BMP activity is necessary for PhRs specification. (A,B) Confocal sections of control (A) and Tg(hs:dnBmpr1a-CFP);Tg(hsp70:noggin3) double transgenic (B) embryo. Pineal glands are double labeled with the Tg(AANAT2:GFP) transgene (in red) and a HuC/D antibody (in green) at 48 hours. Anterior is upwards. Scale bar: 16 μm. (C-E) Average numbers of Isl1+ neurons, Tg(AANAT2:GFP)+ and Tg(HuC:GFP)+ cells per pineal gland in control, Tg(hs:dnBmpr1a-CFP) transgenic, Tg(hsp70:noggin3) transgenic and Tg(hs:dnBmpr1a-CFP);Tg(hsp70:noggin3) double-transgenic embryos at 48 hours. Error bars represent s.d. **P<0.001; ***P<0.0005 using a t-test. The number (n) of embryos analyzed is noted for each case. (F-G′′) Confocal images of the pineal gland of wild-type host embryos that have received cells transplanted from wild type (F-F′′) or Tg(hs:dnBmpr1a-CFP) (G-G′′) donors. Tg(AANAT2:GFP)+ PhRs are in green and the transplanted cells are shown in red. White arrowheads show transplanted cells with a PhR identity; white arrows indicate transplanted cells not expressing GFP from the Tg(AANAT2:GFP) transgene.

Bmp2a is required for PhR specification. (A-D) Expression of bmp2a in the wild-type pineal at 14, 16, 18 and 20 hours. (E-H′) Activation of the BMP pathway as judged by labeling with an α-PSmad 1/5/8 antibody shown on confocal projections. The presumptive pineal territory is delineated by GFP expression from the Tg(flh:gfp) transgenic line (Concha et al., 2003). Anterior is upwards. Scale bars: 16 μm. (I-O) Average numbers of Isl1+ neurons, Tg(AANAT2:GFP)+, FRet43+, HuC/D+, Tg(HuC:GFP)+, Isl1+/Pax6+ and lhx3+ cells per pineal gland at 48 hours in control, bmp2a and smad5 morpholino-injected embryos. The number (n) of embryos analyzed is noted for each case. Error bars represent s.d. **P<0.001, ***P<0.0005 using a t-test.

BMP activity is sufficient to promote the PhR fate. (A-C) Confocal sections of control, Tg(hsp70:bmp2b) transgenic and Tg(hs:Gal4);Tg(UAS:caBmpr1a) double transgenic embryos at 48 hours. Pineal glands are double-stained for Tg(HuC:GFP) (in green) and FRet43 (in red). Anterior is upwards. Scale bar: 16 μm. (D-I) Average numbers of Isl1+ neurons, Tg(AANAT2:GFP)+, lhx3+, FRet43+, HuC/D+ and Isl1+/Pax6+ cells in the pineal gland of control and transgenic Tg(hsp70:bmp2b) embryos heat-shocked at various stages and analyzed at 48 hours. In E and G, Tg(hs:bmp2b) embryos heat-shocked at 21 hours show a relatively mild but statistically significant effect (P=0.03 and P=0.037, respectively). (J-M) Average numbers of Isl1+ neurons, Tg(AANAT2:GFP)+, Fret43+ and HuC/D+ cells in the pineal gland of control and Tg(hs:Gal4); Tg(UAS:caBmpr1a) double-transgenic embryos at 48 hours. Heat shocks were performed at 16 hours. Error bars represent s.d. *P<0.05, **P<0.001, ***P<0.0005 using a t-test.

