Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009 Aug;17(2):222-33.
doi: 10.1016/j.devcel.2009.06.010.

PAR-1 Phosphorylates Mind Bomb to Promote Vertebrate Neurogenesis

Free PMC article

PAR-1 Phosphorylates Mind Bomb to Promote Vertebrate Neurogenesis

Olga Ossipova et al. Dev Cell. .
Free PMC article


Generation of neurons in the vertebrate central nervous system requires a complex transcriptional regulatory network and signaling processes in polarized neuroepithelial progenitor cells. Here we demonstrate that neurogenesis in the Xenopus neural plate in vivo and mammalian neural progenitors in vitro involves intrinsic antagonistic activities of the polarity proteins PAR-1 and aPKC. Furthermore, we show that Mind bomb (Mib), a ubiquitin ligase that promotes Notch ligand trafficking and activity, is a crucial molecular substrate for PAR-1. The phosphorylation of Mib by PAR-1 results in Mib degradation, repression of Notch signaling, and stimulation of neuronal differentiation. These observations suggest a conserved mechanism for neuronal fate determination that might operate during asymmetric divisions of polarized neural progenitor cells.


Figure 1
Figure 1. PAR-1 and aPKC regulate neurogenesis in Xenopus embryos and mammalian neural progenitors
Four-cell embryos were unilaterally injected with LacZ RNA (50–100 pg), indicated RNAs or MOs and subjected to wholemount in situ hybridization with N-tubulin (A–D) or Sox2 (E–H) anti-sense probes at neurula stages. The injected area is identified by β-galactosidase activity (light blue). Dorsal views are shown, anterior is to the left, the injected side is at the top. Arrows point to altered gene expression at the injected side. (A) Three stripes of primary neurons (medial, m, intermediate, i, lateral, l) are indicated. (A, B) PAR-1 RNA (300 pg, doses are per embryo) increased (A), whereas PAR-1BY MO (5 to 10 ng) decreased (B) N-tubulin positive cells on the injected side. (C, D) aPKC-CAAX RNA (30 to 100 pg) inhibited (C), and aPKC-N RNA (4 ng) expanded (D) N-tubulin-positive cells. (A–D) Dotted lines indicate midline. (E–H) Modulation of PAR-1 and aPKC does not influence the pan-neural marker Sox2 (doses are as in A–D). (I, J) Phenotypically active doses of PAR-1 (I, 300 pg) and aPKC-CAAX (J, 100 pg) RNAs do not alter the number of Sox3-positive progenitor cells. GFP-CAAX RNA was used as a lineage tracer. Stage 15 embryos were cryosectioned and immunostained with anti-Sox3C antibodies (red) and GFP-antibodies (green). (I, J) Dotted lines demarcate neural plate (np) and notochord (nc). Scale bar, 100 μm. (K) Summary of data combined for several independent experiments presented in A–D. Frequency of embryos with a change in N-tubulin-positive cells is shown for each experimental group. Numbers of embryos per group are indicated above bars. (L) Sox3+ progenitor analysis in cryosections. Summary of data for embryos injected with aPKC-CAAX (100 pg), PAR-1 (300 pg), tBR (1 ng) RNAs and PAR-1BY MO, COMO (5 to 10 ng). tBR RNA strongly upregulated the number of Sox3-positive progenitors. Mean ratios of Sox3+/DAPI+ cells +/− s. d. are shown. At least 8 embryos were used in each experimental group. *, p<0.01, indicates significant difference from uninjected cells. M, Lack of effect of PAR-1 and aPKC manipulation on Sox3 protein levels in neurula stage (st 15) embryos (doses are as in A–D). BMP7 RNA (500 pg) dramatically reduced Sox3 protein levels.
Figure 2
Figure 2. PAR-1 and aPKC regulate neurogenesis in mammalian neural progenitors
