April 1, 2005;
Xenopus Id3 is required downstream of Myc for the formation of multipotent neural crest progenitor cells.
cells, a population of proliferative, migratory, tissue
-invasive stem cells, are a defining feature of vertebrate embryos. These cells arise at the neural plate border
during a time in development when precursors of the central nervous system
and the epidermis
are responding to the extracellular signals that will ultimately dictate their fates. Neural crest
progenitors, by contrast, must be maintained in a multipotent state until after neural tube
closure. Although the molecular mechanisms governing this process have yet to be fully elucidated, recent work has suggested that Myc
functions to prevent premature cell fate decisions in neural crest
forming regions of the early ectoderm
. Here, we show that the small HLH protein Id3
is a Myc
target that plays an essential role in the formation and maintenance of neural crest
stem cells. A morpholino-mediated ''knockdown'' of Id3
protein results in embryos that lack neural crest
. Moreover, forced expression of Id3
maintains the expression of markers of the neural crest
progenitor state beyond the time when they would normally be downregulated and blocks the differentiation of neural crest
derivatives. These results shed new light on the mechanisms governing the formation and maintenance of a developmentally and clinically important cell population.
[+] show captions
Fig. 1. Id3 is a Myc target in the neural crest. (A) In situ hybridization showing the expression of Xenopus Id3 at stage 14 (left) and stage 19 (right). Strong expression is seen in both the neural crest-forming region of the neural plate border, and in the transverse neural fold, at open neural plate stages. (B) Following MO-mediated depletion of Myc protein, expression of Id3 is greatly diminished in neural crest-forming regions on the injected side of the embryo (arrowheads). (C,D) Id3 rescue of Myc-depleted embryos. (C) In situ hybridization showing loss of Slug expression on Myc-depleted side of the embryo (arrowhead). (D) Expression of Slug is substantially rescued when mRNA encoding Id3 is subsequently injected into the Myc depleted side of the embryo. Red staining denotes expression of the lineage tracer β-galactosidase. Arrowheads indicate the injected side of the embryo. (E) Western blot analysis demonstrating that Myc and Id3 are co-expressed in the human glioma cell line T98G. Neither protein is expressed in adult human brain at detectable levels. (F) ChIP assay demonstrating a direct interaction between Myc and the Id3 promoter. Chromatin was extracted from logarithmically growing T98G cells and protein-DNA complexes were immunoprecipitated with a polyclonal antibody against Myc (α-Myc) or normal rabbit immunoglobulins (NRIg). Precipitated DNA was analyzed by PCR with primers from the Id3 and, as negative control, the OLR1 promoters. PCR was also performed in the absence of DNA (H2O) and from 0.01% of the total protein-DNA complexes used in each immunoprecipitation (input).
Fig. 2. Id3 is required for neural crest precursor formation. (A) Western blot of lysates prepared from control embryos or embryos injected with an epitope tagged Id3 and then further injected with control or Id3 MOs demonstrates successful Id3 depletion. (B) Western blot of lysates from injected embryos demonstrating that although Id3 MOs efficiently deplete Xenopus Id3 protein, they do not deplete the human Id3 used for rescue experiments. The size difference between Xenopus and human Id3 is primarily due to the presence of five rather than six Myc epitopes, respectively. (C,D) In situ hybridization of embryos injected with Id3 MOs shows loss of Slug (C) and Sox10 (D) expression on the injected side of the embryo (arrowheads). (E) Id3-depleted embryos show reduced or absent migratory neural crest cells on the injected side (arrow), as visualized by Twist expression. (F) Loss of Slug expression in Id3-depleted embryos can be rescued by subsequent injection of human Id3, translation of which is not blocked by the MOs. Arrowhead indicates the injected side, red staining is from lineage tracer β-gal.
Fig. 3. Id3-depleted embryos form an excess of CNS progenitors. (A,B) In situ hybridization of Id3-depleted embryos. In the absence of Id3, the injected side of the embryo shows greatly expanded expression of Sox3 (A, arrowhead), a marker of the proliferating multipotent CNS progenitor pool. Similarly, expression of Opl, a marker of early CNS and border cell fates is also greatly expanded (B). (C) Phosphohistone H3 immunocytochemistry of Id3-depleted embryos. No difference is seen in the number of actively cycling cells on the control versus Id3-depleted (arrowhead) side of the embryo. (D) Whole-mount TUNEL staining of Id3-depleted embryos. No change in the numbers of apoptotic cells is noted on the Id3-depleted side of the embryo (arrowhead) as compared with the control side. Red staining is lineage tracerβ -gal.
Fig. 4. (A) Western blot of whole embryo lysates prepared from progressively older embryos that were injected at the two-cell stage with mRNA encoding Id3 isoforms with epitope tags appended to either their N or C terminus. N-terminally tagged Id3 protein persists for a substantially longer period of development. (B) In situ hybridization of embryos injected with mRNA encoding the stabilized N-terminally tagged Id3 protein. Forced Id3 expression of Id3 leads to a modest increase in the expression of Slug in at the neural plate border on the injected side of the embryo (arrowhead). Cyan staining is lineage tracer β-gal. (C) Siblings of the embryos shown in B show substantially increased expression of Slug at migratory neural crest stages on the injected side of the embryo (left panel, arrows). At these stages expression of Slug has normally been downregulated in migrating neural crest cells, as seen on the control side of the embryo (right panel). (D) Phosphohistone H3 immunocytochemistry of Id3-injected embryos. No difference is seen in the number of actively cycling cells on the control versus Id3 injected (arrowhead) side of the embryo, indicating that the increase in Slug expression is unlikely to be secondary to an increase in cell proliferation. (E) Whole-mount TUNEL staining of Id3-injected embryos. No change in the numbers of apoptotic cells is noted on the Id3-depleted side of the embryo (arrowhead). Red staining is lineage tracer β-gal.
Fig. 5. (A) In situ hybridization showing Sox10 expression just after the onset of neural crest migration. At these stages, Sox10 marks the entire neural crest stem cell population. (B) Double in situ hybridization showing expression of Sox10 (cyan) and N-tubulin (magenta) at stage 28. Sox10 expression at these stages is restricted to glia in the peripheral ganglia and CNS (aqua) as well as to melanoblasts (not shown). (C,D) Forced expression of Id3 leads to the persistence of Sox10 expression in neural crest cells migrating to the pharyngeal pouches on the injected side of the embryo (C, arrows). At this stage, expression of Sox10 has normally been downregulated in all neural crest cells except those committing to a glial or melanocyte fate, as seen on the uninjected side of the embryo (D). (E,F) No significant difference is seen in the number of migratory neural crest precursor cells following forced Id3 expression, as visualized by comparing expressing of Twist on the Id3-injected (E) versus uninjected (F) side of the embryo. (G,H) Overexpression of Slug leads to expanded expression of markers of neural crest precursor cells on the injected side of the embryo (arrowheads) at neural plate stages. (I,J) In contrast to forced Id3 expression, Slug expression does not lead to persistent expression of Sox10 in migrating neural crest cells (arrows indicate location of migratory neural crest cells).
Fig. 6. Neural crest derivatives fail to differentiate in Id3-injected embryos. (A) Forced Id3 expression leads to the formation of substantially fewer melanocytes on the injected side (arrowhead) of embryos that were injected in one cell at the two-cell stage. (B,C) Embryos injected bilaterally with Id3 mRNA (C) show more dramatic deficits in melanocyte formation (arrows) and also have reduced or absent dorsal fins as compared to sibling control embryos (B). (D) Following forced expression of Id3, neural crest cells that populate the branchial arches fail to give rise to chondrocytes, as visualized by Sox9 expression (black arrows) on the injected (D) versus control (E) side of the embryo. Expression of Sox9 in the developing ear (red arrows) is unperturbed by Id3 expression. (F) The differentiation of N-tubulin-expressing primary neurons is also inhibited on the Id3-injected side of the embryo (arrowhead). (G) Rohon-Beard sensory neurons, as marked by expression of brn3, are particularly sensitive to Id3 (injected side marked by arrowhead and β-gal staining, blue). (H) The effects of Id3 in these experiments are directly on the ectoderm. In situ hybridization for muscle actin expression shows that the underlying mesoderm is unperturbed on the injected side of the embryo.
Fig. 7. Promotion of neural crest progenitor fate is a conserved activity of Id family proteins. (A,B) Expression of an N-terminally tagged form of Xenopus Id2 phenocopies the effects of Id3, leading to the persistent expression of Sox10 (red arrows) on the injected (A) versus control (B) side of the embryo. (C-F) Expression of N-terminally epitope tagged forms of human Id2 (C versus D) and human Id3 (E versus F) had identical effects.