|
Figure 1. Identification of EFTFs associated with the RISC in the eye field. (a) Construct used for rax1A‐myc‐AGO transgenesis. (b) Q‐RT‐PCR of AGO‐RNP‐associated RNA obtained from rax‐AGO transgenic embryos (red; n = 3 independent experiments) and embryos injected with myc‐Ago RNA injected into the dorsal animal blastomeres (blue; n = 4 independent experiments). Values are presented as fold abundance relative to coimmunoprecipitations from embryos that do not express myc‐AGO, to provide a control for nonspecific binding of RNA. Values represent mean ± S.E.M. AGO, Argonaute; EFTFs, eye field transcription factors; Q‐RT‐PCR, quantitative reverse‐transcriptase‐polymerase chain reaction
|
|
Figure 2. Expression of miR‐199a‐3p during early development. In situ hybridizations were carried out using an LNA probe to evaluate expression of mir‐199 within the eye field. (a) Vegetal view of miR‐199 expression in the circumblastoporal region at st. 11. (b–d) Dorsal, lateral, and frontal views showing expression of miR‐199 at st. 18. (e) Frontal view of miR‐199 expression at st. 23; arrows in (d) and (e) indicate expression in eye‐forming regions. (f, g – boxed) Frontal and dorsal views of st. 18 embryos that have been hybridized in situ with a control LNA probe. All scale bars approximately 250 μm. In situ hybridizations were carried out on embryos from three independent clutches. LNA, Locked nucleic acid
|
|
Figure 3. Overexpression of miR‐199 leads to defects in eye development. Unilateral (a–i) or bilateral (j) injections of miR‐199 were carried out to evaluate the effects of miR‐199 overexpression on eye phenotype and EFTF expression. (a–c) Embryo injected with miR‐199a‐3p in two left anterior dorsal blastomeres at the 16‐cell stage. Eye on injected side (a) is smaller than the eye on the uninjected side (b,c), and shows a coloboma. Injected sides are marked with asterisks. (d–h) In situ hybridizations of selected EFTFs (rax, optx2, tbx3, pax6), and otx2. With the exception of pax6 (h), all show restricted expression on the injected side. Scale bars approximately 200 μm. (i) Frequencies of small eye and coloboma phenotypes; n = 48 embryos from eight independent experiments. (j) Q‐RT‐PCR showing expression of EFTFs and otx2 in embryos following targeted overexpression of miR‐199 or MTT‐199. Embryos were injected at the 8‐cell stage as described and collected for RNA isolation and analysis at st. 18 (N = 3 independent experiments; *p < .05; unpaired Student's t test). EFTFs, eye field transcription factors; Q‐RT‐PCR, quantitative reverse‐transcriptase‐polymerase chain reaction
|
|
|
|
Figure 5. Validation of interactions between the rax1A 3′UTR and miR‐199a‐3p. Luciferase “sensor” constructs carrying the rax1a 3′UTR were used to evaluate the sensitivity of the rax 3′UTR to miR‐199 gain‐ and loss‐of‐function. (a) Luciferase assay of embryos injected with either the wild type (luc‐rax‐3′UTR, left) or mutant (luc‐rax‐MUT, right) luciferase constructs (b), in combination with either the wild type miR‐199 sequence (miR‐199) or a mutant sequence carrying four reversed bases within the seed‐binding sequence (MTT‐199). For both the wild type and the mutant 3′UTR construct, the results for the miR‐199 overexpression are normalized to those of the sample containing MTT‐199. (N = 5 independent experiments; *p < .05; unpaired Student's t test for all panels shown). (b) Comparison of the putative miR‐199 sequences for the luc‐rax‐3′UTR and the luc‐rax‐MUT constructs. The “seed sequence” is underlined. Four bases within the seed sequence are reversed in the MUT construct, and these changes are complementary to the mutations in MTT‐199. (c) Luciferase assay of embryos injected with the luc‐rax‐3′UTR construct in combination with a LNA oligonucleotide inhibitor of miR‐199 (LNA‐199) or a control LNA sequence (ctrl LNA). (N = 5 independent experiments; **p < .01). (D) Luciferase assay of embryos injected with the luc‐rax‐3′UTR construct and a TPMO complementary to the predicted miR‐199 binding sequence within the rax 3′UTR. (N = 4 independent experiments; *p < .05). (e) To evaluate the significance of interactions between miR‐199 and the rax 3′UTR, targeted injections were carried out with the either wild type or mutant forms of miR‐199 in the presence of either the TPMO or a 5‐base mispair MO (TPMM). Embryos were scored for the small eye phenotype. (N = 363 embryos across nine independent experiments). LNA, locked nucleic acid; TPMO, “target protector” morpholino oligonucleotide
|
|
Figure 6. miR‐199 knockdown leads to defects in eye development. (a) Phenotypes of embryos injected with the LNA oligonucleotide inhibitor of miR‐199 (LNA‐199). An embryo injected with a control LNA sequence (ctrl LNA) is shown in (i); (ii)–(v) show the range of phenotypes resulting from LNA‐mediated knockdown of miR‐199. Scale bar, 250 μm. The percentages showing each phenotype are indicated. (N = 76 embryos across seven independent experiments). (b and c) Neural tubes isolated at st. 24 from embryos injected unilaterally with either the control (b) or the miR‐199 (c) LNA oligonucleotide. The injected side is marked with an asterisk. Scale bar, 250 μm. (N = 4 independent experiments). (d) Quantitative RT‐PCR showing expression of EFTFs and otx2 in st. 18 embryos following targeted knockdown of miR‐199. Values shown represent the mean ± S.E.M. of the fold difference between embryos injected with LNA‐199 and the control LNA. (N = 3 independent experiments; *p > .05; **p > .01; unpaired Student's t test). (e–g) In situ hybridization to show expression of the EFTFs rax (e), optx2 (f), and pax6 (g) in embryos injected unilaterally with LNA‐199. The injected side is marked with an asterisk. Scale bars, 100 μm. EFTFs, eye field transcription factors; LNA, locked nucleic acid
|
|
Figure 7. Identification and testing of miR‐199 candidate target genes. (a) Model of miR‐199 function as a positive regulator of eye development, and criteria for identifying relevant candidates among predicted miR‐199 targets. (B) Q‐RT‐PCR to evaluate the effects of miR‐199 knockdown on expression of relevant candidate targets of miR‐199. LNA‐199 or the control LNA were introduced by targeted injection, injected embryos were cultured until st. 18, and collected for RNA isolation and Q‐RT‐PCR. Values represent the mean ± S.E.M of the fold difference relative to controls. N = 4 independent experiments, *p < .05; unpaired Student's t test. Q‐RT‐PCR, quantitative reverse‐transcriptase‐polymerase chain reaction; LNA, locked nucleic acid
|
|
Figure 8. Overexpression of ptk7 inhibits eye development. (a) Representation of predicted miR‐199 binding sites on ptk7 3′UTR. Red bar indicates strongest predicted target sequence. Blue arrows indicate the portion of 3′UTR cloned into the pmiRglo (PMG) plasmid for luciferase assays. (b) Embryos were prepared by bilateral targeted injections (50 pg plasmid +1.5–2 μg LNA) and collected at stage 24 for luciferase assay. (N = 4 independent experiments, *p < .05; unpaired Student's t test). Values represent mean ± S.E.M. (c) Representative images of St 35 tail bud embryos after ptk7 overexpression. Embryos were injected in both dorsal animal blastomeres at the 8‐cell stage with 1.5–2 μg/embryo of mRNA encoding ptk7 or β‐galactosidase. Panel (i) shows results of β‐galactosidase control targeted overexpression at St 35 showing a typical eye. Panels (ii–ix) show the range of phenotypes observed, from the normal phenotype in (ii) to the absence of eye formation observed in panels (viii–ix). Scale bar indicates 500 μm. (D) Quantitative representation of targeted ptk7 overexpression phenotypes from four independent experiments. Ptk7‐overexpressing embryos (n = 43) exhibit a normal eye phenotype in 22% of the population (panel ii) as compared to 91% in uninjected embryos (n = 142) and 72% in β‐galactosidase injected embryos (n = 57.) In the remainder of the embryos overexpressing ptk7, 46% exhibit a small eye (panels v–vii in c), 23% are characterized by a coloboma (panels iii–v in c), and 9% show a loss of eye formation (panels viii–ix in c). (e) The percentage of normal eyes in each population is significantly different between ptk7‐overexpressing embryos and β‐galactosidase‐overexpressing embryos. (**p ≥ .001; unpaired Student's t test). Significance was calculated in R based on the percentage of normal eyes for each population for all biological replicates using ANOVA in conjunction with Tukey's HSD analysis set at a 95% confidence level)
|
|
Figure 9. Roles for miR‐199 in eye development. The microRNA miR‐199 targets rax during neurula stages, contributing to the regulation of eye field specification and cell proliferation. It also targets ptk7, which negatively regulates morphogenesis of the optic cup. Purple shaded areas indicate regions affected by mir‐199 expression through either rax (left) or ptk7 (right)
|
|
|
|
Figure S2 Expression of nontargeted neural transcription factors following gain‐ or loss‐of‐function of miR‐199. Quantitative reverse‐transcriptase‐polymerase chain reaction of several neural transcription factors (geminin, gbx2.1, irx3, sox11, zic2, and foxd4l1.1) following overexpression (a) or knockdown (b) of miR199. These genes were selected because they (a) are transcription factors expressed in neural ectoderm (as opposed to the eye field); and (b) they do not possess high‐confidence target sites for miR‐199 in their 3′UTRs. Embryos were subjected to targeted injection of either miR‐199 (a) or LNA‐199 (b) with corresponding controls (MTT‐199 or control LNA) as described previously, and lysed at st. 18. No significant changes in gene expression were observed in the expression of these genes following either overexpression (a) or knockdown (b) of miR‐199. (a) n = 6 independent experiments; (B) n = 4 independent experiments. Results are presented as fold difference relative to controls; p > .05 for all genes tested.
|
|
Figure S3 Gain‐ or loss‐of‐function for miR‐199 yield morphologically distinct phenotypes. Sections through unilaterally injected embryos reveal differences between the eye phenotypes produced by targeted overexpression of miR‐199 (a), a control LNA (b), or LNA‐199 (c). Embryos were injected in two neighboring dorsal animal blastomeres at the 16‐cell stage; asterisk indicates the injected side. Although overexpression of miR‐199 usually leads to a reduction in eye size, the optic cup forms normally and is fully separate from the neural tube (a). Introduction of LNA‐199 disrupts optic cup morphogenesis with varying degrees of severity; here (c), the optic tissue is continuous with the lateral forebrain.
|