesis precedes mRNA deadenylation presented as ratios of poly(A) and total mRNA, respectively. Each data point
and decay. (A to D) Normalized levels of protein amounts for both miRNA- represents the mean value T SD calculated from three independent
targeted (T) and nontargeted (NT) constructs; normalized mRNA levels for experiments.

238 13 APRIL 2012 VOL 336 SCIENCE www.sciencemag.org

 REPORTS
the protein level. mRNA levels for reporter and related and poorly correlated (5, 1012, 19, 20). enylation and decay were unaffected (fig. S5,
control genes were determined by using quantita- For reporters with natural 3 UTRs, the repression A and B).
tive reverse transcription polymerase chain reaction effects on protein and mRNA levels are corre- We performed a similar set of experiments
(qRT-PCR) from RNA samples isolated with oligo lated at later (but not early) time points (Fig. 2, C with use of actinomycin D to more rapidly shut
(dT)25 resin, as well as from total RNA. Both val- and D), consistent with global studies on miRNA- off transcription (fig. S6, A to D). At a single
ues were normalized with respect to the parallel mediated silencing (11, 12). For reporters with time point after induction, all constructs showed
transfected control genes and were used to calculate synthetic 3UTRs, the repression effects on protein a substantial reduction in protein production
repression ratios between miRNA-targeted and and mRNA levels are uncorrelated at both early but no reduction in mRNA abundance (Fig. 2).
nontargeted constructs (Fig. 2). We emphasize that and late time points (Fig. 2, A and B) (5, 10, 19, 20). Consistent with BCS shut-off data, translation
there is substantial mRNA degradation (for all re- The differences in overall stability of the mRNAs rates for targeted constructs are substantially re-
porters) during the experiment (fig. S4), but here likely reflect the complexity of RNA degrada- duced with respect to nontargeted ones (fig. S6,
we are interested in the relative amount of decay of tion as specified by various 3UTR-located se- A to D).
the targeted and nontargeted mRNAs. The results quence motifs. In all experiments here, repression mRNA polyadenylate [poly(A)] tails are in-
of the pulse-induction experiments were consistent of protein synthesis is consistently seen just 2 volved in translation initiation and thus in de-
and show that the repressive effects of miRNAs on hours after induction, whereas mRNA destabili- termining translational efficiency (22, 23). Potential
synthesis of all four proteins precede any effects on zation comes later. Additionally, in experiments subtle changes in poly(A) tail length not detect-
mRNA deadenylation or decay (Fig. 2). where less bantam miRNA was present (by not able using an oligo(dT)25 isolation procedure were
These four examples include effects on pro- supplying an exogenous source), the observed analyzed for the complete set of RNA samples
tein and mRNA levels that are both well cor- timing of protein repression and mRNA dead- by using a poly(A) tail-length assay that allows

Downloaded from http://science.sciencemag.org/ on July 22, 2016

Fig. 3. mRNA deadenylation is not required for miRNA-mediated trans-
lational repression. (A to D) Length of poly(A) tail was determined by using
G/I tailing PCR-based amplification (materials and methods). Positions
of overamplified products with or without (C) poly(A) tail are indicated.
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA poly(A) tail
length is shown as a control. M lane represents 100base pair markers. (E)
Time-resolved progression of miRNA-mediated gene silencing for histone H3
constructs. Normalized levels of protein and mRNA amounts for miRNA-
targeted (T) and nontargeted (NT) constructs are shown. Each data point
represents an average value T SD from three independent experiments.

www.sciencemag.org SCIENCE VOL 336 13 APRIL 2012 239

REPORTS

Fig. 4. Translational elongation stalling assay in-

dicates that miRNA-mediated gene silencing targets
early steps in translation. (A to E) Time-resolved pro-
gression of miRNA-mediated effects on the stability
of various mRNA constructs containing the lysine-
induced elongation stall. Normalized mRNA amounts
for the miRNA-targeted (T) and nontargeted (NT) con-
structs are shown as the ratio of NT/T; note that

