We model wound healing in several genetically tractable model organism from the fruitfly, Drosophila, through to mice.   Of particular interest to us is the wound inflammatory response which we have shown to be causal of scarring at the wound site.  Our current goal is to use a mix of live imaging and genetic studies to determine how recruited innate immune cells direct aberant deposition of collagen at sites of tissue damage. Recently, we have also begun investigating parallels between wound healing and cancer. We use translucent zebrafish to model cancer surgery, in particular investigating how the wound inflammatory response impacts on immune cell recruitment to nearby transformed cells and what might be the downstream consequences of this.

Wound healing and parallels to embryonic morphogenesis

There are numerous cell and tissue migrations that underpin the morphogenetic and organ building phases of embryogenesis and it seems that when adult tissues heal a wound, much of the same motility machinery is reactivated. Previously, we have analysed how the actin cytoskeletal machinery of wound re-epithelialisation is regulated by Rho family small GTPases and how this actin cable and filopodia recapitulate morphogenetic events that occur during development such as dorsal closure at the end of Drosophila embryogenesis. We now have evidence that these parallels extend beyond the leading edge and also involve polarised contractions and cell:cell intercalation events back from the leading edge that mirror what occurs during germ band extension.

Inflammation and scarring

Besides epithelial sheet movements and fibroblast migration and contraction, adult tissue repair invariably triggers a wound inflammatory response which we believe leads to pathological collagen deposition at the wound site and ultimately to fibrosis and scarring. Embryonic tissue can heal rapidly and efficiently without leaving a scar and we think this is because there is a much reduced inflammatory response. To further study the wound inflammatory response we have established models of inflammation in the Drosophila embryo and in the translucent zebrafish larva, which allow us to make movies of leukocyte migration into the wound and to dissect the genetics of inflammatory cell recruitment towards tissue damage. These studies have revealed how the leukocytic actin and microtubule cytoskeleton are regulated, and what the earliest wound attractant signals are and how immune cells might integrate these “damage” signals with other competing signals (eg developmental guidance signals) that they are exposed to.

We have also utilized array approaches in both flies and PU.1 mice that are deficient in inflammatory cells, to determine inflammation-dependent versus -independent gene inductions upon wounding. These studies are helping guide us as to which aspects of the wound transcriptome might lead to fibrosis and/or impaired repair. One such inflammation-dependent, wound-expressed gene is osteopontin and we find that if this gene is “knocked-down” at the wound site, then repair is faster and scarring is significantly reduced. Our initial observation that osteopontin “knockdown” might reduce fibrosis was made in studies of murine skin healing but we now know this therapeutic approach can work in other tissues also, for example, to block adhesions following abdominal surgeries. We are currently developing other strategies for downregulation of immune cell driven fibrosis signals. One further negative consequence of inflammation at the wound site is hyperpigmentation and we have established a model of this in zebrafish which may enable identification of immune derived hyperpigmentation signals.

Parallels with inflammation and how it might drive cancer progression

Recently we have also begun investigating parallels between wound healing and cancer using the translucent zebrafish model. Just as for wounds, we find H2O2 to be a key attractant enabling immune cells to sense early clones of pre-neoplastic cells before they progress to cancers. Subsequently, our movies reveal neutrophils “nibbling” at the transformed cells, while macrophages phagocytose whole living pre-neoplastic cells. However, despite this feasting by immune cells, we find that clones of pre-neoplastic cells deprived of immune cells proliferate at a slower rate suggesting that trophic signals are delivered to the pre-neoplastic cells by immune cells. We now find that exogenous addition of the stable prostaglandin, dmPGE2, can in part rescue growth of these cells after transient depletion of immune cells; moreover, morpholino knockdown of the rate-limiting prostaglandin synthesizing enzyme, mPGES, mimics loss of leukocytes in reducing pre-neoplastic cell number suggesting that prostaglandins might be a component of the trophic signal. This observation may, in part, explain recent studies showing how low dose aspirin can stave off the onset of gut and other cancers.

Since surgery is one of the most effective means of treating cancer we have begun to use larval and adult zebrafish to model cancer surgery, in particular investigating how the wound inflammatory response impacts on immune cell recruitment to nearby transformed cells and what might be the downstream consequences of this.  We are now curious whether we can model radiotherapy as a cancer treatment in zebrafish also.