BMP activity is necessary and sufficient for the expression of regulators of the PhR fate. (A-D′) Confocal sections of pineal glands, showing expression of neurod1 or otx5 (in red) and GFP (in green) from Tg(AANAT2:GFP) (A,A′,C,C′) and Tg(HuC:GFP) (B,B′,D,D′) transgenic embryos at 24 hours. White arrowheads indicate double-labeled cells. neurod1 and Tg(AANAT2:GFP) are co-expressed in the majority of cells, although single labeled cells are also present. The observation of neurod1+/Tg(AANAT2:GFP)- cells most probably reflects the earlier onset of expression for neurod1 compared with that of the Tg(AANAT2:GFP) transgene (Gothilf et al., 2002). Conversely, neurod1-/Tg(AANAT2:GFP)+ could be PhRs that do not express neurod1 during their life or alternatively more mature cells that have already turned off the gene. Similarly, although all Tg(AANAT2:GFP)+ cells are also otx5 positive, a number of single-labeled otx5+ cells are observed, which is probably due to the early onset of otx5 (see below) compared with the Tg(AANAT2:GFP) transgene. (E-L) Dorsal view of pineal glands stained for neurod1 (E-H) or otx5 (I-L) by in situ hybridization. Both genes start to be expressed at 16 hours. (M-O) Dorsal view of pineal gland from wild-type (M), Tg(hs:dnBmpr1a-CFP); Tg(hsp70:noggin3) double transgenic (N) and Tg(hs:bmp2b) embryos (O) at 24 hours stained for neurod1. (P) Quantification of the data represented in M-O. (Q-S) Dorsal view of pineal glands from wild-type (Q), Tg(hs:dnBmpr1a-CFP); Tg(hsp70:noggin3) double transgenic (R) and Tg(hs:bmp2b) embryos (S) at 24 hours stained for otx5. (T) Quantification of the data represented in Q-S. In M-T, Tg(hs:dnBmpr1a-CFP); Tg(hsp70:noggin3) transgenics were heat shocked at 16 hours, while Tg(hs:bmp2b) were heat shocked at 21 hours. The number (n) of embryos analyzed is noted for each case in P and T. Error bars represent s.d. ***P<0.0005 using a t-test. Anterior is upwards. Scale bars: 16 μm.

BMP regulates the competence to respond to Notch. (A-C) Confocal projections of control (A), Tg(hs:Gal4);Tg(UAS:Notch^intra) transgenic (B) and Tg(hs:Gal4);Tg(UAS:Notch^intra) transgenic bmp2a-morphant (C) embryos at 48 hours. Heat shock was performed at 14 hours. The effects of the various treatments were monitored by GFP immunostaining in the Tg(HuC:GFP) background. (D,E) Average numbers of Tg(HuC:GFP)+ cells in 48 hour control, Tg(hs:Gal4);Tg(UAS:Notch^intra) transgenic, Tg(hs:Gal4);Tg(UAS:Notch^intra) transgenic bmp2a-morphant and bmp2a-morphant (D) embryos or smad5-morphant embryos (E). Heat shocks were performed at 14 hours. The number (n) of embryos analyzed is noted for each case in D and E. Error bars represent s.d. *P<0.05, ***P<0.0005 using a t-test. Anterior is upwards. Scale bar: 16 μm.

BMP regulates Notch target gene expression. (A-I) Dorsal view of pineal glands showing expression of her genes in control (A,D,G), mib (B,E,H) and Tg(hs:dnBmpr1a-CFP);Tg(hsp70:noggin3) transgenic embryos (C,F,I) at 20 hours. For Tg(hs:dnBmpr1a-CFP);Tg(hsp70:noggin3) transgenic embryos, heat shocks were performed at 16 hours. (J-L) Dorsal view of pineal glands from mock-treated (J), DAPT-treated (K) or smad5-morphant (L) embryos. The effects of the various treatments were monitored by in situ hybridization against gfp transcripts in a Tg(Tp1bglob:eGFP) background. Anterior is upwards. Scale bar: 16 μm. (M-N′) Confocal sections of pineal glands of wild-type host having received cells transplanted from wild-type (M-M′) or Tg(hs:dnBmpr1a-CFP) donors (N-N′). Transplanted cells are shown in green, while expression of her4 mRNA is shown in red. White arrowheads highlight transplanted cells expressing her4.

Levels of BMP activity upon induction of the various transgenes. Confocal projections from control (A,A′,E,E′), Tg(hs:dnBMPr1a-CFP) transgenic (B,B′), Tg(hsp70:noggin3) transgenic (C,C′), Tg(hs:dnBMPr1a-CFP);Tg(hsp70:noggin3) double-transgenic (D,D′); Tg(hs:BMP2b) transgenic (F,F′) and Tg(hs:Gal4); Tg(UAS:caBMPr1a) double-transgenic (G-G′) embryos double-labeled with an α-PSmad 1/5/8 antibody (grey) and a Tg(HuC:GFP) transgene at 20 (A-D′) and 24 hours (E-G′). The Tg(HuC:GFP) transgene helps delineating the posterior limit of the pineal gland. The pineal area is indicated with a bracket. Transgenic and control embryos were heat shocked at 16 (A-D′) or 21 hours (E-G′). Anterior is upwards. Scale bars: 16 µm.