| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online October 2, 2003 as doi:10.1096/fj.03-0040fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Cancer Genomics, Hollings Cancer Center, and Department of Cell Biology and Anatomy, Medical University of South Carolina, Charleston, South Carolina, USA
2Correspondence: Medical University of South Carolina, Laboratory of Cancer Genomics, Hollings Cancer Center Rm. 333, 86 Jonathon Lucas St., Charleston SC 29425, USA. E-mail: musehelm@musc.edu or hsut{at}musc.edu
SPECIFIC AIMS
A precisely coordinated expression of growth factors and their elicited signaling pathways in endothelial cells regulate new vessel formation. Although many signaling components and downstream effectors have been identified, their exact functions in promoting cell motility and tubulogenesis are unresolved. Our specific objective in this study was to determine the roles of two angiogenic regulators, the serine threonine kinase Akt and the transcription factor Ets1.
PRINCIPAL FINDINGS
1. Ets1 transcription is regulated by VEGF via an Akt-specific pathway
We sought to determine the mechanism by which VEGF induces the expression of Ets1 in primary endothelial cells (EC), a transcription factor that activates such downstream target genes as metalloproteinases and vimentin, which are important for matrix degradation and cell migration. Using the global transcription inhibitor actinomycin D and nuclear run-on assays, we found that VEGF induces the transcription of ETS1. Since ETS1 activity is generally controlled on the level of phosphorylation, we asked whether VEGF stimulation also caused an increase in its phosphorylation. We found that unlike growth factors in other cellular systems, VEGF did not affect the overall state of ETS1 phosphorylation. This suggests there is a sustained basal level of Ets1 phosphorylation that is maintained in endothelial cells regardless of growth factor stimulation. Inhibition of PI3 kinase (PI3K) by the pharmacological inhibitor wortmannin resulted in a marked decrease in VEGF induction of Ets1 expression, indicating a potential Akt-linked pathway. To test the role of Akt activation in the expression of Ets1, we expressed an activated, membrane-targeted form of Akt (myrAkt) in EC. As shown in Fig. 1
A, activation of this kinase in the absence of VEGF was able to induce the expression of Ets1. These data then suggest a linear pathway resulting from VEGF stimulation: PI3K>Akt1>ETS1. Since Akt is emerging as a central regulator of cellular migration (a role already ascribed to ETS1 and one that angiogenic tubulogenesis requires, a priori, migratory potential of the vascular cells), this linear pathway seems plausible.
|
2. Both Akt and ETS1 are needed for angiogenesis in vitro
Indeed, upon transfection of myrAkt we noticed changes in human umbilical endothelial cell (HUVEC) morphology (data not shown) in the absence of exogenously added VEGF. To determine whether these morphological changes were associated with changes in cell migration, we tested the effect of Akt activation on endothelial cell motility. We chose a Matrigel-based angiogenesis assay because this model system focuses on the processes involved in the formation of cord-like structure and branching morphogenesis. Since cord formation occurs within hours of plating onto Matrigel, these assays measure only guided cell migration, not cell proliferation. Under serum- and growth factor-free conditions, cells expressing myrAkt formed striking cord-like structures within 3 h of plating, whereas in the absence of myrAkt and serum HUVEC remained an even monolayer (Fig. 1D
). These data suggest that activation of Akt can substitute for VEGF in the induction of the cellular migration and cord formation in this type of assay. The action of Akt on cord formation is blocked by antisense ETS1 (Fig. 1D
). Taken together, our results indicate that activation of Akt plays a central role in the cellular reorganization essential for the generation of new capillary vessels and that this process requires the activity of ETS1.
3. Activation of Akt and ETS1 regulate different aspects of tubulogenesis
So far our results indicate that myrAkt causes early induction of ETS1 expression and later events that result in increased migration and cord formation in vitro. We sought to test our linear pathway in an in vivo model. We chose the Drosophila tracheal system in which homologs of Akt (Dakt1) and ETS1 (pointed) play important roles. Mutations in pnt result in severe impairment of branching morphogenesis. The involvement of Dakt1 in tracheal cell migration has not been described, although it has been shown that Dakt1 is required for cell survival during early embryogenesis and is essential for tracheal cell fate determination. We used a reporter transgenic line 1-eve-1 that expresses lacZ in all tracheal cells for visualization. The 1-eve-1 marker was then combined with Dakt1 and pnt alleles and the trachea was visualized by staining with anti-ß-Gal antibody. Both heterozygous and homozygous pnt mutants show a lack of tubule migration of different degrees of severity (Fig. 2
E–G). However, in both heterozygous and homozygous Dakt1q we observed ectopic migration instead of the expected lack of migration. Moderate phenotypes show aberrant linkage between adjacent branches, and ectopic filopodia-like cellular projections can be seen at the tracheal tip (Fig. 2I-J
). In severe cases, the overall tubule network is disrupted with obvious ectopic branching and "breakaway" tracheal cells (Fig. 2K
). These phenotypes suggest that the normal function of Dakt1 in tubule formation is to restrict the movement of migrating cells, a result that appears to be contradictory to our in vitro data.
