The Wnt/β-catenin/TCF/Sp5/Zic4 Gene Network That Regulates Head Organizer Activity in Hydra Is Differentially Regulated in Epidermis and Gastrodermis

Abstract

Hydra head formation depends on an organizing center in which Wnt/β-catenin signaling, that plays an inductive role, positively regulates Sp5 and Zic4, with Sp5 limiting Wnt3/β-catenin expression and Zic4 triggering tentacle formation. Using transgenic lines in which the HySp5 promoter drives eGFP expression in either the epidermis or gastrodermis, we show that Sp5 promoter activity is differentially regulated in each epithelial layer. In intact animals, epidermal HySp5:GFP activity is strong apically and weak along the body column, while in the gastrodermis, it is maximal in the tentacle ring region and maintained at a high level along the upper body column. During apical regeneration, HySp5:GFP is activated early in the gastrodermis and later in the epidermis. Alsterpaullone treatment induces a shift in apical HySp5:GFP expression towards the body column where it forms transient circular figures in the epidermis. Upon β-catenin(RNAi), HySp5:GFP activity is down-regulated in the epidermis while bud-like structures expressing HySp5:GFP in the gastrodermis develop. Sp5(RNAi) reveals a negative Sp5 autoregulation in the epidermis, but not in the gastrodermis. These differential regulations in the epidermis and gastrodermis highlight the distinct architectures of the Wnt/β-catenin/TCF/Sp5/Zic4 network in the hypostome, tentacle base and body column of intact animals, as well as in the buds and apical and basal regenerating tips.

Keywords: HEK293T mammalian cells; Hydra head organizer; Hydra transgenic lines; Sp5 promoter autoregulation; Sp5 transcription factor; Wnt/β-catenin signaling; Zic4 tentacle regulator; Zic4 transcription factor; epidermal and gastrodermal epithelial layers; gene regulatory network (GRN).

Figure 1

Differential regulation of HySp5-3169:GFP expression in the epidermal and gastrodermal layers. (A) Schematized view of Hydra anatomy that includes the apical region or head, formed from a dome-shape named hypostome, centered around the oral opening, surrounded by a ring of tentacles at its basis, the elongated or contracted body column and the basal disc or foot that can attach to substrates. The respective expression levels of Wnt3, Sp5 and Zic4 define three distinct domains in the apical region (see ref. [34]). (B) Structure of the HyAct-1388:mCherry_HySp5-3169:GFP reporter construct used to generate the epidermal and gastrodermal HySp5-3169:GFP transgenic lines, where epithelial cells from the epidermis and gastrodermis, respectively, express GFP and mCherry (sequence in Figure S2). TCF-BS: TCF-binding sites (orange); Sp5-BS: Sp5-binding sites (grey). (C,D) Optical sections of live transgenic animals expressing HySp5-3169:GFP in the epidermis (C) and gastrodermis (D). Brightfield and eGFP channels are shown and the apical region of each animal is magnified on the right (Figure S3A,B). The dashed white line indicates the thin extracellular layer known as the mesoglea, which delimits the epidermal (outer) and gastrodermal (inner) epithelial layers. Scale bar: 100 µm. (E,F) Q-PCR analysis of GFP, Sp5 and Wnt3 transcript levels in the apical, body column (BC) and basal regions of epidermal (E) and gastrodermal (F) HySp5-3169:GFP animals fixed immediately after dissection. p values as indicated in Materials & Methods. (G,H) Live imaging of epidermal (G) and gastrodermal (H) HySp5-3169:GFP animals with eGFP (green), mCherry (red); the apical region of each animal is magnified on the right. White arrows point to the tip of the hypostome. The graphs show the eGFP/mCherry fluorescence intensity ratios (relative GFP intensity) along the animal axis. (I,J) Immunodetection of GFP (green) and mCherry (red) in the apical region (white arrow) of epidermal (I) and gastrodermal (J) HySp5-3169:GFP animals. Scale bars in panels (G–J): 250 µm. (K,L) Graphs showing the relative GFP fluorescence recorded live (K) or after immunodetection (L) in epidermal or gastrodermal HySp5-3169:GFP animals. The apical extremity is on the left (100%), the basal one on the right (0%). See Figure S4.

Figure 2

GFP regulation in regenerating and budding HySp5-3169:GFP transgenic animals. (A,B) GFP expression in regenerating halves from epidermal (A) and gastrodermal (B) HySp5-3169:GFP transgenic animals bisected at t0 and fixed at indicated times. Regen.: regeneration; hpa: hours post-amputation; red arrows point to apical-regenerating (AR) regions, red triangles to the basal-regenerating (BR) regions, vertical bars indicate gastrodermal GFP expression along the body column, asterisks the original basal discs, white arrows outlined red to the regenerated heads, white triangles outlined red to the regenerated basal discs. See Figure S5. (C,D) GFP (green) and mCherry (red) fluorescence in AR and BR halves of HySp5-3169:GFP transgenic animals pictured live at indicated time points. White arrows point to apical regions of original polyps, red arrows to AR regions, white arrows outlined red to regenerated heads; white arrowheads to original mature basal discs, red arrowheads to the BR regions. See Figure S6. (E–H) Live imaging of budding HySp5-3961:GFP transgenic animals, either epidermal (E) or gastrodermal (F), pictured at indicated stages with the Olympus SZX10 microscope ((E,G), GFP fluorescence only) or the Zeiss LSM780 microscope ((F,H), GFP and mCherry fluorescence). On the parental polyp, yellow arrowheads point to the “budding belt” that forms in the budding zone; on the developing buds, red arrows point to the developing apical region, red arrowheads to the differentiating basal region and white arrowheads outlined red to fully differentiated basal discs. Scale bar: 250 µm.

Figure 3

Alsterpaullone-induced modulations of GFP, Wnt3 and Sp5 expression along the epidermis and gastrodermis of HySp5-3169:GFP and HyWnt3-2149:GFP transgenic animals. (A) Schematic representation of the activating effect of ALP on Wnt/β-catenin signaling. (B) Co-detection of GFP (purple) and Wnt3 (red) (left half) or Sp5 (purple) and Wnt3 (red) (right half) in wild-type Hv_AEP animals or in transgenic animals that constitutively express the HySp5-3169:GFP or HyWnt3-2149:GFP constructs, after 2- or 4-day ALP exposure. White arrows: Wnt3 expression at the tip of the hypostome; black arrows: expression immediately below the apical region; black arrowheads: ALP-induced GFP expression in the peduncle zone; blue and grey arrows: ALP-induced circular zones of GFP and Sp5 expression respectively along the body column; yellow arrows: ALP-induced ectopic Wnt3 expression in the apical or basal regions; orange arrows: Wnt3-expressing spots along the body column; vertical black bars: areas of ALP-induced GFP expression along the body column; s.t.: short tentacles; Te: testis. (C) Schematic representation of the ALP-induced modulations of Wnt3 and Sp5 in the epidermal and gastrodermal HySp5-3169:GFP and HyWnt3-2149:GFP transgenic lines. See Figures S7–S9. (D) Live imaging of mCherry and GFP fluorescence in epidermal and gastrodermal HySp5-3169:GFP animals treated for 2, 4 and 7 days with ALP or DMSO. For each condition, GFP fluorescence is shown on the left and the merged GFP (green) and mCherry (red) fluorescence on the right. Vertical white bars indicate areas of ectopic GFP fluorescence. Scale bar: 250 µm.

Figure 4

Impact of β-catenin(RNAi) on Sp5 expression in Hv-Basel and HySp5-3169:GFP transgenic animals. (A) Schematic view of the procedure: After one, two or three electroporations (EP1, EP2 and EP3) with scramble or β-catenin siRNAs, animals were either dissected in apical and body column regions for RNA extraction (grey triangles), or fixed for whole-mount in situ hybridization (ISH, blue triangles) or imaged live (red triangles) at indicated time-points (d pEP: day(s) post-EP). (B) Q-PCR analysis of Sp5, β-catenin and GFP transcript levels in apical (100–80%, left) and body column (80–0%, right) regions of epidermal (left) and gastrodermal (right) HySp5-3169:GFP transgenic animals taken one or two days after EP1 (days 1 and 2), after EP2 (days 3 and 4) or one day after EP3 (day 5). In each panel, the colored line corresponds to the Fold Change (FC) values between β-catenin RNAi animals (continuous grey lines) and control animals exposed to scramble siRNAs (dotted grey lines). Values are each expressed as FC relative to non-electroporated animals at time 0, just before EP1. See Figure S10. (C) Sp5 expression pattern and phenotypic changes in scramble and β-catenin(RNAi) Hv_Basel animals at indicated time-points. Red arrows point to Sp5-expressing patches along the body column, yellow arrows to bud-like structures that express Sp5 and become multiple two days pEP3 without differentiating apical structures. See Figure S11. (D) GFP (green) and mCherry (red) fluorescence in β-catenin(RNAi) HySp5-3169:GFP transgenic live animals. GFP fluorescence in apical regions is either normal (white arrows) or missing (red arrows). Note the transient bud-like structures that develop after β-catenin(RNAi) (red arrowheads) and express GFP in gastrodermal_HySp5-3169:GFP animals (green triangles). See Figure S12. (E) Immunodetection of GFP and mCherry in β-catenin(RNAi) HySp5-3169:GFP animals 6 days post-EP2 (6d pEP2). Apical regions of scramble and β-catenin(RNAi) animals are magnified on the right. White and red arrows as in (D). See Figure S13. Scale bar: 250 µm.

Figure 5

Ectopic GFP/GFP expression in HySp5-3169:GFP animals knocked-down for Sp5. (A) RNAi procedure applied in experiments depicted in panels A-C. At 8 h, 16 h and 24 h post-EP1 (pEP1) and 8 h, 16 h and 24 h post-EP2 (pEP2, red triangles), animals were either fixed for RNA extraction, or imaged live and fixed for whole-mount ISH. Q-PCR analysis of Sp5, β-catenin and GFP transcript levels in apical (100–80%, left) and body column (80–0%, right) regions of epidermal (left) and gastrodermal (right) HySp5-3169:GFP transgenic animals exposed to scramble siRNAs or to Sp5 siRNAs. In each panel, the colored line corresponds to the Fold Change (FC) values between Sp5 RNAi animals (continuous grey lines) and control animals exposed to scramble siRNAs (dotted grey lines), which are each expressed as FC relative to non-electroporated animals at time 0, just before EP1. See Figure S14. (B) GFP expression detected by WM-ISH at indicated time-points after Sp5(RNAi), as depicted in (A). Vertical black bars along the body column and white arrows in the lower body column indicate regions where GFP is up-regulated. See Figure S15. (C) GFP (green) and mCherry (red) fluorescence in Sp5 (RNAi) epidermal (left) or gastrodermal (right) HySp5-3169:GFP animals as depicted in (A). See Figure S16. (D) RNAi procedure applied in experiments depicted in panels D and E. Q-PCR quantification of Sp5, GFP, Wnt3 and β-catenin transcripts in the apical (100–80%), central body column (BC, 80–30%) and basal (30–0%) regions of epidermal (left) and gastrodermal (right) HySp5-3169:GFP animals dissected two days post-EP2 (2d pEP2). ns: non-significant value, other statistical values as indicated in Materials & Methods. (E) GFP (green) and mCherry (red) fluorescence in epidermal (left) or gastrodermal (right) HySp5-3169:GFP animals 2d pEP2. White bars indicate areas of ectopic GFP fluorescence along the body column, white arrows point to spots of ectopic gastrodermal GFP fluorescence in tentacles. See Figure S17. Scale bars correspond to 250 µm, except in (B) where it is 200 µm.

Figure 6

CHIP-qPCR identification of Sp5-binding sites in the Hydra Sp5 promoter. (A) Structure of the Hydra Sp5 protein with the conserved Sp box (green), Buttonhead box (Btd, light purple) and zinc-finger (ZF) domain (light blue); the purple line indicates the 218 AA-long region used to raise the monoclonal anti-Sp5 (mSp5) antibody, the green lines indicated the peptides used to raise the polyclonal anti-Sp5 (pSp5) antibody. (B) Western blot using the mSp5 antibody to detect the TNT-produced HySp5 protein (lanes 1, 2), the HySp5-218 recombinant protein used to raise mSp5 (lane 3), HySp5 expressed in HEK293T cells (lanes 4, 5) or HySp5 present in Hv_AEP2 nuclear extracts prepared from whole animals (lane 6), from apical (100–50%, lane 7) and basal (50–0%, lane 8) halves. (C) Schematic view of the 15 regions (see 15 pairs of primers in Table S2) tested along the 2992 bp-long Sp5 promoter by ChIP-qPCR using Hm105 extracts and the mSp5 antibody. Significant enrichment is noted only in regions PP4 and PP5. (D) Similar ChIP-qPCR enrichment in regions PP4 and PP5 obtained with mSp5 antibody when Hm105 (left) or Hv_AEP2 (right) extracts are used. See Figure S18. (E) Proximal HySp5 promoter sequence (−162 to +29), which contains the transcriptional start sites (TSS) identified in Hv_AEP (TSS1 +1) and Hm105 (TSS2 −149, see Figure S19), five Sp5 binding sites (SP5-BS, light blue), a single TCF binding site (TCF-BS, orange), the PP4 (grey) and PP5 (green) primers used for ChIP-qPCR. In the same region, the PPA (−135 to −67) and PPB (−71 to +2) stretches, underlined with purple (PPA) and pink (PPB) dashed lines, respectively, were used in Electro-Mobility Shift Assay (EMSA). (F) Schematic map of the HySp5 2992 bp promoter region indicating the predicted TSS1, the clustered TCF-BS and Sp5-BS (light purple and orange bars, respectively), the PP1, PP4 and PP5 primer pairs. Sequences of the TCF-BS and Sp5-BS identified in the HySp5 promoter. (G) EMSA showing a shift (arrows) of the PPA and PPB ds-DNAs incubated with Hm105 NEs. Comp.: unlabeled ds-PPA (left) or ds-PPB (right) added as competitor 200 fold in excess during the incubation; mut: mutated.

Figure 7

Functional analysis of the Hydra Sp5, Zic4 and Wnt3 promoters. (A–G) Luciferase reporter assays performed in HEK293T cells to measure the Relative Luciferase Activity (RLA) driven by various promoters when Hydra proteins are co-expressed. Each data point represents one biologically independent experiment. (A) RLA levels driven by the HySp5 promoter either when 2992 bp long (HySp5-2992), or when deleted from its proximal region (HySp5-2828) that contains five Sp5-binding sites (Sp5BS), or when one of these 5 Sp5BSs is mutated (HySp5-2992-mBS1, HySp5-2992-mBS2, …). Each construct was tested either in the absence of any protein co-expressed (CMV-empty), or in the presence of co-expressed proteins, full-length Sp5 (HySp5-420) or Sp5 lacking its DNA-Binding Domain (HySp5-ΔDBD). (B) RLA levels driven by the HySp5-2992 promoter when the full-length HyZic4 (HyZic4-431) or the truncated HyZic4 lacking its DNA-Binding Domain (HyZic4-ΔDBD) are co-expressed. (C,D) RLA levels driven by the HyZic4-3505 promoter when full-length or truncated HySp5 (HySp5–420, HySp5-ΔDBD in (C)) or full-length or truncated HyZic4 (HyZic4-431, HyZic4-ΔDBD in (D)) are co-expressed. (E,F) RLA levels driven by the TOPFlash or FOPFLASH reporter constructs that contain 6× TCF-binding sites either consensus or mutated when HyZic4-431 (E,F) or HyZic4-ΔDBD (E) are co-expressed. (G) RLA levels driven by the Wnt3-2142 promoter when HyZic4-431 or HyZic4-ΔDBD are co-expressed. (H) Diagram showing the regulations detected in HEK293T cells on the Hydra Wnt3, Sp5 and Zic4 upstream sequences when the human β-catenin and/or HySp5 and HyZic4 proteins are co-expressed (this work, [33,34]). (I) Immunoprecipitation (IP) of HA-tagged HyZic4-431 expressed in HEK293T cells together or not with huTCF1. IP was performed with an anti-HA antibody and co-IP products were detected with the anti-TCF1 antibody. Same results were obtained in two independent experiments. (J,K) Zic4 (J) and Sp5 (K) transcript levels measured by qPCR in Hv_Basel animals exposed twice to scrambled (scr) RNAs or Zic4, Sp5, β-catenin (b-cat), Wnt5 or Wnt8 siRNAs. Levels are normalized to those measured in control animals exposed to scr RNAs. In all panels, error bars indicate Standard Deviations and statistical p values are as indicated in Materials & Methods (unpaired t-test).

Figure 8

Schematic summary of the layer-specific Sp5 regulation along the Hydra body axis. (A) Dot plot view of Zic4, Wnt3, Sp5, TCF, Wnt5A, Wnt8 and β-catenin expression in cells from the epithelial lineages, either epidermal or gastrodermal, and the interstitial lineage as deduced from Hydra single-cell transcriptome analysis [35]. Along the central body column, single-cell sequencing has identified two populations of epithelial stem cells in the epidermis (SC1 and SC2) and three in the gastrodermis (SC1, SC2 and SC3). See other abbreviations in Figure S1. (B,C) Schematic representation of the predicted genetic regulation network (GRN) at work in the epidermal and gastrodermal layers of the hypostome (B) and tentacle (C) regions in Hydra. (D,E) Schematic view of GFP fluorescence (green), Sp5 expression and predicted GRNs at work in the epidermal (D) and gastrodermal (E) layers of the body column (BC) of HySp5-3169:GFP transgenic animals, either maintained in homeostatic conditions (left) or ALP-treated, or knocked-down for Sp5 or β-catenin (right). Bold letters, black arrows and thick red bars indicate a stronger activity.