Conserved amino acids at active site of an enzyme

Question

[mvkac4app]

Dear Members and Experts,

I am looking for tutorials on bioinformatic analysis where I want to find out how amino acids at the active site of a certain enzyme (eg. human DHFR) are evolved. I have training in computational biophysics and not in bioinformatics.

With some research, I came across CATH, Pfam servers, and representations as HMM logo. I have not performed such an analysis before and I wish to avoid a black-box approach to solve this problem.

If you can point to or recommend some tutorials to achieve this objective, it will be really helpful.

Thanks in advance, Mandar Kulkarni

[cpad0112]

try Evolutionary trace analysis

[mvkac4app]

Thanks !!!

Answer 1

[Mensur Dlakic]

I explain residue conservation to my students using human hand as an example. Maybe it will work here as well.

Let's say that there are 5 conserved residues in the active site that are primarily responsible for the catalysis. Let's also say that each of them is located individually at tips of your fingers. Now, if I take a ping-pong ball and in one half of it put 5 dots, each person that has 5 fingers would be able to line up tips of their fingers with those dots. It doesn't matter that people have larger or smaller hands, longer or shorter fingers, and slimmer of thicker fingers. All people can line up tips of their fingers so they occupy the same points in space. That's the thing with divergent protein (super)families that may recognize different substrates, but operate with the same type of reaction chemistry. As long as their catalytic residues are in the same spatial orientation (tips of fingers), the exact amino-acid composition and length of the connecting parts (fingers) does not matter. That is why divergent protein families may be difficult to link together as related, because the evolution forces them only to preserve what is absolutely important for catalysis (tips of fingers), while the rest of the protein changes to accommodate different substrates or simply because it can (the fingers and the palm of the hand). More specifically, all methyltransferases must bind S-adenosyl-methionine and transfer its methyl group onto a given substrate, which is why the protein part that does those two functions will be fairly conserved. The rest, and in particular the part that recognizes a specific substrate, is usually different. That is why one may have hard time proving at the sequence level that a caffeine methyltransferase is related to a histone methyltransferase. In most extreme cases, this can't be shown at all until their structures are solved, because the actual fold will reveal what can't be deduced from simple sequence comparisons.

This is depicted below, where a diverse group of nucleases, phosphatases and other signaling proteins are aligned. Beware that this is taken from my own publications, so this qualifies as a shameless self-promotion.

You will notice that there are only several absolutely conserved residues (highlighted in yellow), and some moderately conserved residues immediately around them. The further away one moves from those conserved clusters the more its sequence becomes almost random (white parts without any background color). Also note that the length of connecting parts (numbers in parentheses between sequence blocks) may be completely different, which is again that analogy with finger length.

[mvkac4app]

Dear Dr. Mensur Dlakic, Thanks for this amazing and easy explanation. I am doing this analysis to understand how the evolution of amino acids at or near active site modulates water networks or water dynamics to improve catalytic efficiency. Even though certain reactions do not involve water directly, I am trying to build a hypothesis that water near the active site affects the catalytic reaction. So, as a first step, I am trying to understand the evolution at or near the active sites for certain enzymes. In this context, please can you suggest whether residues are evolved to alter water dynamics? Thanks, Mandar Kulkarni