2001 Apr 10;40(14):4242C52. potency, as well as excellent binding efficiency against hTS. Relative binding orientations for both leads were modeled using AutoDock, and the most likely bound conformations validated using experimentally-derived STD-NMR binding epitope data. These ligands represent novel starting points for fragment-based drug design of non-canonical TS inhibitors, and their binding epitopes highlight important and previously unexploited interactions with conserved residues in the cofactor binding site. activity data show two of these compounds to possess low micromolar affinity and mid-micromolar potency against the enzyme. We applied experimentally derived STD-NMR binding epitope maps to computationally-derived binding models in order to approximate the binding conformations of AG337 and our two micromolar leads. This approach has allowed the rapid discovery of two non-canonical TS antifolates without utilizing any directed synthetic chemistry. With high-resolution complex structures of the target available, new lead compounds can be identified using ligand-based NMR as an alternative to iterative structure determination, as demonstrated in this work. Both micromolar leads obtained through these methods will serve as platforms for further chemical development of novel TS inhibitors. RESULTS & DISCUSSION Fragmentation Studies of AG331 AG331 (1) has low nanomolar activity against hTS (Figure 1). Crystallographic data of a close chemical analog position members of this ligand series precisely in the folate binding pocket (33). The benz[cd]indole moiety of AG331 mimics the pterin folate ring system in a subpocket of TS, sitting in close proximity to the bound substrate dUMP. The benzyl sulfonyl morpholine group of AG331 follows a groove created by hydrophobic sidechains and mimics the p-amino-benzoate moiety of folate. A lactam variant of AG331 (2) has a Kd = 300 nM against hTS (33). By splitting 2 in half, we obtain two small molecules, 3 and 4, which largely conform to the criteria we (and others) use to characterize fragments (Figure 1) (42). Since the two halves Valsartan of 2 bind distinct subpockets of the TS folate site when linked, molecules 3 and 4 seemed suitable controls for developing fragment screening methods against hTS. Open in a separate window Figure 1 Structures of AG331 (1), AG337 (5) and related structures, with proton assignments characterized by 1H NMR. Typical sample preparations for STD-NMR utilize 2 to 100 M protein, with ligand:protein ratios ranging from 10:1 to 100:1 (43-46). After testing a range of conditions, we found 5 to 12.5 M holoenzyme (10 to 25 M monomeric hTS) and 0.2 to 1 1.0 mM ligand concentrations to be suitable for ligand-observe NMR experiments (see Experimental Methods). Using 25 M monomeric hTS and a ligand:protein ratio of 40:1, we first characterized binding of 3 and of 4 to hTS individually by STD-NMR (Figure S1, Supporting Information). Both compounds gave rise to saturation peak difference (STD) intensities, indicative of weak to moderate (ie. micromolar) binding interactions with a macromolecule (47). The raw STD peak intensities of compound 3 were greater than that observed for hTS binding to Valsartan compound 4, with the Valsartan latter generating STD amplification factors (SAFs) ranging from 1.6 to 2.6 (Figure 2). We therefore chose a criterion of SAF 2.5 for the strongest proton signal of a test ligand, as this would be necessary to identify fragment 4 as a hit from our library (see Experimental Methods for SAF calculation). Open in a separate window Figure 2 STD amplification factor (SAF) values for resolvable protons in STD-NMR spectra for compounds 3 and 4 (Figure 1) in aqueous buffer with hTS. (above): Addition of 3 Rabbit polyclonal to INPP1 does not significantly alter SAF values for 4, and addition of 4 does not significantly impact 3. (below): Addition of the native substrate dUMP does not impact binding of 3 or 4 4, while addition of methotrexate effectively competes with both fragments in.