Figure 1.
Sequential activation of Brachyury target genes in the developing notochord of C. intestinalis.
(A–O) WMISH of C. intestinalis embryos at the 110-cell (A,F,K), gastrula (B,G,L), neural plate (C,H,M), neurula (D,I,N), and initial tailbud (E,J,O) stage. Embryos in (A,F,K) are photographed with anterior up, all other embryos are oriented with anterior to the left. Scale bar: approximately 100 µm. Red arrowheads indicate the regions containing stained notochord cells. (P) Schematic representation of the developmental time-courses of Ci-Bra (red horizontal bar) and its target genes (colored bars), plotted against a time-table denoting notochord morphogenetic events and the embryonic stages for Ciona at 18°C [32]. Early-onset genes are depicted as pink bars, middle-onset genes as orange bars, and the two late-onset genes are indicated by blue bars. All bars are dashed on their right side because the expression of the genes that they represent is yet to be determined at the larval stage. hpf, hours post-fertilization.
Figure 2.
Synergistic activity of three Ci-Bra binding sites in the Ci-thrombospondin 3 notochord CRM.
Identification of the minimal sequences necessary for notochord activity of the Ci-thbs3 notochord CRM through sequence-unbiased truncations and site-directed mutations. Red and grey rectangles symbolize genomic fragments displaying or lacking notochord activity, respectively. A pink rectangle indicates the Ci-FoxA-a basal promoter and a blue rectangle the LacZ reporter, both present in the electroporation vector (not drawn to scale; omitted in the truncation and mutation constructs for clarity). Parallel diagonal lines indicate genomic regions that are present in the constructs but are not depicted in the figure. “+” and “−” signs are used to show presence or absence of notochord activity, respectively. Putative binding sites are indicated by the symbols in the figure; their orientation is provided by small arrows on top of each site. Mutagenized sequences are depicted in grey and marked by an “X”. (A) Sequence-unbiased truncations and site-directed mutations. Putative Ci-Bra binding sites are indicated as T1–T4. (B–G) Low-magnification group microphotographs of late tailbud embryos electroporated in parallel with either the wild-type 116-bp Ci-thbs3 notochord CRM (B) or with mutated versions of this construct carrying either individual (C,D), double (E,F), or triple (G) mutations in the Ci-Bra binding sites. The mutations were as follows (mutated sequences are in lower case): T1M, TGGCAC to TGGtct; T2M, TCGCAC to TCGtct; T3M, TAACAC to TAAtct; T4M, TCGCAC to TCGtct. Insets in (B,G) show higher magnification views of representative stained embryos. (H) Quantification of the activity of the constructs shown in (B–G) in different tissues, as detected by X-Gal staining. The y-axis reports the percentage of embryos showing staining in a certain tissue as a fraction of the total number of stained embryos scored. Dark red, activity in notochord only; pink, activity in notochord and in additional tissues; blue, activity only in tissues other than notochord. The number of stained embryos scored (n) for each construct is shown below the x-axis. Results were averaged from three independent experiments carried out on different batches of embryos.
Figure 3.
Notochord CRMs controlled by Ci-Bra through two cooperative binding sites.
Identification of the minimal sequences necessary for notochord activity of the Ci-FCol1 and Ci-Noto5 notochord CRMs, through sequence-unbiased truncation analyses and site-directed mutations. All schematics are as in Figure 2. (A) Sequence-unbiased truncation and site-directed mutation analyses of the Ci-FCol1 notochord CRM. (B–F) Low-magnification group microphotographs of embryos electroporated in parallel with either the wild-type 65-bp Ci-FCol1 minimal notochord CRM (B) or with mutated versions of this construct carrying individual mutations in each Ci-Bra binding site, as shown at the bottom of each panel (C–E) or a double mutation in the two proximal Ci-Bra binding sites (F). (G) Quantification of the activity of the constructs shown in (B–F) in different tissues, as in Figure 2. (H) Sequence-unbiased truncation and site-directed mutation analyses of the Ci-Noto5 notochord CRM. (I–M) Low-magnification group microphotographs of embryos electroporated in parallel with either the wild-type 83-bp Ci-Noto5 minimal notochord CRM (I) or with mutated versions of this construct carrying the individual mutations in each Ci-Bra binding site shown at the bottom of each panel (J–L) or a double mutation in the two proximal Ci-Bra binding sites (M). (N) Quantification of the activity of the constructs shown in (I–M) in different tissues, plotted as described in Figure 2. Insets show higher magnifications of representative stained embryos.
Figure 4.
Notochord CRMs controlled by Ci-Bra through single binding sites.
(A,H,M) Identification of the minimal sequences necessary for notochord activity of the Ci-Noto1, Ci-Noto9, and Ci-Noto4 notochord CRMs, through sequence-unbiased truncation analyses and site-directed mutations. All schematics are as in Figure 2. In the 1.1-kb Ci-Noto1 construct, only the binding sites that were analyzed are depicted. (B) Control late tailbud embryo carrying the 170-bp minimal Ci-Noto1 notochord CRM. (C) Embryo from the same clutch as (B), co-electroporated with the 170-bp construct and the Ci-FoxA-a>Bra construct; misplaced notochord, endoderm, and CNS cells are indicated by an aqua arrowhead. (D) Control late tailbud embryo carrying the 170-bp minimal Ci-Noto1 notochord CRM in which the Ci-Bra site shown in (A) has been mutated as shown at the bottom of the panel. (E) Embryo from the same clutch as (D) co-electroporated with the minimal CRM and the Ci-FoxA-a>Bra construct. (F) Control embryo carrying a larger 1.1-kb Ci-Noto1 CRM truncation. (G) Embryo electroporated in parallel with the embryo in (F), carrying the 1.1-kb Ci-Noto1 CRM truncation with only the Ci-Bra binding site mutated, as shown at the bottom of (A). Note that in the 1.1-kb construct, only the binding sites that were subjected to mutation analysis are depicted. (I,J) Low-magnification group microphotographs of embryos electroporated in parallel with either the 248-bp Ci-Noto9 minimal notochord CRM (I) or with a version of this construct carrying a mutation in the Ci-Bra binding site (J). (K,L) Individual embryos selected from the experiments shown in (I,J). The mutation of the Ci-Bra binding site is shown at the bottom of (L). (M) Sequence-unbiased truncation analysis of the Ci-Noto4 notochord CRM. Additional constructs used for the truncation and mutation analysis are shown in Figure S4. In the 0.88-kb construct, only the sites that were analyzed are depicted. (N) Embryo electroporated with the 0.88-kb Ci-Noto4 CRM fragment shown in (M). (O) Embryo from the same clutch as (N) co-electroporated with the construct in (N) and the Ci-FoxA-a>Bra construct. (P) Embryo electroporated with the 0.88-kb Ci-Noto4 CRM carrying the mutation of a single Ci-Bra binding site, shown at the bottom. (Q) Embryo co-electroporated with the construct in (P) and the Ci-FoxA-a>Bra construct. Representative cells of stained tissues are indicated by arrowheads, which are color-coded as follows: red, notochord; blue, CNS; green, epidermis and tail epidermal neurons; purple, mesenchyme; orange, muscle; yellow, endoderm. Green arrowheads are used also to indicate papillary neurons in (F,G).
Table 1.
Occurrence and functional requirement of different core T-box binding sites found within notochord CRMs.
Table 2.
Comparison of functional Ci-Bra binding sites with the consensus binding site identified for other Brachyury proteins.
Figure 5.
A Ci-Bra-downstream minimal notochord CRM lacking Ci-Bra binding sites.
Identification of the minimal sequences required for notochord activity of the Ci-ACL notochord CRM. (A) Schematic representation of constructs containing serial truncations and site-directed mutations. (B–E) Low-magnification group microphotographs of embryos electroporated in parallel with either the wild-type 215-bp Ci-ACL notochord CRM (B) or 195-bp truncation (C), a mutated version of the 215-bp construct carrying a mutation in the putative Homeodomain (abbreviated as HD) binding site (D), or the 175-bp truncation (E). Red arrowheads in (D) highlight embryos showing residual notochord staining.
Figure 6.
Developmental onsets of activity of notochord CRMs representative of the multiple- and single-site Ci-Bra targets, and of two notochord CRMs lacking Ci-Bra binding sites.
Time-course experiments were carried out by WMISH on transgenic embryos (A–F) carrying the notochord CRMs indicated on the top right corner of (A, C, E), fixed at the 110-cell (A,C,E) and early gastrula stage (B,D,F). Insets in the bottom right corners display high-magnification views of representative stained embryos. The inset in the top right corner of (F) shows an embryo carrying the −0.6 kb Ci-Noto1>LacZ transgene stained with X-Gal. (G) Graph of a representative time-course experiment, reporting the number of transgenic embryos showing LacZ expression in notochord precursors, for the constructs and stages detailed in the panel. n, number of scored embryos showing hybridization signal. (H) Graph of time-course experiments for the notochord CRMs associated with the late-onset genes Ci-ACL and Ci-β4GalT, as determined by X-Gal staining. Results of three representative experiments were averaged. The number of embryos scored for the Ci-ACL CRM was: neural plate, n = 327; neurula, n = 419; initial tailbud, n = 389; early tailbud, n = 237. The number of embryos scored for the Ci-β4GalT CRM was: neural plate, n = 342; neurula, n = 402; initial tailbud, n = 348; early tailbud, n = 220.
Figure 7.
Applicability of the cis-regulatory mechanisms identified.
(A) A 560-bp notochord CRM (red box) containing Ci-Bra binding sites (yellow vertical bars) was predicted and identified in the 5′-flanking region of Ci-ERM and subsequently truncated to a 362-bp sequence containing two Ci-Bra binding sites. (B–E) Late tailbud Ciona embryos electroporated with the wild-type 362-bp Ci-ERM notochord CRM (B) and its individual (C,D) and double (E) mutant versions. Inset in (E): a representative embryo showing staining only in mesenchyme cells. (F–O) Identification of functional Ci-Bra binding sites in notochord CRMs linked to genes that had not been previously associated with Ci-Bra. (F) Genomic location, structure, and mutation analysis of the notochord CRM identified upstream of Ci-laminin gamma-1 (Ci-lamc1). (G,H) Low-magnification microphotographs of embryos electroporated with the 122-bp Ci-lamc1 notochord CRM, wild-type (G) or (H) carrying mutations affecting two Ci-Bra binding sites (bottom construct in (F)). Insets in (G,H): individual embryos shown at a higher magnification. (I,J) Late tailbud embryos from the same clutch hybridized in situ with the Ci-lamc1 antisense RNA probe. (I) Control wild-type embryo. (J) Embryo electroporated with the Ci-FoxA-a>Bra construct. Inset: late tailbud embryo co-electroporated with the construct in (G) and with Ci-FoxA-a>Bra. (K) Genomic location, structure, and mutation analysis of the notochord CRM identified within the Ci-ABCC10 genomic locus. (L,M) Low-magnification microphotographs of embryos electroporated with the 772-bp wild-type Ci-ABCC10 CRM cloned on its endogenous promoter. Insets: individual embryos shown at a higher magnification. Inset in (L) displays a representative stained embryo. (N,O) Embryos hybridized in situ with the Ci-ABCC10 antisense RNA probe. (N) Control wild-type embryo. (O) Embryo electroporated with the Ci-FoxA-a>Bra construct. Inset: individual embryo co-electroporated with the construct in (L) and with the Ci-FoxA-a>Bra construct. The sequences of the additional putative Ci-Bra binding sites found in the 772-bp Ci-ABCC10 CRM are, from left to right: TCTCAC, TTTCAC, TGTCAC, and TGCCAC. Red arrowheads indicate representative notochord cells, aqua arrowheads indicate the territory where Ci-Bra is ectopically expressed.
Figure 8.
Validation of the in vivo occupancy of the Ci-Bra target notochord CRMs through ChIP assays.
(A) Graph of ChIP-qPCR experiments carried out using the antibody shown in (B) on various sequence stretches from the Ci-FCol1 genomic locus and its neighboring gene, KH.C7.121. Orange diamonds indicate peaks of Ci-Bra occupancy, measured as fold change between Ci-Bra-immunoprecipitated and IgG-immunoprecipitated chromatin. A red rectangle shows the location of the CiFCol1 notochord CRM. Colored boxes below the graph symbolize exons (green, gene KH.C7.121; grey, 5′-UTR; pink, Ci-FCol1); the lines connecting them represent introns. (B) Fluorescence microphotograph of a late tailbud Ciona embryo carrying the Ci-Bra>GFP transgene [27], immunostained with the Ci-Bra-specific antibody [46]. The nuclei of the 40 notochord cells are stained by the antibody in red; GFP expression is detected in a subset of the 40 notochord cells, owing to mosaic incorporation of the transgene. Nuclei in tissues other than the notochord are stained with 4′,6-diamidino-2-phenylindole (DAPI) (blue). (C) Graph showing the results of ChIP-qPCR experiments on the notochord CRMs indicated on the x-axis, reporting the enrichment of the immunoprecipitated DNA over the input. All p-values were <0.01, with the exception of Ci-ACL and Ci-18S RNA, which had non-significant p-values (negative controls). The primers used are listed in Table S3. Light blue bars, negative controls, i.e. ChIP assays executed with IgG. Magenta, ChIP performed with the Ci-Bra antibody shown in (B).
Figure 9.
Effects of the removal of functional Ci-Bra binding sites on the developmental onset of a multiple-site notochord CRM.
(A–E) Low-magnification microphotographs of Ciona embryos carrying either the wild-type Ci-lamc1>LacZ notochord CRM transgene (A,D) or a mutated version of this CRM lacking its distal-most Ci-Bra binding site (Ci-lamc1-T1M>LacZ; see Figure 7F) or its proximal Ci-Bra binding site (Ci-lamc1-T4M>LacZ; see Figure 7F), fixed at the early gastrula (A–C) and mid-gastrula (D,E) stages, and stained with X-Gal. Insets in the low right corners show high-magnifications view of representative stained embryos. The inset in the top right corner of (D) shows an embryos carrying the 434-bp Ci-Bra>LacZ transgene, which was used to label the notochord lineage at this stage for comparison. Red and white arrowheads: notochord staining, or lack thereof, respectively; orange: muscle staining. (F) Graph showing the percentage of embryos showing notochord staining as a fraction of the total number of stained embryos scored. Blue bars, embryos at the early gastrula stage; brown, embryos at mid-gastrula; green, embryos at the late tailbud stage (see Figure S6). The number of embryos scored for each construct was: WT, early gastrula, n = 646; mid-gastrula, n = 672; mid-tailbud, n = 502. T1M, early gastrula, n = 705; mid-gastrula, n = 994; mid-tailbud, n = 449. T4M, early gastrula, n = 756; mid-gastrula, n = 700; mid-tailbud, n = 431. Statistically significant p-values are indicated by asterisks.
Figure 10.
Different structural features of notochord CRMs associated with early, middle, and late-onset Ci-Bra targets.
Summary of the developmental onsets of different Ci-Bra targets plotted against the developmental stages indicated at the bottom. Red bars symbolize minimal notochord CRMs; yellow vertical bars indicate functional Ci-Bra binding sites; a blue and a grey square represent binding sites for transcription factors other than Ci-Bra. CRM structures and sizes are approximate, for simplicity. (Left) The notochord CRMs associated with the early-onset genes Ci-thbs3, Ci-FCol1, Ci-ERM, and Ci-lamc1 are controlled by Ci-Bra through multiple functional binding sites. (Center) The notochord CRMs associated with the middle-onset genes Ci-Noto1, Ci-Noto8, Ci-Noto4, Ci-Noto9, and Ci-ABCC10 are controlled by Ci-Bra through one main binding site. (Right) The minimal notochord CRMs associated with the late-onset genes Ci-ACL and Ci-β4GalT are controlled indirectly by Ci-Bra via a relay mechanism.