Actin (red) polarizes to one end of the cell in response to LKB1 (bottom).


Single epithelial cells can polarize when given a healthy dose of activated LKB1, a human homologue of the Par-4 polarity protein characterized in worms and flies, according to Annette Baas, Hans Clevers (Center for Biomedical Genetics, Utrecht, The Netherlands), and colleagues.

The experiment overturns the dogma of epithelial polarity research, which holds that contact between adjacent cells defines boundaries and junctional complexes needed to divide one surface domain from another. By contrast, activated LKB1 creates a polarization triumvirate: cells form an apical brush border; position junctional proteins around the border; and sort membrane proteins to the respective apical and basolateral domains.

The direct consequence of activated LKB1 is probably the activation of Cdc42, a Rho family small GTPase. Cdc42's reorganization of actin should form the brush border, perhaps creating binding sites for junctional proteins (peripherally) and apical proteins.

The brush border forms at the top of the cell. There is no excess of LKB1 at this or any other site in the cell, but the top may be marked by some other protein as proto-apical based on its lack of interaction with integrins.

Gain of function Par mutations have been lacking in the worm embryo system. Baas et al. now have a cell line that can be induced to polarize within 3 to 6 hours, compared with the days or weeks that normal confluent cultures take to polarize. ▪


Baas, A.F., et al.