TNF-alpha, recombinant murine protein Indeed the alkyl group
Indeed, the -alkyl group is not seen in the original crystallographic electron density omit map prior to positioning either inhibitor in DHODH, nor can it be found in the final maps. Rather, the electron density maps are consistent with hydrolysis of the amide to the acid. Once the scaffold was clearly defined, compound was fitted within the electron density, leaving no space for the -alkyl group. The final electron density map is fully consistent with the structure of (shown in a), including the conformations around the ethyl-amide and the N-linked methylpyridine. In contrast to the initially docked precursor amide , where the pyridine nitrogen was predicted to interact with Arg-136 (b), the pyridine nitrogen of is hydrogen-bonded to the side-chain hydroxyl of Thr-63. Both carboxylate oxygen atoms of are hydrogen-bonded to the side-chain of Arg-136; one oxygen atom forms a hydrogen-bond to Gln-47 and the second to a crystallographic water molecule. The carbonyl oxygen of the ethyl-amide is hydrogen-bonded to Tyr-356 hydroxyl. The corresponding NH forms an internal hydrogen-bond to the carboxylate oxygen (2.5Å). The crystallographic water molecule near the side-chain of Arg-136 in is in the same position (W1 in ) as that observed in and . In a comparison with , the acidic group of adopts a different orientation relative to the Arg side-chain () and the second crystallographic water of (W2 in ) is displaced by one of the oxygen atoms of the acidic group in . This altered orientation appears to be the result of steric constraints that prevent the amidoethyl group from burying more deeply into the protein, thus pushing the carboxylate away from the position it adopts in the benzoic acids. The end result is that the acidic group is not as well positioned for interactions with the side-chain of Arg-136 as is the corresponding group in brequinar or other related compounds. Similarly, the -alkyl group of the presumed amide derivative is not seen in the electron density map, which is consistent with the structure of the TNF-alpha, recombinant murine protein . The resulting structure of is the same as that of (including the lengths of the hydrogen-bonds), except for the amidocyclopropyl group in place of amidoethyl. The cyclopropyl ring lies in a hydrophobic pocket and points toward Val-134. It appears that the suboptimal orientation of the carboxylate is partly compensated by hydrogen-bonding interactions of the pyridine with Thr-63 and the amide oxygen with Tyr-356. The near perfect fit of the cyclopropyl ring in a hydrophobic pocket may explain the improved activity of compound relative to compound . shows the superposition of compounds 1–4 with a recently reported amino-benzoic acid inhibitor in complex with DHODH (PDB: 2PRL, with a reported IC of 81nM). The acidic group is seen to span a range of orientations, all of which maintain hydrogen-bonds to the guanidinyl group of Arg-136. The specific orientation appears to be related to the size of the group that buries in the pocket. The acidic group of compounds and is most deeply buried, as no steric hindrance from a substitution on the benzoic acid is present. In compounds and , the acidic group is pushed away from the buried position by the relatively large substitution in the position. 2PRL has a substitution of intermediate size, which results in an orientation intermediate between those reported for the current sanofi-aventis compounds. Although all of the amino-benzoic acid scaffolds occupy the same binding pocket, subtle changes (such as ethyl to cyclopropyl) have dramatic effects on inhibitory activities. The potency of the cyclopropyl analogue may be attributed to an accessible bisected conformation or, simply result from a snug hydrophobic packing of the cyclopropyl group. In summary, the repertoire of carboxylic acid inhibitors for DHODH was expanded by the discovery of and –-substituted benzoic acids by means of structure-based virtual screening and X-ray crystallographic structures of the bound ligand complexes. The structure–activity information gained from these inhibitors is expected to aid the discovery of novel DHODH inhibitors that may be effective therapeutics for cancer and immunological disorders.