XB-ART-47612Development. December 1, 2013; 140 (23): 4763-75.
Par3 controls neural crest migration by promoting microtubule catastrophe during contact inhibition of locomotion.
There is growing evidence that contact inhibition of locomotion (CIL) is essential for morphogenesis and its failure is thought to be responsible for cancer invasion; however, the molecular bases of this phenomenon are poorly understood. Here we investigate the role of the polarity protein Par3 in CIL during migration of the neural crest, a highly migratory mesenchymal cell type. In epithelial cells, Par3 is localised to the cell-cell adhesion complex and is important in the definition of apicobasal polarity, but the localisation and function of Par3 in mesenchymal cells are not well characterised. We show in Xenopus and zebrafish that Par3 is localised to the cell-cell contact in neural crest cells and is essential for CIL. We demonstrate that the dynamics of microtubules are different in different parts of the cell, with an increase in microtubule catastrophe at the collision site during CIL. Par3 loss-of-function affects neural crest migration by reducing microtubule catastrophe at the site of cell-cell contact and abrogating CIL. Furthermore, Par3 promotes microtubule catastrophe by inhibiting the Rac-GEF Trio, as double inhibition of Par3 and Trio restores microtubule catastrophe at the cell contact and rescues CIL and neural crest migration. Our results demonstrate a novel role of Par3 during neural crest migration, which is likely to be conserved in other processes that involve CIL such as cancer invasion or cell dispersion.
PubMed ID: 24173803
PMC ID: PMC3833433
Article link: Development.
Grant support: MR/J000655/1 Medical Research Council , G1000080 Medical Research Council , MRC_G1000080 Medical Research Council
Genes referenced: cdh2 ctnnd1 foxd3 mapk1 pard3 rac1 ran rho rho.2 snai2 sox10 sox9 twist1
Morpholinos referenced: Pard3 MO1 cdh2 MO1 trio MO1
Article Images: [+] show captions
|Fig. 1. Par3 is required for NC migration in Xenopus. (A) Western analysis of Par3 using protein extracts from Xenopus embryos injected with ControlMO or Par3MO. Band intensity is shown relative to ControlMO and normalised to the loading control (MAPK). Arrows indicate individual Par3 and MAPK bands. Error bars show s.d. The experiment was repeated three times; the difference between control and Par3MO was significant (P<0.005). (B) Dorsal view of stage 16 Xenopus embryo injected unilaterally with Par3MO (asterisk) and processed for in situ hybridisation against Snail2, FoxD3, Sox9 and Sox10. (C) Par3MO does not affect NC induction (n=181). n.s., not significant. (D-F) Lateral view of Xenopus embryos showing Twist expression for ControlMO (D), Par3MO (E) or Par3MO co-injected with mRNA for zebrafish Par3GFP (F). Asterisk marks the eye. (G) Percentage of embryos with normal NC migration. Par3MO, n=45; ControlMO, n=24; P<0.001; Par3MO+Par3GFP, n=34; P<0.001. (H) Distance migrated by NC cells relative to mean migration in control embryos. Migration is reduced in Par3MO-injected embryos (P<0.001) but co-injection of Par3GFP with Par3MO rescues migration (P<0.001). (I,J) Single frames from time-lapse movies showing control (I) and Par3MO-injected (J) explants and Delaunay triangulation at 0 and 8 hours. Scale bar: 100 μm. (K) Dispersion between cells increases over time in control but not in Par3MO-injected explants (n=8 explants for each condition, more than 30 cells analysed per explant; *P<0.05, **P<0.01, ***P<0.001). (L) Speed of single NC cells. n=20 per experiment, from three independent experiments; P=0.2146. (M) Persistence of single NC cells. n=20 experiment, from three independent experiments; P=0.4021. (K-M) Error bars indicate s.d.|
|Fig. 2. Par3 is required for NC migration in zebrafish. (A) Dorsal view of 5-somite stage zebrafish embryos injected unilaterally with ControlMO or Par3MO and processed for in situ hybridisation against foxd3; no effect on NC induction was observed (n=72). (B) Schematic representation of zebrafish embryo indicating the regions shown in C-H. (C,D) Lateral views showing foxd3 expression in the trunk of ControlMO-injected (C) or Par3MO-injected (D) 20-somite embryos. Arrows indicate distinct trunk NC streams. (E,F) Lateral views showing heads of 24-hour sox10:egfp embryos injected with ControlMO (E) or Par3MO (F). OV, otic vesicle. Arrows indicate distinct cranial NC streams. Lines indicate that no distinct streams are observed. (G,H) Single frames from time-lapse movies showing one cranial NC stream and corresponding Delaunay triangulation at 0 hours (15 somites) and 6 hours from ControlMO-injected (G) and Par3MO-injected (H) embryos. (I) Distance between cranial NC cells analysed by nearest neighbour computation. n=15 embryos for each condition, with more than 30 cells analysed per explant; **P<0.01, ***P<0.001. Error bars indicate s.d.|
|Fig. 3. Par3 inhibition does not affect cell adhesion in Xenopus or zebrafish. (A-F) Cell adhesion molecules analysed in Xenopus embryos. (A-C) Immunostaining against β-catenin in control (A) or Par3MO-injected NC cells (B). (C) Pixel intensity of β-catenin immunostaining was measured across the contact and normalised to the average value background levels 5 μm away from the contact for each image. There is no difference in pixel intensity between control cells (n=84 contacts) and Par3MO-injected cells (n=44 contacts; P>0.05 at all distances from contact). (D-F) Control (D) or Par3MO-injected (E) NC cells expressing p120-cateninGFP. (F) Pixel intensity analysis as described in C. There is no difference in pixel intensity of p120-catenin immunostaining between control cells (n=24 contacts) and Par3MO-injected cells (n=41 contacts; P>0.05 at all distances from contact). (G-L) N-cadherin immunostaining in control (G-I) or Par3MO injected (J-L) zebrafish embryos. sox10GFP transgenic embryos were used to identify NC cells. Note that N-cadherin staining is not affected by Par3MO in the NC. n=150 embryos, fixed at 24 hpf. (M-P) Cell adhesion is not affected by Par3MO. (M) Schematic representation of the cell reaggregation assay. (N) Control Xenopus NC cells. (O) Control/Par3MO Xenopus NC cells. (P) Control/N-cadherinMO Xenopus NC cells. Scale bars: 10 μm.|
|Fig. 4. Par3 is required for CIL in vitro in Xenopus. (A-D) Explant confrontation assay. (A-C) Single frames from time-lapse movies showing maximum overlap between two control explants (A), two Par3MO-injected explants (B) or one control and one Par3MO-injected explant (C). (D) Overlap area as a percentage of total explant area. Control/control, n=27; Par3MO/Par3MO, n=16; P=0.0026; control/Par3MO, n=9; P=0.0252. (E-G) Protrusions overlap analysis for ControlMO-injected (E) or Par3MO-injected (F) cells expressing mbGFP. Magenta overlay shows the protrusion overlap. (G) Protrusion overlap area as a percentage of cell area. Par3MO, n=40; control, n=50; P=0.0002. (H-J) Cell collision assay. (H) Schematic representation of cell collision. Cell position was analysed at 15-minute intervals to measure velocity following collision between ControlMO-injected (I) or Par3MO-injected (J) cells. Velocity vectors were clustered away from the contact in control cells (n=10; P<0.001) but a change in velocity following collision was not observed in Par3MO-injected cells (n=10; P<0.001). Red arrows indicate the initial velocity vector. *P<0.05, **P<0.01, ***P<0.001. Error bars indicate s.d. Scale bars: 100 μm in A; 10 μm in E.|
