The Power of Fragments: FBLD approach to investigate protein structures
Fragments are small chemical entities with the extraordinary advantage of belonging to a wide variety of functional groups able to explore various chemical spaces, penetrate small protein pockets and allow the mapping of various protein binding sites. Nowadays, fragment-based lead discovery (FBLD) i...
|PDF Full Text
No Tags, Be the first to tag this record!
|Fragments are small chemical entities with the extraordinary advantage of belonging to a wide variety of functional groups able to explore various chemical spaces, penetrate small protein pockets and allow the mapping of various protein binding sites. Nowadays, fragment-based lead discovery (FBLD) is the method of choice to screen fragment libraries and plays an increasingly important role in the drug development process. It becomes therefore particularly important to have a general-purpose fragment library ready to use. In our group a 96-fragment library was validated on eight protein targets, which was designed in a joint project between the HZB MX-group at BESSY II (AG Weiss) and the Institute of Pharmaceutical Chemistry, University of Marburg (AG Klebe) (Part I). Among the eight validated proteins, there are Thermolysin (TLN, Part II) and Trypanosoma cruzi Farnesyl Pyrophosphate Synthase (T. cruzi FPPS, Part III) which were the main focus of the present study. The 96-fragment library was therefore used as primary X-ray crystallographic screening method to search for novel chemical scaffolds. However, a new soaking protocol had to be established prior to the fragment screening. In fact, unlike the typical TLN inhibitor, fragments usually have insufficient affinity to displace on their own the autoproteolitically product Val-Lys dipeptide which occupies the TLN active site. The protein crystals were therefore incubated in an isopropanol solution able to displace the dipeptide and subsequentially screened against the fragment library resulting in seven crystallographic fragment hits. Unlike TLN which is usually used as a model protein, T. cruzi FPPS is a key enzyme involved in the mevalonate pathway and is essential for sterol production. T. cruzi and T. brucei parasites cause Chagas disease (CD) and human African trypanosomiasis (HAT) respectively and inhibition of FPPS in these species is therefore a valid target to fight these two tropical diseases. The present work focused first on optimizing the stability of protein samples in solution using various techniques such as thermal shift assay (TSA) and light scattering (LS). A crystallization and soaking protocol were subsequently established prior to performing a crystallographic fragment screening of the above-mentioned 96 fragment library. In addition, the role of the magnesium ion as cofactor of the enzyme was also elucidated, either alone or in complex with isopentenyl pyrophosphate (IPP) or bisphosphonates (BPs). Based on the discovery by Jahnke et al. of a new allosteric pocket in human FPPS, it was assumed that a similar pocket was also present in the Trypanosoma species which was therefore the main target-site of the present work. The fragment screening project resulted in three crystallographic hits, one of which binding the allosteric pocket while the other two in a new pocket whose function is still unknown. They can serve as good candidates for the design of a new series of lead compounds.
In addition to the 96-fragment library, other series of compounds of fragment-size were screened against Endothiapepsin (EP, Part IV) and different human Carbonic Anhydrase (hCA, Part V) isoforms. In particular, a series of tetrazole compounds were crystallographically screened against EP in order to validate a novel pipeline for the rapid development of novel aspartyl protease inhibitors using an anchoring approach. The hydrazide moiety was here chosen as fragment-anchor able to bind the catalytic dyad in the EP active site. Moreover, in addition to X-ray crystallography which was the dominant screening method used in the present thesis, a series of para-substituted benzenesulfonamide were screened against different hCAs isoforms using Surface Plasmon Resonance (SPR). According to Gaspari et al., the kinetics of binding of hCAII depend on the nature of the substituent that decorate the benzenesulfonamide moiety. In particular, the association rate kon becames faster by increasing the hydrophobic nature of the para-alkyl chain due to a pre-binding event. This finding was also confirmed in the present study. However, it seems that more than one features contribute to the trends in kon and koff. Furthermore, in contrast with hCAII, hCAXII showed a faster dissociation rate by increasing the length of the para-alkyl chain, probably due to the higher surface hydrophilicity. The nature of the surface around the active site of hCAs plays therefore an important role in the kinetics of binding. In conclusion, fragments are powerful molecules able to explore and discover new protein pockets, to evolve into more potent lead compounds and often they are used as starting point in several screening methods in the field of FBLD.