|
We postulate that tubule formation may require two opposing functions, one to mobilize the cells (the ETS1 function) and one to confer the restricted directionality of the cell movement (the Akt function). To test this model, we performed a genetic epistasis analysis in Drosophila. We reasoned that pnt mutation should improve the Dakt1 phenotypes due to the reduced migratory potential (in pnt mutant). This is indeed the case. There was a 60% reduction of the phenocopy number of Dakt1-like phenotypes in Dakt1q/pnt
88 trans-heterozygotes compared with Dakt1q heterozygotes. In the trans-heterozygotes, occasionally a new class of mutants was observed, which we interpret to have dual characteristics of pnt and Dakt1 phenotypes. Primary branches do not form in these mutants, but breakaway cells are detected around the remnants of tracheal placodes (Fig. 2L
) or small, erratic branches sprout from the otherwise clustered tracheal subunits (Fig. 2M, N
). These data suggest that Pointed/ETS1 and Dakt1 do not function in a linear pathway; rather, they control different aspects of cell motility that ultimately result in organized tubule network.
CONCLUSION
Our in vitro results appear to indicate that VEGF, Akt, and ETS1 form a linear epistatic signaling pathway that promotes cell movement and leads to tubulogenesis. However, when tested in the in vivo Drosophila tracheal system, a complex relationship between Akt and ETS1 was revealed. Flies carrying a pnt mutation have the expected tracheal phenotype: disruptions in tracheal branch migration. But mutations in Dakt1 do not result in inhibition of migration; instead, they result in misguided, ectopic branching and breakaway cells. These results indicated that rather than controlling cellular migration per se, Akt controls directed migrations that result in appropriate tracheal branch formation. This interpretation can explain why blocking ETS1 function with antisense ETS1 can counteract the action of constitutively active Akt in promoting tubule formation; there can be no organized cell migration (the Akt function) without a priori the intrinsic ability of the cells to migrate (the ETS1 function). Indeed, we demonstrate that the pnt mutation can rescue the Dakt1 phenotypes.
As with the in vitro angiogenesis model used here, tracheal development in Drosophila does not require cellular proliferation. At no time after the tracheal placodes are formed do the tracheal cells proliferate, nor are other cells recruited into the network. Hence, in both of our model systems, cellular mechanisms that specifically govern cell motility can be analyzed unambiguously. Our comparative analyses using these two systems indicate that simple overexpression of gene functions or epigenetic knockouts commonly used in cell culture studies, although informative, could be misleading. Thus, by cross-referencing the endothelial cell culture and Drosophila tracheal systems, we provide novel insight into the cell migration events underlying capillary tubule formation, which requires the seemingly opposite but coordinated functions of ETS1 and Akt. Although the involvement of Akt in angiogenesis has been noted, to our knowledge this report is the first to indicate that the Akt function in tubulogenesis is an essential but restrictive one. The exact cellular mechanism of this aspect of the Akt function remains to be elucidated. An important clue may lie in the Akt action on cell adhesion and cytoskeletal rearrangement. Akt phosphorylation sites have been found in the Rho family of proteins. These GTPases are known to play important roles in the control of cell shape, adhesion, movement, and actin cytoskeleton organization. Modulating these GTPases may underlie the mechanisms of directed cell migration. While Rac1 is known to promote cell migration, phosphorylation of Rac1 by Akt may be inhibitory. The PI3K-Akt pathway may also direct restricted migration through integrin engagement. Thus, the molecular mechanisms controlled by PI3K and Akt are likely to be complex and involve multiple players. Elucidation of these complex mechanisms will help unravel the process of tubulogenesis.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-0040fje; 
This article has been cited by other articles:
|
|
 |
|
  A. Khokhar, N. Chen, J.-P. Yuan, Y. Li, G. N. Landis, G. Beaulieu, H. Kaur, and J. Tower Conditional Switches for Extracellular Matrix Patterning in Drosophila melanogaster Genetics, March 1, 2008; 178(3): 1283 - 1293. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-M. Hong, Y.-L. Chen, Y.-Y. Wu, A. Yuan, Y.-C. Chao, Y.-C. Chung, M.-H. Wu, S.-C. Yang, S.-H. Pan, J.-Y. Shih, et al. Targeting Neuropilin 1 as an Antitumor Strategy in Lung Cancer Clin. Cancer Res., August 15, 2007; 13(16): 4759 - 4768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pozzi, S. Coffa, N. Bulus, W. Zhu, D. Chen, X. Chen, G. Mernaugh, Y. Su, S. Cai, A. Singh, et al. H-Ras, R-Ras, and TC21 Differentially Regulate Ureteric Bud Cell Branching Morphogenesis Mol. Biol. Cell, April 1, 2006; 17(4): 2046 - 2056. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Mai, D. Chen, T. Ding, I. Kim, S. Park, S.-y. Cho, J. S.F. Chu, D. Liang, N. Wang, D. Wu, et al. Inhibition of Pkhd1 Impairs Tubulomorphogenesis of Cultured IMCD Cells Mol. Biol. Cell, September 1, 2005; 16(9): 4398 - 4409. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
