Substrate requirements for let-7 function in the developing zebrafish
http://www.100md.com
《核酸研究医学期刊》
The Hubrecht Laboratory, Centre for Biomedical Genetics, 3584 CT Utrecht, The Netherlands
* To whom correspondence should be addressed. Tel: +31 30 2121963; Fax: +31 30 2516554; Email: plasterk@niob.knaw.nl
ABSTRACT
MicroRNAs (miRNAs) are involved in the regulation of gene expression at the post-transcriptional level by base pairing to the 3'-UTR (untranslated region) of mRNAs. The let-7 miRNA was first discovered in Caenorhabditis elegans and is evolutionarily conserved. We used zebrafish embryos as a vertebrate in vivo system to study substrate requirements for function of let-7. Injection of a double-stranded let-7 miRNA into the zygotes of zebrafish and frogs causes specific phenotypic defects. Only the antisense strand of the let-7 duplex has biological activity. In addition, co-injected mRNA of gfp fused to the 3'-UTR of a zebrafish lin-41 ortholog (a presumed target of let-7) is silenced by let-7. Point mutant studies revealed that the two let-7 target sites in the lin-41 3'-UTR are both essential and sufficient for silencing. let-7 and mir221 together, but not either of them alone, can silence a construct with one of the let-7 target sites replaced by a target site for mir221, showing that two different miRNAs can provide the required cooperative effect. let-7 target sites can be moved around: they are also functional when positioned in the coding sequence or even in the 5'-UTR of gfp. We took advantage of reporter and phenotypic assays to analyze the activity of all possible point mutant derivatives of let-7 and found that only the 5' region is critical for function of let-7.
INTRODUCTION
Hundreds of microRNAs (miRNAs) have been discovered in eukaryotes (1–11) and they form an abundant class of post-transcriptional regulators . MiRNAs are initially transcribed as longer precursors and subsequently processed into 21–23 nt double-stranded RNAs with 2 nt 3' overhangs by the RNAse III-like endoribonucleases Drosha (14) and Dicer (15–18), respectively. MiRNAs regulate the gene expression by incorporating into a RISC (RNA-induced silencing complex) complex that binds to miRNA complementary elements in the 3'-UTR (untranslated region) of target genes. Targets have been predicted computationally for many miRNAs, based on the conservation of miRNA targets and known miRNA-target interactions (19–23). Despite the numerous targets predicted for many miRNAs, only few studies have addressed a role for distinct miRNAs in animals: In flies, the bantam miRNA was shown to be involved in the control of cell proliferation (24) and miR-14 suppresses apoptosis and is required for fat metabolism (25). Some miRNAs from mouse are implicated in the modulation of hematopoietic lineage differentiation (26). The Caenorhabditis elegans lsy-6 and mir-273 miRNAs regulate chemosensory laterality (27,28). lin-4 is the founding member of the miRNA class of genes. It acts on C.elegans by binding to complementary sites in the 3'-UTR of the heterochronic genes lin-41 and lin-28 (29,30). The let-7 miRNA also regulates developmental timing in C.elegans by inhibiting the expression of heterochronic genes, among which lin-41 (31,32). At least two out of six let-7 target sites in the C.elegans lin-41 gene, together with the 27 bp sequence in between, were shown to be necessary for let-7-mediated gene silencing (33). Both let-7 and lin-41 are conserved in evolution and let-7 target sites are also present in the lin-41 orthologs of Drosophila and zebrafish (31).
Here, we demonstrate that injection of a synthetic let-7 miRNA (in double-stranded form) causes specific defects in the vertebrate embryo. Furthermore, we employ the zebrafish embryo to show that two let-7 target sites from the zebrafish lin-41 gene mediate silencing. Both target sites are essential for silencing; an mRNA with one let-7 target site replaced by a mir221 target site can be silenced by both miRNAs together. Target sites for let-7 are also functional when placed, in the coding sequence or in the 5'-UTR of a reporter gene.
No study systematically determined the importance of every position of a miRNA. The let-7 mutant allele (n2853) in C.elegans harbors a mutation at position 5 from the 5' end of the miRNA (32). This single point mutation abolishes the function of let-7 in C.elegans. Since the let-7 miRNA is strongly conserved, we took this miRNA to derive a complete mutational spectrum using zebrafish as an in vivo vertebrate system.
MATERIALS AND METHODS
Construction of gfp reporters
A 379 bp fragment containing two putative let-7 target sites in the zebrafish lin-41 3'-UTR (Al794385) was amplified from genomic DNA using primers: lin-41 3'-UTR F, GGCATTGAATTCATAAGACTGCTGCAAGCTGAGAG (EcoRI restriction site is underlined) and lin-41 3'-UTR R, GGCATTTCTAGATCAGGGATATAACTTGCGTTC (XbaI restriction site is underlined). This fragment was cloned into pCS2 (Clontech), containing a gfp cDNA sequence cloned between restriction sites BamHI and ClaI (pgfp) resulting in pgfp::lin-41 3'-UTR. pgfp::lin-41 3'-UTR-1, -2, -1 mut1, -1 mut2, pgfp::let-7tar1+2, pgfp::let-7tar1-2x, pgfp::let-7tar2-2x, pgfp::let-7tar1+mir221tar and gfp::let-7tar cds were made by cloning of double-stranded oligonucleotides lin-41 3'-UTR-1 (5'-gaattcataatcttcctgcattacacctacctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctacctcatctaga-3'), lin-41 3'-UTR-2 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctatctcatctaga-3'), lin-41 3'-UTR-1 mut-1 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctacctcatctaga-3'), lin-41 3'-UTR-1 mut2 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctatctcatctaga-3'), let-7tar1+2 (5'-gaattcataatcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcatctaga-3'), let-7tar1-2x (5'-gaattcataatctctcctgcattacacctacctcatctagctgcattacacctacctcatctaga-3'), let-7tar2-2x (5'-gaattcataatcttcgttaaaaaaaaattgtctacctcatctaggttaaaaaaaaattgtctacctcatctaga-3'), let-7tar1+mir221tar (5'-gaattcataatcttcctgcattacacctacctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttaacccagcagtgtatgtagcttctaga-3') and let-7 tar cds (5'-gaattctcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcactaagtctaga-3') in the EcoRI and XbaI sites of pgfp. Construct gfp::let-7tar 5'-UTR was made by cloning of oligonucleotide let-7tar 5'-UTR (ggatcctcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcactaagccatgg) in the NcoI and BamHI sites of pgfp. let-7 and mir221 target sites are underlined, and the mutations are shown in bold.
Microinjections
Wild-type and mutant let-7 deprotected and desalted RNA oligonucleotides (Proligo) were dissolved in RNAse free water at a concentration of 100 μM. Pre-let-7 (UGAGGUAGUAGGUUGUAUAGUUUUAGGGUCACACCCACCACUGGGAGAUAACUAUACAAUCUACUGUCUUUC) was obtained from Biolegio. Oligos were annealed using a 5x buffer containing 30 mM HEPES–KOH, pH 7.4, 100 mM KCl, 2 mM MgCl2 and 50 mM NH4Ac. 2'-O-methyl oligonucleotides (anti-let-7, UCUUAACUAUACAACCUACUACCUCAACCUU and control, UCUUCAGCCUAUCCUGGAUUACUUGAAACCUU; Dharmacon) were dissolved in RNAse free water at a concentration of 500 μM. Anti-let-7 morpholino (AACTATACAACCTACTACCTCA) was dissolved in water at a concentration of 25 ng/nl and injected at a concentration of 10 ng/nl. mRNA derived from Green fluorescent protein (GFP) reporter constructs was obtained by in vitro transcription using SP6 (Boehringer) and SacII linearized plasmid as a template. Injection mixtures contained 10 μM of a let-7 duplex and 100 ng/μl gfp mRNA and 50 μM 2'-O-methyl oligonucleotide where indicated. This solution was injected into the one-cell stage of wild-type embryos derived from the TL line using 1 nl per embryo. Xenopus tropicalis embryos were injected in the two-cell stage using 2 nl per cell.
Northern blot analysis
Total RNA from embryos was isolated using TRIzol Reagent (Invitrogen). GFP mRNA was detected using RNA from 10 embryos (3 μg), separated on 1.5% agarose gels according to the standard procedures. A random primed 32P-dCTP radiolabeled probe covering the complete GFP cDNA sequence was used for hybridization. let-7 was detected using RNA isolated from 5 embryos (1.5 μg). RNA was separated on a 15% polyacrylamide gel. A radiolabeled probe complementary to let-7 was used for hybridization.
Western blot analysis
Protein was isolated by boiling 5 embryos (5 mg) for 10 min in 10 μl loading buffer. Prior to loading lysates were centrifuged for 5 min at 14 000 g. Western blotting was performed according to the standard procedures. GFP was detected using a rabbit polyclonal antibody.
RESULTS AND DISCUSSION
Effects of injected let-7 on development
During the first 48 h of zebrafish development, no endogenous let-7 miRNA is expressed (31,34). We observed a specific phenotype upon injection of a double-stranded let-7 miRNA in one-cell stage zebrafish embryos (Figure 1A). At 26 h post-fertilization (hpf), embryos were generally retarded in development. More pronounced characteristics were a lack of proper eye development and reduced tail and yolk sac extension. The embryos died after 2 days. This phenotype was only induced by a double-stranded version of let-7. Injection of either the sense or the antisense strand alone did not affect development, although these species remain stable in vivo for at least 48 h (data not shown). A control miRNA bearing five mutations (mmlet-7) failed to induce the phenotype, indicating the specificity of the observed phenotype (Figure 1A). Furthermore, an injected pre-let-7 hairpin is processed in vivo and induces a phenotype similar to that of a mature let-7 duplex (Figure 1B).
Figure 1. Misexpression of let-7 in vertebrate embryos. (A) Lateral views of Danio rerio and X.tropicalis embryos injected with let-7 and a mutant version containing five mismatches (underlined). (B) Phenotype induced by injection of a pre-let-7 hairpin. The northern blot shows processing of pre-let-7 in zebrafish embryos. Pictures were taken at 28 h post-fertilization (hpf) and 2 days post-fertilization (dpf) for D.rerio and X.tropicalis, respectively. NIC, not injected control.
Because of its perfect conservation, we also investigated the effects of let-7 microinjection on X.tropicalis embryos. Strikingly, this resulted in similar defects as for zebrafish, i.e. embryos were retarded in development and exhibited a reduced eye size and tail length (Figure 1A). Development was virtually normal in embryos injected with the mmlet-7 duplex. The phenotypic effects caused by let-7 misexpression in zebrafish could be specifically inhibited by co-injection with a 2'-O-methyl oligonucleotide complementary to the let-7 miRNA, but not by a control 2'-O-methyl oligonucleotide unrelated in sequence to let-7 (Figure 3B) (35). Injection of the oligonucleotide alone did not cause any developmental abnormalities. Similarly, a morpholino directed against let-7 inhibited the misexpression phenotype, but did not induce a phenotype when injected alone (data not shown). The results presented here show that the let-7 miRNA is active in the zebrafish embryo and severely affects normal development upon misexpression. Although the biological significance of the observed effects remains unclear, it seems likely that ectopic expression of let-7 causes precocious downregulation of one or more endogenous let-7 targets. Misexpression studies can be important for defining miRNA function, especially in cases where knockout is difficult because of redundancy.
Figure 3. Effects of let-7 injection on gfp::lin-41 fusions and zebrafish development. (A) Structure of let-7 bound to its target sites in the zebrafish lin-41 3'-UTR. The position of the point mutation in the target and let-7 is indicated in red. At these positions the C in the target is changed to a T and the G in let-7 is changed to an A. (B) Lateral views of 28 h zebrafish embryos injected with gfp with and without the lin-41 3'-UTR. The upper panel shows phenotypes and the lower panel GFP fluorescence. (C) Western and northern analysis of GFP in 28 h zebrafish embryos co-injected with gfp:lin-41 3'-UTR. (D) Phenotypes and GFP expression in 28 h embryos injected with mutant (gfp::lin-41 3'-UTR-2, let-7mm5) and wild-type (gfp::lin-41 3'-UTR-1, let-7) let-7 and gfp::lin-41 3'-UTR fusions.
The antisense strand of the let-7 duplex is biologically active and an asymmetric duplex exhibits enhanced activity
We next investigated which strand of the let-7 duplex exerts a biological effect by examining the activity of heteroduplexes (Figure 2). A heteroduplex containing two mutations at the 5' end of the antisense strand of let-7 (heteroduplex 1) fails to induce a phenotype, while a heteroduplex with two mutations at the 3' end of the sense strand (heteroduplex 2) functions extremely well. This indicates that only the antisense strand of the let-7 duplex exerts a biological effect. The phenotype induced by heteroduplex 2 is even stronger than that induced by the let-7 homoduplex. In contrast, injection of a heteroduplex containing two mismatches at the 5' end of the sense strand (heteroduplex 3) resulted in a reduced phenotype. These data support recent work, which shows that both the absolute and the relative stabilities at the 5' ends of siRNA or miRNA duplexes are critical determinants for incorporation of one or the other strand into RISC (36–38). Our experiments demonstrate that the in vivo activity of a miRNA is subject to the same criteria.
Figure 2. Phenotypic effects of injection of 3 let-7 heteroduplexes. (A) Sequences of heteroduplexes. The upper strand is the antisense strand of the let-7 duplex. (B) Phenotypes of 28 h zebrafish embryos injected with heteroduplexes. (C) Bar diagram showing relative length of the yolk sac extension of fish injected with let-7 heteroduplexes.
Silencing of a gfp::lin-41 3'-UTR reporter by let-7
The let-7/lin-41 miRNA-target couple was experimentally verified in C.elegans (32), but conserved sequences are found in Drosophila and zebrafish (31). To show that injected let-7 acts as a miRNA on the zebrafish lin-41 3'-UTR, we fused a 379 bp fragment comprising part of the zebrafish lin-41 3'-UTR and containing two let-7 target sites (Figure 3A) to gfp. Co-injection of let-7 with mRNA derived from this construct resulted in the translational repression of GFP expression (Figure 3B). The gfp::lin41 3'-UTR mRNA levels remained unaffected, whereas GFP protein levels were dramatically reduced upon let-7 overexpression (Figure 3C). Silencing of gfp is dependent on the lin-41 3'-UTR, since gfp mRNA without the lin-41 3'-UTR was not silenced by let-7 (Figure 3B). Furthermore, silencing of gfp::lin-41 3'-UTR could be blocked by the 2'-O-methyl oligonucleotide complementary to let-7, and the mmlet-7 version did not silence gfp::lin-41 3'-UTR. We did not observe downregulation of the gfp::lin-41 3'-UTR reporter by endogenous let-7 as shown by a recent study using mouse embryos transiently expressing lacZ miRNA sensor constructs (39). This is likely due to the fact that there are no detectable amounts of let-7 expressed during these stages of development.
To investigate the interaction between let-7 and its target sites in more detail, gfp was fused to a 100 bp fragment of the lin-41 3'-UTR containing only the two let-7 target sites including intervening sequence (pgfp::lin-41 3'-UTR-1). In addition, a similar construct was made that contains the same fragment, but with a point mutation in both let-7 target sites (gfp lin-41 3'-UTR-2) (Figures 3A and 4A). mRNA derived from gfp lin-41 3'-UTR-1 could only be silenced by wild-type let-7, while expression of gfp::lin-41 3'-UTR-2 was inhibited exclusively by a let-7 duplex with a compensatory mutation (let-7mm5, Figure 3D). In addition, duplex let-7mm5 does not induce a phenotype in fish embryos. Taken together, our data show that injection of a mature let-7 miRNA duplex can specifically induce gene silencing via the lin-41 3'-UTR in the developing zebrafish embryo.
Figure 4. let-7 target site 1 and 2 are sufficient for silencing. (A) Overview of constructs and miRNAs analyzed in this study. + indicates strong silencing, – indicates no silencing (B) GFP fluorescence in 24 h zebrafish embryos injected with let-7 or let-7mm5 and gfp::let-7tar 5'-UTR or gfp::let-7tar cds.
We next asked whether the sequence intervening the let-7 target sites is an essential component for silencing. In C.elegans, it was shown that silencing of a lacZ reporter fused to the C.elegans lin-41 3'-UTR is dependent on two specific let-7 target sites and the 27 bp sequence between the target sites (33). In contrast, we found that a gfp reporter containing both let-7 target sites from the zebrafish lin-41 gene separated by a 5 bp sequence is still silenced by co-injected let-7 (Figure 4A, pgfp::let-7tar1+2), while in the natural situation both let-7 target sites are separated by a 57 bp sequence. Similar constructs containing either two copies of target site 1 or two copies of target site 2 fused to gfp were also silenced by let-7 (Figure 4A, pgfp::let-7tar1-2x and pgfp::let-7tar2-2x), showing that both target sites are equally potent in silencing. However, reporter constructs with a point mutation (Figure 3A) in either let-7 target site 1 or 2 could not be silenced by let-7, demonstrating that both target sites act together to mediate a silencing response (Figure 4A, pgfp::lin-41 3'-UTR-1 mut1 and pgfp::lin-41 3'-UTR-1 mut2). GFP expression from pgfp::lin-41 3'-UTR-1 mut1 and pgfp::lin-41 3'-UTR-1 mut2 cannot be downregulated by co-injection of a mixture of let-7 and let-7mm5. This is probably due to a competition effect (i.e. let-7 binding to the mutant target site and let-7mm5 binding to the wild-type target site), since a mixture of let-7 and mir221 could silence a reporter construct with let-7 target site 2 replaced by an artificially designed mir221 complementary site (Figure 4A, pgfp::let-7tar1+mir221tar), whereas let-7 and mir221 alone did not affect the GFP expression from this construct. These data are in agreement with previous studies in cultured mammalian cells (40), showing that miRNAs can act cooperatively to repress gene expression, although our data differ from these studies because we used a combination of two copies of one target site instead of two copies of two target sites acting together.
Taken together, we demonstrate that injected let-7 can act on the zebrafish homolog of the lin-41 gene, but we have no way of knowing whether endogenous let-7 silences lin-41 in vivo or whether the phenotype we observe is the result of injected let-7 acting on lin-41.
Silencing of reporters containing let-7 targets in the coding region or 5'-UTR
To investigate whether silencing of the GFP reporter is restricted to the presence of let-7 target sites specifically in the 3'-UTR, constructs were made with the let-7 target unit comprising of both target sites with a 5 bp spacer inserted in the 5'-UTR (gfp::let-7tar 5'-UTR) and 12 bp before the stop codon in the gfp coding sequence (gfp::let-7tar cds) (Figure 4A). Surprisingly, both constructs could be specifically silenced by co-injected let-7, but not by let-7mm5 (Figure 4B). These data indicate that GFP is specifically downregulated due to let-7 sites in the coding region or the 5'-UTR. In the latter case, it cannot be ruled out that translation initiation is inhibited due to steric hindrance of a let-7 loaded RISC complex just upstream of the start codon.
In plants, it is known that miRNA targets are also present in the coding sequence of the mRNA, but plant miRNAs often target the mRNA for degradation via single complementary sites (41). No natural examples are known yet of animal miRNAs regulating a target gene via complementary sites in the coding sequence or 5'-UTR, although it was shown that siRNA off-target effects can be mediated through translational repression of mRNAs due to imperfect base pairing of the siRNA with the coding sequence (42).
In our zebrafish system, we find that miRNA-target sites do not necessary need to be located in the 3'-UTR, since let-7 targets both in the 5'-UTR or in the coding sequence could induce a specific silencing response. This is important information for studies aimed at predicting miRNA targets, which are currently only focusing on 3'-UTR sequences. As also suggested by others (12,13,20), these should include 5'-UTR and coding sequences as well.
The 7 bases at the 5' end of let-7 are most important for its activity
To define the most important positions for functionality of the let-7 miRNA, we performed a mutational scan by introducing point mutations at each of the 22 positions of let-7. For all 66 (3 x 22) mutants, we determined phenotypic defects and the influence on gfp::lin-41 3'-UTR mRNA translation (Figure 5A and B) as these assays are good indicators for the activity of let-7. Similar data were obtained by both the analysis of GFP protein level and the examination of phenotypic defects: mutations induced in the region spanning positions 1–7 from the 5' end of the antisense strand of the let-7 duplex eliminate the function of the miRNA (Figure 5). However, small growth defects were observed for let-7 duplexes with mutations at position 1. Versions of let-7 with a mutation at position 8 showed a strong phenotype but only a low reduction in the GFP protein level. Remarkably, the change of an A to a G at positions 3 and 7, enabling the formation of a G-U instead of an A-U base pair with the lin-41 reporter construct, did not inhibit protein expression or affect normal development.
Figure 5. Mutational analysis of the let-7 miRNA. (A) Phenotypes were scored by measuring the length of the yolk sac extension relative to the diameter of the yolk (bar diagram). For each data point, five individuals were measured. Approximate GFP levels, as assessed from western blots (B), are indicated in the table. – indicates that GFP protein level is similar to a control where only the gfp::lin-41 3'-UTR mRNA was injected; + indicates a significant decline of GFP protein level compared with the control; 0 indicates a moderate decline of GFP protein level compared with the control. At each position, the three types of possible mutations are shown in the sequences beneath the bars. The first series of mutants (purple bars and sequence a) has an equal number of hydrogen bonds compared with the wild-type let-7 duplex. Both other series (dark red and gray bars and sequences b and c) have a different number of hydrogen bonds, depending on the nature of the mutation, i.e. U/A to C/G or vice versa. The let-7 sequence is depicted in red. (B) Western blot analysis of GFP protein level for 3 x 22 mutant derivatives of let-7. Protein was extracted from 5 embryos. For each sample, 5 mg of protein was loaded. As an injection control, the level of gfp::lin-41 3'-UTR mRNA was determined by northern blotting. For each sample, 3 μg of RNA, obtained from 10 individuals, was loaded. Numbering is from the 5' end of let-7 and corresponds to that in (A).
Most prominently at positions 10 and 13, GFP silencing and phenotypic effects were different for the 3 mutant let-7 derivatives. These differences probably reflect changes in the let-7 duplex free energy, which is relatively low for positions 9–14 in natural miRNA duplexes (36,37). For example, at position 10 the change of the original A-U base pair in the let-7 duplex into an U-A base pair still causes strong silencing. However, the change to a G-C or C-G base pair increases stability of the duplex and decreases its activity. Similar although less pronounced effects were observed for mutations at position 9. Thus, mutations at certain positions in the miRNA duplex can affect its activity, not because of an altered interaction with the target RNA, but because of a change in the stability of the miRNA duplex itself.
Recently, the specificity of miRNA-target interaction was analyzed in cell culture using luciferase reporter assays (40). In this study, the miRNA target instead of the miRNA itself was analyzed for critical positions and, similar to our observations, these reporter studies indicated that base pairing of the first 8 nt in the 5' region of the miRNA is most important for activity, although the binding of the 3' end of a miRNA also appeared to contribute to gene silencing. These observations were made with reporter constructs containing multiple miRNA-target sites similar to our gfp::lin-41 reporter. Investigation of rules governing the interaction of a miRNA with a single target site showed that interaction of the proximal (i.e. 5') part of miRNA with its target is indeed the most susceptible to mutations (43). Only a target:miRNA mismatch in the 5' part flanked by 4 bp on each side resulted in silencing of the reporter. Similarly, mutational analysis of the plant mir165/166 complementary site of PHABULOSA revealed that disrupting miRNA pairing near the 5' region causes stronger developmental consequences and reduced miRNA-directed cleavage in vitro (44).
For our phenotypic assay, it is unclear whether the affected target mRNA(s) that cause(s) the phenotype contain one or multiple let-7 targets. Our data add to the previous studies that irrespective of the targets, the major sequence determinants for in vivo function of let-7 in zebrafish lie in the 5' 1–7 residues.
This study represents the first complete mutational analysis of a miRNA in a vertebrate model organism. It is striking and unexplained that the let-7 miRNA is perfectly conserved throughout evolution, while only mutations in the first 7 residues affect its activity. It could be that the sensitivity of both assays described here is too low, since they make use of a rather artificial over expression setup. Another explanation might be that the mutational spectrum was derived by injection of mature let-7 miRNA sequences. The processing of the precursor into its mature form might also be affected by mutations in (the 3' part of) let-7, which was not addressed by our screen.
ACKNOWLEDGEMENTS
We thank B. Ason, R. May and M. Tijsterman for critical reading of the manuscript and G. Ro?l for help with Xenopus tropicalis injections.
REFERENCES
Lagos-Quintana,M., Rauhut,R., Yalcin,A., Meyer,J., Lendeckel,W. and Tuschl,T. ( (2002) ) Identification of tissue-specific microRNAs from mouse. Curr. Biol., , 12, , 735–739.
Lagos-Quintana,M., Rauhut,R., Meyer,J., Borkhardt,A. and Tuschl,T. ( (2003) ) New microRNAs from mouse and human. RNA, , 9, , 175–179.
Lagos-Quintana,M., Rauhut,R., Lendeckel,W. and Tuschl,T. ( (2001) ) Identification of novel genes coding for small expressed RNAs. Science, , 294, , 853–858.
Lau,N.C., Lim,L.P., Weinstein,E.G. and Bartel,D.P. ( (2001) ) An abundant class of tiny RNAs with probable regulatory roles in Caenorhabditis elegans. Science, , 294, , 858–862.
