Dual function of zonula occludens proteins

Tight intercellular contacts (tight junctions, TJs) are responsible for the "barrier" function of epithelia and endothelia. A series of pathological conditions has been related to the disintegrity or misfunction of these tissue barriers, such as the opening of lung epithelial TJs by environmental proteolytic allergens which is supposed to be the initial step in the development of asthma. Similarly, the breakdown of the endothelial blood-brain barrier is a major factor in the progression of diabetic retinopathy, brain tumors, meningitis, encephalitis, cerebral malaria and multiple sclerosis. However, in recent years experimental and medical evidence has accumulated suggesting that TJs are much more related to cell growth and proliferation than expected. The structural similarity between the TJ-associated zonula occludens proteins (ZOPs) and the Drosophila tumor suppressor gene dlg-A provided a first clue towards a yet undefined role of these TJ proteins. ZOPs (ZO-1, -2, and -3) are multidomain proteins, with some domains required for structural organization of TJs and other domains capable of functioning in signal transduction pathways. We have observed recently that ZO-2 localizes to the nucleus of highly migratory endothelial cells which lack contact-inhibition of cell proliferation. Although direct experimental evidence is still missing, it is thrilling to speculate that the nuclear targeting of ZO-2 is a crucial step in the signaling cascade leading to increased motility and reduced adhesiveness of endothelial/epithelial cells. Results from our molecular studies have indicated that ZO-2 interacts directly with a DNA/RNA binding protein, termed SAF-B (scaffold attachment factor-B). The biological significance of this interaction is still unknown. In this project we will focus on the nucleo-cytoplasmic shuttling of ZO-2. We will try to identify conserved domains of this TJ-related protein which may be responsible for its nuclear import and export. In addition, we will study possible alterations of gene expression following nuclear accumulation of nuclear ZO-2 and will try to find out whether this effect is dependent on ZO-2/SAF-B binding. A series of in vitro experiments will be performed to detect the influence of increased nuclear ZO-2 on proliferation, migratory activity, invasiveness and barrier features of endothelial and epithelial cells. In this way, we hope to shed light on the question how ZOPs are involved in a signaling cascade regulating epithelial/endothelial cell growth and function.

The establishment and function of polarized epithelial cells requires the asymmetric distribution of cytoplasmic organelles and plasma membrane proteins, to create two functionally distinct apical and basolateral membrane domains commonly referred to as cell "polarity". The two domains are separated by apico-lateral tight junctions in vertebrates or septate junctions in invertebrates. These junctions form a transepithelial barrier and restrict membrane proteins to either the apical or basolateral membrane. Loss of cell polarity and tissue architecture are characteristics of malignant cancers derived from epithelial tissues. Zonula occludens proteins (ZOPs) are multi-domain scaffolding proteins, localizing preferentially at tight junctions. ZOPs, currently comprising ZO-1, ZO-2, and ZO-3, are structurally similar to the membrane-associated guanylate kinase-like (MAGUK) protein family of signaling molecules. The observation that ZOPs also associate with adherens junctions and gap junctions in cells lacking tight junctions suggests a universal role of ZOPs at the cytoplasmic surface of the junctional plaque. The homology of ZOPs with the Drosophila tumor suppressor protein dlgA has provided a first clue towards a yet undefined role of ZOPs in epithelial cancer development and/or progression. This notion was further supported by the observation that ZOPs are delocalized or down-regulated in several types of carcinomas and may even serve as cellular targets for tumorigenic viral proteins. The molecular pathways by which the tight junction-associated ZOPs interfere with cell growth and proliferation are far from being understood. In our study we have shown that ZO-2 localizes to the nucleus of epithelial and endothelial cells. Using a yeast-based Two-Hybrid assay and co-immunoprecipitation we have demonstrated that ZO-2 interacts directly with the nuclear scaffold attachment factor-B (SAF-B), a co-repressor of estrogen receptor- known to regulate transcription of genes involved in proliferation, apoptosis, migration and other cellular processes. Results from our molecular and functional analyses indicate that nuclear accumulation of ZO-2 in epithelial and endothelial cells is accompanied by increased proliferation and adhesion as well as by the expression of tumor- related marker proteins and a decrease in junctional stability.