QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 1 October 2003
doi: 10.1242/dev.00798

This Article Summary Full Text Full Text (PDF) Supplemental Data Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, C. Articles by Klymkowsky, M. W. Search for Related Content PubMed PubMed Citation Articles by Zhang, C. Articles by Klymkowsky, M. W. Social Bookmarking                         What's this?

The ß-catenin/VegT-regulated early zygotic gene Xnr5 is a direct target of SOX3 regulation

Chi Zhang^*, Tamara Basta^*, Eric D. Jensen and M. W. Klymkowsky

Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, Colorado 80309-0347, USA

View larger version (107K): [in a new window]   Fig. 1. Antibody specificity. (A) An autoradiogram of SDS-PAGE-separated TnT-synthesized XTCF4, XLEF and XTCF3 polypeptides. (B) A blot of a parallel gel probed with the anti-XTCF3n antibody; the antibody reacts strongly with XTCF3 and more weakly with XTCF4. (C) A blot of a parallel gel probed with the anti-XTCF3c antibody; the antibody reacts with XTCF3 but not XTCF4. Neither anti-XTCF3 antibody reacts with XLEF. (D-F) Embryonic lysates, prepared from unfertilized eggs (egg), two-cell (2C), 32-cell (32c), 64-cell (64c), stage 7 (st7), stage 8 (st8), gastrula/stage 10 (st10) and neurula/stage 13 (s13) embryos were separated by SDS-PAGE, blotted and probed using the anti-TCF3n (D), anti-TCF3c (E) or anti-XSOX3c (F) antibodies. Each lane is loaded with one embryo equivalent. Both anti-XTCF antibodies reacted with a 67 kDa polypeptide that we presume to be XTCF3 (D,E, arrowhead). The anti-XTCF3n antibody reacted weakly with a 54 kDa polypeptide (D, **) that might be maternal XTCF4 polypeptide. The XTCF3c antibody reacted with distinct sets of lower molecular weight bands (E, *). Anti-XSOX3c (F) reacts strongly with a 35 kDa polypeptide (arrowhead) throughout early development. The level of this polypeptide decreases in neural stage embryos. In the unfertilized egg (egg) anti-XSOX3c reacts with slower migrating bands (arrows). Occasionally, such slower migrating, anti-XSOX3c-reactive polypeptides are seen in embryonic lysates (arrow). (G) Anti-XTCF3n (Tn) and anti-XTCF3c (Tc) antibodies immunoprecipitated (IP) a common anti-XTCF3n-reactive 67 kDa polypeptide (arrowhead) present in embryonic lysates (Lys, arrowhead). This polypeptide was not immunoprecipitated by the anti-XSOX3c antibody (Sx). (H) Anti-XSOX3c antibody precipitates a single 35 kDa, anti-XSOX3c-reactive polypeptide (arrow) that migrates with the 35 kDa polypeptide seen in lysates (Lys, arrowhead). No such band is seen in anti-XTCF3n (TCF3n) or anti-XTCF3c (data not shown) immunoprecipitates. The asterisk (*) marks the heavy and light chains of the precipitating antibody.

View larger version (113K): [in a new window]   Fig. 2. Immunocytochemical analysis of XSOX3. (A) In situ hybridization of fertilized eggs with an antisense probe for XSOX3 RNA reveals that XSOX3 mRNA is localized primarily to the fertilized egg's animal hemisphere (`an' and `vg' mark the animal and vegetal hemispheres, respectively, in all parts). (B) Whole-mount immunocytochemistry of a 64-cell embryo with the anti-XSOX3c antibody reveals a strong cytoplasmic reaction with the animal hemisphere; preincubation of the antibody with the peptide conjugate against which it was raised completely abolished staining (ab). (C) The nuclear nature of the staining becomes more pronounced as development proceeds but can be clearly seen in early stage embryos (128-cell stage). The protein is still primarily localized to the animal hemisphere but nuclei in vegetal blastomeres (arrows) clearly contain the immunoreactive polypeptide. Staining of early stage embryos with either anti-XTCF3n (D) or anti-XTCF3c (E) produced a pattern of staining similar to that seen for anti-XSOX3c. Nuclei are marked by arrows. (F) By mid-blastula stages, the XSOX3 polypeptide appears to be nuclear except in mitotic cells (arrows). Nuclear XSOX3 staining is seen throughout the embryo. (G) During gastrulation, anti-XSOX3 staining can be seen in the nuclei (arrows) of yolk plug (YP) cells. The blastopore (BP) is clearly visible and XSOX3 staining is seen throughout the surface ectoderm.

