[Protein-analysis] Re: Question about enzymes

[Tom Anderson]

On Wed, 18 Oct 2006 lucasea from sbcglobal.net wrote:

> "Tom Anderson" <twic from urchin.earth.li> wrote in message
> news:Pine.LNX.4.62.0610171928500.24308 from urchin.earth.li...
>
>> - the mammalian fatty acid synthase, a homodimer which catalyses no less
>> than four reactions [1]
>
> Just curious (this is an honest question), are these four reactions
> catalyzed by four different active sites?

Yep - as Bob mentioned, they are (and it's six, not four - sorry!). See 
here:

http://www.rpi.edu/dept/bcbp/molbiochem/MBWeb/mb2/part1/fasynthesis.htm#fasyn
http://www.ethlife.ethz.ch/e/articles/sciencelife/banFAShumfung.html

>> Because, having looked at more enzymes than you can shake a 
>> spectrophotometer at, biochemists have observed that most do only have 
>> one active site. And why is that? Good question; i'm not aware that 
>> there's a definitive answer, and not sure that there even could be one.
>
> Again, not the opinion of a biochemist, but I suspect it's basically a 
> probability/entropic argument.  I suspect that the chances of a new 
> function arising in a pre-existing enzyme the has some other function 
> without destroying that original function is less than the function 
> arising in a new protein

A what?

> or than the new function arising in the original enzyme but at the same 
> time destroying the original function of that enzyme.

I'd certainly agree with that - provided that the new activity is pretty 
similar to the old one. If it's wildly different, then i don't see how the 
old active site would be any more likely to lead to the new one than some 
other random cleft on the molecule.

Not that i have a better explanation. There must be work on the evolution 
of new active sites, though - time to hit pubmed ...

> Intuition also tells me that it would be a low entropy configuration to 
> have two active sites in a single enzyme, as compared to the same two 
> active sites in two separate enzymes.  Maintaining this lower-entropy 
> configuration costs the organism more free energy, and that's usually 
> not in the best interests of the organism, evolutionarily speaking.

Hmm. In terms of literal free energy, i don't think that works - it's not 
as if the parts of an enzyme other than its active site are random! 
There's still a specific sequence and fold, it's just not catalytically 
active. Modulo the costs of the actual amino acids involved, making an 
active site and a chunk of inert protein cost the same amount of energy 
(not that i'm saying the rest of the enzyme is inert - the bit that's not 
the active site is where crucial regulatory and structural interactions 
happen).

If you're talking in a slightly more metaphorical sense, about the 
constraints having two active sites would impose on the configuration 
space explored by evolution, then yes, that's an interesting idea. Loading 
more function into a protein means it's more vulnerable to mutations, and 
so less robust over the generations.

My theory is, as student_A suggested, and i rather rudely dismissed, that 
it allows finer control over activities - with each activity on a separate 
protein, the cell can control the level of each individually, making just 
as much as it needs. Where this falls down is where you have enzymes which 
are needed with fixed stoichiometry, such as where they form a pathway 
with no branches. But then this is exactly the sort of situation where we 
*do* see multi-activity enzymes, like fatty acid synthase.

So, er, yes, i don't know.

tom

-- 
packaheomg sogma's