On 16 Jan 2002, Athel Cornish-Bowden wrote:
> Good question, and one that we've tended to be a bit vague in answering. If
> you mean how do the tools of metabolic control analysis (control
> coefficients, elasticities, etc.) help in designing drugs, then I suspect
> the answer is "not much", though there are certainly those who argue that
> knowing which enzymes have the highest flux control coefficients gives a
> guide as to which enzymes you need to hit with inhibitors or activators. My
> feeling is that that is a bit optimistic and that in practice you will
> always need to do some computer simulation to get a clear idea of what will
> happen if you vary the activity of an enzyme.
With the state of MCA is it then possible to look at a problem such as
viral infection and pull apart the features that make up the replication
cycle, call this "the system" and then design therapeutics against targets
with the highest flux control coefficients? How do you correctly define
such a system since much of your parameters will be defined by
experimentals where aspects of this system are removed from the system and
so will not behave properly? (An unrelated question I have is does MCA
remove the issues of non-ideality?).
> Nonetheless, if the question refers more to the philosophy of metabolic
> control analysis than to its nuts and bolts then I continue to think it
> very important for drug design, because it forces you to start thinking in
> systemic terms; only by focussing on the whole system can you begin to
> grasp what pharmacology is all about.
I'll check out your references and get back to you, thanks. In terms of
what you are saying above, I practise "focused combinatorial chemistry".
Where I work from a protein structure and look at all the parameters such
as pocket elasticity and enzyme plasticity to define maximum and minimum
changes, then work off a fixed template to "probe" the pocket in a
combinatorial manner. The combinatorial aspect will help us with issues
such as bioavailability and resistance through mutation (enzyme
plasticity), but the results are sometimes wild, and this is where
automated approaches and "classical" combichemistry falls down. Sometimes
a subtle change can completely alter the binding mode, and if you are
careful and observe well, sometimes this provides very good information.
I imagine computer based and combichem guys would throw out much of this
sort of data as a "non-hit" and miss some of the most beautiful
information the enzyme is telling them. The other problem is good assay
inhibitors are often non-bioavailable. It is a compromise. Perhaps if MCA
could tell you at which point to stop (because it won't get better) we'd
be better off.
I have two papers on this approach and another I'm preparing if you're
interested. I'm not 100% happy with these papers though (the problems of
being one voice amongst many others - and chemists can be so damn thick!).
Glenn et al., J Med Chem 2002 Jan 17;45(2):371-81
beta-Strand Mimicking Macrocyclic Amino Acids: Templates for Protease
Inhibitors with Antiviral Activity.
Tyndall et al., J Med Chem 2000 Sep 21;43(19):3495-504
Synthesis, stability, antiviral activity, and protease-bound structures of
substrate-mimicking constrained macrocyclic inhibitors of HIV-1 protease.
Len...
