The work of the laboratory is focused on signal transduction pathways linked to key phenotypic properties of transformed cells. Specifically, interests lie in the function of the Protein Kinase C (PKC) superfamily. This family of serine/threonine protein kinases have long been implicated in controlling transformation-associated phenotypic properties. In line with this, we are investigating the roles of members of this family in the control of cell division, cell migration/invasion and survival. Ourwork spans the molecular eventsassociated with inputs and outputs of the different PKC isoforms, (classical, novel, atypical and 'related' isoforms) through their cell-based actions, to their roles in disease models.
Cell migration
Underlying the invasive properties of cancers are cellular functions associated with migratory behaviours. Emerging evidence indicates that there are multiple contributory factors influencing migration, particularly speed of movement and direction. Our continued work in this area has focused on three broad issues related to the PKC superfamily. Does migratory dependence on specific PKC family members reflect unique patterns of expression, function or regulatory 'wiring'? How does the PKC superfamily influence the spatial constraints on signals involved in migratory behaviour? Can we map migratory signalling pathways through siRNA screening?
The question of PKC family redundancy would appear to have multiple answers. In a migratory model displaying dependence on atypical PKC (aPKC) isoforms, knock-down of either PKCζ or PKCι partially inhibits migration while the combination knock-down has a greater effect. For the PKN1,2,3 isoforms the effect of knock-down appears to reflect expression patterns such that where abundantly coexpressed, combination knock-down is required to suppress migration. However this is not simple redundancy as it appears that there are selective regulatory inputs peculiar to the particular cell. Protein arrays and substrate specificity screens have been exploited to generate a series of direct targets of PKNs that are candidates for mediating the PKN role(s) in migratory responses.
The nature of PKC family outputs is in part to control the subcellular distribution of signalling events. This has been investigated extensively in relation to PKCζ/ι action in migration and in PKCα/ε dependent HGF/cMet induced migration. For the aPKCs, they regulate the delivery of signals to the leading edge, in a monolayer wound-healing model. The extent to which these events account for aPKC action in this model is being probed with genetically encoded, drug responsive tools. For HGF/cMet, we have shown previously that PKCε influences a related signal delivery process. We have now demonstrated that the PKCα control acting on the delivery of activated cMet to a perinuclear compartment is instrumental in determining HGF-dependent activation and nuclear accumulation of STAT3, an event which is required for HGF-induced migration.
To further elucidate roles in these migratory responses, HGF-dependent monolayer migration and a distinctive transwell migratory model have been employed in siRNA library screens to identify relevant regulators. Validation of hits is ongoing. In parallel an siRNA screen to map a specific PKC-dependent pro-invasive property has been established.
Survival
PKCα is upregulated in multiple glioblastoma cell lines. To assess whether this is involved in the phenotype and how intervention might influence glioblastoma behaviour, we have compared the effects of PKCα knock-down with PKC(α) inhibition. Notably, loss of PKCα in U87MG cells was associated with induction of apoptosis. This exhibited a strong threshold effect requiring efficient knock-down to elicit this response. Consistent with a survival role for PKCα, expression of a kinase inactive, dominant negative PKCα also induced apoptosis. Interestingly, the requirement for PKCα is not corroborated by catalytic site directed inhibitors. It appears that PKCα has a scaffolding function which is activity independent and dominates behaviour in these glioblastoma cells.
The distinctions between siRNA knock-down, expression of a dominant negative kinase inactive allele and inhibition of PKCα are of some significance. One critical aspect of this is the altered conformation associated with the kinase inactive mutation of the conserved lysine in the ATP binding pocket. This mutant is not phosphorylated in the kinase domain and retains a non-functional conformation, unlike the inhibited WT-PKCα. The nature of these states, their inter-conversion and their inhibition are the subject of ongoing functional and structural work with Dr Neil McDonald (Structural Biology Laboratory) and the Development Laboratory (Cancer Research Technology).
Division
PKC isoforms have been implicated in controls acting at various steps in the cell cycle. We have recently demonstrated a requirement for PKCε in the completion of cytokinesis. Inactive forms of PKCε accumulate at the midbody and are associated with a slow down or complete failure of cytokinesis - the latter leading to binucleation, a prelude to aneuploidy (Figure 1). Analysis of the effectors located at the furrow alongside inhibited PKCε has identified RhoA. It appears that PKC&epsilon's role in this final step is to regulate the switch off of RhoA, an event required for the disassembly of the contracted actin ring. The trigger(s) for recruitment of PKCε to the furrow, the dynamics of its association with the midbody and the proximal effector(s) responsible for RhoA control are the subject of ongoing work.
PKCε is engaged in this final irreversible step through a series of phosphorylations that enable the assembly of a complex with 14-3-3, which in turn confers constitutive activity on PKCε. The mapping of the docking sites has defined interaction within the V3 domain and the structure of the V3 domain/14-3-3 complex has been solved in collaboration with Dr Neil McDonald's Laboratory (Structural Biology Laboratory).
Figure 1. Failed cytokinesis following induction of an inactive PKCε mutant. Cells (293T) harbouring an inducible, kinase inactive PKCε mutant were induced for 24h and then filmed. Shown is a series of phase images illustrating the progression of mitosis, through typical metaphase, anaphase and furrowing stages, but in this context ultimately leading to a failure of cytokinesis. The final image depicts the binucleate cell derived from this failed division. For these cells, 24h postinduction of the kinase inactive PKC¿ mutantthe majority of cells are binucleate and a further 24h later most cells are dead.
For a list of refereed research papers, see Publications (in navigation on left).
