Discovery of a potent and orally bioavailable benzolactam-derived inhibitor of Polo-like kinase 1 (MLN0905).

This article describes the discovery of a series of potent inhibitors of Polo-like kinase 1 (PLK1). Optimization of this benzolactam-derived chemical series produced an orally bioavailable inhibitor of PLK1 (12c, MLN0905). In vivo pharmacokinetic-pharmacodynamic experiments demonstrated prolonged mitotic arrest after oral administration of 12c to tumor bearing nude mice. A subsequent efficacy study in nude mice achieved tumor growth inhibition or regression in a human colon tumor (HT29) xenograft model.

Substrate-assisted inhibition of ubiquitin-like protein-activating enzymes: the NEDD8 E1 inhibitor MLN4924 forms a NEDD8-AMP mimetic in situ.

The NEDD8-activating enzyme (NAE) initiates a protein homeostatic pathway essential for cancer cell growth and survival. MLN4924 is a selective inhibitor of NAE currently in clinical trials for the treatment of cancer. Here, we show that MLN4924 is a mechanism-based inhibitor of NAE and creates a covalent NEDD8-MLN4924 adduct catalyzed by the enzyme. The NEDD8-MLN4924 adduct resembles NEDD8 adenylate, the first intermediate in the NAE reaction cycle, but cannot be further utilized in subsequent intraenzyme reactions. The stability of the NEDD8-MLN4924 adduct within the NAE active site blocks enzyme activity, thereby accounting for the potent inhibition of the NEDD8 pathway by MLN4924. Importantly, we have determined that compounds resembling MLN4924 demonstrate the ability to form analogous adducts with other ubiquitin-like proteins (UBLs) catalyzed by their cognate-activating enzymes. These findings reveal insights into the mechanism of E1s and suggest a general strategy for selective inhibition of UBL conjugation pathways.

Mapping the binding site of melanocortin 4 receptor agonists: a hydrophobic pocket formed by I3.28(125), I3.32(129), and I7.42(291) is critical for receptor activation.

The melanocortin 4 receptor is involved in the control of the feeding behavior and energy homeostasis. It is regulated by internal agonist (alpha-MSH) and antagonists (Agouti). Peptide agonists bind in a beta-turn conformation that organizes the characteristic message sequence (His-L/DPhe-Arg-Trp) in an optimal arrangement for binding and activation of the receptor. Our goal is to determine the most likely binding modes of peptide and small molecule agonists to use this information to guide our structure-based drug design efforts. Previous studies have identified some residues that are likely to be involved in peptide agonist binding, giving an initial estimate of the main contacts between peptides and receptor. However, a more detailed description of the orientation of the peptide in a beta-turn conformation in the binding site, as well as of the small molecule agonists, and it is commonalities with the peptide agonist binding modes is necessary to serve as the basis for structure-based drug design. In the current study we combine site-directed mutagenesis with molecular modeling studies to determine the most likely binding mode of peptide and small molecule agonists, and we found that Y6.58(268), Y7.38(287), I3.28(125), I3.32(129), and I7.42(291) also line the binding site and are likely to have direct contacts with the MC4R agonists. Of particular interest are residues I3.28(125), I3.32(129), and I7.42(291), which form a hydrophobic pocket where I7.42(291), on top of the NPXXY motif, is likely to act as a new rotamer switch implicated in the activation of the receptor.

A second site rescue mutation partially restores functional expression to the serotonin transporter mutant V382P.

Transmembrane span 7 (TM7) of the serotonin transporter (SERT) was previously subjected to random mutagenesis, and the mutation V382P was found to abolish transport activity. Val-382 lies next to a threonine residue in the native sequence, creating a TP motif in this mutant. On the basis of molecular modeling studies, which have shown that the presence of a TP motif produces a very large kink in an alpha-helix, it was hypothesized that this motif could be the source of V382P's deleterious effects. We tested this hypothesis by producing second site mutations in the V382P construct that removed the TP motif: T381A-V382P and T381V-V382P. These mutants were tested for the recovery of serotonin transport and binding activities and for expression at the cell surface. The TM7 alpha-helix was modeled computationally, using Biased Monte Carlo simulations to quantify the conformational preferences of the wild type and mutant helices. The double mutation T381A-V382P, which was predicted by modeling to produce a smaller perturbing bend in TM7, was indeed found to allow partial rescue of transport activity. The double mutation T381V-V382P, on the other hand, did not rescue transport activity. Computational analysis of this mutant predicted a markedly different conformational preference from either the V382P or the T381A-V382P mutants. These studies show that changes in the structure of TM7 exert a strong influence on SERT's ability to achieve a mature, properly folded, cell surface conformation.

Molecular basis of partial agonism: orientation of indoleamine ligands in the binding pocket of the human serotonin 5-HT2A receptor determines relative efficacy.

Based on experiment and computational simulation, we present a structural explanation for the differing efficacies of indole agonists at the human serotonin 5-HT2A receptor (5HT2AR). We find that serotonin [5-hydroxytryptamine (5-HT)] forms hydrogen-bonds with Ser3.36 in helix 3 and Ser5.46 in helix 5. Disruption of these hydrogen bonds by methyl-substitution of the cationic primary amine or of the backbone N1-amine, respectively, leads to a reduction in agonist efficacy. Computational simulation predicts that mutation of Ser3.36 to Ala should allow a similar interaction with helix 3 both for agonists that have unmodified cationic amine side chains and for those with substituted amines. Experimentally, this mutation was found to largely eliminate the differences in efficacy caused by cationic amine substitution for a series of indole congeners. Similarly, substitution of the N1-amine, which interacts with Ser5.46, reduced efficacy more markedly at the wild-type (WT) than at the Ser5.46Ala mutant receptor. Computational modeling of binding pocket interactions of ligands with WT and mutant receptor constructs demonstrate how the Ser3.36 and Ser5.46 interactions serve to modify the agonist's favored position in the binding pocket. A striking correlation was found between differences in the position assumed by the indole ring and differences in agonist activity. These data support the hypothesis that the position of the agonist interacting with the receptor is influenced by specific interactions in helices 3 and 5 and determines the degree of receptor activation by agonist through a mechanism that is likely to be shared by other G-protein coupled receptors in this class.

The local environment at the cytoplasmic end of TM6 of the mu opioid receptor differs from those of rhodopsin and monoamine receptors: introduction of an ionic lock between the cytoplasmic ends of helices 3 and 6 by a L6.30(275)E mutation inactivates the mu opioid receptor and reduces the constitutive activity of its T6.34(279)K mutant.

Activation of rhodopsin and monoamine G protein-coupled receptors (GPCRs) has been proposed to involve in part the disruption of a conserved E6.30-R3.50 ionic interaction between transmembrane segments (TMs) 3 and 6. However, this interaction does not occur in the opioid receptors, which have L275 at 6.30. On the basis of our findings that mutations of T6.34(279) to K and D produced, respectively, a constitutively active and an inactive form of the mu opioid receptor, we previously suggested that the functional role of the 6.30(275) residue could be assumed by T6.34(279), but the interplay between residues at positions 6.30 and 6.34 remained unresolved. In this study, we examined the effects of introducing an E in position 6.30(275) of the wild type (WT) and of the T6.34(279) mutants of the mu opioid receptor to compare the participation of the 6.30 locus in molecular events during activation in this receptor with its role in other GPCRs. The L6.30(275)E and the L6.30(275)E/T6.34(279)D mutants displayed no constitutive activity and could not be activated by the agonist DAMGO or morphine. The L6.30(275)E/T6.34(279)K mutant had some constitutive activity, but much less than the T6.34(279)K mutant, and could be activated by both agonists. The rank order of affinity for the agonist DAMGO is as follows: T6.34(279)K > WT congruent with L6.30(275)E/T6.34(279)K > L6.30(275)E congruent with T6.34(279)D > L6.30(275)E/T6.34(279)D; however, all constructs have a similar affinity for the antagonist [(3)H]diprenorphine. These data are interpreted in the context of interactions with the conserved R3.50(165) in TM3. When L6.30(275) is mutated to E, the favorable E6.30(275)-R3.50(165) interaction stabilizes an inactive state, as in rhodopsin, and hence reduces the activities of T6.34(279) mutants. Thus, the mu opioid receptor is shown to be different from rhodopsin and monoamine GPCRs, of which the WTs with native E6.30 can be activated, and the 6.34D or 6.34K mutants display enhanced constitutive activities. Our molecular modeling results suggest that some specific differences in local geometry at the cytoplasmic ends of TM5 and TM6 may account in part for the observed differences in the molecular mechanisms of receptor activation.

Conserved helix 7 tyrosine acts as a multistate conformational switch in the 5HT2C receptor. Identification of a novel "locked-on" phenotype and double revertant mutations.

Studies in many rhodopsin-like G-protein-coupled receptors are providing a general scheme of the structural processes underlying receptor activation. Microdomains in several receptors have been identified that appear to function as activation switches. However, evidence is emerging that these receptor proteins exist in multiple conformational states. To study the molecular control of this switching process, we investigated the function of a microdomain involving the conserved helix 7 tyrosine in the serotonin 5HT2C receptor. This tyrosine of the NPXXY motif was substituted for all naturally occurring amino acids. Three distinct constitutively active receptor phenotypes were found: moderate, high, and "locked-on" constitutive activity. In contrast to the activity of the other receptor mutants, the high basal signaling of the locked-on Y7.53N mutant was neither increased by agonists nor decreased by inverse agonists. The Y7.53F mutant was uncoupled. Computational modeling based on the rhodopsin crystal structure suggested that Y7.53 interacts with the conserved aromatic ring at position 7.60 in the recently identified helix 8 domain. This provided a basis for seeking revertant mutations to correct the defective function of the Y7.53F receptor. When the Y7.53F receptor was mutated at position 7.60, the wild-type phenotype was restored. These results suggest that Y7.53 and Y7.60 contribute to a common functional microdomain connecting helices 7 and 8 that influences the switching of the 5HT2C receptor among multiple active and inactive conformations.