Lee,R.C. and Ambros,V. ( (2001) ) An extensive class of small RNAs in Caenorhabditis elegans. Science, , 294, , 862–864.
Lai,E.C., Tomancak,P., Williams,R.W. and Rubin,G.M. ( (2003) ) Computational identification of Drosophila microRNA genes. Genome Biol., , 4, , R42.
Lim,L.P., Glasner,M.E., Yekta,S., Burge,C.B. and Bartel,D.P. ( (2003) ) Vertebrate microRNA genes. Science, , 299, , 1540.
Lim,L.P., Lau,N.C., Weinstein,E.G., Abdelhakim,A., Yekta,S., Rhoades,M.W., Burge,C.B. and Bartel,D.P. ( (2003) ) The microRNAs of Caenorhabditis elegans. Genes Dev., , 17, , 991–1008.
Griffiths-Jones,S. ( (2004) ) The microRNA Registry. Nucleic Acids Res., , 32, (Database issue), D109–D111.
Pfeffer,S., Zavolan,M., Grasser,F.A., Chien,M., Russo,J.J., Ju,J., John,B., Enright,A.J., Marks,D., Sander,C. et al. ( (2004) ) Identification of virus-encoded microRNAs. Science, , 304, , 734–736.
Reinhart,B.J., Weinstein,E.G., Rhoades,M.W., Bartel,B. and Bartel,D.P. ( (2002) ) MicroRNAs in plants. Genes Dev., , 16, , 1616–1626.
Bartel,D.P. ( (2004) ) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell, , 116, , 281–297.
Ambros,V. ( (2004) ) The functions of animal microRNAs. Nature, , 431, , 350–355.
Lee,Y., Ahn,C., Han,J., Choi,H., Kim,J., Yim,J., Lee,J., Provost,P., Radmark,O., Kim,S. et al. ( (2003) ) The nuclear RNase III Drosha initiates microRNA processing. Nature, , 425, , 415–419.
Bernstein,E., Caudy,A.A., Hammond,S.M. and Hannon,G.J. ( (2001) ) Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, , 409, , 363–366.
Grishok,A., Pasquinelli,A.E., Conte,D., Li,N., Parrish,S., Ha,I., Baillie,D.L., Fire,A., Ruvkun,G. and Mello,C.C. ( (2001) ) Genes and mechanisms related to RNA interference regulate expression of the small temporal RNAs that control C.elegans developmental timing. Cell, , 106, , 23–34.
Hutvagner,G., McLachlan,J., Pasquinelli,A.E., Balint,E., Tuschl,T. and Zamore,P.D. ( (2001) ) A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA. Science, , 293, , 834–838.
Ketting,R.F., Fischer,S.E., Bernstein,E., Sijen,T., Hannon,G.J. and Plasterk,R.H. ( (2001) ) Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C.elegans. Genes Dev., , 15, , 2654–2659.
Enright,A.J., John,B., Gaul,U., Tuschl,T., Sander,C. and Marks,D.S. ( (2003) ) MicroRNA targets in Drosophila. Genome Biol., , 5, , R1.
Lai,E.C. ( (2004) ) Predicting and validating microRNA targets. Genome Biol., , 5, , 115.
Lewis,B.P., Shih,I.H., Jones-Rhoades,M.W., Bartel,D.P. and Burge,C.B. ( (2003) ) Prediction of mammalian microRNA targets. Cell, , 115, , 787–798.
Rhoades,M.W., Reinhart,B.J., Lim,L.P., Burge,C.B., Bartel,B. and Bartel,D.P. ( (2002) ) Prediction of plant microRNA targets. Cell, , 110, , 513–520.
Stark,A., Brennecke,J., Russell,R.B. and Cohen,S.M. ( (2003) ) Identification of Drosophila microRNA targets. PLoS Biol., , 1, , 1–13.
Brennecke,J., Hipfner,D.R., Stark,A., Russell,R.B. and Cohen,S.M. ( (2003) ) bantam encodes a developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell, , 113, , 25–36.
Xu,P., Vernooy,S.Y., Guo,M. and Hay,B.A. ( (2003) ) The Drosophila microRNA Mir-14 suppresses cell death and is required for normal fat metabolism. Curr. Biol., , 13, , 790–795.
Chen,C.Z., Li,L., Lodish,H.F. and Bartel,D.P. ( (2004) ) MicroRNAs modulate hematopoietic lineage differentiation. Science, , 303, , 83–86.
Johnston,R.J. and Hobert,O. ( (2003) ) A microRNA controlling left/right neuronal asymmetry in Caenorhabditis elegans. Nature, , 426, , 845–849.
Chang,S., Johnston,R.J.,Jr, Frokjaer-Jensen,C., Lockery,S. and Hobert,O. ( (2004) ) MicroRNAs act sequentially and asymmetrically to control chemosensory laterality in the nematode. Nature, , 430, , 785–789.
Lee,R.C., Feinbaum,R.L. and Ambros,V. ( (1993) ) The C.elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell, , 75, , 843–854.
Wightman,B., Ha,I. and Ruvkun,G. ( (1993) ) Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C.elegans. Cell, , 75, , 855–862.
Pasquinelli,A.E., Reinhart,B.J., Slack,F., Martindale,M.Q., Kuroda,M.I., Maller,B., Hayward,D.C., Ball,E.E., Degnan,B., Muller,P. et al. ( (2000) ) Conservation of the sequence and temporal expression of let-7 heterochronic regulatory RNA. Nature, , 408, , 86–89.
Reinhart,B.J., Slack,F.J., Basson,M., Pasquinelli,A.E., Bettinger,J.C., Rougvie,A.E., Horvitz,H.R. and Ruvkun,G. ( (2000) ) The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature, , 403, , 901–906.
Vella,M.C., Choi,E.Y., Lin,S.Y., Reinert,K. and Slack,F.J. ( (2004) ) The C.elegans microRNA let-7 binds to imperfect let-7 complementary sites from the lin-41 3'UTR. Genes Dev., , 18, , 132–137.
Wienholds,E., Koudijs,M.J., van Eeden,F.J., Cuppen,E. and Plasterk,R.H. ( (2003) ) The microRNA-producing enzyme Dicer1 is essential for zebrafish development. Nature Genet., , 35, , 217–218.
Hutvagner,G., Simard,M.J., Mello,C.C. and Zamore,P.D. ( (2004) ) Sequence-specific inhibition of small RNA function. PLoS Biol., , 2, , E98.
Schwarz,D.S., Hutvagner,G., Du,T., Xu,Z., Aronin,N. and Zamore,P.D. ( (2003) ) Asymmetry in the assembly of the RNAi enzyme complex. Cell, , 115, , 199–208.
Khvorova,A., Reynolds,A. and Jayasena,S.D. ( (2003) ) Functional siRNAs and miRNAs exhibit strand bias. Cell, , 115, , 209–216.
Tomari,Y., Matranga,C., Haley,B., Martinez,N. and Zamore,P.D. ( (2004) ) A protein sensor for siRNA asymmetry. Science, , in press.
Mansfield,J.H., Harfe,B.D., Nissen,R., Obenauer,J., Srineel,J., Chaudhuri,A., Farzan-Kashani,R., Zuker,M., Pasquinelli,A.E., Ruvkun,G. et al. ( (2004) ) MicroRNA-responsive ‘sensor’ transgenes uncover Hox-like and other developmentally regulated patterns of vertebrate microRNA expression. Nature Genet, ., 36, , 1079–1083.
Doench,J.G. and Sharp,P.A. ( (2004) ) Specificity of microRNA target selection in translational repression. Genes Dev., , 18, , 504–511.
Hake,S. ( (2003) ) MicroRNAs: a role in plant development. Curr. Biol., , 13, , R851–R852.
Saxena,S., Jonsson,Z.O. and Dutta,A. ( (2003) ) Small RNAs with imperfect match to endogenous mRNA repress translation. Implications for off-target activity of small inhibitory RNA in mammalian cells. J. Biol. Chem., , 278, , 44312–44319.
Kiriakidou,M., Nelson,P.T., Kouranov,A., Fitziev,P., Bouyioukos,C., Mourelatos,Z. and Hatzigeorgiou,A. ( (2004) ) A combined computational-experimental approach predicts human microRNA targets. Genes Dev., , 18, , 1165–1178.
Mallory,A.C., Reinhart,B.J., Jones-Rhoades,M.W., Tang,G., Zamore,P.D., Barton,M.K. and Bartel,D.P. ( (2004) ) MicroRNA control of PHABULOSA in leaf development: importance of pairing to the microRNA 5' region. EMBO J., , 23, , 3356–3364.(Wigard P. Kloosterman, Erno Wienholds, R)
* To whom correspondence should be addressed. Tel: +31 30 2121963; Fax: +31 30 2516554; Email: plasterk@niob.knaw.nl
ABSTRACT
MicroRNAs (miRNAs) are involved in the regulation of gene expression at the post-transcriptional level by base pairing to the 3'-UTR (untranslated region) of mRNAs. The let-7 miRNA was first discovered in Caenorhabditis elegans and is evolutionarily conserved. We used zebrafish embryos as a vertebrate in vivo system to study substrate requirements for function of let-7. Injection of a double-stranded let-7 miRNA into the zygotes of zebrafish and frogs causes specific phenotypic defects. Only the antisense strand of the let-7 duplex has biological activity. In addition, co-injected mRNA of gfp fused to the 3'-UTR of a zebrafish lin-41 ortholog (a presumed target of let-7) is silenced by let-7. Point mutant studies revealed that the two let-7 target sites in the lin-41 3'-UTR are both essential and sufficient for silencing. let-7 and mir221 together, but not either of them alone, can silence a construct with one of the let-7 target sites replaced by a target site for mir221, showing that two different miRNAs can provide the required cooperative effect. let-7 target sites can be moved around: they are also functional when positioned in the coding sequence or even in the 5'-UTR of gfp. We took advantage of reporter and phenotypic assays to analyze the activity of all possible point mutant derivatives of let-7 and found that only the 5' region is critical for function of let-7.
INTRODUCTION
Hundreds of microRNAs (miRNAs) have been discovered in eukaryotes (1–11) and they form an abundant class of post-transcriptional regulators . MiRNAs are initially transcribed as longer precursors and subsequently processed into 21–23 nt double-stranded RNAs with 2 nt 3' overhangs by the RNAse III-like endoribonucleases Drosha (14) and Dicer (15–18), respectively. MiRNAs regulate the gene expression by incorporating into a RISC (RNA-induced silencing complex) complex that binds to miRNA complementary elements in the 3'-UTR (untranslated region) of target genes. Targets have been predicted computationally for many miRNAs, based on the conservation of miRNA targets and known miRNA-target interactions (19–23). Despite the numerous targets predicted for many miRNAs, only few studies have addressed a role for distinct miRNAs in animals: In flies, the bantam miRNA was shown to be involved in the control of cell proliferation (24) and miR-14 suppresses apoptosis and is required for fat metabolism (25). Some miRNAs from mouse are implicated in the modulation of hematopoietic lineage differentiation (26). The Caenorhabditis elegans lsy-6 and mir-273 miRNAs regulate chemosensory laterality (27,28). lin-4 is the founding member of the miRNA class of genes. It acts on C.elegans by binding to complementary sites in the 3'-UTR of the heterochronic genes lin-41 and lin-28 (29,30). The let-7 miRNA also regulates developmental timing in C.elegans by inhibiting the expression of heterochronic genes, among which lin-41 (31,32). At least two out of six let-7 target sites in the C.elegans lin-41 gene, together with the 27 bp sequence in between, were shown to be necessary for let-7-mediated gene silencing (33). Both let-7 and lin-41 are conserved in evolution and let-7 target sites are also present in the lin-41 orthologs of Drosophila and zebrafish (31).
Here, we demonstrate that injection of a synthetic let-7 miRNA (in double-stranded form) causes specific defects in the vertebrate embryo. Furthermore, we employ the zebrafish embryo to show that two let-7 target sites from the zebrafish lin-41 gene mediate silencing. Both target sites are essential for silencing; an mRNA with one let-7 target site replaced by a mir221 target site can be silenced by both miRNAs together. Target sites for let-7 are also functional when placed, in the coding sequence or in the 5'-UTR of a reporter gene.
No study systematically determined the importance of every position of a miRNA. The let-7 mutant allele (n2853) in C.elegans harbors a mutation at position 5 from the 5' end of the miRNA (32). This single point mutation abolishes the function of let-7 in C.elegans. Since the let-7 miRNA is strongly conserved, we took this miRNA to derive a complete mutational spectrum using zebrafish as an in vivo vertebrate system.
MATERIALS AND METHODS
Construction of gfp reporters
A 379 bp fragment containing two putative let-7 target sites in the zebrafish lin-41 3'-UTR (Al794385) was amplified from genomic DNA using primers: lin-41 3'-UTR F, GGCATTGAATTCATAAGACTGCTGCAAGCTGAGAG (EcoRI restriction site is underlined) and lin-41 3'-UTR R, GGCATTTCTAGATCAGGGATATAACTTGCGTTC (XbaI restriction site is underlined). This fragment was cloned into pCS2 (Clontech), containing a gfp cDNA sequence cloned between restriction sites BamHI and ClaI (pgfp) resulting in pgfp::lin-41 3'-UTR. pgfp::lin-41 3'-UTR-1, -2, -1 mut1, -1 mut2, pgfp::let-7tar1+2, pgfp::let-7tar1-2x, pgfp::let-7tar2-2x, pgfp::let-7tar1+mir221tar and gfp::let-7tar cds were made by cloning of double-stranded oligonucleotides lin-41 3'-UTR-1 (5'-gaattcataatcttcctgcattacacctacctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctacctcatctaga-3'), lin-41 3'-UTR-2 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctatctcatctaga-3'), lin-41 3'-UTR-1 mut-1 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctacctcatctaga-3'), lin-41 3'-UTR-1 mut2 (5'-gaattcataatcttcctgcattacacctatctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttgttaaaaaaaaattgtctatctcatctaga-3'), let-7tar1+2 (5'-gaattcataatcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcatctaga-3'), let-7tar1-2x (5'-gaattcataatctctcctgcattacacctacctcatctagctgcattacacctacctcatctaga-3'), let-7tar2-2x (5'-gaattcataatcttcgttaaaaaaaaattgtctacctcatctaggttaaaaaaaaattgtctacctcatctaga-3'), let-7tar1+mir221tar (5'-gaattcataatcttcctgcattacacctacctcatctagcttatgtatgaatgtactcgcgtttgtgcagagacctagtcggtgaagttttaacccagcagtgtatgtagcttctaga-3') and let-7 tar cds (5'-gaattctcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcactaagtctaga-3') in the EcoRI and XbaI sites of pgfp. Construct gfp::let-7tar 5'-UTR was made by cloning of oligonucleotide let-7tar 5'-UTR (ggatcctcttcctgcattacacctacctcatctaggttaaaaaaaaattgtctacctcactaagccatgg) in the NcoI and BamHI sites of pgfp. let-7 and mir221 target sites are underlined, and the mutations are shown in bold.
Microinjections
Wild-type and mutant let-7 deprotected and desalted RNA oligonucleotides (Proligo) were dissolved in RNAse free water at a concentration of 100 μM. Pre-let-7 (UGAGGUAGUAGGUUGUAUAGUUUUAGGGUCACACCCACCACUGGGAGAUAACUAUACAAUCUACUGUCUUUC) was obtained from Biolegio. Oligos were annealed using a 5x buffer containing 30 mM HEPES–KOH, pH 7.4, 100 mM KCl, 2 mM MgCl2 and 50 mM NH4Ac. 2'-O-methyl oligonucleotides (anti-let-7, UCUUAACUAUACAACCUACUACCUCAACCUU and control, UCUUCAGCCUAUCCUGGAUUACUUGAAACCUU; Dharmacon) were dissolved in RNAse free water at a concentration of 500 μM. Anti-let-7 morpholino (AACTATACAACCTACTACCTCA) was dissolved in water at a concentration of 25 ng/nl and injected at a concentration of 10 ng/nl. mRNA derived from Green fluorescent protein (GFP) reporter constructs was obtained by in vitro transcription using SP6 (Boehringer) and SacII linearized plasmid as a template. Injection mixtures contained 10 μM of a let-7 duplex and 100 ng/μl gfp mRNA and 50 μM 2'-O-methyl oligonucleotide where indicated. This solution was injected into the one-cell stage of wild-type embryos derived from the TL line using 1 nl per embryo. Xenopus tropicalis embryos were injected in the two-cell stage using 2 nl per cell.
Northern blot analysis
Total RNA from embryos was isolated using TRIzol Reagent (Invitrogen). GFP mRNA was detected using RNA from 10 embryos (3 μg), separated on 1.5% agarose gels according to the standard procedures. A random primed 32P-dCTP radiolabeled probe covering the complete GFP cDNA sequence was used for hybridization. let-7 was detected using RNA isolated from 5 embryos (1.5 μg). RNA was separated on a 15% polyacrylamide gel. A radiolabeled probe complementary to let-7 was used for hybridization.