C17.2 cells infected with lentiviruses expressing GFP or different aPKC and PAR-1 constructs were allowed to differentiate for five days. (A) Representative images of cells stained for TuJ1 (red) and DAPI (blue) before and after differentiation. Neuronal differentiation is revealed by bipolar morphology of cells, containing long processes, and TuJ1 staining. PAR-1, PAR-1T560A and aPKC-N stimulated, whereas aPKC-CAAX suppressed neuronal differentiation. Scale bar, 20 μm. (B) Summary of experiments shown in (A). *, p<0.01, indicates significant differences from the control GFP group. (C) Phospho-histone H3 positive cells in C17.2 cultures expressing indicated lentiviral constructs. PAR-1, PAR-1T560A and aPKC-N decreased, whereas aPKC-CAAX increased the number of mitotic cells. Data are from three independent experiments expressed as ratio of PH3+/DAPI+ cells. *, p<0.01 indicates significant differences from the control GFP group. (D) aPKC and PAR-1 regulate Notch reporter activity in neural progenitor cells. C17.2 cells were transfected with aPKC and PAR-1 constructs as indicated, and the pGL3-11CSL-Luc reporter and harvested 24 hrs later for luciferase activity measurement. Relative luciferase activity was normalized to Renilla enzyme activity. (B–E) Data are presented as the means ± s. d. of three independent experiments. *, p<0.05. (E) PAR-1 inhibits Dll1 activity in signal-sending cells. HEK293T cells were transfected with Dll1 and aPKC or PAR-1 constructs, then cocultured for 24 hrs with NIH3T3 cells expressing Notch and pGL3-11CSL-Luc reporter. pGL3-Luc is a control reporter. Each group contained triplicate samples. **, p<0.01, or *, p<0.05, indicate significant differences from the control- or the Dll1-transfected group, respectively.
Figure 3
Figure 3. PAR-1 phosphorylates Mib and regulates its protein levels
(A) PAR-1 interacts with Mib. Levels of Flag-Mib, GFP-PAR-1 and GFP-PAR-1KD were assessed in HEK293T cell lysates 18 hrs after transfection with indicated plasmids, followed by immunoprecipitation (IP) with anti-Flag beads and immunoblotting (IB) analysis. (B) Phosphorylation of Mib and MibC1001S by PAR-1 in the immune complex kinase assay. Autoradiography (top) demonstrates the autophosphorylation of PAR-1 and reveals its kinase activity towards Mib and Mib C1001S. No significant kinase activity is detectable in lysates containing PAR-1KD. Anti-Myc antibodies (bottom) detect similar protein levels in immunoprecipitates. (C) PAR-1 downregulates Mib in Xenopus embryos. Coexpressed GFP is not decreased by PAR-1. Western analysis of lysates of embryos (stage 10.5) that were injected with Myc-Mib (top band) and Myc-PAR-1 and GFP RNAs. Lower panel, a ratio of Myc-Mib/GFP levels; α-tubulin controls loading. (D) Flag-Mib levels are downregulated by myc-PAR-1 in HEK293T cells. Western analysis of cell lysates prepared 48 hrs after cotransfection of cells with indicated plasmids and pEGFP-C3. Lower panel, a ratio of Flag-Mib/GFP levels. (E) PAR-1 stimulates Mib degradation by the proteasome pathway. Transfected HEK293T cells were treated with 10 μM MG132 or DMSO (control) for 6 hrs. MG132 stabilizes Mib and rescues PAR-1 effects. Lower panel, a ratio of Flag-Mib/GFP levels. (F) Endogenous PAR-1 downregulates Mib in vivo. Embryos were injected at the two-cell stage with indicated MOs, Myc-Mib (0.3 ng) and GFP-CAAX (0.45 ng) RNAs. Western analysis of stage 10.5 embryo lysates with indicated antibodies. PAR-1 MO, but not control MO (CO MO) upregulates Mib at gastrula stage, but has no effect on GFP-CAAX or β-catenin levels. Lower panel, a ratio of Myc-Mib/GFP levels.
Figure 4
Figure 4. Regulation of Dll1 ubiquitination by PAR-1
(A) Dll1 ubiquitination is inhibited by PAR-1 in Xenopus ectoderm. Embryos were injected with RNAs at indicated doses per embryo. At stages 10–11, embryo lysates were collected for immunoprecipitation (IP) of Dll1-myc with anti-Myc antibodies and immunoblotted with anti-HA or anti-Myc to visualize ubiquitinated (brackets) or total Dll1, respectively. (B) Mib-dependent Dll1 ubiquitination is inhibited by PAR-1 in HEK-293T cells. Cells were transfected with Dll1-myc, Flag-Mib or Flag-Mib C1001S, HA-ubiquitin and PAR-1 DNAs as indicated. Ubiquitinated Dll1 (brackets) was precipitated from lysates of cells 48 hr after transfection with anti-Myc antibodies and detected with anti-HA antibodies. Lower panels: ratio of ubiquitinated (HA-tagged) Dll1 to total (myc-tagged) Dll1 in immunoprecipitates.
