What is in a structure? Cell Permeability and Solubility of Series of Macrocycles and Linear Matched Pairs
There is a growing list of drugs and clinical candidates for previously-thought “undruggable” targets, demonstrating how new technologies, new modalities and, simply, new approaches to a problem can reveal the solution. The value of “fresh eyes” can never be understated! One of the reasons that the now-suspicious adjective “undruggable” came about, was the observation of target active sites that did not resemble classic “pockets”, where an (ant)agonist small molecule could bind and potentially become a drug if the stars aligned. Kinase domain targets are highly worked-over for this reason: the ATP binding sites therein are generally perfect examples of deep, “feature-rich” pockets, where molecules mimicking ATP in the most important ways also bind well, sometimes leading to potent inhibitors.
In tricky targets, you might see a wholly different situation, where the target domain presents a flat, featureless, smooth, rolling surface or groove, and no matter what classical small molecules you throw at it, nothing seems to bind! Here we appear to have reached an impasse, assuming that the rules of drug discovery from the past apply to the present and also the future. It turns out that you can bind to these seemingly intractable domains, if you break one or more of these rules Such as, using “extended rule-of-5” (eRo5) or “beyond rule-of-5” (bRo5), larger, compounds that display complementary molecular shapes to these smoother protein surfaces. In particular, macrocyclic compounds are preorganised into sphere-like or disc-like shapes that can happily occupy such sites.
So, we have solved the problem of inhibiting the undruggable target by using macrocyclic compounds, but can we claim to have drugged the undruggable if the resulting macrocycle has low solubility and 0% oral bioavailability? This is often the case, but there are exceptions and a deeper understanding of the causes of this has been the product of the work of Jan Kihlberg and others, who describe phenomena such as “molecular chameleonicity” for eRo5 and bRo5 compounds, where the properties of a given compound change (adapt, if you will) according to its surroundings. The main application of chameleonicity is the enhanced cell permeability given to macrocyclic compounds thanks to dynamic intramolecular interactions that arise before passing through a cellular membrane.
In this publication, the authors studied a family of unusual bis-aryl-ether macrocyclic compounds, interesting due to their anti-leishmanial properties in a prior report. Non-canonical amino acid building blocks were used to generate a series of analogues, changing pendant R groups and macrocycle ring size. These analogues were tested for their LogD(pH=7.4), aqueous solubility and passive permeability through CACO-2 cell monolayers. Their 3D structures were interrogated using molecular dynamics, NMR spectroscopy and X-Ray crystallography. The sparsity of such high-quality data for permeable macrocycles in the literature was noted by the authors, and the following analysis was carried out systematically in the form of Matched Molecular Pairs (MMPs), which makes us all very happy.
Some great insights into otherwise non-obvious improvements in permeability are given, particularly when the latter are not explained by the concomitant increase in LogD. Property changes are taken apart in the context of particular intramolecular interactions (NH-π, NH-n) alongside solvent-accessible polar and nonpolar surface area, all based upon the 3D structural data mentioned above. Of particular note is the comparative analysis of macrocyclic vs linear compounds, for which, 7 examples are presented.
If you would like to read more, click here