Western blot analysis
Protein was isolated by boiling 5 embryos (5 mg) for 10 min in 10 μl loading buffer. Prior to loading lysates were centrifuged for 5 min at 14 000 g. Western blotting was performed according to the standard procedures. GFP was detected using a rabbit polyclonal antibody.
RESULTS AND DISCUSSION
Effects of injected let-7 on development
During the first 48 h of zebrafish development, no endogenous let-7 miRNA is expressed (31,34). We observed a specific phenotype upon injection of a double-stranded let-7 miRNA in one-cell stage zebrafish embryos (Figure 1A). At 26 h post-fertilization (hpf), embryos were generally retarded in development. More pronounced characteristics were a lack of proper eye development and reduced tail and yolk sac extension. The embryos died after 2 days. This phenotype was only induced by a double-stranded version of let-7. Injection of either the sense or the antisense strand alone did not affect development, although these species remain stable in vivo for at least 48 h (data not shown). A control miRNA bearing five mutations (mmlet-7) failed to induce the phenotype, indicating the specificity of the observed phenotype (Figure 1A). Furthermore, an injected pre-let-7 hairpin is processed in vivo and induces a phenotype similar to that of a mature let-7 duplex (Figure 1B).
Figure 1. Misexpression of let-7 in vertebrate embryos. (A) Lateral views of Danio rerio and X.tropicalis embryos injected with let-7 and a mutant version containing five mismatches (underlined). (B) Phenotype induced by injection of a pre-let-7 hairpin. The northern blot shows processing of pre-let-7 in zebrafish embryos. Pictures were taken at 28 h post-fertilization (hpf) and 2 days post-fertilization (dpf) for D.rerio and X.tropicalis, respectively. NIC, not injected control.
Because of its perfect conservation, we also investigated the effects of let-7 microinjection on X.tropicalis embryos. Strikingly, this resulted in similar defects as for zebrafish, i.e. embryos were retarded in development and exhibited a reduced eye size and tail length (Figure 1A). Development was virtually normal in embryos injected with the mmlet-7 duplex. The phenotypic effects caused by let-7 misexpression in zebrafish could be specifically inhibited by co-injection with a 2'-O-methyl oligonucleotide complementary to the let-7 miRNA, but not by a control 2'-O-methyl oligonucleotide unrelated in sequence to let-7 (Figure 3B) (35). Injection of the oligonucleotide alone did not cause any developmental abnormalities. Similarly, a morpholino directed against let-7 inhibited the misexpression phenotype, but did not induce a phenotype when injected alone (data not shown). The results presented here show that the let-7 miRNA is active in the zebrafish embryo and severely affects normal development upon misexpression. Although the biological significance of the observed effects remains unclear, it seems likely that ectopic expression of let-7 causes precocious downregulation of one or more endogenous let-7 targets. Misexpression studies can be important for defining miRNA function, especially in cases where knockout is difficult because of redundancy.
Figure 3. Effects of let-7 injection on gfp::lin-41 fusions and zebrafish development. (A) Structure of let-7 bound to its target sites in the zebrafish lin-41 3'-UTR. The position of the point mutation in the target and let-7 is indicated in red. At these positions the C in the target is changed to a T and the G in let-7 is changed to an A. (B) Lateral views of 28 h zebrafish embryos injected with gfp with and without the lin-41 3'-UTR. The upper panel shows phenotypes and the lower panel GFP fluorescence. (C) Western and northern analysis of GFP in 28 h zebrafish embryos co-injected with gfp:lin-41 3'-UTR. (D) Phenotypes and GFP expression in 28 h embryos injected with mutant (gfp::lin-41 3'-UTR-2, let-7mm5) and wild-type (gfp::lin-41 3'-UTR-1, let-7) let-7 and gfp::lin-41 3'-UTR fusions.
The antisense strand of the let-7 duplex is biologically active and an asymmetric duplex exhibits enhanced activity
We next investigated which strand of the let-7 duplex exerts a biological effect by examining the activity of heteroduplexes (Figure 2). A heteroduplex containing two mutations at the 5' end of the antisense strand of let-7 (heteroduplex 1) fails to induce a phenotype, while a heteroduplex with two mutations at the 3' end of the sense strand (heteroduplex 2) functions extremely well. This indicates that only the antisense strand of the let-7 duplex exerts a biological effect. The phenotype induced by heteroduplex 2 is even stronger than that induced by the let-7 homoduplex. In contrast, injection of a heteroduplex containing two mismatches at the 5' end of the sense strand (heteroduplex 3) resulted in a reduced phenotype. These data support recent work, which shows that both the absolute and the relative stabilities at the 5' ends of siRNA or miRNA duplexes are critical determinants for incorporation of one or the other strand into RISC (36–38). Our experiments demonstrate that the in vivo activity of a miRNA is subject to the same criteria.
Figure 2. Phenotypic effects of injection of 3 let-7 heteroduplexes. (A) Sequences of heteroduplexes. The upper strand is the antisense strand of the let-7 duplex. (B) Phenotypes of 28 h zebrafish embryos injected with heteroduplexes. (C) Bar diagram showing relative length of the yolk sac extension of fish injected with let-7 heteroduplexes.
Silencing of a gfp::lin-41 3'-UTR reporter by let-7
The let-7/lin-41 miRNA-target couple was experimentally verified in C.elegans (32), but conserved sequences are found in Drosophila and zebrafish (31). To show that injected let-7 acts as a miRNA on the zebrafish lin-41 3'-UTR, we fused a 379 bp fragment comprising part of the zebrafish lin-41 3'-UTR and containing two let-7 target sites (Figure 3A) to gfp. Co-injection of let-7 with mRNA derived from this construct resulted in the translational repression of GFP expression (Figure 3B). The gfp::lin41 3'-UTR mRNA levels remained unaffected, whereas GFP protein levels were dramatically reduced upon let-7 overexpression (Figure 3C). Silencing of gfp is dependent on the lin-41 3'-UTR, since gfp mRNA without the lin-41 3'-UTR was not silenced by let-7 (Figure 3B). Furthermore, silencing of gfp::lin-41 3'-UTR could be blocked by the 2'-O-methyl oligonucleotide complementary to let-7, and the mmlet-7 version did not silence gfp::lin-41 3'-UTR. We did not observe downregulation of the gfp::lin-41 3'-UTR reporter by endogenous let-7 as shown by a recent study using mouse embryos transiently expressing lacZ miRNA sensor constructs (39). This is likely due to the fact that there are no detectable amounts of let-7 expressed during these stages of development.
To investigate the interaction between let-7 and its target sites in more detail, gfp was fused to a 100 bp fragment of the lin-41 3'-UTR containing only the two let-7 target sites including intervening sequence (pgfp::lin-41 3'-UTR-1). In addition, a similar construct was made that contains the same fragment, but with a point mutation in both let-7 target sites (gfp lin-41 3'-UTR-2) (Figures 3A and 4A). mRNA derived from gfp lin-41 3'-UTR-1 could only be silenced by wild-type let-7, while expression of gfp::lin-41 3'-UTR-2 was inhibited exclusively by a let-7 duplex with a compensatory mutation (let-7mm5, Figure 3D). In addition, duplex let-7mm5 does not induce a phenotype in fish embryos. Taken together, our data show that injection of a mature let-7 miRNA duplex can specifically induce gene silencing via the lin-41 3'-UTR in the developing zebrafish embryo.
Figure 4. let-7 target site 1 and 2 are sufficient for silencing. (A) Overview of constructs and miRNAs analyzed in this study. + indicates strong silencing, – indicates no silencing (B) GFP fluorescence in 24 h zebrafish embryos injected with let-7 or let-7mm5 and gfp::let-7tar 5'-UTR or gfp::let-7tar cds.
We next asked whether the sequence intervening the let-7 target sites is an essential component for silencing. In C.elegans, it was shown that silencing of a lacZ reporter fused to the C.elegans lin-41 3'-UTR is dependent on two specific let-7 target sites and the 27 bp sequence between the target sites (33). In contrast, we found that a gfp reporter containing both let-7 target sites from the zebrafish lin-41 gene separated by a 5 bp sequence is still silenced by co-injected let-7 (Figure 4A, pgfp::let-7tar1+2), while in the natural situation both let-7 target sites are separated by a 57 bp sequence. Similar constructs containing either two copies of target site 1 or two copies of target site 2 fused to gfp were also silenced by let-7 (Figure 4A, pgfp::let-7tar1-2x and pgfp::let-7tar2-2x), showing that both target sites are equally potent in silencing. However, reporter constructs with a point mutation (Figure 3A) in either let-7 target site 1 or 2 could not be silenced by let-7, demonstrating that both target sites act together to mediate a silencing response (Figure 4A, pgfp::lin-41 3'-UTR-1 mut1 and pgfp::lin-41 3'-UTR-1 mut2). GFP expression from pgfp::lin-41 3'-UTR-1 mut1 and pgfp::lin-41 3'-UTR-1 mut2 cannot be downregulated by co-injection of a mixture of let-7 and let-7mm5. This is probably due to a competition effect (i.e. let-7 binding to the mutant target site and let-7mm5 binding to the wild-type target site), since a mixture of let-7 and mir221 could silence a reporter construct with let-7 target site 2 replaced by an artificially designed mir221 complementary site (Figure 4A, pgfp::let-7tar1+mir221tar), whereas let-7 and mir221 alone did not affect the GFP expression from this construct. These data are in agreement with previous studies in cultured mammalian cells (40), showing that miRNAs can act cooperatively to repress gene expression, although our data differ from these studies because we used a combination of two copies of one target site instead of two copies of two target sites acting together.
