RACK1 is a novel interaction partner of PTK7 that is required for neural tube closure.
RACK1 is an evolutionarily conserved intracellular adaptor protein that is involved in a wide range of processes including cell adhesion and migration; however, its role in vertebrate development is largely unknown. Here, we identify RACK1 as a novel interaction partner of PTK7, a regulator of planar cell polarity that is necessary for neural tube closure. RACK1 is likewise required for Xenopus neural tube closure. Further, explant assays suggest that PTK7 and RACK1 are required for neural convergent extension. Mechanistically, RACK1 is necessary for the PTK7-mediated membrane localization of Dishevelled (DSH). RACK1 facilitates the PTK7-DSH interaction by recruiting PKCδ1, a known effector of DSH membrane translocation. These data place RACK1 in a novel signaling cascade that translocates DSH to the plasma membrane and regulates vertebrate neural tube closure.
PubMed ID: 21350015
Article link: Development.
Genes referenced: dvl2 gnb2l1 myc ptk7
Morpholinos referenced: gnb2l1 MO1 gnb2l1 MO2
Article Images: [+] show captions
|Fig. 1. RACK1 interacts with PTK7 in co-precipitation assays and is required for Xenopus neural tube closure. (A) RACK1 co-precipitates a deletion mutant of PTK7 that lacks the extracellular domain (ΔEPTK7) in lysates of embryos injected with 300 pg HA-tagged RACK1 RNA and 500 pg myc-tagged ΔEPTK7 RNA. WB, western blot. (B) In vitro translated myc-tagged ΔEPTK7 protein precipitates in vitro translated HA-tagged RACK1 protein. (C) RACK1 MO verification in radioactive in vitro translation assays. MO1 targets the ATG and MO2 the 5′UTR of the Xenopus RACK1 mRNA. Both MOs inhibited the in vitro translation of full-length RACK1 containing the 5′UTR (xRACK1), whereas a control MO (co MO) did not affect protein translation. The translation of an HA-tagged RACK1 construct lacking the 5′UTR (xRACK1-HA), which contains only the MO1 binding site, is inhibited by MO1 but not MO2. The radioactive, translated RACK1 proteins (S35 RACK1) are shown. (D) Embryos injected with 20 ng MO1 or MO2 show neural tube closure defects, whereas an embryo injected with 20 ng co MO displays a closed neural tube. (E,G,H) Bar charts that summarize the average percentage of neural tube closure phenotypes from three independent experiments in D. The total number of injected embryos from all three batches and the s.e.m. are indicated for each condition/column. Neural tube closure defects were determined at stage 17/18. ce, uninjected control embryos. (E) Neural tube closure defects caused by 20 ng RACK1 MO1 or 20 ng MO2 are significantly different from phenotypes in embryos injected with 20 ng control MO (*, P<0.005, Student's t-test). (G) Human RACK1 (hRACK1) RNA (300 pg) rescues the neural tube closure defects caused by 10 ng RACK1 MO1. Asterisk indicates that neural tube closure defects are significantly different from effects in embryos injected with MO1 (P<0.005, Student's t-test). (H) Injection of 300 pg xRACK1-HA RNA lacking the MO2 binding site rescues the neural tube closure defects caused by 20 ng MO2. Asterisk indicates neural tube closure defects significantly different from effects in embryos injected with MO2 (P<0.01, Student's t-test). (F) The translation of an HA-tagged Xenopus RACK1 construct (xRACK1) was inhibited by RACK1 MO1 but not by a control MO in Xenopus lysates. By contrast, the translation of an HA-tagged human RACK1 construct (hRACK1-HA) was not affected by RACK1 MO1 or the control MO. Embryos were injected at the one-cell stage with 10 ng of the respective MOs in combination with 300 pg RACK1 RNA. At stage 11, cells were lysed and protein expression was detected by western blotting using anti-HA antibodies. Actin expression served as a loading control.|
|Fig. 2. RACK1 loss of function leads to similar neural tube closure defects as PTK7 loss of function. (A) Time-lapse images of Xenopus embryos injected with 10 ng control MO, RACK1 MO1 or PTK7 MO analyzed at stages 15, 18 and 23 show a similar delay in neural tube closure in the RACK1 and PTK7 morphants and the embryos remain shorter than the control. Embryos were injected at the one-cell stage and images were taken at 30-minute intervals from stage 13 to stage 23. (B) The expression of SOX2 (a pan-neural marker) and PAX3 (a lateral neural plate marker) is expanded in stage 17-18 embryos injected with 10 ng RACK1 MO1 or PTK7 MO compared with control MO. (C,D) Transverse sections of stage 18 embryos injected with 20 ng RACK1 MO or PTK7 MO. Embryos were injected in one blastomere at the two-cell stage; the dashed line demarcates the injected from the control side of the embryo. Phalloidin staining (red) marks the actin cytoskeleton. Actin accumulation, indicating hinge point formation (white arrows), is seen on the injected and control sides. However, the neural plate remains wide on the injected side (yellow arrow), indicating a defect in convergent extension. (E-H) Neural convergent extension was analyzed in ectodermal explants (animal caps) injected with 450 pg XBF-2 RNA. Explant elongation was analyzed when control embryos reached stage 20. (E) Uninjected control caps did not elongate. (F) Explants injected with XBF-2 and 20 ng control MO showed various degrees of elongation. (G) Co-injection of 20 ng RACK1 MO inhibited explant elongation. (H) Co-injection of 20 ng PTK7 MO inhibited elongation of explants. (I) The percentage of elongated explants from the experiments illustrated in E-H. Black indicates a strong elongation, gray a mild elongation. The number of explants is indicated for each condition.|