|Fig. 5. Par3 is required for CIL in vivo in zebrafish. Single frames from time-lapse movies illustrating a collision between two cranial NC cells (1 and 2) in a zebrafish embryo injected with ControlMO (A,B) or Par3MO (D,E). Images are overlapped with tracks of migratory paths (A,D) or the difference in protrusion between current and previous frames (B,E). Red area indicates new protrusions. (C,F) Velocity vectors of colliding NC cells in ControlMO-injected (C) or Par3MO-injected (F) embryos. Cell position was recorded at 5-minute intervals. Velocity change was clustered in ControlMO-injected cells (n=10; P<0.05), but no difference in the velocity change was observed in Par3MO-injected cells (n=9; P>0.1). Red arrow indicates the initial velocity vector.|
|Fig. 6. Par3 is localised at the contact site of Xenopus NC cells and does not affect stable microtubules. (A-F) Par3GFP localises to contacts between NC cells in clusters (A) and between colliding single cells (B). (C-F) z-projection of confocal images showing immunostaining against endogenous Par3 in cryosections of Xenopus embryos. (C) Par3 is located at contacts between NC cells. (D) mbRFP identifies NC cells and cell contacts; each asterisk indicates a different cell. (E) Merge of I and J. (F) Colocalisation mask of I and J. (G-J) Analysis of microtubules in NC cells. (G,H) Immunostaining against α-tubulin shows the microtubule array in ControlMO-injected (G) and Par3MO-injected (H) cells. (I,J) Immunostaining against acetylated tubulin stains a stable subpopulation of microtubules in ControlMO-injected (I) and Par3MO-injected (J) cells. Scale bars: 10 μm in A; 20 μm in C.|
|Fig. 7. Par3 affects microtubule dynamics at the cell-cell contact in Xenopus. (A-C) EB3GFP-labelled cells were used to analyse microtubule catastrophe. (A,B) ControlMO-injected (A) and Par3MO-injected (B) cells undergoing collision; consecutive frames are presented from time-lapse movies showing one microtubule (time is given in seconds). The boxed area is shown at higher magnification at the cell-cell contact. Arrowhead indicates the highlighted microtubule. Arrow indicates microtubule in contact with the membrane. Circle indicates site of microtubule collapse. (C) Summary of microtubule collapse at the free edge or cell contact. Control cells: n=150 from six independent experiments; *P=0.0167. Par3MO-injected cells: n=100, from six independent experiments; P=0.3. (D-K) plusTipTracker analysis of microtubules at cell-cell contacts. (D-F) Microtubule tracks in ControlMO-injected (D) and Par3MO-injected (E) cells were separated into four groups based on growth speed (slow <26 μm/minute <fast) and lifetime (short-lived <20 seconds <long-lived). (F) Par3MO-injected cells have a smaller percentage of slow, short-lived microtubules at the contact than control cells (*P=0.0276) and a greater percentage of slow, long-lived microtubules (*P=0.0238). (G-K) No difference was found between ControlMO-injected and Par3MO-injected cells in terms of growth rate (H; P>0.05), depolymerisation rate (I; P>0.05), rescue rate (J; P>0.05) or pause duration (K; P>0.05). However, Par3MO-injected cells exhibited a lower catastrophe rate at the contact compared with control cells (G; *P=0.035). A minimum of ten cells and 25 microtubules per cell were analysed from three independent experiments. Error bars indicate s.d.|
