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

First published online 31 January 2007
doi: 10.1242/dev.02796

This Article Summary Full Text Full Text (PDF) Supplementary Material 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 Chao, A. T. Articles by Bejsovec, A. Search for Related Content PubMed PubMed Citation Articles by Chao, A. T. Articles by Bejsovec, A. Social Bookmarking                         What's this?

The HMG-box transcription factor SoxNeuro acts with Tcf to control Wg/Wnt signaling activity

Department of Biology, Duke University, Durham, NC 27708, USA.

View larger version (71K): [in this window] [in a new window]   Fig. 1. SoxN loss of function suppresses wg mutant phenotypes. (A) Hypomorphic wgNE2 mutation reduces the zones of naked cuticle, which separate denticle belts, on the ventral side and disrupts dorsal patterning, resulting in strong curvature of the embryonic cuticle. (B) SoxNNC14 mutation rescues wgNE2 ventral patterning to almost wild type (compare with Fig. 2A), without rescuing dorsal patterning. (C,D) The RNA-null wgCX4 allele produces a `lawn of denticles' phenotype (C), which is partially suppressed by the SoxNNC14 mutation (D). (E) wgCX4 mutant embryos lose epidermal expression of the Wg target gene en before stage 10 (compare with wild-type pattern in Fig. 2G). (F) wgCX4, SoxNNC14 double-mutant homozygotes retain some epidermal en expression (arrows) even at late stages. SoxNNC14 is linked to wg on the second chromosome; single and double homozygotes are recognized by the absence of GFP from a marked balancer chromosome. Embryos are oriented with anterior to the left and dorsal side up.

View larger version (92K): [in this window] [in a new window]   Fig. 2. SoxN regulates Wg pathway activity. (A) Wild-type cuticle pattern shows a normal expanse of Wg-specified naked cuticle separating denticle belts. (B) SoxNNC14 single mutants produce excess naked cuticle. (C) Wild-type stripes of en expression span 2-3 cells in each segment. (D) en-expressing stripes in SoxNNC14 are broadened, similar to known phenotypes produced by ectopic Wg signaling. (E) Ubiquitous expression of wild-type SoxN with the arm-Gal4 driver rescues the excess-naked-cuticle phenotype in SoxNNC14 homozygotes. This treatment does not rescue embryonic lethality. (F) Overexpressing SoxN at higher levels, using the E22C-Gal4 driver in an otherwise wild-type embryo, affects Wgmediated cuticle patterning. Ectopic denticles replace some of the ventral and ventrolateral naked cuticle (arrows), and dorsal patterning is disrupted, leading to curvature of the embryo. We observe an average of 12 ectopic denticles within the naked cuticle zone of a typical abdominal segment (n=100). (G) Stripes of en expression extend evenly from the ventral midline to the edge of the dorsal epidermis in wild-type stage-10 embryos. (H) These stripes are narrowed, particularly in the dorsolateral regions (arrow), when UAS-SoxN is driven with E22C-Gal4. Ventrally, expansion of en expression in the underlying central nervous system (bracket) can be seen; this en expression is not under the control of Wg (Bejsovec and Martinez Arias, 1991; Heemskerk et al., 1991) and presumably reflects the role of SoxN in specifying neuronal fates. (I) Side-by-side comparison showing no difference in anti-Arm staining between a SoxNGA1192 mutant (bottom) and a wild-type sibling (top). (J) Anti-GFP staining reveals the presence of the twist-GFP balancer chromosome in a wild-type sibling and its absence in the homozygous mutant embryo. (K) Quantitative immunoblot of lysates from hand-selected embryos shows equivalent Arm levels in homozygous mutants for SoxNGA1192 and SoxNNC14 compared with their wild-type CyO-GFP-bearing siblings. When normalized to the tubulin loading control, there is no detectable difference among the first three lanes. By contrast, Arm levels are 25% higher in RacGap50CAR2 mutant homozygotes. Embryos are oriented with anterior to the left and dorsal side up.

View larger version (147K): [in this window] [in a new window]   Fig. 3. Epistatic relationships of SoxN with the Wg pathway. (A) The SoxNNC14 mutation placed in trans with a small deficiency for the region, Df(2L)Exel7040, shows no change from the homozygous mutant phenotype (compare with Fig. 2B), indicating that SoxNNC14 behaves as a null allele. (B) Removing maternal SoxN does not increase the severity of the mutant phenotype; therefore, SoxN acts zygotically. (C) Dorsal patterning elements show mild disruptions in some segments of SoxNNC14/Df-mutant embryos (arrow; compare with more anterior segments, which have normal dorsal pattern elements). (D) Overexpressing SoxN in embryos derived from mothers that were heterozygous for groBX22, a deficiency removing the locus, produces milder pattern disruptions both dorsally and ventrally (compare with Fig. 2F). (E,F) Double homozygotes for arm4 and either SoxN allele show the arm `lawn of denticles' phenotype (E), but embryos are smaller and have stronger dorsal pattern disruptions than do arm4 single mutants (F) (data shown in Table 2). (G) Mutants homozygous for SoxN that were derived from groBX22 heterozygous mothers show increased naked cuticle (n=140). (H,I) SoxN; Tcf2 double-mutant embryos also show increased naked cuticle (H). Thus, the SoxN mutant phenotype is epistatic to the Tcf `lawn of denticles' phenotype (I) (data shown in Table 2). (J) E22C-Gal4-driven ubiquitous expression of dominant-negative Tcf produces a `lawn of denticles' phenotype and severely reduces the size of the embryonic cuticle. (K) Segmental patterning and body size of TcfDN-expressing embryos are partially rescued when SoxN dosage is reduced. The SoxNNC14 mutation is linked to the E22C-Gal4 insertion in this experiment, so that all embryos ubiquitously expressing UAS-TcfDN are also heterozygous for SoxN. All show a milder phenotype regardless of whether the SoxN mutation was introduced from the mother or the father (n=205). Embryos are oriented with anterior to the left and dorsal side up.

View larger version (97K): [in this window] [in a new window]   Fig. 4. Tcf acts synergistically with SoxN. (A,B) Overexpressing the wild-type form of Tcf, using the arm-Gal4 driver, produces no effect on en expression (A) or on cuticle pattern (B), and most such embryos hatch into larvae (data shown in Table 1). (C,D) arm-Gal4 drives a lower level of expression than E22C-Gal4, producing milder effects of overexpressed SoxN. Under these conditions, en expression is only slightly narrowed (C) and cuticle pattern is mostly normal (D), with only occasional ectopic denticle formation (arrow) and little dorsal curvature. (E,F) When Tcf and SoxN are expressed together, using arm-Gal4 to drive a double-homozygous UAS stock, en expression is more severely narrowed (arrows, E), and ectopic denticles (arrows, F) appear with greater frequency and broader distribution across the ventral naked cuticle domain; cuticles are strongly curved due to defective dorsal patterning. The severity of the phenotype is comparable to expressing higher levels of UAS-SoxN alone with the stronger E22C-Gal4 driver (see Fig. 2F,H). Arrowheads in A, C and E indicate the posterior end of the ventral midline. Embryos are oriented with anterior to the left and dorsal side up.

View larger version (43K): [in this window] [in a new window]   Fig. 5. SoxN represses Wnt signaling in HEK293T cells. (A) Increasing the amount of SoxN decreases the amount of TOPflash activity that is detected. Fold activation is defined as the ratio between relative luciferase units under induced versus uninduced conditions. For comparison between independent experiments, values were normalized to the 0 ng SoxN data point. (B) Schematic diagram of constructs used for Tcf-bindingsite competition assay. FOP-L is identical to TOP-L in structure but carries mutated Tcf-binding sites. (C) Increasing amounts of TOP-L, but not of FOP-L, decrease the amount of TOPflash activity that is detected. Fold activation is defined as in A. (D) Extra Tcf-binding sites do not affect SoxN repression of TOPflash activity, expressed as the fold activation with 10 ng pcDNA-SoxNflag/fold activation with 0 ng pcDNA-SoxNflag. Each competitor DNA was present at 2.5x the amount of TOPflash reporter. No statistically significant difference in the degree of repression can be detected. (E) SoxN and Tcf4 at low doses synergistically repress TOPflash activity. Cell cultures containing 10 ng pcDNA-SoxNflag and 0, 5, 25, 50 or 100 ng of pcDNA-Tcf4myc were assayed. Empty pcDNA3.1 vector was used to hold constant the total amount of DNA added. Each bar represents relative activation, defined as fold activation with 10 ng pcDNA-SoxNflag/fold activation with 0 ng pcDNA-SoxNflag. (F) Beta-catenin, but not SoxN, is co-immunoprecipitated with Tcf in TOPflash cell extracts. pcDNA-TCF4myc (0.5 µg) and pcDNA-SoxNflag (0.05 µg) were co-transfected into HEK293T cells and grown under induced and uninduced conditions. Control cultures contained only pcDNA3.1 to reveal non-specific cross-reacting proteins. Cell extracts (e) were subjected to immunoprecipitation using Myc-antibody-conjugated ProtG beads (b). Immunoblot was stained with anti-Flag and anti-beta-catenin. SoxNflag is not detected in the Tcf4-myc-bound fraction under conditions where beta-catenin is found in the induced Tcf4-myc bound fraction (third lane). (G) SoxN and beta-catenin are not co-immunoprecipitated in TOPflash cell extracts. pcDNA-SoxNflag (0.5 µg) was transfected into HEK293T cells and grown under induced conditions. Cell extracts (e) were subjected to immunoprecipitation using flag-antibody-conjugated ProtG beads (b). Immunoblot was stained with anti-flag and anti-beta-catenin. Beta-catenin is not detected in the SoxNflag-bound fraction (first lane).

View larger version (14K): [in this window] [in a new window]   Fig. 6. Model for SoxN interaction with the Tcf-responsive promoter. SoxN may be able to bind DNA sequences adjacent to, or overlapping with, Tcf-binding sites and may contribute to the repressive capacity of Tcf, for example, by helping to recruit an unidentified scaffolding molecule that stabilizes the Tcf-Gro interaction.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter