Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
Dev Cell
2013 Aug 12;263:237-49. doi: 10.1016/j.devcel.2013.06.023.
Show Gene links
Show Anatomy links
MiR-142-3p controls the specification of definitive hemangioblasts during ontogeny.
Nimmo R
,
Ciau-Uitz A
,
Ruiz-Herguido C
,
Soneji S
,
Bigas A
,
Patient R
,
Enver T
.
???displayArticle.abstract???
Hematopoietic stem cells (HSCs) emerge during embryogenesis from hemogenic endothelium, but it remains unclear how the HSC lineage is initially established from mesoderm during ontogeny. In Xenopus, the definitive hemangioblast precursors of the HSC lineage have been identified in dorsal lateral plate (DLP) mesoderm, and a transcriptional gene regulatory network (GRN) controlling hemangioblast programming has been elucidated. Herein, we identify an essential role for microRNAs (miRNAs) in establishing the mesodermal lineage leading to both HSC emergence and vasculogenesis and determine that a single miRNA, miR-142-3p, is primarily responsible for initiation of definitive hemangioblast specification. miR-142-3p forms a double-negative gate unlocking entry into the hemangioblast program, in part by inhibiting TGFβ signaling. Our results table miR-142-3p as a master regulator of HSC lineage specification, sitting at the apex of the hierarchy programming the adult hemangioblast, thus illustrating that miRNAs can act as instructive determinants of cell fate during development.
Figure 1. Identification of a Role for miRNAs in Specification of the HSC Lineage
(A) Schematic diagram showing the stepwise differentiation of the blood stem cell lineage in Xenopus, from DLP hemangioblast precusors through to HSCs via hemogenic endothelium. The DLP hemangioblasts give rise to the DA, including the hemogenic endothelium in the ventral wall from which stem cells emerge as part of hematopoietic clusters. The hematopoietic clusters in the DA at stages 42–44 of the Xenopus embryo are analogous to the clusters found in the E11.5 AGM of the mouse.
(B) miRNAs are required for HSC lineage specification. dgcr8 morphants do not produce hemogenic endothelium because of failure to correctly specify hemangioblasts in the DLP. Top: WISH showing Runx1 expression in the DA (arrowheads) at stage 39 is dependent on Dgcr8. Bottom: WISH showing loss of Flk1 expression in the DLP (arrows) of dgcr8 MOs at stage 24. Numbers of embryos represented by each panel, out of the number analyzed, are indicated in the top right corner. The anterior of the embryo is to the left and dorsal to the top in all images.
(C) Heatmap showing hierarchical clustering of miRNA expression profile of sorted cell populations from E11.5 mouse AGM. E: Ter119 CD31+ c-KitCD45 endothelial cells (including hemogenic endothelium), presumptive pre-HSCs: Ter119 CD31+ c-Kit+ CD45HSC precursors, and HSPC: Ter119 CD31+ c-Kit+ CD45+ hematopoietic stem and progenitors. Red indicates high expression and green indicates low expression as represented by the Z score. The ‘‘HSPC-miRs’’ cluster is expanded to show miRNA names. All miRNA qRT-PCR array data are available in Table S1. See Figure S1D for gating strategy of the sort and Figure S1E for schematic of AGM populations.
(D) Ranking of miRNAs by scoring peaks within 5 kb up- and downstream of the genomic location of the pre-miRNA in ChIP-seq data for ten hematopoietic TFs from hematopietic progenitor cells (Wilson et al., 2010). Graph showing rankings for the ‘‘HSPC-miRs’’ cluster; only miRNAs with a score greater than zero are shown.
(E and F) Relative expression of the 50 and 30 strands of mir-142, miR-142-5p, and miR-142-3p. (E) qRT-PCR expression analysis on sorted mouse BM adult HSPC (Flt3-KLS) and mouse AGM E, pre-HSC, and HSPC populations. (F) qRT-PCR expression analysis on dissected stage 26 DLP and, for comparison, stage 14 whole embryos. DCt values relative to U6 are shown on a log scale. (Error bars represent mean ± SD.)
See also Figure S1 and Table S1.
Figure 2. A Combination of Genetic Approaches to Investigate miRNA Function in Xenopus Embryos
Schematic diagram summarizing the different approaches used to show that miR-142-3p is necessary and sufficient for hemangioblast specification during establishment of the HSC lineage. A combination of loss-of-function (LOF), gain-of-function (GOF), and rescue experiments were used to investigate whether miR-142-3p controls definitive hemangioblast specification.
Figure 3. miR-142-3p Is Required for Definitive Hemangioblast Specification and Is Sufficient to Rescue Hemangioblast Programming in dgcr8 Morphants
(A–C) LOF experiment: miR-142-3p is required for formation of hemogenic endothelium and for hemangioblast specification in the DLP. Embryos injected with 100 ng miR-142-3p MO (A), but not a mismatch control MO (B), phenocopy hematopoietic defects of Dgcr8 MO embryos. Top: WISH showing runx1 expression in hemogenic endothelium (arrowheads) at stage 39. Bottom: WISH showing flk1 expression in DLP hemangioblasts (red arrows) at stage 24. Endothelial cells in VBI are indicated by black arrows. (C) WISH showing flk1 expression is normal in the DLP (arrows) at stage 24 in miR-142-5p (passenger strand) morphants.