Taken together, we demonstrate that injected let-7 can act on the zebrafish homolog of the lin-41 gene, but we have no way of knowing whether endogenous let-7 silences lin-41 in vivo or whether the phenotype we observe is the result of injected let-7 acting on lin-41.
Silencing of reporters containing let-7 targets in the coding region or 5'-UTR
To investigate whether silencing of the GFP reporter is restricted to the presence of let-7 target sites specifically in the 3'-UTR, constructs were made with the let-7 target unit comprising of both target sites with a 5 bp spacer inserted in the 5'-UTR (gfp::let-7tar 5'-UTR) and 12 bp before the stop codon in the gfp coding sequence (gfp::let-7tar cds) (Figure 4A). Surprisingly, both constructs could be specifically silenced by co-injected let-7, but not by let-7mm5 (Figure 4B). These data indicate that GFP is specifically downregulated due to let-7 sites in the coding region or the 5'-UTR. In the latter case, it cannot be ruled out that translation initiation is inhibited due to steric hindrance of a let-7 loaded RISC complex just upstream of the start codon.
In plants, it is known that miRNA targets are also present in the coding sequence of the mRNA, but plant miRNAs often target the mRNA for degradation via single complementary sites (41). No natural examples are known yet of animal miRNAs regulating a target gene via complementary sites in the coding sequence or 5'-UTR, although it was shown that siRNA off-target effects can be mediated through translational repression of mRNAs due to imperfect base pairing of the siRNA with the coding sequence (42).
In our zebrafish system, we find that miRNA-target sites do not necessary need to be located in the 3'-UTR, since let-7 targets both in the 5'-UTR or in the coding sequence could induce a specific silencing response. This is important information for studies aimed at predicting miRNA targets, which are currently only focusing on 3'-UTR sequences. As also suggested by others (12,13,20), these should include 5'-UTR and coding sequences as well.
The 7 bases at the 5' end of let-7 are most important for its activity
To define the most important positions for functionality of the let-7 miRNA, we performed a mutational scan by introducing point mutations at each of the 22 positions of let-7. For all 66 (3 x 22) mutants, we determined phenotypic defects and the influence on gfp::lin-41 3'-UTR mRNA translation (Figure 5A and B) as these assays are good indicators for the activity of let-7. Similar data were obtained by both the analysis of GFP protein level and the examination of phenotypic defects: mutations induced in the region spanning positions 1–7 from the 5' end of the antisense strand of the let-7 duplex eliminate the function of the miRNA (Figure 5). However, small growth defects were observed for let-7 duplexes with mutations at position 1. Versions of let-7 with a mutation at position 8 showed a strong phenotype but only a low reduction in the GFP protein level. Remarkably, the change of an A to a G at positions 3 and 7, enabling the formation of a G-U instead of an A-U base pair with the lin-41 reporter construct, did not inhibit protein expression or affect normal development.
Figure 5. Mutational analysis of the let-7 miRNA. (A) Phenotypes were scored by measuring the length of the yolk sac extension relative to the diameter of the yolk (bar diagram). For each data point, five individuals were measured. Approximate GFP levels, as assessed from western blots (B), are indicated in the table. – indicates that GFP protein level is similar to a control where only the gfp::lin-41 3'-UTR mRNA was injected; + indicates a significant decline of GFP protein level compared with the control; 0 indicates a moderate decline of GFP protein level compared with the control. At each position, the three types of possible mutations are shown in the sequences beneath the bars. The first series of mutants (purple bars and sequence a) has an equal number of hydrogen bonds compared with the wild-type let-7 duplex. Both other series (dark red and gray bars and sequences b and c) have a different number of hydrogen bonds, depending on the nature of the mutation, i.e. U/A to C/G or vice versa. The let-7 sequence is depicted in red. (B) Western blot analysis of GFP protein level for 3 x 22 mutant derivatives of let-7. Protein was extracted from 5 embryos. For each sample, 5 mg of protein was loaded. As an injection control, the level of gfp::lin-41 3'-UTR mRNA was determined by northern blotting. For each sample, 3 μg of RNA, obtained from 10 individuals, was loaded. Numbering is from the 5' end of let-7 and corresponds to that in (A).
Most prominently at positions 10 and 13, GFP silencing and phenotypic effects were different for the 3 mutant let-7 derivatives. These differences probably reflect changes in the let-7 duplex free energy, which is relatively low for positions 9–14 in natural miRNA duplexes (36,37). For example, at position 10 the change of the original A-U base pair in the let-7 duplex into an U-A base pair still causes strong silencing. However, the change to a G-C or C-G base pair increases stability of the duplex and decreases its activity. Similar although less pronounced effects were observed for mutations at position 9. Thus, mutations at certain positions in the miRNA duplex can affect its activity, not because of an altered interaction with the target RNA, but because of a change in the stability of the miRNA duplex itself.
Recently, the specificity of miRNA-target interaction was analyzed in cell culture using luciferase reporter assays (40). In this study, the miRNA target instead of the miRNA itself was analyzed for critical positions and, similar to our observations, these reporter studies indicated that base pairing of the first 8 nt in the 5' region of the miRNA is most important for activity, although the binding of the 3' end of a miRNA also appeared to contribute to gene silencing. These observations were made with reporter constructs containing multiple miRNA-target sites similar to our gfp::lin-41 reporter. Investigation of rules governing the interaction of a miRNA with a single target site showed that interaction of the proximal (i.e. 5') part of miRNA with its target is indeed the most susceptible to mutations (43). Only a target:miRNA mismatch in the 5' part flanked by 4 bp on each side resulted in silencing of the reporter. Similarly, mutational analysis of the plant mir165/166 complementary site of PHABULOSA revealed that disrupting miRNA pairing near the 5' region causes stronger developmental consequences and reduced miRNA-directed cleavage in vitro (44).
For our phenotypic assay, it is unclear whether the affected target mRNA(s) that cause(s) the phenotype contain one or multiple let-7 targets. Our data add to the previous studies that irrespective of the targets, the major sequence determinants for in vivo function of let-7 in zebrafish lie in the 5' 1–7 residues.
This study represents the first complete mutational analysis of a miRNA in a vertebrate model organism. It is striking and unexplained that the let-7 miRNA is perfectly conserved throughout evolution, while only mutations in the first 7 residues affect its activity. It could be that the sensitivity of both assays described here is too low, since they make use of a rather artificial over expression setup. Another explanation might be that the mutational spectrum was derived by injection of mature let-7 miRNA sequences. The processing of the precursor into its mature form might also be affected by mutations in (the 3' part of) let-7, which was not addressed by our screen.
ACKNOWLEDGEMENTS
We thank B. Ason, R. May and M. Tijsterman for critical reading of the manuscript and G. Ro?l for help with Xenopus tropicalis injections.
REFERENCES
Lagos-Quintana,M., Rauhut,R., Yalcin,A., Meyer,J., Lendeckel,W. and Tuschl,T. ( (2002) ) Identification of tissue-specific microRNAs from mouse. Curr. Biol., , 12, , 735–739.
Lagos-Quintana,M., Rauhut,R., Meyer,J., Borkhardt,A. and Tuschl,T. ( (2003) ) New microRNAs from mouse and human. RNA, , 9, , 175–179.
Lagos-Quintana,M., Rauhut,R., Lendeckel,W. and Tuschl,T. ( (2001) ) Identification of novel genes coding for small expressed RNAs. Science, , 294, , 853–858.
Lau,N.C., Lim,L.P., Weinstein,E.G. and Bartel,D.P. ( (2001) ) An abundant class of tiny RNAs with probable regulatory roles in Caenorhabditis elegans. Science, , 294, , 858–862.
