Of course you do not need a lens to see.  The nautilus does just fine without one.  But the benefits of a lens are strong enough that they have evolved across the animal phyla, and even occur in at least one protist.  The function of the lens is to provide a dense structure that will refract light on its way to the retina.  A common solution for making a lens is to express proteins at very high concentrations, especially in aquatic eyes that do not get the benefit of refraction from the cornea.  But interestingly, different organisms use different proteins to do the job.  And as Darwin would have predicted, these proteins are not created de novo, but are borrowed from other parts of the body.  While debates rage about whether the developmental mechanisms to make an eye evolved once or multiple times, the lens clearly has multiple origins.  And each time, organisms have drawn from their biochemical toolbox when producing proteins at high concentration in the lens.  These densely packed lens proteins have diverse evolutionary sources, but they are all referred to as lens crystallins.

Where did lens crystallins come from?  Some are basic housekeeping enzymes, others are protective proteins produced by cells when they are stressed and others have more mysterious origins.  But many of the proteins recruited to build lenses maintain their original function in other parts of the body, and are encoded by the same gene.  That means that one DNA blueprint can simultaneously make a metabolic enzyme like lactate dehydrogenase throughout the body, but when this gene is used in the lens of a duck, its protein product becomes a structural material for bending light.  This concept of gene-sharing, where one gene contains the code for making protein with multiple functions, was first proposed to explain the evolution of lens crystallins.

Were these lens crystallins recruited because of their original enzymatic functions, or were they simply convenient building blocks for the dense packing required to make a lens?  For many crystallins the answer is – both.  My lab does research on an abundant lens crystallin family found in the vertebrate lens, the alpha crystallins.  All vertebrates contain at least two closely related alpha crystallins that resulted from a gene duplication event near the beginning of vertebrate evolution.  When the gene sequences for the alpha crystallins were resolved in the late 1980′s it was clear that they were small heat shock proteins.  This family of proteins is produced by cells that are under stress – perhaps because they are too hot, or are encountering dangerously high oxygen levels.  

So why was a stress-induced protective protein being used to make up to 30% of a vertebrate lens?  There are two reasons.  First, it turns out that alpha crystallins make great building blocks.  You can pack them in at very high concentrations and still maintain the necessary protein fluidity needed in the lens.  But second, the same protective function that they serve in other parts of the body comes in very handy in the lens.  The central cells of the lens destroy their own nuclei and other cellular machinery to prevent the scattering of light as it passes through the retina.  No nucleus means no new protein, so our lens cells must make do with the same proteins for their entire life.  Old proteins get shabby, fall apart, and then start sticking to each other.  This sticky mess interferes with the passage of light, and voila – you now have a cataract.  But alpha crystallins use their stress protective function to prevent this aggregation of old, decrepit proteins, preserving lens transparency until they are used up, generally starting at around age 50 for humans.  

At the macroevolutionary scale a wide array of proteins have been co-opted as lens building blocks because of their structural and enzymatic properties.  My lab is currently investigating the microevolutionary story.  How do small changes in alpha-crystallins alter their ability to function at different temperatures, for example?  Darwin perhaps sensed that the eye would yield excellent examples of his two great ideas: descent with modification and natural selection.