In the 1990s, Suzanne Rutherford and Susan Lindquist were studying fruit flies with a mutated version of the Hsp90 gene and found that the absence of this single gene led to a wide range of developmental defects.  This was surprising not only because Hsp90 isn’t directly related to development, but also because of the remarkable breadth of its impact.  Uncovering how this gene affects so many aspects of development has led to an intriguing story linking responses to environmental stress with the evolution of developmental pathways.

To figure out what processes a gene is involved in, geneticists often study the changes in individuals with a mutated copy of the gene.  For example, by observing that individuals with a mutation in the PAX6 gene have malformed or missing eyes, researchers discovered that this gene plays an important role in the developing eye, earning it the alternative name eyeless.  In their 1998 paper, Rutherford and Lindquist reported that fruit flies with a mutant version of Hsp90 had deformed eyes and legs, extra antennae, changes in the wing veins, altered patterning and partial or complete transformation of one organ into another.  They also discovered that they could selectively breed the deformed flies to keep the deformity even if they had a normal, unmutated copy of Hsp90.  How could one gene affect so many processes?  How do the deformities become independent of Hsp90 and persist even in offspring with a functional copy of the gene? Researchers have since reported similar findings in zebrafish and in the plant Arabidopsis thaliana.  Given its influence over so many processes in distantly related organisms, it seems clear that, whatever role it plays, Hsp90 must be of some evolutionary significance.

Like most genes, the Hsp90 gene simply encodes a protein, the Hsp90 protein.  The role of genes is to carry the information needed to make proteins, which are the molecular machines that actually do the work in our cells; genes have “an effect” because of the activity and interactions of the protein they encode.   The Hsp90 protein is a “heat shock protein”, a class of proteins that protect cells from heat damage by interacting with other proteins to stabilize their shape. The shape of a protein, which depends on its composition and environmental conditions, determines its ability to function; misshapen proteins work poorly, if at all.  Many human-built machines also accomplish their task thanks to their shape; wheels would be much less efficient if they weren’t round and a key has to be a specific shape to open a lock.  Increases in temperature cause proteins to change their shape and stop working correctly.  When you cook an egg, the heat changes the shape of the proteins which causes the yolk to harden and the white to solidify and turn opaque.  By acting as a kind of scaffolding for other proteins, the heat shock proteins help them keep their shape at higher temperatures.

Hsp90 is a special heat shock protein.  In addition to stabilizing proteins during heat stress, it also interacts with a diverse set of unstable proteins under normal conditions and helps them keep the correct shape.  Mutations which would change the shape of these proteins are effectively masked by Hsp90, since it maintains their “normal” shape despite the mutations.  When Hsp90 is absent or inhibited, these proteins don’t get stabilized and mutations in them can change their shape and function, affecting the development of the organism.  The same thing can happen if developing embryos are exposed to higher temperatures at the right stage.  Hsp90 gets recruited to respond to the heat stress, leaving the proteins it normally stabilizes without any support; if this happens at a point when those proteins are needed, their malfunction results in developmental defects.  Since Hsp90 interacts with proteins involved in a wide range of processes, the defects will appear in a variety of tissues and take many different forms.

Hsp90 also stabilizes groups of proteins that act in the same pathway, which is why the researchers were able to breed flies that were deformed even with a normal version of Hsp90.  Once the mutated version of the proteins were exposed, they could be increased by selection to the point that too many components of the pathway were defective for Hsp90 to cope with.  By making several parts of a developmental pathway robust against genetic changes under normal conditions, Hsp90 also makes the entire pathway vulnerable to drastic changes in the environment.

Its dual role in stabilizing proteins under normal conditions and responding to heat stress positions Hsp90 to act as a link mediating between development and changes in the environment, giving it an important evolutionary role.  Hsp90 acts as a buffer, allowing mutations to accumulate unseen, free from the pressures of selection.  When conditions change and Hsp90 is no longer able to shield these mutations, they become exposed and lead to a diverse array of developmental changes.  Although most of the changes are likely to cause problems, any that happen to help will be preserved by natural selection; if selection for them is strong enough, they can even escape the control of Hsp90.  Thanks to its dual function, Hsp90 stores up mutations when they’re not needed and then exposes them all at once, making it possible for rapid bursts of change to occur during evolution.

Update 17-12-2012: During this morning’s commute I read the following editorial note in my copy of D’Arcy Thompson’s On Growth and Form (an excellent book!):

"Another possibility has been suggested by various authors, but 
stated most explicitly by Waddingoton [The Strategy of Genes, 1957] 
who bases his conclusions on a series of particularly interesting 
experiments.  For instance, he subjected a certain strain of fruit 
flies to a temperature shock during their development and found that 
a low percentage of the flies produced wings that lacked a cross 
vein.  If he now selected these cross-veinless flies and repeated 
the experiment, the percentage of cross-veinless individuals 
increased, and they continued to do so over a series of successive 
generations, each with a shock treatment.  The surprising thing is 
that after a while some of the flies appeared cross-veinless without 
the temperature shock; the environmental prodding was no longer 
needed."

These are the very same observations which Rutherford & Lindquist later made and which form the core of this story, though it would be another four decades before the mechanism behind them was unravelled.

Refs

Chen, B., & Wagner, A. (2012). Hsp90 is important for fecundity, longevity, and buffering of cryptic deleterious variation in wild fly populations BMC Evolutionary Biology, 12 (1) DOI: 10.1186/1471-2148-12-25
(This paper is open access and the introduction provides an excellent overview. It’s a great resource if you’re interested in this subject and want to read about it in more detail.)

Queitsch, C., Sangster, T., & Lindquist, S. (2002). Hsp90 as a capacitor of phenotypic variation Nature, 417 (6889), 618-624 DOI: 10.1038/nature749

Rutherford SL, & Lindquist S (1998). Hsp90 as a capacitor for morphological evolution. Nature, 396 (6709), 336-42 PMID: 9845070

Sangster, T., Salathia, N., Lee, H., Watanabe, E., Schellenberg, K., Morneau, K., Wang, H., Undurraga, S., Queitsch, C., & Lindquist, S. (2008). HSP90-buffered genetic variation is common in Arabidopsis thaliana Proceedings of the National Academy of Sciences, 105 (8), 2969-2974 DOI: 10.1073/pnas.0712210105

Yeyati PL, Bancewicz RM, Maule J, & van Heyningen V (2007). Hsp90 selectively modulates phenotype in vertebrate development. PLoS genetics, 3 (3) PMID: 17397257