Lee,R.C. and Ambros,V. ( (2001) ) An extensive class of small RNAs in Caenorhabditis elegans. Science, , 294, , 862–864.
Lai,E.C., Tomancak,P., Williams,R.W. and Rubin,G.M. ( (2003) ) Computational identification of Drosophila microRNA genes. Genome Biol., , 4, , R42.
Lim,L.P., Glasner,M.E., Yekta,S., Burge,C.B. and Bartel,D.P. ( (2003) ) Vertebrate microRNA genes. Science, , 299, , 1540.
Lim,L.P., Lau,N.C., Weinstein,E.G., Abdelhakim,A., Yekta,S., Rhoades,M.W., Burge,C.B. and Bartel,D.P. ( (2003) ) The microRNAs of Caenorhabditis elegans. Genes Dev., , 17, , 991–1008.
Griffiths-Jones,S. ( (2004) ) The microRNA Registry. Nucleic Acids Res., , 32, (Database issue), D109–D111.
Pfeffer,S., Zavolan,M., Grasser,F.A., Chien,M., Russo,J.J., Ju,J., John,B., Enright,A.J., Marks,D., Sander,C. et al. ( (2004) ) Identification of virus-encoded microRNAs. Science, , 304, , 734–736.
Reinhart,B.J., Weinstein,E.G., Rhoades,M.W., Bartel,B. and Bartel,D.P. ( (2002) ) MicroRNAs in plants. Genes Dev., , 16, , 1616–1626.
Bartel,D.P. ( (2004) ) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell, , 116, , 281–297.
Ambros,V. ( (2004) ) The functions of animal microRNAs. Nature, , 431, , 350–355.
Lee,Y., Ahn,C., Han,J., Choi,H., Kim,J., Yim,J., Lee,J., Provost,P., Radmark,O., Kim,S. et al. ( (2003) ) The nuclear RNase III Drosha initiates microRNA processing. Nature, , 425, , 415–419.
Bernstein,E., Caudy,A.A., Hammond,S.M. and Hannon,G.J. ( (2001) ) Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, , 409, , 363–366.
Grishok,A., Pasquinelli,A.E., Conte,D., Li,N., Parrish,S., Ha,I., Baillie,D.L., Fire,A., Ruvkun,G. and Mello,C.C. ( (2001) ) Genes and mechanisms related to RNA interference regulate expression of the small temporal RNAs that control C.elegans developmental timing. Cell, , 106, , 23–34.
Hutvagner,G., McLachlan,J., Pasquinelli,A.E., Balint,E., Tuschl,T. and Zamore,P.D. ( (2001) ) A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA. Science, , 293, , 834–838.
Ketting,R.F., Fischer,S.E., Bernstein,E., Sijen,T., Hannon,G.J. and Plasterk,R.H. ( (2001) ) Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C.elegans. Genes Dev., , 15, , 2654–2659.
Enright,A.J., John,B., Gaul,U., Tuschl,T., Sander,C. and Marks,D.S. ( (2003) ) MicroRNA targets in Drosophila. Genome Biol., , 5, , R1.
Lai,E.C. ( (2004) ) Predicting and validating microRNA targets. Genome Biol., , 5, , 115.
Lewis,B.P., Shih,I.H., Jones-Rhoades,M.W., Bartel,D.P. and Burge,C.B. ( (2003) ) Prediction of mammalian microRNA targets. Cell, , 115, , 787–798.
Rhoades,M.W., Reinhart,B.J., Lim,L.P., Burge,C.B., Bartel,B. and Bartel,D.P. ( (2002) ) Prediction of plant microRNA targets. Cell, , 110, , 513–520.
Stark,A., Brennecke,J., Russell,R.B. and Cohen,S.M. ( (2003) ) Identification of Drosophila microRNA targets. PLoS Biol., , 1, , 1–13.
Brennecke,J., Hipfner,D.R., Stark,A., Russell,R.B. and Cohen,S.M. ( (2003) ) bantam encodes a developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell, , 113, , 25–36.
Xu,P., Vernooy,S.Y., Guo,M. and Hay,B.A. ( (2003) ) The Drosophila microRNA Mir-14 suppresses cell death and is required for normal fat metabolism. Curr. Biol., , 13, , 790–795.
Chen,C.Z., Li,L., Lodish,H.F. and Bartel,D.P. ( (2004) ) MicroRNAs modulate hematopoietic lineage differentiation. Science, , 303, , 83–86.
Johnston,R.J. and Hobert,O. ( (2003) ) A microRNA controlling left/right neuronal asymmetry in Caenorhabditis elegans. Nature, , 426, , 845–849.
Chang,S., Johnston,R.J.,Jr, Frokjaer-Jensen,C., Lockery,S. and Hobert,O. ( (2004) ) MicroRNAs act sequentially and asymmetrically to control chemosensory laterality in the nematode. Nature, , 430, , 785–789.
Lee,R.C., Feinbaum,R.L. and Ambros,V. ( (1993) ) The C.elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell, , 75, , 843–854.
Wightman,B., Ha,I. and Ruvkun,G. ( (1993) ) Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C.elegans. Cell, , 75, , 855–862.
Pasquinelli,A.E., Reinhart,B.J., Slack,F., Martindale,M.Q., Kuroda,M.I., Maller,B., Hayward,D.C., Ball,E.E., Degnan,B., Muller,P. et al. ( (2000) ) Conservation of the sequence and temporal expression of let-7 heterochronic regulatory RNA. Nature, , 408, , 86–89.
Reinhart,B.J., Slack,F.J., Basson,M., Pasquinelli,A.E., Bettinger,J.C., Rougvie,A.E., Horvitz,H.R. and Ruvkun,G. ( (2000) ) The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature, , 403, , 901–906.
Vella,M.C., Choi,E.Y., Lin,S.Y., Reinert,K. and Slack,F.J. ( (2004) ) The C.elegans microRNA let-7 binds to imperfect let-7 complementary sites from the lin-41 3'UTR. Genes Dev., , 18, , 132–137.
Wienholds,E., Koudijs,M.J., van Eeden,F.J., Cuppen,E. and Plasterk,R.H. ( (2003) ) The microRNA-producing enzyme Dicer1 is essential for zebrafish development. Nature Genet., , 35, , 217–218.
Hutvagner,G., Simard,M.J., Mello,C.C. and Zamore,P.D. ( (2004) ) Sequence-specific inhibition of small RNA function. PLoS Biol., , 2, , E98.
Schwarz,D.S., Hutvagner,G., Du,T., Xu,Z., Aronin,N. and Zamore,P.D. ( (2003) ) Asymmetry in the assembly of the RNAi enzyme complex. Cell, , 115, , 199–208.
Khvorova,A., Reynolds,A. and Jayasena,S.D. ( (2003) ) Functional siRNAs and miRNAs exhibit strand bias. Cell, , 115, , 209–216.
Tomari,Y., Matranga,C., Haley,B., Martinez,N. and Zamore,P.D. ( (2004) ) A protein sensor for siRNA asymmetry. Science, , in press.
Mansfield,J.H., Harfe,B.D., Nissen,R., Obenauer,J., Srineel,J., Chaudhuri,A., Farzan-Kashani,R., Zuker,M., Pasquinelli,A.E., Ruvkun,G. et al. ( (2004) ) MicroRNA-responsive ‘sensor’ transgenes uncover Hox-like and other developmentally regulated patterns of vertebrate microRNA expression. Nature Genet, ., 36, , 1079–1083.
Doench,J.G. and Sharp,P.A. ( (2004) ) Specificity of microRNA target selection in translational repression. Genes Dev., , 18, , 504–511.
Hake,S. ( (2003) ) MicroRNAs: a role in plant development. Curr. Biol., , 13, , R851–R852.
Saxena,S., Jonsson,Z.O. and Dutta,A. ( (2003) ) Small RNAs with imperfect match to endogenous mRNA repress translation. Implications for off-target activity of small inhibitory RNA in mammalian cells. J. Biol. Chem., , 278, , 44312–44319.
Kiriakidou,M., Nelson,P.T., Kouranov,A., Fitziev,P., Bouyioukos,C., Mourelatos,Z. and Hatzigeorgiou,A. ( (2004) ) A combined computational-experimental approach predicts human microRNA targets. Genes Dev., , 18, , 1165–1178.
Mallory,A.C., Reinhart,B.J., Jones-Rhoades,M.W., Tang,G., Zamore,P.D., Barton,M.K. and Bartel,D.P. ( (2004) ) MicroRNA control of PHABULOSA in leaf development: importance of pairing to the microRNA 5' region. EMBO J., , 23, , 3356–3364.(Wigard P. Kloosterman, Erno Wienholds, R)