Figure 5
Figure 5. PAR-1 does not affect signaling of a Mib-independent form of Dll1
(A, B) Embryos were injected with indicated RNAs and processed for in situ hybridization with N-tubulin probe as described in Fig. 1. RNAs were injected at 0.3 ng. Dorsal views are shown, with anterior to the left, the injected side is up (light blue). Arrows point to changes in N-tubulin expression. (A) PAR-1 suppressed the inhibitory effect of untagged Dll1 on neurogenesis, but had no effect on the activity of Dll1ΔC-Ub. (B) Quantification of PAR-1 effects on Dll1 and Notch signaling. Numbers of embryos per group are indicated above bars. Data are pooled from several independent experiments. RNA doses are indicated per embryo. (C) PAR-1 does not affect Notch reporter activation by Dll1ΔC-Ub. Signal-sending cells were transfected with PAR-1, Dll1 or Dll1ΔC-Ub constructs, and cocultured with signal-receiving cells that were transfected with Notch and pGL3-11CSL-Luc, as described in Fig. 2D. *, p<0.01, and #, p<0.01, indicate significant differences from control- or the Dll1-transfected group, respectively. Results are presented as the means ± s. d. of three independent experiments, each carried out with triplicate samples per group.
Figure 6
Figure 6. PAR-1 influences Dll1 ubiquitination by phosphorylating Mib
(A) Putative PAR-1 phosphorylation sites in the Mib protein. The HERC2 domain (grey box), zinc finger regions, ZZ-type (ZZ) and ring-type (R), and the ankyrin repeat regions are shown. Alignment of the M2 and M8 Mib region that are critical for PAR-1-dependent degradation. Asterisk marks positively charged residue upstream of S804/S806. Sequence accession numbers: Mm Mib1, NM_144860; Hs Mib1, NP_065825; Dr Mib, AF537301; Xt Mib1, BC167461, Xt Mib2, NP_001123444; Xl Mib, NP_001085805. (B) Mib M2/M8 is insensitive to downregulation by PAR-1. Embryos were injected with indicated RNAs at the two-cell stage for western analysis at stage 10.5. Mib (top band) and PAR-1T560A (bottom band) are detected with anti-Myc antibodies. α-tubulin is a control for loading. (C, D) Decreased phosphorylation of Mib proteins with mutated M2 (C) and M8 (D) sites as compared to wild-type Mib. Arrow points to lack of decrease in M2/M8 levels in presence of PAR-1. (C) M2 mutations are in the context of the full-length protein. (D) Phosphorylation of Mib-M8 mutant is analyzed in the context of a carboxy-terminal fragment of Mib (amino acids 691–1031) which migrates faster than PAR-1. Carboxy-terminal fragments of the wild-type Mib or M8 mutants are compared. The details of the immune complex kinase assay are as in Fig. 3B legend. (E) M8* Mib promotes Dll1 ubiquitination in a PAR-1-independent manner. Embryos were injected with Dll1-myc (0.3 ng), HA-Ub (1 ng), PAR-1 (0.4 ng), Flag-Mib or Flag-M8* Mib (0.3 ng) RNAs as indicated. Dll1 ubiquitination was analyzed as described in Fig. 4A. Compared to Fig. 4A, image was acquired at a different exposure time, since Delta is strongly ubiquitinated by Flag-Mib. Lower panel shows ratios of ubiquitinated (HA-tagged) Dll1 to total (myc-tagged) Dll1 in immunoprecipitates. (F) A model for regulation of neurogenesis by PAR-1 and aPKC. aPKC promotes Notch signaling and negatively regulates PAR-1 activity in neural progenitors. PAR-1 phosphorylates Mib leading to its degradation, suppressed Dll1 monoubiquitination, inhibited Notch signaling and promoted neurogenesis.

Comment in

Similar articles

See all similar articles

Cited by 40 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources