Proliferation and patterning are mediated independently in the dorsal spinal cord downstream of canonical Wnt signaling

Abstract

Canonical Wnt signaling can regulate proliferation and patterning in the developing spinal cord, but the relationship between these functions has remained elusive. It has been difficult to separate the distinct activities of Wnts because localized changes in proliferation could conceivably alter patterning, and gain and loss of function experiments have resulted in both types of defects. To resolve this issue we have investigated canonical Wnt signaling in the zebrafish spinal cord using multiple approaches. We demonstrate that Wnt signaling is required initially for proliferation throughout the entire spinal cord, and later for patterning dorsal progenitor domains. Furthermore, we find that spinal cord patterning is normal in embryos after cell division has been pharmacologically blocked. Finally, we determine the transcriptional mediators of Wnt signaling that are responsible for patterning and proliferation. We show that tcf7 gene knockdown results in dorsal patterning defects without decreasing the mitotic index in dorsal domains. In contrast, tcf3 gene knockdown results in a reduced mitotic index without affecting dorsal patterning. Together, our work demonstrates that proliferation and patterning in the developing spinal cord are separable events that are regulated independently by Wnt signaling.

Figure 1

Canonical Wnt signaling regulates proliferation and patterning in the developing spinal cord. (A) Expression of top:dgfp, a β-catenin dependent reporter, is widespread in the neural keel (dotted line) at 15 hpf, with a dorsal bias. (B) Expression of top:dgfp is restricted to the dorsal spinal cord at 24 hpf. (C) Immunofluorescence of unphosphorylated β-catenin is localized to the dorsal spinal cord at 24 hpf. (D) Expression of wnt1 is found in the roof plate of the spinal cord at 24 hpf. (E,F) Immunostaining for phospho-histone H3 is reduced at 24 hpf following Dkk1 overexpression at 12 hpf. (G) The mitotic index is reduced in dorsal, intermediate and ventral domains following Dkk1 overexpression. *=p<0.05 by T-test; error bars indicate +/− SEM. (H–I) Expression of radar and (J–K) msxc is eliminated following Dkk1 overexpression. (L–M) Expression of pax3 is reduced in Dkk1-expressing embryos. (N–O) Expression of dbx2 shifts into the dorsal domain in Dkk1-expressing embryos. (P–Q) Expression of nkx6.2 is unaffected in Dkk1-expressing embryos. (R–S) HuC/D immunostaining is unaffected in Dkk1-expressing embryos. (A–D, R–S) show 12 μm cross-sections, and (E–F, H–Q) show lateral mounts. Scale bars = 10μm.

Figure 2

Reduction of cell proliferation does not cause spinal cord patterning defects. (A–B) Anti-phospho histone H3 staining reveals a decrease in proliferative cells in the HUA-treated spinal cord. (C–D) Expression of pax3 is unaffected in embryos treated with HUA for 12 hours. (E–F) The dbx2 expression domain is unchanged in HUA embryos although an overall decrease in signal was observed. Expression in the dorsal epidermis is non-specific. (G–H) Expression of nxk6.2 is unaffected in HUA treated embryos. Lateral mounts are shown. Scale bar = 10μm.

Figure 3

Tcf7 is required for dorsal progenitor patterning. (A) At 15 hpf tcf7 is expressed widely throughout the neural keel (dotted line), with a dorsal bias. (B) At 24 hpf tcf7 is expressed in the dorsal spinal cord as well as the median fin fold. (C) There is no significant decrease in the mitotic index of the dorsal and intermediate domains in tcf7 morphants, but mitosis in the ventral domain is significantly reduced. *=p<0.05 by T-test, error bars indicate +/− SEM. (D–E) Expression of wnt1 and (F–G) radar mRNA is unaffected in tcf7 morphants. (H–I) Expression of msxc and (J–K) pax3 is reduced in tcf7 morphants. (L–M) Expression of dbx2 shifts into the dorsal domain in tcf7 morphants. (N–O) Ventral expression of nkx6.2 is unaffected in tcf7 morphants. (A–B) 12 μm cross-sections are shown. (D–O) Lateral mounts are shown. Scale bars = 10μm.

Figure 4

Tcf3 is required for proliferation but not for dorsal progenitor patterning. (A–D) Both tcf3a (A,C) and tcf3b (B,D) are expressed at higher levels in tcf7 morphants. (E) The mitotic index in dorsal, intermediate and ventral domains is significantly reduced in tcf3a/tcf3b double morphants. *=p<0.05 by T-test, error bars indicate +/− SEM. (F–G) Expression of msxc and (H–I) pax3 is unaffected in tcf3a/tcf3b double morphants. (J–K) Expression of dbx2 is eliminated in tcf3a/tcf3b double morphants, but the intermediate progenitor domain is not shifted dorsally as indicated by iro3 expression (L–M). (A–D) 12 μm cross-sections are shown. (F–M) Lateral mounts are shown. Scale bars = 10μm.

Figure 5

Transcriptional activation by Lef/Tcf factors mediates progenitor patterning in the spinal cord. (A–B) Expression of pax3 and (C–D) msxc is reduced in ΔTcf-expressing embryos. (E–F) Expression of dbx2 shifts into the dorsal domain of the spinal cord in ΔTcf-expressing embryos, and is ectopically present in the somites. (G–H) Expression of nkx6.2 is unaffected in ΔTcf-expressing embryos. (A–D, G–H) Lateral mounts are shown. (E–F) 12 μm cross-sections are shown. Scale bars = 10μm.