About UCSF A-Z Web Listing UCSF Search Campus Directory
  PSPG program information
  Admissions
 
 
 
 
  Events
 
  Links
  Program Support
  Contact Us
  Information for:
   
   
 
 
  Home
 


Keith E. Mostov,
M.D., Ph.D.

Professor of Anatomy, Biochemistry and Biophysics; Cardiovascular Research Institute



Contact Information:
keith.mostov@ucsf.edu
Tel: (415) 476-6048
Fax: (415) 514-0169
Box 2140, Genentech Hall, Room N212B

Links:
lab website
Anatomy page
PIBS
Biomedical Sciences
UCSF Cancer Center

Publications
Complete
Selected

 

Epithelial polarity and morphogenesis

The most fundamental type of organization cells in metazoa is that of epithelia. Epithelial cells form sheets of cells that line surfaces and internal cavities. The simplest metazoa, such as hydra, consist largely of two concentric cylinders of epithelial cell sheets. Most internal organs in higher animals, such as the respiratory, digestive, genito-urinary and vascular systems are lined by a single layer of epithelial cells. We are studying both the structure of individual epithelial cells and how they are organized into multicellular tissues and organs.

The central feature of epithelial cells is that they are polarized. Their plasma membrane is divided into an apical surface, which faces the outside world or the lumen of the cavity, and a basolateral surface, which faces adjacent cells and the underlying connective tissue. These plasma membrane domains perform completely different functions and therefore have quite distinct protein compositions. Cells send proteins to the correct surface by two different pathways. In the direct pathway, newly made membrane proteins are sorted in the trans-Golgi network into vesicles that carry them directly to the apical or basolateral surface. In the indirect pathway, vesicles are sent first to one surface (usually the basolateral) and then endocytosed and sorted into vesicles that transcytose them to the apical surface.

A major question in the membrane traffic field is how vesicles fuse only with the correct target membrane and not with incorrect targets. Membrane fusion is mediated by assembly of SNARE proteins. We have found that target SNARE syntaxin 3 is exclusively on the apical plasma membrane, while syntaxin 4 is exclusively basolateral. Very recently, we found that syntaxin 3 can be mistargeted to the basolateral surface by mutation of just a few amino acids. This mislocalized syntaxin 3 causes the mistargeting to the basolateral surface of some vesicles carrying cargo proteins that are normally sent to the apical surface. This provides the first evidence that the SNARE machinery is involved in the specificity of membrane targeting in vivo.

Transcytosis from the basolateral to the apical surface is a vital process for establishing and maintaining epithelial polarity. A model transcytotic protein, the polymeric immunoglobulin receptor (pIgR) is endocytosed from the basolateral surface and very efficiently transcytosed to the apical surface, with almost none of the receptor being diverted to the lysosomal degradative pathway. This is due to three processes. First, when the ligand of the pIgR, polymeric immunoglobulin A (pIgA), binds to the receptor, the receptor initiates a signaling pathway. This pathway involves the Src-family tyrosine kinase, p62Yes, elevation of intracellular calcium and activation of protein kinase C. This pathway promotes transcytosis. Second, , the small GTPase rab3b, binds directly to the pIgR. Binding of pIgA to pIgR causes the rab3b to dissociate from the pIgR, which decreases recycling of the receptor to the basolateral surface. Third, the pIgR also interacts with the retromer complex. The retromer is known from work in yeast, where it promotes retrieval of proteins from late endosomes. We have found that the retromer prevents the pIgR from entering the degradative pathway; this is the first known function of the retromer in higher eukaryotes.

Epithelial cells are organized into higher order tissue and organ structures. To study this process in a simple cell culture system, we grow Madin-Darby canine kidney (MDCK) cells in 3 dimensional (3D) collagen gels. Under these conditions, the MDCK cells form hollow cysts lined by a monolayer of epithelial cells. This is a simple model of multicellular organ structure. Not only are all of the cells polarized, but they their polarity is orientated in the same direction, with their apical surfaces all facing the same direction. How is this coordination achieved? We found that by expressing a dominant negative form of the small GTPase Rac (DN rac), the cells invert the orientation of their polarity, so that they are all polarized with their apical surface facing the outside of the cyst. This indicates that polarization and the orientation of that polarization can be experimentally separated. Although much has been learned by many groups about how cells become polarized, much less is known about how the cells orientate their polarity in a specific direction. We have found that orientation is controlled by a pathway whereby Rac activity is needed for cells to assemble a laminin network around the cyst. This assembled laminin then signals back to the cell to orientate its polarity with the apical surface at the opposite side of the cell from the laminin. This signaling pathway involves another small GTPase, Cdc42, as well as a complex consisting of atypical protein kinase C, Par3 and Par6.

Many epithelial organs consist of branching tubules, which often end in cyst-like alveoli or acini. Treatment of MDCK cysts grown in collagen gels with hepatocyte growth factor (HGF) causes the cysts to develop branching tubules, which resemble the branching tubulogenesis seen in many organs in vivo. We have found that tubulogenesis can be divided into two phases. In the first, some cells undergo a partial epithelial-mesenchymal transition, lose their apical-basolateral polarity and migrate out of the cyst wall. These cells form chains, which retain some cell-cell contact. This process requires PI3 kinase and PIP3 lipid accumulates at the leading edge of the extending cell. Moreover, instead of cells ordinarily dividing in plane of the monolayer, the cells change the axis of their mitotic spindle, so that one daughter leaves the cyst wall to initiate the cord. As more cells divide, the chains become thicker cords. Cells begin to re-establish apical-basolateral polarity and form lumens between the cells. This re-establishment of polarity probably uses the same pathway for orientation of polarity that we uncovered using DN Rac expressed in cysts. Eventually these lumens link up with the lumen of the cyst, to form a mature tubule.

The MDCK-collagen gel system provides a powerful approach to study epithelial morphogenesis and to link basic cellular processes such as membrane traffic, polarization, migration and orientation of cell division with the higher order formation of multicellular organs.

Selected Publications:
O’Brien, L.E., Jou, T.-S., Pollack, A.L., Zhang, Q., Hansen, S.H., Yurchenco, P., and Mostov, K.E. Rac1 orients epithelial apical polarity through effects on basolateral laminin assembly. Nat. Cell Biol., 3:831-838, 2001
PDF File of this Publication

van IJzendoorn, S.C.D., Tuvim, M.J., Weimbs, T., Dickey, B.F., and Mostov, K.E. Direct interaction between Rab3b and the polymeric immunoglobulin receptor controls ligand-stimulated transcytosis in epithelial cells. Develop. Cell, 2:219-228, 2002
PDF File of this Publication

O’Brien, L.E., Zegers, M.M.P., and Mostov, K.E. Opinion: Building epithelial architecture: insights from three-dimensional culture models. Nat. Rev. Mol. Cell Biol. 3:531-537, 2002
PDF File of this Publication

Yu, W., O’Brien, L.E., Wang, F., Bourne, H., Mostov, K.E., and Zegers, M.M.P. Hepatocyte growth factor switches orientation of polarity and mode of movement during morphogenesis of multicellular epithelial structures. Mol. Biol. Cell, 14:748-763, 2003
PDF File of this Publication


Copyright © 2008 The Regents of the University of California. Individual Copyrights for select images. Contact Us.
Last updated:
August 4, 2008