One might expect as much interaction between innexins and the cytoskeleton as there is between connexins and the cytoskeleton. Gap junction plaques are relatively large structures that would theoretically require some rigidity in the plasma membrane wherever they exist. The transport and targeting of new subunits and the endocytosis of old plaques would also probably require dynamic interaction with cytoskeletal components. Trying to visualise this interaction in Drosophila cells is complicated by their small size. The large salivary glands cells are also problematic - their size requires structural support provided by thick actin cables which stain very well using a fluorescent F-actin-binding phalloidin-rhodamine conjugate marker - but unfortunately, the strong fluoresence from these cables generally obscures the finer strands of actin that might associate with innexin structures. For this reason it was only possible to identify general positional relationships between thick accumulations of F-actin and innexin immunopositive regions. At the site of synthesis (presumably the golgi complex structure) where innexin subunits theoretically oligomerise and become incorporated into transport vesicle membranes, innexin-staining and F-actin are closely associated (Image: Innexin in putative golgi complex). From the site of synthesis, it is not known exactly how innexins are transported to the plasma membrane - two alternative hypothetical routes are presented on the innexin lifecycle page. If innexins are transported directly towards the basal end of the cell where large numbers of big (Image: Annular junctions) and small vesicles can be detected then they might be transported there along the thick actin cables that run within the cell (Image at top of page). The region of the salivary gland cell just below the basal surface appears to have a meshwork of actin fibres and may function as a sorting zone for transport vesicles on their way to the plasma membrane, and endocytosed vesicles away from it. Innexin-immunopositive vesicles are found in close apposition to some of the actin fibres in this basal zone. If newly synthesised innexin-containing vesicles are transported laterally (not towards the basal end of the cell) to be incorporated into the cell membrane, we have found the area next to the plasma membrane in this region (at the apical end of the cluster of gap junction plaques - which are all found in the basal 1/2 of the cell) is rich in actin fibers which seem to envelope small innexin-immunopositive vesicles (Inx2 and actin distribution in the lateral cell membrane). The arrangement of innexins and actin near gap junction plaques in the plasma membrane is variable, but, often plaques are located next to accumulations of actin (Image: F-actin near plaques) - perhaps providing membrane rigidity necessary for plaques to form? Other observations of potential interactions between innexins and the cytoskeleton include the linear distribution of Inx2 near the surface of some salivary gland cells and the fact that UAS-innexin mediated overexpression can led to alterations in cell shape (Bauer et al, 2004) - possibly a secondary effect as a result of defective cell polarity - but until more is known about the direct and indirect interactions between gap junction proteins and the cytoskeleton it's not possible to be certain. Insect haemocytes infected with viruses that encode innexins (-called vinnexins) exhibit a disrupted cytoskeleton (Turnbull et al, 2005), although it's not known whether vinnexins play a direct role in this process. Connexin-based channels are known to be associated with the cytoskeleton via accessory proteins such as drebrin (Butkevich et al. 2004). There is no obvious Drosophila homologue of drebrin; presumably, as yet unidentified molecules perform a similar role for innexin-based channels.
Innexins and the actin cytoskeleton
Figure legend: Innexins are often detected in close proximity to actin cables. Left image panel- Shows Inx2 (red) protein at the lateral plasma membrane in an optical section taken between the nucleus and the basal cell surface of a third instar larval salivary gland cell. A few innexin-immunopositive vesicles can be seen within the cell. Right panel Showing F-actin (yellow, phalloidin-rhodamine) staining in the same optical section. Arrows indicate where Inx2 and actin cables exhibit a close positional relationship. Not all Inx2 staining is close to thick actin cables however.