Cousin H and Alfandari D (2004), A PTP-PEST-like protein affects alpha5beta1-int...

XB-ART-4080

Dev Biol 2004 Jan 15;2652:416-32. doi: 10.1016/j.ydbio.2003.09.038.

Show Gene links Show Anatomy links

A PTP-PEST-like protein affects alpha5beta1-integrin-dependent matrix assembly, cell adhesion, and migration in Xenopus gastrula.

Cousin H .

???displayArticle.abstract???
During amphibian gastrulation, mesodermal cell movements depend on both cell-cell and cell-matrix interactions. Ectodermal cells from the blastocoel roof use alpha5beta1 integrins to assemble a fibronectin-rich extracellular matrix on which mesodermal cells migrate using the same alpha5beta1 integrin. In this report, we show that the tyrosine phosphatase xPTP-PESTr can prevent fibronectin fibril formation when overexpressed in ectodermal cells resulting in delayed gastrulation. In addition, isolated ectodermal cells overexpressing xPTP-PESTr are able to spread on fibronectin using the alpha5beta1 integrin in the absence of activin-A induction and before the onset of gastrulation. We further show that while the inhibition of fibrillogenesis depends on the phosphatase activity of xPTP-PESTr, induction of cell spreading does not. Finally, while cell spreading is usually associated with cell migration, xPTP-PESTr promotes ectodermal cell spreading on fibronectin but also reduces cell migration in response to activin-A, suggesting an adverse effect on cell translocation. We propose that xPTP-PESTr overexpression adversely affect cell migration by preventing de-adhesion of cells from the substrate.

???displayArticle.pubmedLink??? 14732402
???displayArticle.link??? Dev Biol
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: adm ag1 bcar1 bcr chrd fn1 myc odc1 pacsin2 ptpn12 ptpru pxn sox2 tbxt
???displayArticle.antibodies??? Bcar1 Ab1 Fn1 Ab3 Itgb1 Ab3 Pacsin2 Ab1

???attribute.lit??? ???displayArticles.show???

	Fig. 2. Expression of xPTP-PESTr. RT-PCR on RNA extracted from groups of 10 embryos at various stages of development (A), 20 explants (B), or the heart and liver of an adult male (C) using primer for xPTP-PESTr and ODC. The negative control ( ) corresponds to RNA from Stage 36 embryos treated in the absence of reverse transcriptase. (A) xPTP-PESTr-specific amplification products appear as a double band at every embryonic stages tested (arrowheads 1 and 2). (B) The two amplification products are expressed in the ectoderm (Cap), the dorsal (DMZ) and ventral mesoderm (VMZ), and the endoderm (Veg) of Stage 10.5 gastrula. (C) PCR performed on liver and heart cDNA display the same amplification products than the embryonic stage 10.5. The lower band (arrowhead 1) corresponds to the size amplified from the EST, clone. The upper band (arrowhead 2) corresponds to the splicing variant presented in Fig. 1A.
	Fig. 3. Biochemical interactions of xPTP-PESTr. (A, B) Immunoprecipitation of protein extracted from Stage 10.5 embryos either uninjected (NI) or injected with myc-tagged-PTP-PEST (Wt) using antibodies to p130CAS, paxillin, or PACSIN2. (A) The presence of myc-PTP-PEST protein is visualized using the myc-specific mAb 9E10 on either total extract (extract) or the various immunoprecipitates (IP). Myc-PTP-PEST is coprecipitated with p130CAS and Paxillin but not PACSIN2. (B) The presence of each protein in the immunoprecipitates is controlled by reprobing the nitrocellulose membrane from Awith antibodies to p130CAS, Paxillin, and PACSIN2. The arrowhead to the right indicates the expected bands for each antibody. (C) Protein extracts from Stage 20 embryos injected with either the wild-type or C231S form of PTP-PEST were immunoprecipitated using the p130CAS antibody and compared to immunoprecipitates from control embryos (NI). Immunocomplexes were probed using an antibody to phosphorylated tyrosine (PY20) or the p130CAS antibody. The presence of the overexpressed Wt- or C231S-PTP-PEST was confirmed using the 9E10 mAb. Results from quantification of three independent experiments is plotted on the right. Signal intensities have been normalized to the non injected control embryos (100%). The error bars correspond to the standard deviation from the mean. The decrease of p130CAS phosphorylation is significant in the wild-type-expressing embryos (81 F 5.4%, P = 0.019) but not in embryos expressing the C231S mutant (95 F 11.8%, P = 0.31).
	Fig. 4. Overexpression of xPTP-PESTr inhibits gastrulation. Embryos injected in the two blastomeres at two-cell stage were grown to late gastrula (Stage 13) and treated by whole-mount in situ hybridization with the makers Sox2, brachyury (Xbra), and chordin. Embryos expressing the wild-type or the point mutant C231S fail to close their blastopores (white double arrows) and their notochords (n) are shorter than the control (black vertical bar), indicating that gastrulation did not occur properly. In some embryos, two notochord-like structures form around the yolk plug. The neural plate is present (Sox2) but the anterior part is narrower (horizontal black bar) and the posterior portion splits around the yolk plug (y).
	Fig. 5. xPTP-PESTr overexpression induce ectoderm cell spreading on fibronectin. Animal cap ectoderm cells expressing GFP, the wild-type, or the C231S mutant forms of Xenopus PTP-PEST were seeded on FN (10Ag/ml). Photographs were taken before (Stage 9) and after (Stage 10.5) sibling embryos initiated gastrulation movements. Control cells expressing GFP were plated in the absence ( ) or presence (+) of the mesoderm inducing factor activin-A. Before gastrulation, cells expressing either form of PTP-PEST spread on FN while control cells do not. During gastrulation, cells expressing Wt- or the C231S mutant are fully spread and resemble activin-A-treated cells expressing GFP.
	Fig. 6. The CCBD is sufficient to mediate xPTP-PESTr-induced cell spreading. (A) Schematic representation of FN (FN) and the various GSTFN fusion proteins used in the adhesion assay. The black rectangles represent the EIIIA and EIIIB domains. The two heparin-binding domains (HepI and HepII), the V region (V), and the Central Cell-Binding domain (CCBD) are indicated. The RGD sequence in the 10th type III repeat and the synergy domain (Syn) in the 9th type III repeats are also marked. The point mutations in the 9.11a and 9.11e are indicated and the domains mutated are represented in dark gray. (B) Adhesion assays were performed as described in Fig. 4 on 10 Ag/ml of fibronectin (FN) or 6 Ag/ml of the fusion proteins 9.11, 9.11e, and 9.11a. For counting purposes, cells are considered spread when they display an asymmetrical shape with two or more lamelliform protrusions. Wt- and C231S drastically increased cell spreading when compared to GFP, on both fibronectin (FN) and the central cell binding domain fusion protein (9.11). Cell spreading is never observed on the mutated fusion protein 9.11a or 9.11e. (C) Histogram representing a quantitative analysis from four independent experiments for FN and 9.11, and two experiments for 9.11a and 9.11e. Error bars represent the standard deviation from the mean. Statistical analysis shows that spreading in response to wt- and C231S-PTP-PEST is significantly increased compared to GFP on both FN and 9.11 ( P < 0.01).
	Fig. 7. a5h1 integrin is responsible for xPTP-PESTr-induced cell spreading on fibronectin. (A) Photographs of gastrula-stage animal cap ectoderm cells on FN (10 Ag/ml). Cells expressing GFP, the Wt- or C231S-PTP-PEST were seeded in the presence or absence of a function blocking mAb directed against integrin a5h1 (P8D4). (B) Histogram representing a quantitative analysis of the experiments presented in A. The error bars represent the standard deviation from the mean. The presence of P8D4 inhibits cell spreading induced by both the wild-type and C231S mutant forms of xPTPPEST to the same level as the GFP control ( P = 0.36).
	Fig. 8. xPTP-PESTr overexpression prevents activin-A-induced cell migration on fibronectin. Animal cap ectoderm cells expressing GFP, Wt-, or C231S-PTP-PEST proteins were seeded on FN (10 Ag/ml) with or without activin-A. Measure of cell motility was performed on 10 cells for each condition taking the nucleus as reference. Three independent experiments were performed. The histogram represents the average migration speed in Am/h for each condition. The error bars represent the standard deviation from the mean. Activin-A treatment of control cells (GFP) increases cell migration, while it has no effect on cells expressing either forms of PTP-PEST.
	Fig. 9. xPTP-PESTr overexpression inhibits fibronectin fibrillogenesis and perturbs orientation of cell division. Blastocoel roofs (BCR) from embryos expressing GFP, Wt-, or C231S-PTP-PEST were dissected at Stage 12, fixed, and processed for immunofluorescence. The FN fibrils of the BCR were detected using a rabbit polyclonal antibody (32F; Aâ€“C). (Dâ€“F) Nuclei staining of the same fields as in Aâ€“C using DAPI. (A) BCR of GFP injected embryo shows well-developed FN fibrils that connect each cell with its neighbors. (B) The BCR overexpressing wild-type xPTP-PESTr have FN only at cell â€“cell contacts (arrowhead). (C) The BCR overexpressing C231S displays three types of fibrils: long fibrils that cross over up to three cells (arrowhead), very short fibrils with more branching (arrow) and groups of parallel fibrils (double arrow and inset). (D) Nuclei staining of control cells reveals anaphase figures exclusively in the plane of the BCR (double arrows). (E) BCR cells overexpressing Wt-PTP-PEST presents random cell division polarity: the metaphase plates show angles of 90j (arrow 1), 45j (arrow 2), or 0j (arrow 3) to the blastocoel roof plane. (F) In BCR overexpressing the C321S point mutant, cell divisions occur, as for the control, along the BCR plane (double arrows).
	Fig. 10. xPTP-PEST perturbation does not affect cell intercalation in vivo. (A) Median optical section of early gastrula stage embryos (Stages 10 to 11) expressing GFP (top), Wt- (middle), or the C231S-xPTP-PESTr mutant (bottom). The right panels correspond to higher magnification of the blastocoele (b) roof. The various myc-tagged protein are detected using mAb 9E10 (green) while fibronectin is detected using 32F (red). The yellow color correspond to regions of overlap. Thinning of the blastocoele roof at the onset of gastrulation down to three cell layers is observed in all experimental cases. (B) Animal cap expressing either GFP, Wt-, or the C231S-PTP-PEST were cut at Stage 8 induced or not with 5 ng/ml of activin-A and grown until control embryos reached Stage 15. While all animal cap expressing GFP extend in response to the activin treatment (GFP + activin, black arrowhead), elongation is dramatically reduced in animal cap expressing either the Wt- or the C231S-PTP-PEST mutant. No extension is observed in the absence of activin ( activin). (C) Optical sections of tailbud stage embryos (Stage 22) injected in one dorsal vegetal blastomere at the eight-cell stage. Anterior is to the left (Ant) and ventral is down (Vent). The overexpressed proteins are detected using mAb 9E10 (green), while fibronectin is detected using 32F (red). Cells expressing either the Wt- (left panels) or the C231S-PTP-PEST (right panels) can participate in dorsal and anterior structures (upper panels), including mesoderm from the head (h), somites (s), and notochord. The lower panels represent a higher magnification of the notochord (n). The white arrowheads point to the fibronectin staining around the notochord (left) and the somites (s, right). The staining pattern of green cells alternating with dark cells in the notochord indicates that cells expressing either Wt- (left) or the C231S-PTP-PEST (right) can intercalate with nonexpressing cells to form the notochord.