Downloaded from http://science.sciencemag.org/ on July 22, 2016

values are less than one and decreasing. Each data
point represents an average T SD from three inde-
pendent experiments.

for specific amplification of poly(A)-containing er mRNAs over time (Fig. 4, A to D, and fig. 8. R. S. Pillai et al., Science 309, 1573 (2005); 10.1126/
mRNAs. Subtle changes in poly(A) tail length S11, A to C). Ratios of the targeted and nontar- science.1115079.
9. C. P. Petersen, M. E. Bordeleau, J. Pelletier, P. A. Sharp,
could be documented with this assay when cells geted proteins and mRNAs were normalized to Mol. Cell 21, 533 (2006).
were treated with puromycin (fig. S7) (24). How- the amount of a parallel control reporter. In all 10. X. C. Ding, H. Grosshans, EMBO J. 28, 213 (2009).
ever, we did not observe any shortening of poly(A) cases, mRNAs of nontargeted reporters were 11. D. G. Hendrickson et al., PLoS Biol. 7, e1000238
tails in reporter constructs responding to bantam rapidly cleaved and degraded, whereas mRNAs (2009).
12. H. Guo, N. T. Ingolia, J. S. Weissman, D. P. Bartel,
or mir-9b and mir-279 (Fig. 3, A to D). Because of the targeted reporters were relatively stabilized. Nature 466, 835 (2010).
no intermediates are observed, these data suggest Similar results were obtained in a stalling exper- 13. L. Wu, J. Fan, J. G. Belasco, Proc. Natl. Acad. Sci. U.S.A.
that deadenylation in vivo is processive and close- iment with histone H3 constructs, confirming 103, 4034 (2006).
ly coupled to mRNA decay. These data argue that that the poly(A) tail is not essential for miRNA- 14. M. Wakiyama, K. Takimoto, O. Ohara, S. Yokoyama,
Genes Dev. 21, 1857 (2007).
translational inhibition is not triggered by dead- mediated translational repression (Fig. 4E). These 15. G. Mathonnet et al., Science 317, 1764 (2007).
enylation, either partial or complete, in the system data establish that miRNA-mediated translational 16. A. Zdanowicz et al., Mol. Cell 35, 881 (2009).
that we have established [in contrast with earlier silencing happens in the Drosophila system dur- 17. M. R. Fabian et al., Mol. Cell 35, 868 (2009).
studies (6, 12, 25)]. ing the initiation or early elongation phase of pro- 18. H. Johansen et al., Genes Dev. 3, 882 (1989).
19. A. Nahvi, C. J. Shoemaker, R. Green, RNA 15, 814
We also evaluated the timing and extent of tein synthesis.
(2009).
silencing of reporter gene pairs carrying histone We find that miRNA-mediated gene silencing 20. I. Behm-Ansmant et al., Genes Dev. 20, 1885 (2006).
H3 terminal stem loops [and no poly(A) tail] (26) in Drosophila S2 cells is first manifested through 21. A. Selvaraj et al., Genes Dev. 19, 891 (2005).
(fig. S8). This reporter pair is translationally effects on translation, and in particular the early 22. D. R. Gallie, Genes Dev. 5, 2108 (1991).
repressed rapidly (Fig. 3E) and at levels compa- events thereof, and is subsequently consolidated 23. T. H. Beilharz, T. Preiss, RNA 13, 982 (2007).
24. M. Muckenthaler, N. Gunkel, R. Stripecke, M. W. Hentze,
rable to those observed with equivalent constructs by mRNA deadenylation and decay. Although it RNA 3, 983 (1997).
carrying a poly(A) tail (Fig. 2A versus Fig. 3E). is possible that the order of events is different in 25. S. Iwasaki, T. Kawamata, Y. Tomari, Mol. Cell 34, 58
These data provide further support for the idea other systems or in a fashion that is mRNA- (2009).
that deadenylation is not required for translational specific, our data in Drosophila are consistent. 26. W. F. Marzluff, E. J. Wagner, R. J. Duronio, Nat. Rev.
Genet. 9, 843 (2008).
repression by miRNAs (5). However, deadenyl- Moreover, these observations are consistent with 27. J. E. Braun, E. Huntzinger, M. Fauser, E. Izaurralde,
ases may play a direct role in translational repres- earlier studies on miRNA-mediated silencing Mol. Cell 44, 120 (2011).
sion independent of their deadenylation activity in vitro (15, 17) and with previous studies of trans- 28. M. Chekulaeva et al., Nat. Struct. Mol. Biol. 18, 1218
(2729), and deadenylation may consolidate the lation as affected by iron levels (24). With these (2011).
29. M. R. Fabian et al., Nat. Struct. Mol. Biol. 18, 1211
observed translational repression. insights into the relative timing of the events in- (2011).
To define where in the translation cycle the volved in miRNA-mediated gene silencing, we 30. K. Kuroha et al., EMBO Rep. 11, 956 (2010).
miRNA-mediated stalling occurs, we developed can now focus subsequent molecular mechanistic
an approach that uses as a read-out the specific analysis on these earliest triggering steps. Acknowledgments: We thank K. Wehner, J. Doudna,
M. Jinek, N. Guydosh, and J. Coller for helpful comments.
cleavage of mRNAs when ribosomes stall during
Funding is from HHMI.
translation elongation (30) (fig. S9). We inserted
12 lysine codons at position 6 in both the natural References and Notes Supplementary Materials
1. D. P. Bartel, Cell 136, 215 (2009). www.sciencemag.org/cgi/content/full/336/6078/237/DC1
and synthetic reporters; as a control, we inserted 2. S. Bagga et al., Cell 122, 553 (2005).
Materials and Methods
12 arginines and 12 more neutral glutamine res- 3. L. P. Lim et al., Nature 433, 769 (2005).
Figs. S1 to S11
idues (fig. S10). By using qRT primers that en- 4. A. J. Giraldez et al., Science 312, 75 (2006); 10.1126/
Tables S1 to S4
science.1122689.
compassed the predicted cleavage site as well as References
5. A. Eulalio et al., RNA 15, 21 (2009).
the stall sequence, we determined the extent of 6. T. H. Beilharz et al., PLoS ONE 4, e6783 (2009). 24 October 2011; accepted 13 March 2012
cleavage of both targeted and nontargeted report- 7. P. H. Olsen, V. Ambros, Dev. Biol. 216, 671 (1999). 10.1126/science.1215691

240 13 APRIL 2012 VOL 336 SCIENCE www.sciencemag.org

 miRNA-Mediated Gene Silencing by Translational Repression
Followed by mRNA Deadenylation and Decay
Sergej Djuranovic, Ali Nahvi and Rachel Green (April 12, 2012)
Science 336 (6078), 237-240. [doi: 10.1126/science.1215691]

Editor's Summary

Translation Block
MicroRNAs (miRNAs) are small, noncoding RNA genes that are found in the genomes of most
eukaryotes, where they play an important role in the regulation of gene expression. Although whether
gene activity is repressed by blocking translation of messenger RNA (mRNA) targets, or by promoting
their deadenylation and then degradation, has been open to debate. Bazzini et al. (p. 233, published
online 15 March) and Djuranovic et al. (p. 237) looked at early points in the repression reaction in the
zebrafish embryo or in Drosophila tissue culture cells, respectively, and found that translation was

Downloaded from http://science.sciencemag.org/ on July 22, 2016

blocked before target mRNAs were significantly deadenylated and degraded. Thus, miRNAs appear to
interfere with the initiation step of translation.

This copy is for your personal, non-commercial use only.

Article Tools Visit the online version of this article to access the personalization and
article tools:
http://science.sciencemag.org/content/336/6078/237

Permissions Obtain information about reproducing this article:

http://www.sciencemag.org/about/permissions.dtl

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week
in December, by the American Association for the Advancement of Science, 1200 New York
Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for the
Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.