Shadow functions

August 11, 2010 § 3 Comments

There’s increasing evidence from Drosophila that genes that are important in development are not satisfied with having just one set of enhancer sequences to drive their transcription.  Why is this?  Sometimes, two different enhancer sequences have obviously different functions: they respond to different sets of transcription factors, for example, and create different patterns of transcription.  But often the patterns produced by the different enhancers — so-called “shadow” enhancers — are apparently redundant.  Are these shadow enhancers just “junk”, or are they doing something useful?

Angela DePace pointed me to a recent paper (Frankel et al. 2010. Phenotypic robustness conferred by apparently redundant transcriptional enhancers. Nature 466 490-3 PMID: 20512118) that provides strong evidence that the apparently redundant enhancers are actually very important.  Frankel et al. discovered two new shadow enhancers, which contribute to the expression of the gene shavenbaby.  Before this work, we felt we had a pretty reasonable understanding of the control of shavenbaby expression: three enhancer modules were known, and these three modules working together were able to recapitulate the proper pattern of shavenbaby expression in the Drosophila embryo.  As the name suggests, shavenbaby  [wouldn’t science be less interesting without the wacky gene names created by Drosophila geneticists?] is required to make bristles — trichomes — on the abdomen of the Drosophila larva.  Not all species of Drosophila make such bristles, for example Drosophila sechellia has lost them somewhere in evolution; reassuringly, D. sechellia also shows a loss of function in the three enhancer modules we originally knew about.  So our original three enhancers seemed to explain essentially everything we know about the expression patterns of shavenbaby.   Why does the fly need more?

One of the theories that has been floating around is that these “shadow” enhancers are important for robustness: they help maintain the proper pattern of transcription even when conditions aren’t ideal.  Frankel et al. set out to test this idea by knocking out the chunk of the Drosophila X chromosome that carries the shadow enhancers, and looking to see what happens to bristle development.  As expected, losing the shadow enhancers doesn’t do much when the embryos are reared under normal lab conditions; the number of bristles might be reduced, but not by a lot.  But when you change the temperature of the room the flies are living in, the situation is very different: embryos lacking the shadow enhancers have only about 2/3 the number of bristles as normal when they’re grown either at 17 °C (too cold) or at 32 °C (too hot; Goldilocks temperature for flies is 25 °C).  So these apparently redundant enhancers do have a function after all.  They — or at least, something that was lost when Frankel et al. knocked out the bit of the chromosome that they’re carried on — are responsible for making it possible for flies to develop normal bristle patterns at a wide range of temperatures.

OK, let’s deal with that caveat before we go any further.  Is it actually the enhancers that are responsible, or is it something else that happened to be nearby on the chromosome?  To check, Frankel et al. constructed a cDNA with the shavenbaby gene under the control of one of the shadow enhancers — called Z — and put it back into the knockout flies, on a different chromosome from where it normally sits.  This completely rescued the expression of bristles in the relevant patch of embryo skin, at both high and low temperatures.  (The two shadow enhancers seem to contribute to temperature robustness in different bits of the embryo.)  Since this works even when the Z/shavenbaby combination is introduced into a different chromosome, it’s pretty clear that Z doesn’t have to interact with the other enhancers that control shavenbaby expression — the ones we knew about before — to have its effect.  Instead, the authors propose that Z‘s effect is simply to boost the expression of shavenbaby in certain cells.

Don’t go away thinking that the authors were super-lucky to try the one form of manipulation that would show the importance of these enhancers: this effect is not limited to temperature.  Frankel et al. chose another gene that’s important in bristle development, wingless, and checked to see whether the presence or absence of the shavenbaby shadow enhancers makes a difference to a fly’s sensitivity to wingless gene reduction.  Flies that are heterozygous for wingless (carrying one normal gene and one non-functional gene) develop perfectly normal bristles when the shavenbaby shadow enhancers are present.  But flies that lack the shadow enhancers and have only one functional wingless gene show a significant decrease in bristle number, even at 25 °C.  So losing the shadow enhancers makes the fly’s developmental program more fragile in at least two different ways.  The role of the shadow enhancers — now better named “secondary” enhancers — appears to be to canalize the pattern of development, making sure that the normal phenotype of the fly emerges even in the face of significant genetic or environmental perturbations.

It has been proposed that shadow enhancers might play another important role in evolution by, essentially, being unnecessary and therefore being able to evolve more rapidly than the core enhancers that drive the gene expression necessary for life.  It could still be true that having multiple enhancers makes it easier to evolve new functions.  But at least for the shavenbaby enhancers, there is now a more direct explanation for why shadow enhancers exist: they’re needed to deal with the fact that we live in a profoundly imperfect world.

Frankel N, Davis GK, Vargas D, Wang S, Payre F, & Stern DL (2010). Phenotypic robustness conferred by apparently redundant transcriptional enhancers. Nature, 466 (7305), 490-3 PMID: 20512118

Tagged: , ,

§ 3 Responses to Shadow functions

  • Ik says:

    As a scientist who uses Drosophila and as a theoretician who understands the limitations of current theoretical offerings – and I mean no disrespect here – the proposed ideas are embedded with fundamental assumptions and faulty logic.

    Without the ability to understand the origin, emergence, and ordering of RNA, protein, and DNA in all living cells, all current ideas – even those in Nature are, unfortunately, speculation.



  • Angela says:

    I thought this paper had really exciting implications for regulatory evolution – specifically models of duplication and divergence of regulatory elements. We still don’t know what happens mechanistically when an enhancer duplicates. The authors speculate that duplication increases expression of the target, and that this leads to robustness to temperature variation. Why would this be the case? Other mechanisms are possible, such as differential activity of the two enhancers at different temperatures.

    It’s also not clear how many ways there are to get a higher dose of target gene expression. Could an existing enhancer be tuned to express more strongly? Or are some enhancers already at their production limit and therefore require duplication to increase the level?

    @lk I always enjoy papers that describe a phenomenon clearly, and allow the authors the fun of speculating on the underlying mechanisms. There’s always more science to do!

  • Ik says:

    I, too, enjoyed the paper. But I should note that the current model to understand how genetic information is transmitted is incomplete and incorrect. The underlying phenomenon is not mechanistic, nor, alas, can it be reduced. That’s what I have been working on for the last five years.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

What’s this?

You are currently reading Shadow functions at It Takes 30.


%d bloggers like this: