Friday Feature: Calling for help
July 9, 2010 § 1 Comment
This amazing movie, from Niethammer P, Grabher C, Look AT, Mitchison TJ. 2009 A tissue-scale gradient of hydrogen peroxide mediates rapid wound detection in zebrafish. Nature 459 996-9 PMCID: PMC2803098, shows leukocytes (the white blobs) rushing to the site of a wound in response to a hydrogen peroxide signal (fluorescence in upper panel).
We’ve known for a while that leukocytes rapidly (within minutes) home to the sites of wounds. What hasn’t been clear is what signal attracts them. We’ve also known for a while that hydrogen peroxide is generated in wound sites: but until now, the general belief has been that it comes from the leukocytes that are attracted into the wound. Hydrogen peroxide has a role in killing bacteria, at least under some conditions, and so this all seemed to make sense. Until this movie. You can’t really tell by eye, but the quantitative analysis clearly shows that the hydrogen peroxide production starts before the first leukocyte arrives. In fact, the timing is such that it seems very plausible that it is the hydrogen peroxide that calls in the leukocytes: soon after the hydrogen peroxide reaches the nearest blood vessel, you start seeing purposeful leukocytes making tracks towards the wound.
Is the hydrogen peroxide gradient we see causal, or is it a byproduct of something else? There are five enzymes in the zebrafish genome that can produce hydrogen peroxide, directly or indirectly (four NADPH oxidases (Nox-1, -2, -4 and -5), and Dual Oxidase, abbreviated Duox). Niethammer et al. showed that small molecule inhibitors that inhibit all 5 enzymes prevented the hydrogen peroxide gradient from forming in response to a wound, and also strongly reduced leukocyte recruitment to the wound. Using antisense morpholinos and quantitative PCR, they then narrowed down which of the five possible enzymes could be responsible: none of the Nox enzymes seems to be involved, but knocking out Duox blocked both gradient formation and leukocyte recruitment. Problem solved: the leukocyte recruitment signal has been discovered!
Well — there may still be more to the story.
The one remaining fly in the ointment (no disrespect to Drosophila intended) is that in humans Duox is only known to be expressed in epithelia that face liquid environments, such as the gut or lung epithelia. Of course, zebrafish epithelia face liquid environments too. This raises several questions. Are dry epithelia different from wet ones in their response to wounding? If so, why? Or, do dry epithelia produce hydrogen peroxide using a different enzyme? Or have we just missed the presence of Duox in dry epithelia until now, not knowing that it could be important? Philipp Niethammer says that it’s not yet completely clear how relevant this is to mammalian wounds — but he plans to find out.
If hydrogen peroxide is indeed the signal we’ve been looking for, why would evolution have chosen this from all the possible options? Possibly important factors: hydrogen peroxide is quick to make, diffuses rapidly and can be rapidly broken down by catalases: it can get to where it’s needed very fast, and once the signal is delivered it can be removed just as quickly. Because it’s highly reactive, it can modulate the activity of many proteins, for example by reversible oxidation of cysteine residues to sulfenic acid. Proteins known to be modulated by hydrogen peroxide include phosphatases, tyrosine kinases, transcription factors, lipids, and complement factors. Philipp speculates that this ability to alter the states of many intracellular signaling pathways at once may be important in mounting a rapid, “full-on” response throughout a tissue; and he points out that the Nox/Duox family of enzymes arose at around the same time as the invention of multicellularity. It’s interesting to speculate that perhaps redox signaling has some kind of general function in regulating the collective behaviors of groups of cells, as in the well-known story of nitric oxide in vascular relaxation. More to come on this story in future, no doubt.
Niethammer, P., Grabher, C., Look, A., & Mitchison, T. (2009). A tissue-scale gradient of hydrogen peroxide mediates rapid wound detection in zebrafish Nature, 459 (7249), 996-999 DOI: 10.1038/nature08119
Really interesting post! I remember covering leukocyte recruitment to wounds in lectures (and associated H2O2) so it was really nice to hear some more to the story. You can *sort* of see by eye on the video if you squint, but it is one of those cases where the computer analysis makes things a lot easier.