May 22, 2007
Technological Assists To Accelerate Evolution?

Some scientists at Arizona State University in Tempe are experimenting around with faster ways to find better gene and protein designs.

Nature, through the trial and error of evolution, has discovered a vast diversity of life from what can only presumed to have been a primordial pool of building blocks. Inspired by this success, a new Biodesign Institute research team, led by John Chaput, is now trying to mimic the process of Darwinian evolution in the laboratory by evolving new proteins from scratch. Using new tricks of molecular biology, Chaput and co-workers have evolved several new proteins in a fraction of the 3 billion years it took nature.

Their most recent results, published in the May 23rd edition of the journal PLoS ONE, have led to some surprisingly new lessons on how to optimize proteins which have never existed in nature before, in a process they call ‘synthetic evolution.’

I expect technological methods for creating new variations will some day allow scientists and bioengineers to search large solution spaces to come up with protein design improvements that will allow us to introduce huge improvements in human body designs. We'll not just take out the harmful mutations that we carry. We'll also replace beneficial mutations with even more beneficial mutations.

Natural selection is slow at selecting better designs because species have to go through many generations of reproduction in order to select for small improvements. Faster ways to find better designs will come in the future with testing of large numbers of gene and protein changes in microfluidic devices.

The scientists generated random mutations and tested them for the desired qualities.

Chaput’s group decided to speed up protein evolution once again by randomly mutating the parental sequence with a selection specially designed to improve protein stability. The team upped the ante and added increasing amounts of a salt, guanidine hydrochloride, making it harder for the protein fragment to bind its target (only the top 10 percent of strongest ATP binders remained). After subjecting the protein fragments to several rounds of this selective environmental pressure, only the ‘survival of the fittest’ ATP binding protein fragments remained.

The remaining fragments were identified and amino acid sequences compared with one another. Surprisingly, Chaput had bested nature’s designs, as the test tube derived protein was not only stable, but could bind ATP twice as tight as anything nature had come up with before.

To understand how this information is encoded in a protein sequence, Chaput and colleagues solved the 3-D crystal structures for their evolutionary optimized protein, termed DX, and the parent sequence.

In a surprising result, just two amino acids changes in the protein sequence were found to enhance the binding, solubility and heat stability. "We were shocked, because when we compared the crystal structures of the parent sequence to the DX sequence, we didn’t see any significant changes," said Chaput. "Yet no one could have predicted that these two amino acids changes would improve the function of the DX protein compared to the parent.

My guess is that the search for useful variations could be accelerated by introducing mutations that are not random. Rules could be developed based on a growing body of knowledge about amino acid sequences and protein functionality in order to come up with genetic variations that are more likely to improve a protein for some purpose. Also, as computers become more powerful we'll reach a point where testing of variations will become possible in computer models rather than by creating real proteins.

We are going to optimize our genomes just as computer programmers optimize and improve computer code. New types of flaws will get introduced in the process. But many more existing flaws will get fixed.

You can read the full paper for free: Structural Insights into the Evolution of a Non-Biological Protein: Importance of Surface Residues in Protein Fold Optimization.

Share |      Randall Parker, 2007 May 22 11:26 PM  Evolution Acceleration


Comments
Post a comment
Comments:
Name (not anon or anonymous):
Email Address:
URL:
Remember info?

                       
Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©