Keedy et al. 2018: An expanded allosteric network in PTP1B by multitemperature crystallography, fragment screening, and covalent tethering.

Allostery is defined as a conformational/activity change of an active site due to a binding event at a distant (allosteric) site.

The paper I presented in the journal club tried to decipher the underlying mechanics of allostery in PTP1B. It is a protein tyrosine phosphatase (the counter parts of kinases) and a validated drug target. Allosteric binding sites are known but so far neither active site nor allosteric site inhibitors have reached clinical use. Thus, an improved mechanistic understanding could improve drug discovery efforts.

I choose the paper because it contains two interesting crystallography methods. Multitemperature crystallography (for further reading see another blog post on room temperature crystallography) and fragment screening which displayes a practical use case of the PanDDA method.

Crucial for PTP1B’s catalytic function is the WPD loop. The transition from an open (inactive) to closed (active) conformation is rate-limiting (Figure 1 displays both conformations).

Multitemperature crystallography

Apo structures of PTP1B have been solved before at 100K. But in the only structure (PDB ID 1SUG) where the WPD is free of crystal lattice contacts (which means that PTP1B molecules of the protein crystal that was used for solving the structure did not touch each other at WPD loops) the authors noticed unexplained electron density. Re-refining the structure by explaining electron density with a two-conformer model of open and closed WPD loop, revealed WPD loop populations of 67 % in the closed conformation and 33 % in the open one.

Apo structures were then also solved at 180K, 240K and 278K (“room temperature”). Figure 2 shows the fitted two-conformer models with the respective population sizes of the open and closed conformations. Figure 3 plots the occupancy of the open WPD loop against the temperature range, showing a trend towards open conformations at non-cryogenic temperatures.

Additionally to the changes of the WPD loop itself, major structural changes were also observed at three allosteric sites: BB (benzbromarone) site, Loop 16 site and 197 site. Minor changes occured in the hydrophobic core („lubricant“).

Structural comparison with complex of known allosteric inhibitor BB revealed that pre-existing conformations of apo PTP1B are stabilised by the allosteric inhibitor.

Fragment screening crystallography

To assess the ligandability of PTP1B, 1918 crystals which were soaked with 1627 (small-molecule) fragments and 48 apo crystals were attempted to be structurally resvolved.
1774 of these experiments yielded diffraction data that could be processed. Data of fragment soaked crystals was then run through PanDDA which identified 381 putative binding events. Manual analysis yielded 110 structures with fragments that could be structurally resolved. Figure 4 shows the identified 11 bindinghot spots (+ active site).

Interestingly, the three allosteric sites with major structural changes identified by multitemperature crystallography had the most fragments bound of all identified hotspots (see Figure 5).

The comparison of apo and holo structures could not show that fragment binding shifts the equilibrium of open/closed WPD loop conformations because basically all WPD loops of the screening study were in an open conformation. The authors accounted this result to a difference in crystalisation conditions. The apo structures of the multitemperature study were all solved in the presence of glycerol whereas all fragment soaked (and accompanying apo) structures were solved without glycerol.

Covalent tethering

To infer functional linkage between the identified allosteric binding sites one would like to see inhibition of the protein’s catalytic function. 20 fragments were selected and enzymatic assays performed but no inhibition could be seen. Therefore, covalent tethering was used to enhance a potential inhibiting effect. For that K197 was mutated into cysteine and a fragment with an disulfid linker was designed. Successful partial inhibition could then be shown. See Figure 6 for an image with the covalently tethered fragment at the 197 site.