Epithelial Cell Polarity

The epithelium was probably the first tissue type to have arisen in the evolution of multicellular animals, and is the first differentiated structure to emerge during embryogenesis.  Its organization depends on three attributes: intercellular adhesion to form a super-cellular structure; mitotic spindle orientation to ensure the formation of a sheet rather than a blob of cells; and apical/basal polarity to distinguish the outside surface from the inside surface of the sheet.  Elaboration of the basic epithelial characteristics in animals has led to the organization of epithelia into complex structures such as stratified epithelia and branched tubules and alveolae that can contain multiple cell types, arising from stem cells or lineage-specific progenitors.

Picture 3.png

We are interested in the molecular machinery that drives and maintains apical/basal polarity and spindle orientation, and use several different systems to study this - MDCK cells grown as acini in 3D culture; and mouse mammary glands, which provide a wonderful model of branching morphogenesis.  We are using the new CRISPR/Cas9 gene editing system to generate knockout cell lines and mice, to facilitate the study of the cell polarity machinery.  We are also interested in how defects in cell polarity can promote breast cancer.


The Cell Polarity Machinery

A conserved set of proteins, found throughout the animal kingdom, drives apical/basal polarization of epithelial cells.  One group of proteins - the PAR proteins - were originally identified in worms. Two other groups -  Crumbs/Pals1/Patj, and Scribble/Lgl/Dlg - were first discovered in flies.  Crumbs expression determines the size of the apical domain and is essential for tight junction formation.  PAR3 recruits atypical protein kinase C, aPKC, to the apical surface where aPKC binds to Crumbs.  It acts to prevent basolateral proteins from invading apical territory.
We are interested in the roles of PAR3 and other polarity proteins during epithelial polarization and in stem cell function, particularly in the context of mammary gland development.  Unexpectedly, loss of PAR3 in mammary cells promotes both cell proliferation and apoptosis.  The apoptotic mechanism is currently being studied in the laboratory.
PAR proteins play central roles in asymmetric stem cell divisions in Drosophila, but it is unclear whether they do so in vertebrates.  Moreover, not all vertebrate stem cells undergo autonomous asymmetric divisions.  It is unclear whether the mammary gland contains multipotent stem cells, and current evidence suggests that it is maintained by unipotent progenitors.  However, these cells can revert to a stem cell state when isolated from the gland and transplanted into a recipient mouse.  We are studying the underlying mechanism for this change in fate.

Breast Cancer

Most human cancers arise from epithelial cells or their progenitors.  A universal feature of cancers is a loss of tissue organization, even when individual cells retain normal epithelial characteristics.  As cancers progress, the cells sometimes also lose apical/basal polarity and cell-cell adhesion.  They become invasive and can migrate to ectopic sites. 

A major question is whether the cell polarity machinery is involved in tumorigenesis.  We found that PAR3 can function as a potent tumor and invasion suppressor in mouse models of breast cancer.  Loss of PAR3 increases tumor growth rates, increases the level of disorganization of the tumor, and promotes metastasis.  Interestingly, these different phenotypes are triggered through different mechanisms.  Activation of Rac drives tumor growth, while metastasis results from an activation of the JAK/STAT3 pathway, which results in the induction of MMP9, a metalloproteinase, and increased invasive migration. 

Figure shows lung metastases (GFP fluorescence) from mice implanted in the mammary fat pads with primary mammary epithelial cells expressing the NICD oncogene +/- an shRNA against the PAR3 polarity gene.


The DNA Damage Response

Some years ago we discovered that SOCS7, which was thought to be involved in attenuation of insulin signaling.  also shuttles in and out of the nucleus, and carries the adapter protein, NCK, in and out of the nucleus too.  UV irradiation causes an accumulation of NCK in the nucleus.  We more recently found that loss of NCK or SOCS7 results in very rapid (< 3 hrs) apoptosis in response to UV irradiation.  Somehow, nuclear NCK normally reduces the phosphorylation of p53, thereby inhibiting the apoptotic response to DNA damage.  In the absence of NCK this attenuation is missing.  The nuclear accumulation of NCK is ATR-dependent, but we do not yet know the underlying mechanism.

One particularly toxic form of damage is inter-strand cross linking (ICL).  We have discovered  a novel nuclease that is important for efficient repair of ICLs.  We have named this nuclease SAN1.  Interestingly, it functions as a 5' exonuclease specific for single-stranded DNA.  We are now working to understand its function at a molecular level, and to ask if it is important in tumorigenesis.  We have created a conditional SAN1 KO mouse and used CRISPR to make a SAN1-knockout HeLa cell line, which will facilitate these studies.  Two-hybrid screens and proteomics have revealed several interesting binding partners.