(D) Rescue experiment: Coinjection of the miR-142-3p mimic is sufficient to restore hemangioblast formation in dgcr8 morphants. Expression of flk1 (top) and scl (bottom) in the DLP (arrows) analyzed by WISH at stages 24–25.
(E) GOF experiment: Enhanced hematopoietic programming by overexpression of miR-142-3p. Left: WISH of runx1 at stage 39 shows that the hemogenic endothelium (arrowheads) is expanded posteriorly out of the trunk, into the postcloacal tail region of the DA (arrows) in the miR-142-3p mimic-injected embryos. Dashed line indicates the location of the cloaca. Right: WISH analysis of flk1 expression in the DLP hemangioblasts (arrows) at stages 24–25. The number of embryos represented by each panel, out of the number analyzed, is indicated in the top right corner. The anterior of the embryo is to the left and dorsal to the top in all images. See also Figure S2.
Figure 4.
miR-142-3pActsattheTopoftheRegulatoryHierarchyControllingSpecificationofHSCandEndothelialCellPrecursorsintheDLP
(A) A provisional GRN controlling hemangioblast specification (Ciau-Uitz et al., 2010, 2013; Liu et al., 2008; Walmsley et al., 2002).
(B and C) Initiation of the hemangioblast program is defective in miR-142-3p MO embryos. (B) WISH showing that expression of early regulators, fli1, flk1, and etv2, is already highly reduced or absent in the DLP (arrows) in miR-142-3p morphants at stage 22 (initiation phase). (C) WISH analysis showing that the expression of hematopoietic gata2 and scl) and endothelial (fli1 and etv2) TFs in the DLP (arrows), are all dependent on miR-142-3p at stages 25–26. Boxed region of the fli1 panels is magnified to show that the expression of fli1 in the pronephric duct (arrowheads) abutting the DLP is unaffected in the morpants, whereas the expression of fli1 in the DLP (white dashed outline and arrows) is abrogated by the miR-142-3p MO. Formation of endothelial cells in the VBI is also disrupted by the MO as shown by reduced etv2 staining (arrowhead).
(D and E) Hemangioblast specification is normal in embryos injected with the control mismatch MO and miR-142-5p MO showing that the effect is specific to the miR-142-3p MO. (D) WISH analysis of etv2 and scl expression in the DLP (arrows) of control mismatch morphants and (E) gata2 and etv2 expression in the DLP (arrows) of miR-142-5p morphants.
(F) Formation of primitive blood is defective in miR-142-3p morphants. Top: Erythroid differentiation is absent from the pVBI region as shown by WISH analysis of aT4-globin expression. Bottom: reduction in formation of myeloid cells as shown by WISH analysis of elas1 expression.
The number of embryos represented by each panel, out of the number analyzed, is indicated in the top right corner. The anterior of the embryo is to the left and dorsal to the top in all images.
See also Figure S3.
Figure 5. Vasculogenesis Defects in miR-142-3p Morphants
(A) Hemogenic endothelium is not formed in miR-142-3p morphants. In situ hybridization analysis of gfi1a and spib expression in the midline (arrowheads) at stage 39. (B) Endothelial differentiation and migration defects in miR-142-3p morphants. WISH analysis of flt4, flk1, and flt1 expression in DLP-derived angioblasts in the PCV (arrows) and in cells migrating to the midline to form the DA (arrowheads) at stage 30.
(C) Defective expression of endothelial markers in miR-142-3p morphants. WISH analysis of AA4.1, CD31, dll4, and cx37 in the DA (arrowhead) and PCV (arrows) at stage 34.
(D) A residual DA is formed in miR-142-3p morphants (as shown by WISH analysis of CD31, arrowhead) but fails to switch on the hematopoietic TFs gata2 at stage 34 or runx1 at stage 36, as shown by WISH. Position of the DA is indicated by arrowheads.
The number of embryos represented by each panel, out of the number analyzed, is indicated in the top right corner. The anterior of the embryo is to the left and dorsal to the top in all images.
See also Figure S4.
Figure 6. miR-142-3p Restricts Inhibitory TGFb Signaling and Forms a Double-Negative Gate, Thereby Controlling the Initiation of the Hemangioblast Program
(A) Alignments of the target sites in tgfbr1, acvr2a, and acvr1b (blue) with miR-142-3p miRNA (red).
(B) WISH analysis at stages 25–26 showing that tgfbr1 expression is elevated in the DLP region (arrows) in miR-142-3p morphants.
(C) Schematic diagram summarizing the rescue experiments performed with the TGFb type I receptor kinase inhibitors, SB431542 and LY2157299, in miR-142-3p morphants.
Figure S1. Hemogenic endothelium is not formed in dgcr8 morphants due to a failure to specify hemangioblastic precurors in the DLP, related to Figure 1.
Figure S2. miR-142-3p mimic is sufficient to rescue definitive hemangioblast specification but fails to restore the formation of the DA and hemogenic endothelium in dgcr8 morphants, related to Figure 3.
Figure S3. miR-142-3p specifically promotes hemangioblast programming and does not have a more widespread role in mesodermal induction or patterning, related to Figure 4.
Figure S4. Control mismatch MO has no effect on HSC specification or endothelial differentiation, related to Figure 5.