|Fig. 8. The Par3 morphant phenotype is rescued with nocodazole or NSC23766 in Xenopus. (A) Dispersion between cells over time under the following treatments: control, Par3MO, Par3MO/nocodazole, Par3MO/NSC23766. n=10 explants for each condition, more than 30 cells analysed per explant; *P<0.05, **P<0.01, ***P<0.001. (B) Speed of cell migration. Control, n=110; Par3MO, n=120; P=0.0503; Par3MO/nocodazole, n=70; ***P=3.751x10-11; Par3MP/NSC23766, n=50; P=0.0660. (C) Persistence. Control, n=100; Par3MO, n=120; P=0.3131; Par3MO/NSC23766, n=50; P=0.7696; Par3MO/nocodazole, n=70; **P=0.0064. (D-H) Explant confrontation assay. Single frames from time-lapse movies showing maximum overlap between control explants (D), Par3MO-injected explants (E), Par3MO-injected explants in nocodazole (F) and Par3MO-injected explants in NSC23766 (G). Scale bar: 100 μm. (H) Overlap area as a percentage of total explant area (n=23; *P<0.05). (I-L) Cell collision assay. Cell positions were analysed at 15-minute intervals to measure the velocity following collision. Velocity change was clustered in control cells (I; n=10; P<0.001) but no velocity change was observed in Par3MO-injected cells (J; n=10; P<0.005). However, velocity vectors were clustered away from the cell contact in Par3MO-injected cells in nocodazole (K; n=10; P<0.001) and Par3MO-injected cells in NCS23766 (L; n=10; P<0.001). Red arrows indicate the initial velocity vector.|
|Fig. 9. Trio interacts with Par3 and TrioMO rescues Par3 morphants in Xenopus. (A) Co-immunoprecipitation of Trio and Par3 in cells expressing TrioHA and Par3GFP. Following pull-down with the HA tag, western blots against HA showed a band at 320 kDa corresponding to TrioHA and western blots against GFP showed a band at 250 kDa corresponding to Par3GFP. Controls using no transfected cells or control IgG antibody showed no HA or GFP staining. Input lanes show that TrioHA and Par3GFP were present and ran at 320 kDa and 250 kDa, respectively. (B-E) Colocalisation analysis in Xenopus NC cells. Par3RFP (B) and TrioGFP (C) are both found at cell-cell contacts, where they colocalise (D). (E) Magnification and colocalisation mask area indicated in D. Scale bar: 20 μm. (F) Dispersion analysis in control (black), Par3MO (red) or Par3Mo/TrioMO (white). n=10 explants for each condition, more than 30 cells analysed per explant; *P<0.05, **P<0.01, ***P<0.001. (G) Speed of migration. Par3MO, n=70; P=0.4642; Par3MO+TrioMO, n=90; P=0.2713. (H) Persistence. Par3MO, n=70; P=0.4637; Par3MO+TrioMO, n=90; P=0.1126. (I-K) Cell collision assay. CIL was analysed with the cell collision assay previously described. (I) Normal CIL behaviour in control cells. n=10; P<0.001. Red arrows indicate the initial velocity vector. (L-P) Twist expression in control embryos (L) or those injected with Par3MO (M) or Par3MO+TrioMO (N). Asterisk indicates the eye. (O) Percentage of embryos with normal NC migration. Control, n=9; Par3MO, n=13; **P<0.01; Par3MO+TrioMO, n=18; P<0.05. (P) Distance migrated by NC cells relative to mean migration in control embryos. Migration is reduced in Par3MO-injected embryos (***P<0.001) but co-injection of TrioMO with Par3MO increases migration (***P<0.001). Error bars indicate s.d.|
|Fig. 10. TrioMO rescues the Par3MO-induced decrease in microtubule catastrophe and increase in Rac activity at contacts in Xenopus. (A-D) EB3GFP-labelled cells were manually analysed for microtubule catastrophe. (A-C) ControlMO-injected (A), Par3MO-injected (B) and Par3MO+TrioMO-injected (C) cells undergoing collision; consecutive single frames are presented from time-lapse movies showing one microtubule. The boxed area is shown at higher magnification at the cell-cell contact. Arrowhead indicates the highlighted microtubule. Arrow indicates microtubule in contact with the membrane. Circle indicates site of microtubule collapse. (D) Microtubule catastrophe. Controls versus Par3MO, n=5, *P=0.0258; control versus Par3MO+TrioMO, n=5, P=0.3070. (E) Analysis of EB3GFP-labelled cells using plusTipTracker. *P<0.05. (F,G) FRET analysis of Rac activity in NC. (F) Rac FRET efficiency for control, Par3MO- and Par3MO/TrioMO-injected cells. (G) Rac FRET efficiency at the cell contact. Fifteen cells were analysed per condition from a minimum of four individual explants. *P<0.05. Error bars indicate s.d.|
|Fig. 11. Par3 controls CIL by inhibiting Trio/Rac and promoting microtubule catastrophe at cell-cell contacts. (A) CIL is required during NC development for directional and collective cell migration. (B) Rac1-Trio activation in the lamellipodia of migrating cells, promoting microtubule stability. (C) Upon contact, Par3 and other proteins such as N-cadherin are localised to the cell-cell contact. Par3 sequestration of Trio at the cell-cell contact to inhibits Rac1-Trio, leading to microtubule catastrophe. Rac1-Trio activation is biased toward another part of the cell, promoting microtubule stability and lamellipodia extension there. (D) The change in Rho-GTPase activity alters cell polarity. A new lamellipodium develops and the cell changes its direction of migration to move away from the contact, completing the process of CIL. (E) Activation of Par3 at the cell contact leads to microtubule depolymerisation. (F) In the absence of Par3, Trio and Rac remain active and microtubule catastrophe is impaired at the cell contact.|
|Fig. S1. Par3 is required in neural crest cells. (A-B) Grafts of fluorescently labelled NC cells into control host embryos. NC were injected with controlMO (A) or Par3MO (B). (C) Percentage of embryos with migratory grafted NC cells. Par3MO, n=18; ControlMO, n=19; p<0.05. (D) Average distance of NC migration for each grafted embryo. ControlMO, n=19; Par3MO, n=18; p=0.0035, error bars show mean and s.d.|
|Fig. S2. Effect of nocodazole on microtubule array in NC cells. (A-C) Immunostaining against α-tubulin shows the microtubule array in control NC cells (A) or NC cells treated with 13nM (B) or 65nM (C) of nocodazole. Note that in cells treated with the nocodazole concentration used to rescue Par3MO (13nM, panel B) microtubules are not completely depolymerized.|
|Fig. S3. Nocodazole and NSC23766 can restore normal dispersion in Par3MO NC cells. (A-D) Dispersion assay in control NC cells (A), Par3MO NC cells (B), Par3MO NC cells treated with nocodazole (C) and Par3MO NC cells treated with Rac inhibitor NSC23766 (D). (E-G) Dispersion assay in control NC cells (E), control NC cells treated with nocodazole (F) and control NC cells treated with Rac inhibitor NSC23766 (G). Note that nocodazole alone does not have an effect on NC dispersion. (H) Quantification of NC dispersion from the experiment depicted in E-G.|
|Fig. S4. Efficiency of Trio MO. Embryos were injected unilaterally with 5�UTRTrio-GFP and membraneCherry mRNA (A, D, G), together with TrioMO (B, E, H) or a control MO (C, F, I). Pictures are shown for bright field (A-C), GFP (D-F) or RFP (G-I) fluorescence. Note that injection of Trio MO leads to a strong inhibition of GFP fluorescence, indicating an efficient decrease of Trio- GFP protein.|
|Fig. S5. Specificity of Trio MO. Embryos were injected at the two cell stage as indicated, fixed at stage 25 and migration of neural crest cells was analysed by in situ hybridization against Xtwist. (A, B) Injection of TrioMO leads to a clear inhibition of neural crest migration at the injected side. (C, D) This inhibition of neural crest migration was rescued by co-injection of human Trio that is not recognized by Trio MO. (E) Quantification of inhibition in neural crest migration produced by TrioMO and its rescue by co-injection of human Trio DNA. This efficient rescue of TrioMO by Trio DNA shows the specificity of Xenopus TrioMO used in this work.|
|Fig. S6. Trio inhibition can restore normal dispersion in Par3MO cells. (A-D) Dispersion assay in control NC cells (A), Par3MO NC cells (B), Par3MO NC cells co-injected with TrioMO (C). (D-E) Dispersion assay in control NC cells (D) and TrioMO NC cells (E). Note that inhibition of Trio alone does not affect NC dispersion. (F) Quantification of NC dispersion from the experiment depicted in D-E.|