|
Fig. 1. Sulf1 is co-expressed with Shh in the floor plate of the neural tube in Xenopus tropicalis and is required for normal Shh signalling. in situ hybridisation shows the normal expression pattern of Sulf1 (A,C,E, G, I, K) and Shh (B, D, F, H, J, L) at stages 15 (A–D) 23 (E–H), and 35 (I–L). Arrows in (A, B, E, F, I, J) indicate level where the vibratome sections were taken at these same stages and are shown in (C, D, G, H, K, L). Sulf1 is expressed in the paraxial mesoderm at Stage 15 (A, C) while Shh is expressed in the floor plate and notochord (B, D). By stage 23, both Sulf1 and Shh are expressed in the floor plate (E–H), and this co-expression is still apparent at stage 35 (I–L). Knockdown of Sulf1 using a splice blocking morpholino oligo shows a reduced level of ptc expression in the neural tube (N, P) as compared to those injected with an equal amount of a control morpholino (M, O).
|
|
Fig. 2. Sulf1 is required for correct DV patterning in the vertebrate neural tube. Immunostaining for Nkx2.2 (A–D), Nkx6.1 (E–H), HB9 (I–L) and Isl1 (M–P) in control (A, B, E, F, I, J, M, N) and Sulf1 knockdown (C, D, G, H, K, L, O, P) X. tropicalis embryos at stages 23 (A–P) and stage 40 (A′–P′). At stage 23, the expression of Nkx2.2 is shifted ventrally in Sulf knockdown embryos (C, D; n=12, 75%) compared with controls (A, B; n=8, 100%). At stage 40, the expression of Nkx2.2 is also shifted ventrally in Sulf knockdown embryos (C′, D′; n=18, 67%) compared with controls (A′, B′; n=11, 100%). Sulf1 knockdown only leads to a small change in Nkx6.1 expression at stage 23 (G, H; n=8, 63%), compared to controls (E, F; n=8, 100%). At stage 40 the expression of Nkx6.1 is shifted ventrally (G′, H′; n=20, 70%), compared to controls (E,′ F′; n=12, 100%). HB9 staining similarly reveals differences between the stages, at stage 23, the expression of HB9 is reduced (K, L; n=12, 75%) compared with controls (I, J; n=7, 100%). Cells positive for HB9 in Sulf1 knockdown embryos increases at stage 40 (K′, L′; n=19, 69%). At stage 23, the expression of Isl1 reduced in the MN domain of Sulf knockdown embryos (O, P; n=8, 63%) compared with controls (M, N; n=12, 100%); whereas at stage 40 Isl1 is not reliably detectable in this domain in Sulf1 knockdowns (O′, P′; n=2, 100%) or in controls (M′, N′; n=5, 100%).
|
|
Fig. 3. Sulf1 is required for normal proliferation in the neural tube. (A, C) Antibody staining for the mitotic cell marker phosphor-Histone3 (pH3) reveals the normal number of cells in mitosis in the neural tube at NF stage 22 (n=11, range from 2 to 4 pH3 positive cells, average 3, median 3, standard deviation 0.77). (B, D) Antibody staining for pH3 when Sulf1 is knocked-down (n=9, range 0 to 2 cells, average 1, median 1, standard deviation 0.71). (E) Graph of the data where the reduction in pH3 positive cells in embryos lacking Sulf1 compared to controls is significant (Student׳s t-test P<0.0001).
|
|
Fig. 4. Sulf1 affects the diffusion of Shh-GFP in embryo explants. Shh-GFP is shown as green, and is co-expressed with the membrane tethered lineage marker, CFP-GPI which is shown as magenta. Sulf1 mRNA was co-injected with membrane RFP, which is shown in yellow. The cartoons in Fig. 4 illustrate the experiments done in the panels below. (A–D) Shh-GFP expressed in a subset of cells labelled with CFP-GPI (magenta) is able to diffuse away from its site of synthesis forming discrete puncta around cells at a distance from its source. (E–F) When Sulf1 is expressed globally, Shh-GFP is less able to diffuse away from its source. When Sulf1 (yellow) is expressed in cells adjacent to a source of Shh-GFP (magenta) (G–J), diffusion of Shh-GFP is completely abolished within the Sulf1 expressing region (K, L). Shh-GFP displays a reduction in its diffusion when it is co-expressed with Sulf1 (M–R). 50⎕M squares are shown at a 10 fold magnification in adjacent panels (D, F, J, L, P, R) revealing that while Shh-GFP forms discrete puncta in controls (D, P), large aggregates of Shh-GFP form when Sulf1 is expressed either globally (F) or co-expressed with Shh-GFP (R). Magnified images show only Shh-GFP which is depicted in white to improve contrast. (S) Fluorescence levels were quantified using the plot profile function in ImageJ and fluorescent intensity is shown as a function of the distance from the source cells in embryos co-injected with Shh-GFP and control mRNA (blue) versus embryos co-expressing both Shh-GFP and Sulf1 (red). Scale bar is 20 μM.
|
|
Fig. 5. Sulf1 is required for the normal distribution of Shh protein in vivo. (A–D) Immunostaining embryos unilaterally injected with a control morpholino with 5E1reveals Shh protein in the notochord (NC) and the neural tube (NT), with highest levels in the floor plate (FP). In these control embryos the level of Shh protein in the neural tube drops off sharply away from the FP and is not detected within the dorsal neural tube (C). (E–H) Unilateral knockdown of Sulf1 in vivo leads to a change in the distribution of Shh protein on the injected side (*) where there is reduced Shh protein detected which is detected much further dorsally in a more diffuse pattern than in controls (compare G with C). (I) Quantification of the immunohistochemistry reveals a marked change in the diffusion of Shh away from its source in the absence of Sulf1 expression. The level of Shh in controls embryos (blue) immediately adjacent to Shh producing cells is high; this drops to almost zero within ~20 μm. In the absence of Sulf1(red), the level of ligand adjacent to producing cells is lower than in controls, but this level remains higher much further from the source, only dropping to zero ~50 μm from the source cells. Graph represents average grey level across the width of the neural tube on the injected side. Area measured for quantification shown. Mean values from a number of samples are shown (CMO n=5, AMO n=7).
|
|
Figure S1.
The Sulf1 antisense morpholino oligonucleotide blocks the splicing of Sulf1 exons 2/3
A) Schematic showing splicing of Sulf1 between exons2 and 3.The Sulf1 antisense morpholino oligonucleotide ((S1MO3) is complimentary to a sequence which spans the splice junction betweeen exon2 and exon3 and acts to inhibit splicing of the pre-mRNA at this site. Inhibiton of splicing results in retention of intron2 within the mature RNA, which introduces a premature a stop codon and results in the formation of a truncated protein. Inclusion or exclusion of intron2 can be detected by PCR using the primer pairs shown by the black and green arrows. Sulf1 CDS shown in blue.
B) The inhibiton of splicing by the Sulf1 antisense morpholino oligonucleotide (S1MO3) can be detected by using PCR primers which span intron 2. The retention of intron2 following blockage of splicing can be detected by a change in the size of the amplicon from primers which span the exon2/3 boundary (black primer pair shown above) (arrow heads, gel lane 3,4 increase in size from 146 to 992 bp). Gel lanes 1,3 and 5 show PCR products from control samples while lanes 2,4 and 6 show the PCR procucts from Sulf1 morphants. Lanes 1 and 2 show amplification of exon2 (green primer pairshown above), lanes 3 and 4 show the products from the primer pair which span exons 2 and 3 (black primer pair shown above) while lanes 5 and 6 show amplification of ODC.
C) The expression of ptc2 is reduced in the neural tube when Sulf1 is knocked down by unilateral injection of the Sulf1 AMO (S1MO3) and rescued by Sulf1 mRNA. The injected side is indicated with an asterix. (a) Embryos injected with a control morpholino oligo into one blastomere at the two-cell stage show normal expression ptc2 in the neural tube and somites at stage 23. (b) Embryos injected with the Sulf1 antisense morpholino oligonucleotide into one blastomere at the two-cell stage show reduced expression ptc2 in the neural tube at stage 23 (61%, N = 77). (c) Embryos injected with the Sulf1 antisense morpholino oligonucleotide together with 1ng of mRNA coding for Sulf1 into one blastomere at the two-cell stage show nearly normal expression ptc2 in the neural tube at stage 23 (27% show reduced ptc2 expression, n = 83). (d) Thick sections show reduced dorsal extent of ptc2 expression in the neural tube (arrows) on the side injected with Sulf1 antisense morpholino oligonucleotide. (e) The dorsal extent of ptc2 expression is rescued when the Sulf AMO is co-injected with 1ng of mRNA coding for Sulf1.
|
|
Cyclopamine treatment
When Shh signalling is blocked using the smoothened inhibitor cyclopamine, the expression of Nkx2.2 is completely lost (n = 20, 100%).The expression of Nkx6.1 is shifted ventrally (n = 14, 93%) in cyclopamine treated embryos, and there are fewer motor neurons progenitors as shown by a loss of HB9+ (n = 15, 93%) and Isl1+(n = 15, 93%) cells in the ventral neural tube. Embryos were mechanically de-membraned at NF stage 9 and cultured in MRS/20 containing 100μM cyclopamine. Embryos were rinsed in series of MRS/20 washes before fixation. Sectioning and immunohistochemistry was carried out as described in the methods.
|
|
Sulf treated heparin has a lower affinity for Shh than untreated heparin.
Shh was expressed in Xenopus embryos and then bound to heparin agarose beads. Shh was competed off the agarose beads either with heparin (at 12.5, 25, 50, or 100 mg/ml) or Sulf treated heparin at the same concentrations. Shh protein that remained bound to the heparin agarose beads was analysed by Western blotting with an anti-body against Shh. This heparin competition assay shows that Sulf treated heparin is less effective at competing for Shh immobilised on heparin agarose and likely has less affinity for Shh then fully sulphated heparin.
Methods:
Shh protein was prepared by injecting Xenopus embryos with 4ng of mRNA encoding for Shh and lysing them in PBS + 0.1% Triton. The lysate was then clarified by centrifuging and subsequently incubated with heparin agarose beads (Sigma) for 30 mins at room temperature. The beads were then washed 5 times with ice cold PBS + 0.1% Triton.
Purified Sulf protein was prepared by injecting Xenopus embryos with 4ng of mRNA encoding for Sulf1-Myc and lysing them in PBS +0.1% Triton. The lysate was then clarified by centrifuging and was incubated with anti-Myc Agarose beads (Sigma) at 5 C for 2 hours. Beads were then washed 5 times with ice cold PBS +0.1% Triton. Control beads were treated in the same way, but were incubated with lysate from uninjected embryos. A sample of beads was retained, and used in a western blot with anti myc (9E10, DSHB) to ensure efficient immunoprecipitation of Sulf1-Myc.
Sulf treated soluble heparin was prepared by dissolving 100mg/ml of Heparin (Sigma) in PBS + 0.1% Triton and incubating this with either Sulf1-Myc beads or control anti-myc beads prepared as above at room temperature for 2-4 hours. After incubation the soluble heparin was serially diluted to the concentrations indicated in the figure.
The heparin competition assay was performed by incubating Shh-heparin beads with either Sulf1-treated or untreated soluble heparin for 30mins at room temperature. After incubation, the Shh-heparin beads were washed 5 times in ice cold PBS+0.1% Triton. The beads were then boiled for 5 min in 5X sample buffer, microfuged, and analysed by PAGE. For the Western blot, goat anti-Shh (N-19, Santa Cruz) was used at 1:1000, ECL conjugated mouse anti goat/sheep (Sigma) was used at 1:2000.
|
|
10E4 immunohistochemistry
Clones of cells in which Sulf1 has been knocked down (green) display an increase in 10E4 immunoreactivity (red)
10E4 antibody staining is used to detect highly sulphated forms of HS and has previously shown to be decreased in cell lines over-expressing Sulf1, and increased in Sulf1 mutant cells. Here, we show that when Sulf1 protein is knocked down there is an increase in 10E4 immunoreactivity can be seen. The panels above show cryosections through the neural tube at stage 22. The top set of panels show a section through an embryo that has been unilaterally injected with SulfMO and GFP mRNA as a lineage tracer; the bottom set was unilaterally injected with control MO and GFP mRNA. The injected clones which in normal embryos would express Sulf1 protein (inset, arrows) are shown at higher magnification. Embryos were injected in one dorsal blastomere at the 4 cell stage with mRNA encoding for nGFP and AMO targeted towards Sulf1, or Control MO.
Methods:
Immunohistochemistry was carried out on stage 24 embryos, injected in one dorsal blastomere at the 4 cell stage with AMOs targeted against Sulf1, or CMO and 100pg of nGFP mRNA. Embryos were fixed for 10mins at room temperature in Sainte-Marie’s fixative (1% acetic acid in Ethanol), rehydrated in PBS, and processed immediately for cryosectioning.
Once sectioned, samples were washed in PBS, and blocked for 1hr at room temperature in PBS+ 0.5%BSA. Samples were incubated in 1:200 anti-10E4 (US Biological) for 24 hours and washed in PBS. Alexa 555 conjugated goat anti mouse (Invitrogen) was used at 1:500 and Alexa488 conjugated anti-GFP (Invitrogen) was used at 1:1,000. Samples were mounted in Vectashield mounting medium with DAPI.
|
