|
Fig. 1. Tadpoles depleted for Pax 2 or Pax8 display pronephric defects. (A) Morphology of stage 45 tadpoles injected at the 4-cell stage in all four blastomeres with MoPax8 (a), MoPax2 (b) or MoC (c). A large edema is developed in the antero-ventral region of the MoPax8 and MoPax2 injected tadpoles. (B) Whole-mount immunofluorescence with 3G8 and 4A6 antibodies. Embryos were injected with MoPax8 (d, h), MoPax2 (e, i, f, j) or MoC (g, k) at the 4-cell stage in the equatorial region of both left blastomeres. Embryos were cultured to the tadpole stage 40 and processed for immunostaining. The right uninjected side is used as control (h, i, j, k). Pax8 knockdown leads to the absence of 3G8 and 4A6 reactivity. In MoPax2 injected tadpoles, 3G8 staining is strongly reduced on the injected side compared to the control side; 4A6 signal is absent. (C) Double immunofluorescence on tadpole (stage 40) transverse sections with 4A6 (l–q, revealed in green) and laminin-1(l, o, revealed in red), collagen IV (m, p, revealed in red) or aPKC (n, q, revealed in red) antibodies. DAPI staining is shown in blue. Embryos were injected with MoPax2 on the left side. The right uninjected side is used as control. A tubular structure, lacking 4A6 staining, but surrounded by laminin-1 and collagen IV is present on the MoPax2 injected side. aPKC is also correctly localized at the apical side of the cells forming this tubule. Insets show higher magnification of the tubule (one focal plane). Scale bar in B: 500 μm, Scale bar in C: 40 μm.
|
|
Fig. 2. Pax 8 and Pax2 loss of function differentially affects tubule but not glomus marker genes expression at late tailbud stage. (A, B) Whole mount in situ hybridization for tubule marker genes. Embryos were injected with MoPax8 (a–d) or MoPax2 (e–m) at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until the tailbud stage 33–35, fixed and analyzed by in situ hybridization with specific probes for the indicated genes. For each embryo, injected (a–m) and uninjected (a׳–m׳) sides are shown. Pax8 depletion prevents expression of marker genes all along the entire tubule expression domain. In Pax2 morphants, gene expression in the proximal part of the tubule, including nephrostomes, is affected. Expression in the distal tubule is not modified by the MoPax2 except for the terminal differentiation marker clcnkb1 whose expression is completely inhibited. Insets show higher magnification of the anterior pronephric territory. (C) Whole mount in situ hybridization for the glomus marker genes wt1 and nephrin. Embryos were injected with a mixture of MoPax8 and MoPax2 in the equatorial region of both left blastomeres at the 4-cell stage and cultured until the tailbud stage 33. Injected sides are shown in (n, o), uninjected sides in (n׳, o׳). Insets show higher magnification of the anterior pronephric territory. (D) RT-qPCR analysis of wt1 and nephrin. Embryos were injected with the indicated Mo at the 4-cell stage in all four blastomeres. They were cultured until the late tailbud stage 33 and processed for RT-qPCR analysis. Average values from three independent experiments. Neither MoPax2 nor MoPax8 injected alone or together significantly affects wt1 and nephrin expression compared to the control (MoC).
|
|
Fig. 3. Pax8 but not Pax2 loss of function inhibits tubule marker genes expression and pronephric tubule anlage formation at the early tailbud stage. (A) Whole mount in situ hybridization for tubule marker genes. Embryos were injected with MoPax8 (a–e), MoPax2 (f–j) or MoC (k–o) at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until tailbud stage 25, fixed and analyzed by in situ hybridization with the indicated probes. For each embryo, injected (a–o) and uninjected (a׳, o׳) sides are shown. Pax8 but not Pax2 depletion strongly reduces expression of lhx1, hnf1b, pax2, hrt1, and evi1 in the pronephric area. (B) Immunofluorescence with an anti-laminin-1 antibody on transverse sections of stage 25 embryos injected with MoPax8 (p, q) or MoPax2 (r, s) on one side. For each embryo, injected (p–s) and uninjected (p׳–s׳) sides are shown. DAPI staining is shown in blue (q, q׳, s, s׳). On uninjected sides and MoPax2 injected side, the pronephric tubule anlage is visualized as a group of somatic cells radially organized, surrounded by laminin-1 immunostaining (p׳, q׳, r, s, r׳, s׳). No such cell arrangement is observed on the MoPax8 inejcted side. (p–q). Laminin-1 is only detected in the extracellular matrix between the epidermis and the somatic mesoderm, as well as between the splanchnic mesoderm and the endoderm. Scale bar: 140 μm.
