The mechanisms of planar cell polarity, growth and the Hippo pathway: some known unknowns

Abstract

Planar cell polarity (PCP) is a small but important area of research. In this review we discuss a limited number of topics within the PCP field, chosen because they are difficult, unsolved, controversial or just because we find them interesting. Because Drosophila is the best studied and technically most amenable system we have concentrated on it, but also consider some examples from work on vertebrates. Topics discussed include the number of genetic pathways involved in PCP, as well as the causal relationship between embryonic axes, gradients of morphogens and PCP itself. We consider the vexed question of the roles of the Wnt genes in PCP in both vertebrates and Drosophila. We discuss whether the proteins involved in PCP need to be localised asymmetrically in cells in order to function. We criticise the way the Hippo pathway is described in the literature and ask what its wildtype function is. We explore afresh how the Hippo pathway might be linked both to growth and to PCP through the gigantic cadherin molecule Fat. We offer some new ways of making sense of published results, particularly those relating to the Frizzled/Starry night and Dachsous/Fat systems of PCP.

Fig. 1

Summary of the axes and morphogens of the eye, wing and abdomen of Drosophila, the main organs used to study PCP. The colours indicate the zones of expression of genes encoding the three morphogens, showing their radically different dispositions. Note the structures that indicate PCP: ommatidia in the eye, hairs of the wing and hairs and bristles of the abdomen. Note that while the wing hairs are aligned parallel to the A/P axis, the abdominal hairs and bristles are orthogonal to the same axis. The ommatidia are aligned along the A/P axis but have two chiral forms depending if they are dorsal or ventral in the disc. Do=dorsal, Di=distal, Pr=proximal, Ve=ventral.

Fig. 2

The polarity effects of ptc^—en^— clones. This model is puzzling, but it illustrates some important facts and principles about the two PCP systems operating in the Drosophila abdomen. Anterior is at the top. In the wildtype, we envisage two gradient systems, one (Ds activity) from the Ds/Ft system shown in grey and one (Fz activity) from the Fz/Stan system shown in red. The Ds/Ft gradient peaks at the A/P compartment border (dashed line) and runs downwards from there in both anterior and posterior directions. The Fz gradient may have a more complex topography than shown but most likely peaks at the front of the A compartment, declining posteriorwards from there to the back or middle of the P compartment. In both the A and P compartments the hairs point down the Fz activity gradient but the two compartments read the Ds gradient with opposite sign, pointing up the Ds slope in the A compartment, and down the Ds slope in the P (Casal et al., 2002). In the A compartment the slopes of both the Fz and Ds activity gradients act together to point the hairs backwards while in the P compartment the two gradients act with opposite sign. In both the A and the P compartments, clones that lack the ptc and en genes make cuticle that shows the identity of cells that are located in the wildtype just anterior to the A/P compartment boundary; within these clones the levels of the gradients are indicated by the colour intensities. These cell identities correspond to high levels of Ds activity (dark grey) and medium levels of Fz activity (pale pink). The arrows indicate the directions of the local slopes of Ds activity (white) and Fz activity (blue) near the clones; in the A compartment these local slopes orient hairs near the clones down the Fz activity gradient and up the Ds activity gradient. However in the P compartment the P cells orient their hairs down the Ds slope (Casal et al., 2002). Thus in the P compartment of the wildtype, hairs near ptc^—en^— clones are subject to two opposing influences; explaining perhaps why these clones have no consistent effect on the cells around them.