Saturday, May 28, 2011

Math nerds identify possible cause of persistent inflammation in eczema

Researchers in London have developed a mathematical model of a molecular feedback loop thought to cause persistent inflammation in eczema patients.

Normal skin is maintained in a steady state in which old skin cells slough off and are continually replaced by new ones pushing from beneath. Flattened skin cells called keratinocytes are initially held together with bonds that are digested by enzymes called "kallikreins," according to a recent paper in the journal PLoS One. Kallikreins also cleave a receptor called PAR2 that, in normal skin, controls the skin steady state but, in eczematous skin, turns on a type of autoimmune response.

The authors of the PLoS One paper observe that, in eczema, inflammation that can start with a mild or short-lived stimulus (such as scratchy clothing) persists far longer than it does in healthy skin. They make the hypothesis that this is the result of a feedback loop between PAR2 and kallikreins. Even when the initial stimulus is gone, this feedback loop--like a screaming speaker connected to a microphone in front of it--keeps eczematous skin inflamed.

The authors ran their model on a computer (actually, they probably ran many models thousands of times on different computers, and tweaked tens or hundreds of parameters until they got the answers they wanted) and showed that it behaved consistently with data from real patients. So it's consistent with reality that there may be a positive feedback between PAR2 and kallikreins, or a negative feedback between PAR2 and a molecule called LEKTI that inhibits kallikreins.

They don't say this right out, but a potential benefit of this work--way, way, way down the line--is that if further modelling and experiments show that this feedback loop is real, then we may be able to break the loop with a custom-designed or -discovered drug, and stop inflammation quickly in eczema.


  1. I did ask myself after writing this why I chose to call the researchers "math nerds," which is hardly a term you'd pick if you wanted to carry on a civil discussion.

    I think research of this type is worth doing, but only in very small amounts. The models bear little resemblance to reality and are usually tweaked so that they match experimental results. Models are only valuable if they make useful predictions that can be tested, and in this case the researchers don't make any predictions for experimentalists to test. They do indeed develop a way to quantify the levels of PAR2, which is apparently near-impossible to measure, but their method is not simple and I have trouble believing a biologist is going to use principle component analysis when they are more used to firing a few sample tubes through a fluorescence-activated cell sorter.

    I do appreciate, though, that there are a lot of very bright people out there looking for problems to solve, and some of them have chosen to look at eczema. Keep it up, people; just give us something more useful next time.

  2. Why not title it "Asshat doesn't understand how research works"? Just as civil and meaningful.