[Nescafe]

Hi,

I really want to open a post about this. I am in the medicinal chemistry stream and am suppose to know a lot about these sort of interactions between lets say ligand + protein. I do have quite a bit of knowledge on the subject but I really need to know more.

Does anyone here have a great understanding of entropic and enthalpic contributions of binding, what is usually more favored, how is one to optimize such interactions etc.

Let's start discussing this cause its confusing
Thanks,

Nescafe.

[Doc Oc]

I read an interesting review a while ago talking about this.  The gist of it was that if you run an assay and collect a bunch of hits, your time is best spent working on the ones with the best enthalpic contribution.  The entropic part is the one that is more easily controlled (ie; cyclizing a structure, adding functional groups to control conformation or increase lipophilicity, etc).  The enthalpy, however, is something that doesn't have as good of a roadmap to optimization, so it's best to pick a compound with a strong contribution there.  I'll see if I can dig that up and I'll post the reference here.

[Nescafe]

I read an interesting review a while ago talking about this.  The gist of it was that if you run an assay and collect a bunch of hits, your time is best spent working on the ones with the best enthalpic contribution.  The entropic part is the one that is more easily controlled (ie; cyclizing a structure, adding functional groups to control conformation or increase lipophilicity, etc).  The enthalpy, however, is something that doesn't have as good of a roadmap to optimization, so it's best to pick a compound with a strong contribution there.  I'll see if I can dig that up and I'll post the reference here.

Interesting, I'd definitely would like to have a look if you don't mind going through the trouble of finding it.

[Doc Oc]

The reference is Freire, E. Drug Discovery Today  2008, 13, 869.

The Pubmed link has the entire article available.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2581116/

[orgopete]

I read an interesting review a while ago talking about this.  The gist of it was that if you run an assay and collect a bunch of hits, your time is best spent working on the ones with the best enthalpic contribution.  The entropic part is the one that is more easily controlled (ie; cyclizing a structure, adding functional groups to control conformation or increase lipophilicity, etc).  The enthalpy, however, is something that doesn't have as good of a roadmap to optimization, so it's best to pick a compound with a strong contribution there.  I'll see if I can dig that up and I'll post the reference here.

I agree with the point being made here. I too had reached that conclusion. I was in the ag area and it was common to see a relatively small number of modes of action to be found from screening compounds. This is how I reasoned it.

Linus Pauling has advanced a model that enzymes catalyze reactions by reducing the activation energy of reactions. (I am presuming this is largely an enthalpic effect.) He then argued that mimics of transition states may serve as inhibitors of a reaction. I consider this to be the working paradigm for the modern pharmaceutical industry.

If you put those two facts together, then you should conclude that when you screen for inhibitors, you are actually screening for high activation energy reactions. That is, enzymes that catalyze high activation energy reactions could have the tightest binding constants. The reduction in the activation energy should be proportional to the degree to which the compound is bound to the enzyme. That would explain why a small number of modes of action should turn up most frequently, they must catalyze high activation energy reactions, and therefore have great affinity for their substrates and inhibitors. Therefore, even relatively poor mimics should still inhibit a reaction. It also follows that hard working enzymes will be more promiscuous than enzymes for reactions with lower activation energies.

This line of thinking also raises a temporal aspect of enzyme catalysis. This could be important as one encounters, slow tight binding enzymes. This Pauling model enables one to understand enzyme turnover. The transition state is bound more tightly that substrate or product. Since that is the case, then the relatively lower affinity of substrate and product allows their exchange. Tight binding of a substrate or product should inhibit an enzyme also. It would seem then that a slow reaction should appear to have a high affinity while a fast reaction should appear to have a lower affinity. This is something I was interested in modeling, but I don't know if that is actually true or useful. That is, tight binding may simply be tight binding and velocity may not matter at all.

[Nescafe]

Does anyone know a good book to read on this subject?

What does an enthlpic gain really mean? Like a new hydrogen bond between your drug and protein is considered an enthalpic gain, wouldn't this in turn mean an entropic loss. To me it just seems like you'll never win! Always one step forward one step back when it comes to enthalpic/entropic optimization of drug/target interaction.


I'd appreciate it if we could keep this topic alive for a while, I'm really interested and confused at the same time,

Cheers,

Nescafé.

[fledarmus]

Yes, enthalpic gains usually mean entropic losses. The challenge is to find enthalpic contributions that are stronger than the entropic losses, or to find other ways of tying the entropic losses into the structure of your inhibitor rather than in the inhibitor-enzyme interaction.

One example of this that drives the molecular modelers nuts is the structured water molecules. These are water molecules which are hydrogen bonded into the binding site of the enzyme. If your molecular modeling program is not sophisticated enough to handle these water molecules, it can appear that simply adding a hydrogen bond donor (for example) to your molecule to reach a previously unused hydrogen bond acceptor on the enzyme would be a great way to pick up activity. In many cases, however, what you gain in enthalpy by adding the hydrogen bond donor to your inhibitor is lost by breaking the hydrogen bond to water, and the additional entropy kills you.

There are ways to make entropy work for you. If you can build the appropriate geometry into your inhibitor by using ring systems or stereochemistry to fix two sites of interaction into exactly the right geometry, then that entropy becomes part of the formation of the inhibitor and you don't have to make up for it when binding your inhibitor into the enzyme, as a floppier inhibitor with more degrees of freedom would.

Of course, the really strong enthalpic contributors would be those that actually make chemical bonds to the enzyme (irreversible inhibitors), but it is difficult to work in the appropriate amounts of chemical stability and selectivity into these types of molecules and many drug discovery programs avoid them. I would say "avoid them like the plague", but that wouldn't account for those drug discovery programs that are actually working on the plague and still wouldn't consider an irreversible inhibitor.

[Babcock_Hall]

According to G. Zubay, Biochemistry, 4th edition (page 31), the binding of an ester substrate to the enzyme pepsin occurs with an entropy increase of 20.6 eu/mole, and for urea binding to urease, the increase is 13.3 ue/mole.  One entropy unit is one calorie per degree Kelvin.  Therefore, the binding of an enzyme to a small molecule does not necessarily lead to a loss in entropy.

[Nescafe]

According to G. Zubay, Biochemistry, 4th edition (page 31), the binding of an ester substrate to the enzyme pepsin occurs with an entropy increase of 20.6 eu/mole, and for urea binding to urease, the increase is 13.3 ue/mole.  One entropy unit is one calorie per degree Kelvin.  Therefore, the binding of an enzyme to a small molecule does not necessarily lead to a loss in entropy.

I'm confused.

So as a result of ester binding the system is now more disordered? If so, then there is no rule of thumb when it comes to explaining with this stuff is there?

[fledarmus]

I don't have that textbook, but my guess would be that the addition of a ligand to that enzyme is done at the cost of disrupting multiple enzyme-water or enzyme-enzyme interactions.

The rule of thumb would be the same as in any other chemical system - the strongest interactions will occur where enthalpic and entropic contributions are both favorable, but a strong enough favorable enthalpic contribution will override a weaker unfavorable entropic contribution and vice versa.

[Babcock_Hall]

I know just enough about this to be dangerous.  However, when two macromolecules (like DNA and a protein) come together, sometimes the binding is also entropically favored.  This may be because of the release of waters of hydration and counterions that otherwise would surround each molecule on its own.

[qw098]

I know just enough about this to be dangerous.  However, when two macromolecules (like DNA and a protein) come together, sometimes the binding is also entropically favored.  This may be because of the release of waters of hydration and counterions that otherwise would surround each molecule on its own.

Yes, correct.

In reality... hydrophobic interactions are not about two non-polar molecules being attracted to one another. By two non-polar molecules coming together, it creates the release of water molecules which is entropically favored.

[Babcock_Hall]

I agree with respect to hydrophobic interactions.  However, when a protein binds DNA, cations (sodium, potassium, or magnesium) are released from being bound to the negatively charged phosphodiester bonds.  I seem to recall that about 14 sodium ions were released per mole of protein, but I could easily be wrong.  This review (which I used to have as a photocopy, IIRC) might be pertinent:  Q Rev Biophys. 1978 May;11(2):103-78.  Thermodynamic analysis of ion effects on the binding and conformational equilibria of proteins and nucleic acids: the roles of ion association or release, screening, and ion effects on water activity.  Record MT Jr, Anderson CF, Lohman TM.  http://www.ncbi.nlm.nih.gov/pubmed/353875

[Nescafe]

I don't have that textbook, but my guess would be that the addition of a ligand to that enzyme is done at the cost of disrupting multiple enzyme-water or enzyme-enzyme interactions.

The rule of thumb would be the same as in any other chemical system - the strongest interactions will occur where enthalpic and entropic contributions are both favorable, but a strong enough favorable enthalpic contribution will override a weaker unfavorable entropic contribution and vice versa.

Did you mean  "will override a weaker unfavorable loss(?) in entropic contributions and vice versa"

[fledarmus]

I don't have that textbook, but my guess would be that the addition of a ligand to that enzyme is done at the cost of disrupting multiple enzyme-water or enzyme-enzyme interactions.

The rule of thumb would be the same as in any other chemical system - the strongest interactions will occur where enthalpic and entropic contributions are both favorable, but a strong enough favorable enthalpic contribution will override a weaker unfavorable entropic contribution and vice versa.

Did you mean  "will override a weaker unfavorable loss(?) in entropic contributions and vice versa"

Sorry, I was trying to avoid terms like "gain" and "loss", "positive" and "negative", because they really depend on which part of the system you are trying to view and how you are writing the reaction. I thought "contributions [to total energy] which favor binding" and "contributions which are unfavorable to binding" would be clearer. Apparently not