|
Figure 1.
Isolation of XSox17β
(A) Functional screen to identify genes regulating axis formation in Xenopus. Synthetic mRNA was prepared from pools of a dorsal lip cDNA library, and 2–5 ng was injected into 2–4 cell stage Xenopus embryos. Embryos were assayed for changes in morphology at the tail bud stage. A pool with ventralizing activity was serially subdivided and retested until a single active sequence was isolated. (B) Uninjected control embryos. (C) Dorsal injection of the active RNA (1 ng) resulted in ventralized embryos, while (D) ventral injection (1 ng) of the same RNA resulted in relatively normal development. Histological section of an uninjected control embryo (E) shows typical dorsal structures such as neural tube (nt), notochord (n), and somites (s), which are absent in the ventralized embryos (F). (G) The active clone encodes the HMG box protein, XSox17β.
|
|
Figure 2.
XSox17β Can Inhibit Wnt Signaling Downstream of β-catenin
(A) The left panels show representative tail bud stage embryos with a second axis induced by the injection of one ventral blastomere at the 4 cell stage with mRNA encoding dnBMPR (250 pg), bVg1 (100 pg), Xwnt8 (5 pg). Panels on the right show representative embryos similarly injected with the addition of XSox17β (500 pg). XSox17β only inhibits the second axis induced by Xwnt8 (lower right).
(B) Summary of second axis assays from three to five separate experiments. Injection of mRNA encoding dnBMPR (250 pg), bVg1 (100 pg), Xwnt8 (5 pg), β-catenin (250 pg), or Siamois (10 pg), resulting in a high frequency of embryos with second axes (gray bars). Embryos similarly injected with the same RNAs + XSox17β mRNA (500 pg) were scored for percent second axis (black bars).
(C) Ventralization assay. RNA encoding XSox17β (1 ng) or XSox17β (1 ng) + Xwnt8 (10 pg); XSox17β (1 ng) + β-catenin (250 pg); XSox17β (1 ng) + XTcf-3 (2 ng); XSox17β (1 ng) + Siamois (20 pg) was injected into both dorsal blastomeres at the 4 cell stage (black bars). Control injections (gray bars) without XSox17β were Xwnt8 (10 pg), β-catenin (250 pg), XTcf3 (2 ng), ΔN-XTcf3 (500 pg), ΔN-XTcf3 (500 pg) + XTcf3 (2 ng), and Siamois (20 pg). The dorsal development of the resulting embryos were scored by a dorsal anterior index (DAI; Kao and Elinson 1988). A score of 5 indicates normal development, and a score of 0 indicates severely ventralized embryos. Only Siamois rescued the ventralized phenotype caused by dorsal XSox17β injection.
(D) XSox17β operates downstream of β-catenin but upstream from Siamois.
|
|
Figure 3.
XSox17β Represses Wnt/β-catenin Stimulated Transcription and Activates Endodermal Genes
(A and B) Human 293T cells were cotransfected with OT reporter construct containing TCF consensus binding sites (Korinek et al. 1997) and expression plasmids encoding (A) S37A-β-catenin (0.2 μg), S37A-β-catenin (0.2 μg) + XSox17β (2 μg and 4 μg) and (B) Wnt1 (2 μg), Wnt1 (2 μg) + XSox17β (2 μg and 4 μg). XSox17β inhibits β-catenin and Wnt1 activation of the OT reporter.
(C) Schematic of XSox17β:GAL4 DNA-binding domain fusion constructs and the 5x GAL DNA binding site–luciferase construct.
(D) Human 293T cells cotransfected with 5x GAL reporter (0.5 μg) and 1 μg of the indicated fusion construct, with (gray bars) or without (black bars) 1 μg of S37A-β-catenin. The C-terminal region of XSox17β has transactivation capacity. Transfection experiments were repeated two to four times, and a representative experiment is shown.
(E) Embryos were injected with mRNA encoding XSox17β (125 pg, 250 pg, 500 pg, 1000 pg), pt-β-catenin (30 pg), and pt-β-catenin (30 pg), + XSox17β (125 pg, 250 pg, 500 pg, 1000 pg). Animal caps dissected at blastula stage were cultured until gastrula stage and (F) analyzed by RT-PCR for expression of the Wnt/β-catenin responsive genes Siamois and Xnr3, the endodermal gene Endodermin, and Ef-1α as a loading control.
|
|
Figure 4.
XSox17β Does Not Bind to a TCF/LEF Consensus DNA Binding Site
(A) Electrophoretic mobility shift assays (EMSAs) performed using in vitro translated TCF4-V5, XSox17β-V5, LEF1-V5, and XTCF3-HA epitopeâtagged proteins, without and with anti-epitope tag antibody and TCF DNA consensus sequence probe.
(B) EMSAs performed with TCF4-V5, XSox17β-V5, and LEF1-V5, without and with anti-epitope tag antibody and a murine Sox17 DNA consensus sequence probe.
|
|
Figure 5.
The C Terminus of XSox17β Is Essential for Repressing Wnt Signaling
(A) Schematic of the XSox17β deletion mutants.
(B) Secondary axis assay. mRNA encoding mutant stabilized pt-β-catenin (Yost et al. 1996) (0.1 ng), XSox17β (1 ng), or pt-β-catenin (0.1 ng) + each of the dXSox17β mutants (1 ng) was injected into one ventral blastomere at the 4 cell stage. Tailbud stage embryos were scored for the presence of second axes, and the compiled results of four separate injection experiments are shown.
(C) OT transcription assay. 293T cells were cotransfected with the OT reporter construct (1 μg) and expression plasmids encoding S37A-β-catenin (0.3 μg), wtXSox17β-V5 (2 μg and 4 μg), S37A-β-catenin (0.3 μg) + each dXSox17β-V5 deletion construct (2 μg and 4 μg; black triangles indicate increasing dose). Experiments were repeated two to four times, and a representative experiment is shown. The C-terminal region of XSox17β is essential for repression of β-catenin transactivation.
|
|
Figure 6.
The C Terminus of XSox17β Binds to β-catenin
(A) XSox17β binds β-catenin in vitro. 35S-labeled XTcf-3, ΔNXTcf-3, and XSox17β proteins were incubated with 1 μg of GST-β-catenin fusion protein (β-cat) or 1 μg of GST alone, immobilized to agarose beads. Ten percent of the input proteins (i) and bound proteins were resolved on SDS-PAGE and visualized by fluorography.
(B) β-catenin binds to the C terminus of XSox17β. The indicated 35S-labeled XSox17β deletion proteins were incubated with GST-β-catenin agarose. Ten percent of the input proteins (i) and the GST-β-catenin bound proteins (b) were resolved on SDS-PAGE and visualized by fluorography. A representative binding experiment is shown. Relative binding of each mutant, summarized from three to four experiments is shown; +++, 60%–100% of wild-type binding; ++, 30%–60%; +, 10%–30%; ±, ∼10%; and −, <10%.
(C) XSox17β associates with β-catenin in vivo. 293T cells were transfected with (+) or without (−) S37A-β-catenin-HA (2 μg) and expression plasmids (8 μg) encoding mLEF1-V5, hTCF4-V5, ΔN-hTCF4-V5, XSox17β-V5, or dSox17β 1–150. The top panel shows an anti-V5 Western of the resulting input lysates. Cell lysates were subjected to immunoprecipitation with anti-HA antibody, followed by anti-HA Western to show levels of β-catenin-HA (middle panel) and an anti-V5 epitope Western (bottom panel) to visualize proteins coprecipitating with β-catenin-HA.
|
|
Figure 7.
Functional and Structural Analysis of the Interaction between XSox17β and β-catenin
(A) Second axis assay. Embryos were injected with the indicated doses of XSox17β and pt-β-catenin RNA, and the frequency of secondary axis was scored.
(B) OT reporter assay. Human 293T cells were cotransfected with OT reporter construct containing TCF consensus binding sites (Korinek et al. 1997) and the indicated amounts of expression plasmids encoding S37A-β-catenin and XSox17β.
(C and D) The indicated 35S-labeled Xβ-catenin deletion proteins were incubated with nickel-agarose resin or nickel-resin coupled with 1 μg of His-XSox17β protein (C). The relative binding of each deletion protein to His-XSox17β (based on three separate experiments) is indicated (D), and a representative experiment is shown. The minimal XSox17β binding site overlaps with the established TCF binding site.
|
|
Figure 8.
XSox17α and XSox3 Bind to β-catenin and Interfere with Its Signaling
(A) Secondary axis assay. mRNA encoding mutant stabilized pt-β-catenin (Yost et al. 1996) (0.1 ng), or pt-β-catenin (0.1 ng) + XSox17β, XSox17α, and XSox3 (0.5 ng) was injected into one ventral blastomere at the 4 cell stage. Tail bud stage embryos were scored for second axes.
(B) OT transcription assay. 293T cells were cotransfected with the OT reporter construct (1 μg) and expression plasmids encoding S37A-β-catenin (0.3 μg), XSox17β-V5 (4 μg), XSox17α-V5 (4 μg), XSox3-V5 (4 μg), or S37A-β-catenin (0.3 μg) + each XSox17β-V5, XSox17α-v5, XSox3-V5 (2 μg and 4 μg; black triangles indicate increasing dose).
(C) XSox17α and XSox3 bind β-catenin in vitro. 35S-labeled XSox17β, dXSox17β 1–150, XSox17α, and XSox3 proteins were incubated with 1 μg of GST-β-catenin fusion protein immobilized to agarose beads. Ten percent of the input proteins (i) and bound proteins (b).
(D) XSox17α and XSox3 associate with β-catenin in vivo. 293T cells were cotransfected with S37A-β-catenin (2 μg) and expression plasmids (8 μg) encoding ΔN-hTCF4-V5 (lane 1), mLEF1-V5 (lane 2), hTCF4-V5 (lane 3), XSox17β-V5 (lane 4), XSox3-V5 (lane 5), or XSox17α-V5 (lane 6). The top panel shows an anti-V5 Western of the resulting extracts. Cell lysates were divided and subjected to immunoprecipitation with anti-β-catenin (+) or a negative control anti-mER antibody (−), followed by an anti-β-catenin Western to show levels of β-catenin (middle panel) and an anti-V5 epitope Western (bottom panel) to visualize proteins coprecipitating with β-catenin.
|
