Kinase Chemistry – Just a year and a half behind the times.

Re-Engineering Gleevec

Posted by kinasepro on December 5, 2007

Alex Fernandez thinks he has a dehydron in action with Gleevec & c-Kit:

The interest of this goes way beyond this particular drug and this particular side effect. The idea is we could demonstrate for the first time that you can take a drug with side effects and re-engineer it to curb those side effects, Fernandez said in a telephone interview. (Reuters)

Adding a methyl group alpha to the the pyridine will make the activity of most kinases go away given a steric clash with the neighboring carbonyl. It’s interesting that the c-Kit activity stays behind, but the evidence given for the dehydron effect is through molecular dynamics and while as a rule I’ll try not to hold that against them, I don’t believe that the authors give a satisfactory answer as to why this new methyl group does not negate the compounds c-Kit activity through a simple steric interaction.


J Clin Invest 2007, 117, 4044
related: Cancer Res. 2007, 67 (9): 4028-33

Nippon Shanyaku took the alternate tack of making the core pharmacophore more potent and made it hit other things… In the discovery of the ph1 compound NS-187 (Inno-406):


Read all about it in the ’06 BMCL, the ’07 BMCL, the 2E2B, or through an Open Access offering here.


7 Responses to “Re-Engineering Gleevec”

  1. weirdo said

    The ignorance of academics who dabble in drug discovery is truly astounding.

    This is the “first time” a minor change has resulted in the side effects of a drug being reduced? What does this guy think medicinal chemists at big and small pharma do all day? What a dolt.

    Also, wouldn’t this qualify as a “me-too” drug, and therefore raise the ire of all of the anti-drug company hacks out there?


    ‘Pro — love the site. The effort you put in to making this one of the premier chemistry-related blogs is admirable.

  2. While even I don’t really understand why there won’t be a steric issue, a methyl group that kicks out an unhappy water molecule could confer considerable gain under optimum conditions, at least 2 kcal or so. MD is not a bad method for assessing such dehydration if well-parametrized.

  3. Were you aware of this?:

    A monkey wrench in the kinase machine:
    Last year Zhang, Kuriyan and colleagues demonstrated that an asymmetric dimer interaction between EGF receptor kinase domains is a key element of receptor activation. They now show that a cellular protein that inhibits receptor activity targets this dimer interface, not only uncovering an important regulatory mechanism but also opening a new route to therapeutic kinase inhibition.
    Daniel J Leahy

  4. rosko said

    If you look at Fig. 3B in the original J. Clin. Invest. paper, the methyl group definitely appears to be pushed out of the plane of the pyridine ring. In the real complex, either the ring must twist about the bond linking the pyridine and pyrimidine, or (as sounds more likely assuming the dehydron theory is correct), there is a change in the local backbone torsion angles in the hinge that rotates the carbonyl out of the binding pocket and toward Gly 676 to form a sort of beta-hairpin. If I remember correctly, peptide bond flips in the hinge region of kinases upon ligand binding have been observed before.

  5. kinasepro said

    The kind of flip I think your describing is observed in Jnk1 and p38 where the alpha residue is leucine:2G01, 2I0H and others. No argument there, but in c-Kit, and Abl the residue is Tyr, and Phe respectively.

    I’m always happy to proved wrong, but I think its much less likely for that residue to flip wherein the neighboring hinge residue bears an aryl ring like that.

    Lets keep an eye on the big picture here though. Tasigna is a re-engineered Gleevec. NS-187 is kind of a a re-engineering in a patent-busting, more potent, less toxeriffic sort of way. Methyl-Gleevec? Which has to be dosed in a special formulation just to get it into solution for a cellular assay?? Which is not statistically significantly more potent but you put that in the title anyways??? This does not a re-engineering make. And I wouldn’t be nearly so critical of this offering if they had chosen a more humble title, and not called a press release to tell the world about it.

    The only important question here to my mind is still un-answered: Why is it so easy to inhibit c-Kit. Any reasonable person who has poured over a few SAR tables of multi-target inhibitors I believe will attest to this. If its due to a specific ‘dehydron’? I’m all ears, but you’d better bring a little more evidence for me to listen.

  6. BRAF.V600E said

    On a related note, I just read this paper on in silico inverse screening:

    DOI 10.1016/j.chembiol.2007.10.010

    The answer to all those uncomfortable kinase selectivity questions?

  7. rosko said

    “The kind of flip I think your describing is observed in Jnk1 and p38 where the alpha residue is leucine:”

    Maybe this is what I was thinking of when I said I remember seeing a peptide bond flip in kinases before. What it seems would happen in the case of the dehydron theory, though, is a rotation in the opposite direction. In other words, if you look down the calpha–C=O bond with the C=O in front (as in a Newman projection), in Jnk1 and p38 the bond rotates counterclockwise, whereas in c-kit it would hypothetically rotate clockwise. This would put it more in plane with the peptide bond 2 residues farther along the sequence. I’m not saying that happens, it’s just that’s what would need to happen for the dehydron theory to make sense.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: