07 July 2014

Halogen Hydrogen Bonding...Designable or Not?

The use of brominated fragments for X-ray screening is well known; it was the basis for former company SGX (now part of Lilly).  The purported advantage of brominated fragment is that you can identify the fragment unambiguously using anomolous dispersion.  In this paper, they are focused on using fragments to identify surface binding sites on HIV protease.  Prior work has focused on creating a new crystal form (complexed with TL-3, a known active site inhibitor) that has four solvent accesible sites: the exosite, the flap, and the two previous identified sites.  They took 68 brominated fragments and soaked these crystals: 23 fragments were found.  However, most of these actives were uninteresting.  Two compounds were found to be interesting, one bound in the exosite and one in the flap site. 
So, what's interesting in this paper?  Well, they (re)discover that brominated fragments can bind all over with a variety of affinities.  However, the bromine allows you to unambiguously identify those fragments through anomolous dispersion.  This is NOT interesting.  They discover that although it is a subject of much debate lately: specific interactions of the ligand with the target dominate the "bromine interaction".  This IS interesting.  They do not discuss this in much detail, but their grand extrapolations of this method to general applicability I don't buy. 

 I think the key take away from this paper is whether the halogen hydrogen bond undesignable and just a subject of serendipity? 


Dan Erlanson said...

I think what strikes me most here is the high fraction of promiscuous binders. I wonder if this reflects this target in particular, whether brominated fragments are more prone to non-specific binding, or whether anomalous scattering is just more likely to detect such fragments.

Unknown said...

In addition to any specific interactions the bromine can form, it is a big greaseball. Increase lipophilicity and you are going to increase promiscuity. Especially doing soaking experiments at high concentration. Basically any hydrophobic site will have a compound binding to it and the bromine allows you to see it more easily.

Regarding designing halogen bonds in, this is of course very feasible. The thing is that you need to know the geometry requirements and this will likely have to be done through manually modifying your pose in the binding site. I doubt that most docking tools currently available will adequately assess a halogen bonding interaction.

Dr. Teddy Z said...

I agree with Lewis. Big ole greaseball yields promiscuous hits. I vaguely recall this was one of the problems with SGX and they in the end ended up dropping the bromines largely, only using them when they need the anomolous dispersion. Can someone confirm or deny this?

Joe Bauman said...

We found a high fraction of halogenated compounds binding to HIV-1 reverse transcriptase (reviewed last year http://practicalfragments.blogspot.com/2013/05/a-slew-of-sites-for-fragments-in-hiv.html) and this was part of the inspiration for this protease screen. If you count every time they saw an anomalous peak then you get a very high 33% hit rate but the hit rate for compound where they could actually see the fragment in electron density (besides the bromine) the rate became a more appropriate 6%. This is still a substantial increase from the earlier screen with the ActiveSite library (0.8 %). I would say that if the fragment is clear in electron density then you must be seeing additional interactions that can be taken advantage of through fragment evolution.

One issue with brominated fragments is that they are such great chemical handles that they can make life difficult for medicinal chemists when the bromine is also forming significant interactions with the protein.

When we have found bromo containing fragments we would usually see if similar/better results can be obtained with chloro, fluoro, methyl - often times they can be but certainly not always. I wonder if they used fluorinated/chlorinated fragments would a comparable hit rate been achieved.

I would argue that halogen containing fragments have promise in triaging otherwise failed fragment screening campaigns by giving the compounds a bit more propensity for binding wether promiscuously or not. Sort of like adding a mobile tether to your compound. So we like having 10-20% of the library halogenated. If all you see is the halogen in the electron density then you might as well soak in NaBr and you would gain the same information. But if you see the full fragment and it is making sensible contacts then you might have something informative.
Once you have the hit then you can get analogs without the halogen (if it is really an issue) and move on from there.