|
|
Fig. 4. Inhibition of lhx1 expression in response to MoPax8 is rescued by expressing Pax8. Embryos were first injected at the 4-cell stage in both left blastomeres with MoPax8. They were split into two groups. Embryos from one group were injected again with RNA Pax8-GR. Dexamethasone was added to the culture medium at the end of gastrulation (stage 13). Phenotype was assessed by studying lhx1 expression by in situ hybridization at tailbud stage 25. Phenotypes were classified into three categories: strong inhibition (a), weak inhibition (b) and normal expression (c). Inhibition of pronephric lhx1 expression caused by MoPax8 was partially rescued by inducing Pax8 function just after completion of gastrulation. Histogram shows quantitative results from three independent experiments (d). n=number of analyzed embryos in total.
|
|
Fig. 5. Hnf1b but not osr2 and lhx1 expression is inhibited in the KF of Pax8 morphant at neurula stage. (A) In situ hybridization for lhx1, osr2 and hnf1b genes. Embryos were injected with MoPax8 at the 4-cell stage, in the equatorial region of both left blastomeres. At neurula stage 18, they were fixed, transversely bisected at the level of the pronephric area and analyzed by in situ hybridization with the indicated probes. For each embryo, the dotted line indicates the frontier between the injected side (on the left) and the uninjected side (on the right). Hnf1b expression is inhibited in the pronephric area (arrow) on the injected side. (B) RT-qPCR analysis of lhx1, osr2 and hnf1b gene expression in KF explants dissected at neurula stage 18 from MoC and MoPax8 injected embryos. Unlike lhx1 and osr2 expression which is not modified, hnf1b expression is downregulated in response to Pax8 depletion. Average values from four independent experiments. *P<0.05.
|
|
Fig. 6. Pax8 ectopic expression is sufficient to up-regulate hnf1b in ectoderm animal caps in absence of protein synthesis. (A and B) RT-qPCR analysis of hnf1b expression in blastula animal caps. (A) Embryos were injected with pax8 RNA in the animal pole of all four blastomeres at the 4-cell stage. Animal caps from injected and uninjected embryos were dissected at blastula stage 9, cultured for 3 h and processed for RT-qPCR. A significant increased of Hnf1b expression is observed in response to Pax8. Average values from five independent experiments. (B) Embryos were injected with pax8-GR RNA in the animal pole of all four blastomeres at the 4-cell stage. Animal caps from injected and uninjected embryos were dissected at blastula stage 9. They were pre-incubated with CHX for 30 min, or with 1×MBS alone. CHX pre-incubated explants were then cultured for 3 h in CHX alone or in CHX+DEX; the others were cultured without CHX for 3 h in the presence or in absence of DEX. CHX alone causes an approximately fivefold increase of hnf1b expression. DEX treatment still causes a threefold increase of hnf1b expression relatively to CHX alone. Average values from three independent experiments. *P<0.05.
|
|
Fig. 7. Expression of somitic, lateral plate, blood and endothelial marker genes in Pax8 morphants. (A) Expression of somitic (myoD and mlc) and lateral plate markers (foxf1). Embryos were injected with MoPax8 at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until tailbud stage 25 (a, a׳, b, b׳) or neurula stage 18 (c, c׳), fixed and analyzed by whole-mount in situ hybridization with the indicated probes. For each embryo, injected (a, b, c) and uninjected (a׳, b׳, c׳) sides are shown. Interference with Pax8 function does not modify the expression pattern of neither myoD, mlc and foxf1. (B) Expression of blood (scl and lmo2) and endothelial markers (msr and flk1). Embryos were injected with MoPax8 (d–g) or MoC (h–k) at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until tailbud stage 27 and processed for whole-mount in situ hybridization with the indicated probes. For each embryo, injected (d–j, h–k) and uninjected (d׳–g׳, h׳–k׳) sides are shown. Double in situs with pax8 and lmo2 probes are shown in whole mount (d, d׳) as well as in section performed at the level of the anterior (l) and posterior (m) part of the pronephros anlage and in the posterior region of the embryo (n). The dotted line delineates injected (left) and uninjected (right) halves. In Pax8 morphants, the expression domain of scl, lmo2, msr and flk1 is enlarged in the anterior region where the tubule anlage normally forms. Double in situ hybridization with pax8 and lmo2 probes shows a much stronger lmo2 signal dorsal to pax8 expression domain on the injected side at the level of the pronephros anlage (l, m). In A and B, pax8 is revealed in light blue while the other genes are revealed in dark blue. Scale bar: 200 μm, (C) Expression of lmo2 in Pax8 morphants in the absence of cell division. Embryos were injected with MoPax8 at the 4-cell stage, in the equatorial region of both left blastomeres. Half of them were cultivated in the presence of HUA from neurula stage 18 onwards. They were processed for whole-mount in situ hybridization at stage 27. The enlargement of the lmo2 expression domain in the anterior region in response to the loss of Pax8 (o) is maintained in HUA treated embryos (p).
|
|
Fig. 8. MoPax8 knockdown results in cell proliferation defect in the KF that is rescued by activation of the canonical wnt pathway. (A) Embryos were injected at the 4-cell stage in the equatorial region of both left blastomeres and cultured until stage 18. They were then injected with BrdU into the archenteron, cultured for an additional one hour and fixed for in situ hybridization and immunostaining. On transverse sections, we combined in situ hybridization for pax8 (a, a׳) with immunolocalization for BrdU (b, b׳) and Pax8 (d, d׳), and DAPI staining (c, c׳). MoPax8 injected side (a–d), uninjected side (a׳–d׳). The outlined area in a and a׳ corresponds to the KF determined by the expression of pax8 mRNA. The number of BrdU stained nuclei was counted in this region Scale bar: 80 μm. (B) Histogram showing the BrdU labeling index (LI) in MoC and MoPax8 injected embryos. For each embryo, uninjected side is compared with the injected side. The LI represents the proportion of BrdU-labeled cells over the total number of cells in the KF area. BrdU LI is significantly decreased by Pax8 depletion. (C) Histogram showing the BrdU LI in embryos co-injected with MoPax8 and TCF-VP16-GR. Dexamethasone was added (+DEX) or not at stage 15. Activation of TCF-VP16-GR rescues the effect of MoPax8 on the BrdU LI. n=number of analyzed embryos in total. For each embryo, 4 sections were analyzed. Average values from three independent experiments. **P<0.005, ***P<0.001.
|
|
Fig. 9. Pax8 regulates the expression of components of the wnt signaling pathway. (A) Expression of genes of the wnt signaling pathway analyzed by RT-qPCR in KF explants dissected at neurula stage 18 from uninjected or injected embryos with MoC or MoPax8. We considered the difference as significative (*) when the p Value between MoPax8 and MoC as well as between MoPax8 and uninjected was <0.05 when both β-actine and ODC are used as endogenous reporter. dvl1 is downregulated and sfrp3 is upregulated in response to Pax8 depletion. In contrast, the expression of the secreted wnt inhibitors crescent, dkk1, dkk3, sfrp1, sfrp2, sfrp5, sizzled, of the wnt ligands wnt4, wnt5A, wnt8, wnt11b, wnt6, wnt9a, wnt9b as well as dvl2 and dvl3 is not modified. Average values from five independent experiments. (B) In situ hybridization with a dvl1 probe. Embryos were injected with MoPax8 or MoC at the 4-cell stage, in the equatorial region of both left blastomeres. At neurula stage 18, they were fixed, transversely bisected at the level of the pronephric area and analyzed by in situ hybridization. For each embryo, the dotted line delineates injected (left) and uninjected halves (right). Dvl1 is expressed in the intermediate mesoderm and its expression is inhibited in response to MoPax8 (arrow). (C, D) RT-qPCR analysis of dvl1 (C) and sfrp3 (D) expression in blastula animal caps. Embryos were injected with the indicated RNA (pax8, pax8-VP16, pax8δO) in the animal pole of all four blastomeres at the 4-cell stage. Animal caps from injected and uninjected embryos were dissected at blastula stage 9, cultured for 3 h and processed for RT-qPCR. A significant increased of dvl1 expression is observed in response to Pax8-VP16. Sfrp3 is upregulated in response to Pax8δO. Average values from four independent experiments. *P<0.05.
|
|
Supplementary Material. Fig. S1 MoPax8 and MoPax2 block endogenous embryonic Pax8 and Pax2 expression, respectively. (A) Immunofluorescence with anti-Pax8 (a-b) and anti-Pax2 (c, d) antibodies performed on transverse sections of embryos injected with MoPax8.2 (a, a׳), MoC (b, b׳) or MoPax2 (c, d) in both left blastomeres at the 4-cell. The KF region of injected (a, b) and uninjected side (b, b׳) is shown at the late neurula stage 18. The dorsal part of a stage 25 embryo containing the neural tube (NT) and the medial part containing the pronephros (arrow) are shown in c and d, respectively. The dotted line delineates injected (left) and the uninjected (right) halves. Pax8 is detected both in the splanchnic and somatic layers of the KF on the uninjected side, while it is absent on the MoPax8 injected side (a, a׳). MoC has no effect on the endogenous Pax8 expression (b׳). Pax2 expression in the neural tube and in the pronephros is inhibited on the MoPax2 injected side (c, d). Scale bar in a and b: 100 μm, scale bar in c and d: 400 μm (B) location of the Mo target sequences used in this study on pax2 and pax8 pseudoalleles. Nucleotides that differ between the two pseudoalleles are in blue. The start codon is in red.
|
|
Supplementary Material. Fig. S2 Pronephric phenotype induced by injection of MoPax2.2. Whole mount in situ hybridization for tubule and glomus marker genes. Embryos were injected with MoPax2.2 at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until the tailbud stages 33–35 (A) or 25 (B) and analyzed by in situ hybridization with specific probes for the indicated genes. For each embryo, injected (a–q) and uninjected (a׳–q׳) sides are shown. Pax2 depletion does not lead to any change in the expression of pronephric marker genes at stage 25. At the late tailbud stage, gene expression in the proximal part of the tubule, including nephrostomes, is affected by MoPax2.2 while expression in the distal tubule is not modified except for the terminal differentiation marker clcnkb1 whose expression is completely inhibited. Expression of the glomus marker genes wt1 and nephrin is not significantly affected in MoPax2.2 injected embryos. Insets show higher magnification of the anterior pronephric territory. Numbers indicate how many embryos showed the phenotype among the total number of injected embryos.
|
|
Supplementary Material. Fig. S3 Pronephric phenotype induced by co-injection of MoPax8 and MoPax8.B. Whole mount in situ hybridization for tubule and glomus marker genes. Embryos were injected with MoPax8.B alone or in co-injection with MoPax8 at the 4-cell stage, in the equatorial region of both left blastomeres. They were cultured until the tailbud stages 33–35 (A), early tailbud stage 23–25 (B) and analyzed by in situ hybridization with specific probes for the indicated genes. For each embryo, injected (a–v) and uninjected (a׳–v׳) sides are shown. A strong reduction of the tubule marker genes (pax2, odf3, smp30, lhx1, hnf1b, pax2, hrt1, and evi1) expression is observed in response to the injection while the glomus markers wt1 and nephrin are still expressed. Insets show higher magnification of the anterior pronephric territory.
|
|
Supplementary Material. Fig. S4Hnf1bbut notosr2andlhx1expression is inhibited in the neurula KF in response to the co-injection of MoPax8 and MoPax8.B. In situ hybridization for lhx1, osr2 and hnf1b genes. Embryos were injected with MoPax8.B alone or in co-injection with MoPax8 at the 4-cell stage, in the equatorial region of both left blastomeres. At neurula stage 18, they were fixed, transversely bisected at the level of the pronephric area and analyzed by in situ hybridization with the indicated probes. For each embryo, the dotted line indicates the frontier between the injected side (on the left) and the uninjected side (on the right). Hnf1b expression is inhibited in the pronephric area (arrow) on the injected side.
|
|
Supplementary Material. Fig. S5 Characterization of the kidney field explants. RT-qPCR analysis of lhx1, pax8, myf5, sox17beta and cytokeratin expression in KF explants and whole embryos. KF explants were dissected at neurula stage 18 from uninjected or injected embryos with MoC or MoPax8. The mRNAs renal markers Pax8 and lhx1 were enriched in the explants in contrast to the endodermal sox17beta and the somitic myf5 mRNAs. Explants including the fidney field and the overlying ectoderm, the epidermal gene cytokeratin is also expressed. Average values from four independent experiments. âŽP<0.05, âŽâŽP<0.005, âŽâŽâŽP<0.001.
|