View larger version (43K): [in a new window]   Fig. 3. Mutations in XSOX3. (A) The HMG box (pfam00505.6) is characterized by three -helices and nine conserved residues. In this image, generated using the Cn3D viewer, the N- and C-termini are indicated and conserved residues are marked in red. (B) The conserved amino acids of the HMG box are indicated using the single-letter amino acid code. With the exception of m68, which is outside the conserved core region, the mutations we generated in the XSOX3 HMG box are indicated. (C) The sequence of the XSOX3 HMG box is displayed and the mutations generated for this analysis are indicated. The first residue of this sequence, D, corresponds to amino acid 38 of the full-length XSOX3 sequence. (D) XSOX3-V5H6 polypeptides (wild type, m7, m8, m17, m40, m55 and m68) generated by in vitro transcription/translation (TnT) were analysed by SDS-PAGE/immunoblot using the anti-V5 antibody. All accumulated to similar levels (arrow). The nature of the extraneous bands (*) are not known. (E) TnT-synthesized proteins were used in oligonucleotide gel-shift studies with the DC5 SOX-binding oligonucleotide. Unprogrammed lysate (Un) showed no shift and no effect upon the addition of antiV5 antibody (+). Oligonucleotide gel shift and antibody-induced supershift were observed upon addition of XSOX3 (wt) and m55 (m55) polypeptides, but not with m7, m8, m17, m40 or m68 polypeptides.

View larger version (37K): [in a new window]   Fig. 4. SOX3 mutant activities. (A) HeLa cells were transfected with plasmids encoding the various XSOX3 mutant polypeptides, along with a plasmids encoding a mutationally stabilized form of Xenopus ß-catenin, the pRL-TK plasmid for the normalization of transfection efficiency and the TOPFLASH reporter. Co-expression of ß-catenin (CAT +) activated TOPFLASH and this activation was suppressed by the co-expression of XSOX3-V5H6 (WT). The co-expression of the mutant XSOX3 polypeptides (m68 was not tested) led to the inhibition of the ß-catenin-induced activation of TOPFLASH. (B) Embryos were injected equatorially into the two dorsal blastomeres of four-cell embryos with RNA encoding V5H6-tagged forms of XSOX3. At stage 12, embryos were homogenized and the lysates immunoprecipitated using the mouse antiV5 antibody and then analyzed by SDS-PAGE/immunoblot using the mouse antiV5 antibody. Two distinct experiments are displayed. In each case, similar amounts of the exogenous polypeptides were found to accumulate. No signal was detected in uninjected (Un) embryos. (C) Control and RNA-injected embryos were allowed to grow out to stage 25. A wild-type (DAI 5) embryo is show at the top of the panel; three embryos displaying various levels of ventralization are shown below. (D) The proportion of embryos ventralized by the injection of XSOX3-V5H6 RNA or its mutated variants is displayed. The exact numbers and extent of ventralization observed are given in Table 2. Wild-type embryos are 0% ventralized.

View larger version (31K): [in a new window]   Fig. 5. XSOX3 effects on Siamois. (A) The 800bp Siamois/firefly-luciferase reporter together plasmid encoding Renilla luciferase were injected into either the dorsal or ventral side of fertilized eggs. Embryos were homogenized at stage 9 and luciferase activities were measured. Activity was normalized using Renilla luciferase levels and the dorsal activity was set to 1. (B) Fertilized eggs were injected with the Siamois/luciferase reporter plasmid together with 1 ng of XSOX3-V5H6 RNA. Reporter activity on the dorsal or ventral sides of the embryo in the absence of exogenous XSOX3 RNA (UN) was set independently to 1. Co-injection of XSOX3 wild-type or m7 RNAs activated the Siamois reporter on the dorsal but not on the ventral sides of embryos; the m8 polypeptide had no effect on either side of the embryo. (C) To determine the effect of XSOX3 RNA injection on the endogenous Siamois gene, embryos were ventralized by UV illumination during the first cell cycle. The dorsal axis was rescued by the injection of mutationally stabilized ß-catenin RNA (1 ng). RNA was isolated from wild-type, UV-treated, UV-treated, ß-catenin-rescued and UV-treated, ß-catenin- and XSOX3-RNA-injected embryos at stage 9, and RT-PCR was performed to visualize expression of the dorsalizing gene Siamois; EF-1 was used as a control. Siamois is expressed in intact embryos, greatly reduced in UV-ventralized embryos and returned to nearly wild-type levels in ß-catenin RNA-injected embryos. The co-injection of ß-catenin and either wild type or m7 XSOX3 RNA (1 ng) suppressed the reappearance of Siamois expression; co-injection of m8 RNA (1 ng) did not suppress ß-catenin-induced Siamois expression. (D) Streptavidin-agarose bound biotinylated mutant Siamois promoter fragment (biot. Sia null), biotinylated wild-type Siamois promoter fragment (biot. Sia), biotinylated DC5 (biot. DC5') or biotinylated TCF (biot. TCF) DNAs were incubated with stage-8 embryo lysates (Lys). Bound proteins were eluted and analysed by immunoblot using the anti-XTCF3n and anti-XSOX3c antibodies. Neither XTCF3 nor XSOX3 bound to the mutated Siamois sequence. XTCF3, but not XSOX3, bound to the wild-type Siamois and TCF sequences, and this binding was blocked by incubation with a 10- to 20-fold excess of unbiotinylated TCF oligonucleotide. XSOX3, but not XTCF3, bound to the DC5 sequence and this binding was blocked by incubation with a tenfold excess of unbiotinylated DC5 oligonucleotide. No binding of XTCF3 or XSOX3 was observed when biotinylated DNAs were omitted from the assay (No DNA, SA beads).

View larger version (49K): [in a new window]   Fig. 6. XSOX3 effects on Xnr5. The 200bp Xnr5/firefly-luciferase reporter together plasmid encoding Renilla luciferase were injected into either the dorsal or ventral (A) or animal or vegetal (B) hemispheres of fertilized eggs. Embryos were homogenized at stage 9 and luciferase activities were measured. Activity was normalized using Renilla luciferase levels and the dorsal activity was set to 1 in (A), whereas vegetal activity was set to 1 in (B). (C) Coinjection of XSOX3 wild-type or m7 RNA (1 ng) activated the Xnr5 reporter to a similar extent on both dorsal and ventral sides of the embryo; m8 RNA (1 ng) had no effect on either side of the embryo. (D) Injection of XSOX3 wild-type or m7 RNA led to activation of the Xnr5 reporter in both animal and vegetal hemispheres; the m8 polypeptide produced no significant activation of the Xnr5-luciferase promoter. (E) WT Xnr5' is the sequence of the distal Lef/TcfA site of the Xnr5 promoter identified by Hilton and Old (E. Hilton and R. Old, unpublished). It contains two SOX binding sites (blue boxes marked SOXa and SOXb) and a LEF/TCF site (red box marked LEF/TCF). MUT1 removes the SOXa site, leaving the SOXb and LEF/TCF sites intact. MUT2 removes the SOXb and LEF/TCF sites, leaving the SOXa site intact. MUT3 removes both SOX sites with no effect on the LEF/TCF site. MUT4 removes both SOX sites, and would remove the LEF/TCF site were it oriented TTGTTTG rather than GTTTGAT. (F) DNA fishing of stage-8 embryonic lysates was used to analyse these mutations. After SDS-PAGE and blotting, the blots were cut. The upper XTCF3 containing region was probed with anti-XTCF3n, the lower XSOX3-containing region was probed with anti-XSOX3c. Neither polypeptide bound to streptavidin beads in the absence of biotinylated DNA (SA beads). Both XTCF3 and XSOX3 were bound to wild-type Xnr5 DNA (wt). Both XTCF3 and XSOX3 bound to the MUT1 DNA (M1), which eliminates the SOXa site. Binding of XTCF3 was eliminated by the MUT2 mutation (M2), but XSOX3 binding remained. Little or no XSOX3 bound to MUT3 (M3) or MUT4 (M4), which eliminate both SOX sites, although both bound XTCF3. XSOX3 but little XTCF3 bound to DC5, whereas the TCF sequence bound XTCF3 but little XSOX3. (G) In whole embryos, the wild-type Xnr5p reporter is activated by co-expression of XSOX3. Removal of SOXa and SOXb binding sites (MUT4) abolishes this activation, whereas removal of the TcfA and TcfB sites, either alone or together leaves the reporter responsive to XSOX3, although the TcfA mutation alone or together with the TcfB mutation reduces responsiveness to XSOX3, presumably because this mutation also removes the SOXb binding site.

View larger version (92K): [in a new window]   Fig. 7. Wild-type and mutant XSOX3 effects on endogenous Xnr5. (A) Fertilized eggs were injected with RNAs encoding wild-type (wt), m7 (m7) or m8 (m8) V5H6-tagged forms of XSOX3. All three polypeptides accumulated to similar levels in stage-8 embryonic lysates (lysates). The binding wild-type and m7 polypeptides to both wild-type Xnr5 (Xn) and DC5 (DC) sequences (fishing) were much stronger than the binding of m8 under these conditions. (B) Xnr5 is normally expressed in the vegetal region of the embryo in response to the maternal factor VegT. Embryos were injected in the animal hemisphere with VegT RNA (1 ng) at the one-cell stage, either alone or together with RNA encoding XSOX3 (1 ng). At stage 8, animal caps were prepared. After a 2 hour incubation, they were homogenized and analysed by RT-PCR. Animal caps from uninjected embryos (AC/Un) did not express Xnr5 RNA; Xnr5 RNA was expressed in animal caps derived from embryos injected with VegT RNA (AC/VegT). Co-injection of VegT and wild-type (AC/VegT/+Sx3) or m7 (AC/VegT/+m7) XSOX3 RNAs suppressed the accumulation of Xnr5 RNA, whereas m8 RNA (AC/VegT/+m8) had no effect on Xnr5 RNA accumulation in response to VegT.

View larger version (37K): [in a new window]   Fig. 8. Quantitative RT-PCR analyses. To determine the effects of XSOX3 expression on endogenous genes, fertilized eggs were injected with wild-type or m8 XSOX3 RNAs (1 ng), and allowed to develop to stage 9, when they were homogenized and RNA was isolated and subjected to quantitative RT-PCR analysis. Compared with uninjected (un) and m8 (m8) injected controls, injection of wild-type XSOX3 RNA (wt) led to a decrease in the level of Xnr5 (A), Xnr6 (B), Siamois (C), Twin (D), Xnr3 (E) and Xbra (F) mRNAs. (G) The HMG boxes of XSOX3 and XSOXD differ at several positions [conservative changes are marked with an asterisk (*)]. (H) Compared with the ventralization of embryos following the injection of wild-type XSOX3 RNA (SOX3), injection of 650 pg mRNA encoding XSOXD-V5H6 (SOXD) failed to ventralize embryos [uninjected embryos (un)]; immunocytochemistry revealed that the XSOXD-V5H6 polypeptide had accumulated (data not shown). (I) Fertilized eggs were injected with 1 ng of either XSOX3-V5H6 wild-type (wt) or XSOXD-V5H6 RNA. At stage 9, embryos were homogenized and subjected to quantitative RT-PCR analysis. Injection of SOXD-V5H6 RNA had no apparent effect on Xnr5 RNA levels.

View larger version (30K): [in a new window]   Fig. 9. Manipulating XSOX3 accumulation. (A) Morpholinos against the 5' untranslated region (5' UTR) and coding sequence of XSOX3 and XSOX7 were generated and compared with the XSOX3 RNA sequence. (B) Upon injection into fertilized eggs, the XSOX3 morpholino led to a decrease in XSOX3 polypeptide level in embryos analysed at stage 9; the injection of the XSOX7 morpholino had no apparent effect on XSOX3 accumulation. Neither morpholino effected the accumulation of XTCF3 (recognized by the anti-XTCF3n antibody). (C) Embryo lysates were prepared and subject to quantitative RT-PCR analysis. Depletion of XSOX3 led to an increase in Xnr5 RNA compared with uninjected and XSOX7-morpholino-injected embryos.

View larger version (43K): [in a new window]   Fig. 10. Manipulating maternal XSOX3 activity. (A) Stage 8 embryo lysates (300 µl) were incubated alone (Lys) or together with 0.5 µg anti-XSOX3c (Lys + Sx3') or anti-XTCF3n (Lys + Tf3') antibodies. Lysates were then incubated with either streptavidinagarose beads (SA alone) or biotinylated DC5-streptavidin-agarose beads (DC5-SA). Both lysates and DC5-bound proteins were analysed by immunoblot with anti-XSOX3c. XSOX3 (arrowhead) was bound to DC5 in control and anti-XTCF3n-containing lysates, but its binding was greatly reduced by the addition of the anti-XSOX3c antibody; no binding was seen in the absence of DC5 DNA. (B) Fertilized eggs were injected with anti-XSOX3c antibody (10 nl of a 7.5 mg ml–1 solution) or anti-XTCF3n antibody (7.8 mg ml–1 solution). At stage 9, the embryos were homogenized and analysed by DNA fishing with DC5-streptavidin-agarose. Anti-XSOX3c-injected (Sx3), anti-XTCF3n-injected (Tf3) and uninjected (Un) lysates were analysed by immunoblot using anti-XSOX3c. The total amount of XSOX3 was unchanged upon antibody injection, but anti-XSOX3c dramatically inhibited the binding of XSOX3 to DC5-streptavidin-agarose (DC5 fishing). (C) Fertilized eggs injected with either anti-XSOX3c antibody (10 nl of a 7.5 mg ml–1 solution) (antiSX3), anti-XTCF3n antibody (10 nl of a 7.8 mg ml–1 solution) (anti-TCF3), XSOX3 RNA (1 ng) (Sx3 RNA), XSOX3C-VP16 RNA (1 ng) (Sx3CVP16) or XSOX3C-EnR RNA (1 ng) (Sx3-EnR) were allowed to develop to stage 9 and then homogenized and analysed by quantitative RT-PCR. XSOX3 and XSOX3C-EnR RNA injection reduced Xnr5 RNA levels; injection of either anti-XSOX3c and XSOX3C-VP16 RNA increased Xnr5 RNA levels; injection of anti-XTCF3n did not alter Xnr5 RNA accumulation.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter