Fragment-based screening by X-ray crystallography, MS and isothermal titration calorimetry to identify PNMT (phenylethanolamine N-methyltransferase) inhibitors.

CNS (central nervous system) adrenaline (epinephrine) is implicated in a wide range of physiological and pathological conditions. PNMT (phenylethanolamine N-methyltransferase) catalyses the final step in the biosynthesis of adrenaline, the conversion of noradrenaline (norepinephrine) to adrenaline by methylation. To help elucidate the role of CNS adrenaline, and to develop potential drug leads, potent, selective and CNS-active inhibitors are required. The fragment screening approach has advantages over other lead discovery methods including high hit rates, more efficient hits and the ability to sample chemical diversity more easily. In the present study we applied fragment-based screening approaches to the enzyme PNMT. We used crystallography as the primary screen and identified 12 hits from a small commercial library of 384 drug-like fragments. The hits include nine chemicals with two fused rings and three single-ring chemical systems. Eight of the hits come from three chemical classes: benzimidazoles (a known class of PNMT inhibitor), purines and quinolines. Nine of the hits have measurable binding affinities (~5-700 µM) as determined by isothermal titration calorimetry and all nine have ligand efficiencies of 0.39 kcal/mol per heavy atom or better (1 kcal≈4.184 kJ). We synthesized five elaborated benzimidazole compounds and characterized their binding to PNMT, showing for the first time how this class of inhibitors interact with the noradrenaline-binding site. Finally, we performed a pilot study with PNMT for fragment-based screening by MS showing that this approach could be used as a fast and efficient first-pass screening method prior to characterization of binding mode and affinity of hits.

Molecular recognition of physiological substrate noradrenaline by the adrenaline-synthesizing enzyme PNMT and factors influencing its methyltransferase activity.

Substrate specificity is critically important for enzyme catalysis. In the adrenaline-synthesizing enzyme PNMT (phenylethanolamine N-methyltransferase), minor changes in substituents can convert substrates into inhibitors. Here we report the crystal structures of six human PNMT complexes, including the first structure of the enzyme in complex with its physiological ligand R-noradrenaline. Determining this structure required rapid soak methods because of the tendency for noradrenaline to oxidize. Comparison of the PNMT-noradrenaline complex with the previously determined PNMT-p-octopamine complex demonstrates that these two substrates form almost equivalent interactions with the enzyme and show that p-octopamine is a valid model substrate for PNMT. The crystal structures illustrate the adaptability of the PNMT substrate binding site in accepting multi-fused ring systems, such as substituted norbornene, as well as noradrenochrome, the oxidation product of noradrenaline. These results explain why only a subset of ligands recognized by PNMT are methylated by the enzyme; bulky substituents dictate the binding orientation of the ligand and can thereby place the acceptor amine too far from the donor methyl group for methylation to occur. We also show how the critical Glu(185) catalytic residue can be replaced by aspartic acid with a loss of only 10-fold in catalytic efficiency. This is because protein backbone movements place the Asp(185) carboxylate almost coincident with the carboxylate of Glu(185). Conversely, replacement of Glu(185) by glutamine reduces catalytic efficiency almost 300-fold, not only because of the loss of charge, but also because the variant residue does not adopt the same conformation as Glu(185).

Structure-Based Drug Design of Bisubstrate Inhibitors of Phenylethanolamine N-Methyltransferase Possessing Low Nanomolar Affinity at Both Substrate Binding Domains1.

The enzyme phenylethanolamine N-methyltransferase (PNMT, EC 2.1.1.28) catalyzes the final step in the biosynthesis of epinephrine and is a potential drug target, primarily for the control of hypertension. Unfortunately, many potent PNMT inhibitors also possess significant affinity for the a2-adrenoceptor, which complicates the interpretation of their pharmacology. A bisubstrate analogue approach offers the potential for development of highly selective inhibitors of PNMT. This paper documents the design, synthesis, and evaluation of such analogues, several of which were found to possess human PNMT (hPNMT) inhibitory potency <5 nM versus AdoMet. Site-directed mutagenesis studies were consistent with bisubstrate binding. Two of these compounds (19 and 29) were co-crystallized with hPNMT and the resulting structures revealed both compounds bound as predicted, simultaneously occupying both substrate binding domains. This bisubstrate inhibitor approach has resulted in one of the most potent (20) and selective (vs the a2-adrenoceptor) inhibitors of hPNMT yet reported.

Time-dependent inactivation of human phenylethanolamine N-methyltransferase by 7-isothiocyanatotetrahydroisoquinoline.

Inhibitors of phenylethanolamine N-methyltransferase [PNMT, the enzyme that catalyzes the final step in the biosynthesis of epinephrine (Epi)] may be of use in determining the role of Epi in the central nervous system. Here we describe the synthesis and characterization of 7-SCN tetrahydroisoquinoline as an affinity label for human PNMT.

Synthesis of 4,5,6,7-tetrahydrothieno[3,2-c]pyridines and comparison with their isosteric 1,2,3,4-tetrahydroisoquinolines as inhibitors of phenylethanolamine N-methyltransferase.

A series of substituted 4,5,6,7-tetrahydrothieno[3,2-c]pyridines (THTPs) was synthesized and evaluated for their human phenylethanolamine N-methyltransferase (hPNMT) inhibitory potency and affinity for the alpha(2)-adrenoceptor. The THTP nucleus was suggested as an isosteric replacement for the 1,2,3,4-tetrahydroisoquinoline (THIQ) ring system on the basis that 3-thienylmethylamine (18) was more potent as an inhibitor of hPNMT and more selective toward the alpha(2)-adrenoceptor than benzylamine (15). Although the isosterism was confirmed, with similar influence of functional groups and chirality in both systems on hPNMT inhibitory potency and selectivity, the THTP compounds proved, in general, to be less potent as inhibitors of hPNMT than their THIQ counterparts, with the drop in potency being primarily attributed to the electronic properties of the thiophene ring. A hypothesis for the reduced hPNMT inhibitory potency of these compounds has been formed on the basis of molecular modeling and docking studies using the X-ray crystal structures of hPNMT co-crystallized with THIQ-type inhibitors and S-adenosyl-L-homocysteine as a template.

Enzyme adaptation to inhibitor binding: a cryptic binding site in phenylethanolamine N-methyltransferase.

Shape complementarity is a fundamental principle of inhibitor design. Here we show that an enzyme for which the crystal structure has been determined (phenylethanolamine N-methyltransferase, PNMT) conceals a cryptic binding site. This site is revealed upon binding of inhibitors that are double the size of the physiological substrate. These large inhibitors are not predicted to bind in that they protrude through the accessible surface calculated from a PNMT/7-aminosulfonyl-1,2,3,4-tetrahydroisoquinoline (SK&F 29661) crystal structure, yet they are potent inhibitors of PNMT. We determined structures of the enzyme complexed with large inhibitors and found that the volume of the active site increases by 140 A3 upon binding. Changes in active site size and shape are brought about by unfavorable side chain conformations and rigid body helix motions. The energetic cost is modest, estimated at 2-3 kcal/mol from mutational analyses. Our findings further underline the importance of protein flexibility in structure-based inhibitor design studies.

Disulfide-linked dimers of human adrenaline synthesizing enzyme PNMT are catalytically active.

The crystal structure of human phenylethanolamine N-methyltransferase (hPNMT) reveals a disulfide-linked dimer, despite the presence of reducing agent in the crystallisation conditions. By removing the reducing agent, hPNMT crystals grow more rapidly and at lower protein concentrations. However, it was unclear whether the disulfide bonds are only present in the crystal form or whether these affect enzyme activity. The solution oligomeric state of hPNMT was investigated using biochemical techniques and activity assays. We found that in the absence of reducing agent, hPNMT forms dimers in solution. Furthermore, the solution dimer of hPNMT incorporates disulfide bonds, since this form is sensitive to reducing agent. The C48A and C139A mutants of hPNMT, which are incapable of forming the disulfide bond observed in the crystal structure, have a decreased propensity to form dimer in solution. Those dimers that do form are also sensitive to reducing agent. Further, the C48A/C139A double mutant shows only monomeric behaviour. Both dimeric and monomeric hPNMT, as well as mutants have wildtype enzyme activity. These results show that a variety of disulfides, including those observed in the crystal structure, can form in solution. In addition, disulfide-linked dimers are as active as the monomeric enzyme indicating that the crystal structure of the protein is a valid target for inhibitor design.

Application of the Goldilocks effect to the design of potent and selective inhibitors of phenylethanolamine N-methyltransferase: balancing pKa and steric effects in the optimization of 3-methyl-1,2,3,4-tetrahydroisoquinoline inhibitors by beta-fluorination.

3-Methyl-1,2,3,4-tetrahydroisoquinolines (3-methyl-THIQs) are potent inhibitors of phenylethanolamine N-methyltransferase (PNMT), but are not selective due to significant affinity for the alpha(2)-adrenoceptor. Fluorination of the methyl group lowers the pK(a) of the THIQ amine from 9.53 (CH(3)) to 7.88 (CH(2)F), 6.42 (CHF(2)), and 4.88 (CF(3)). This decrease in pK(a) results in a reduction in affinity for the alpha(2)-adrenoceptor. However, increased fluorination also results in a reduction in PNMT inhibitory potency, apparently due to steric and electrostatic factors. Biochemical evaluation of a series of 3-fluoromethyl-THIQs and 3-trifluoromethyl-THIQs showed that the former were highly potent inhibitors of PNMT, but were often nonselective due to significant affinity for the alpha(2)-adrenoceptor, while the latter were devoid of alpha(2)-adrenoceptor affinity, but also lost potency at PNMT. 3-Difluoromethyl-7-substituted-THIQs have the proper balance of both steric and pK(a) properties and thus have enhanced selectivity versus the corresponding 3-fluoromethyl-7-substituted-THIQs and enhanced PNMT inhibitory potency versus the corresponding 3-trifluoromethyl-7-substituted-THIQs. Using the "Goldilocks Effect" analogy, the 3-fluoromethyl-THIQs are too potent (too hot) at the alpha(2)-adrenoceptor and the 3-trifluoromethyl-THIQs are not potent enough (too cold) at PNMT, but the 3-difluoromethyl-THIQs are just right. They are both potent inhibitors of PNMT and highly selective due to low affinity for the alpha(2)-adrenoceptor. This seems to be the first successful use of the beta-fluorination of aliphatic amines to impart selectivity to a pharmacological agent while maintaining potency at the site of interest.

Comparison of the binding of 3-fluoromethyl-7-sulfonyl-1,2,3,4-tetrahydroisoquinolines with their isosteric sulfonamides to the active site of phenylethanolamine N-methyltransferase.

3-Fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines (14, 16, and 18-22) are highly potent and selective inhibitors of phenylethanolamine N-methyltransferase (PNMT). Molecular modeling studies with 3-fluoromethyl-7-(N-alkyl aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines, such as 16, suggested that the sulfonamide -NH- could form a hydrogen bond with the side chain of Lys57. However, SAR studies and analysis of the crystal structure of human PNMT (hPNMT) in complex with 7 indicated that the sulfonamide oxygens, and not the sulfonamide -NH-, formed favorable interactions with the enzyme. Thus, we hypothesized that replacement of the sulfonamide -NH- with a methylene group could result in compounds that would retain potency at PNMT and that would have increased lipophilicity, thus increasing the likelihood they will cross the blood brain barrier. A series of 3-fluoromethyl-7-sulfonyl-1,2,3,4-tetrahydroisoquinolines (23-30) were synthesized and evaluated for their PNMT inhibitory potency and affinity for the alpha2-adrenoceptor. A comparison of these compounds with their isosteric sulfonamides (14, 16, and 18-22) showed that the sulfones were more lipophilic but less potent than their corresponding sulfonamides. Sulfone 24 (hPNMT Ki = 1.3 microM) is the most potent compound in this series and is quite selective for PNMT versus the alpha2-adrenoceptor, but 24 is less potent than the corresponding sulfonamide, 16 (hPNMT Ki = 0.13 microM). We also report the crystal structure of hPNMT in complex with sulfonamide 15, from which a potential hydrogen bond acceptor within the hPNMT active site has been identified, the main chain carbonyl oxygen of Asn39. The interaction of this residue with the sulfonamide -NH- is likely responsible for much of the enhanced inhibitory potency of the sulfonamides versus the sulfones.

3-hydroxymethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinoline inhibitors of phenylethanolamine N-methyltransferase that display remarkable potency and selectivity.

Six 3-hydroxymethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines (16-21) were synthesized and evaluated for their phenylethanolamine N-methyltransferase (PNMT) inhibitory potency and affinity for the alpha(2)-adrenoceptor. The addition of nonpolar substituents to the sulfonamide nitrogen of 9 (3-CH(2)OH-7-SO(2)NH(2)-THIQ) led to inhibitors (16-21) that have high PNMT inhibitory potency and high selectivity, and most of these (16-21) are predicted, on the basis of their calculated log P values, to be able to penetrate the blood-brain barrier. Compounds N-trifluoroethyl sulfonamide 20 (PNMT K(i) = 23 nM) and N-trifluoropropyl sulfonamide 21 (PNMT K(i) = 28 nM) are twice as potent at inhibiting PNMT compared to 9 and display excellent selectivity (alpha(2) K(i)/PNMT K(i) > or = 15,000).

Nanomolar inhibitors of CNS epinephrine biosynthesis: (R)-(+)-3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines as potent and highly selective inhibitors of phenylethanolamine N-methyltransferase1.

A series of (R)-(+)-3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines has been synthesized and evaluated as inhibitors of PNMT and for their affinity for the alpha(2)-adrenoceptor. Compounds (R)-8 and (R)-9 are remarkably potent and selective inhibitors of PNMT and are predicted to penetrate the blood-brain barrier on the basis of their calculated log P values. Conformational analysis and docking studies were performed in order to examine why the (R)-enantiomer of these 3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines is more potent than the (S)-enantiomer and to determine the likely bound ring conformer of the (R)-enantiomer. It appears that the (R)-enantiomer participates in a water-mediated hydrogen bond in which the (S)-enantiomer cannot. The likely favored ring conformation for (R)-3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines in the PNMT active site is similar to the ring conformation of (R)-5a as determined by gas-phase ab initio calculations.

Structural, mutagenic, and kinetic analysis of the binding of substrates and inhibitors of human phenylethanolamine N-methyltransferase.

The X-ray structure of human phenylethanolamine N-methyltransferase (hPNMT) complexed with its product, S-adenosyl-L-homocysteine (4), and the most potent inhibitor reported to date, SK&F 64139 (7), was used to identify the residues involved in inhibitor binding. Four of these residues, Val53, Lys57, Glu219 and Asp267, were replaced, in turn, with alanine. All variants had increased Km values for phenylethanolamine (10), but only D267A showed a noteworthy (20-fold) decrease in its kcat value. Both WT hPNMT and D267A had similar kcat values for a rigid analogue, anti-9-amino-6-(trifluoromethyl)benzonorbornene (12), suggesting that Asp267 plays an important role in positioning the substrate but does not participate directly in catalysis. The Ki values for the binding of inhibitors such as 7 to the E219A and D267A variants increased by 2-3 orders of magnitude. Further, the inhibitors were shown to bind up to 50-fold more tightly in the presence of S-adenosyl-L-methionine (3), suggesting that the binding of the latter brings about a conformational change in the enzyme.

Inhibitors of phenylethanolamine N-methyltransferase that are predicted to penetrate the blood-brain barrier: design, synthesis, and evaluation of 3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines that possess low affinity toward the alpha2-adrenoceptor.

(+/-)-7-Aminosulfonyl-3-fluoromethyl-1,2,3,4-tetrahydroisoquinoline (7) is one of the most potent and selective inhibitors of phenylethanolamine N-methyltransferase (PNMT) reported to date, but a blood-brain barrier (BBB) model indicates that it cannot penetrate the BBB. To increase the lipophilicity of 7 by addition of a nonpolar substituent to the sulfonamide nitrogen, a small library of (+/-)-3-fluoromethyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquinolines was synthesized and evaluated as inhibitors of PNMT and for affinity at the alpha2-adrenoceptor. In addition, this library probed the PNMT active site surrounding the sulfonamide nitrogen of 7. Bulky substituents on the sulfonamide nitrogen are disfavored at the PNMT active site more so than at the alpha2-adrenoceptor (thus reducing selectivity). On the other hand, alkyl chains on the sulfonamide nitrogen that contain an electron dense atom, such as a fluorine, are favored in the PNMT active site and possess little alpha2-adrenoceptor affinity, thereby conferring good selectivity (>500). Several members of the library (8, 14, 17, and 18) have excellent PNMT inhibitory potency and selectivity and are predicted, on the basis of their ClogP value (>0.5), to penetrate the BBB to a significant extent. Compounds 17 and 18 are the most potent and selective PNMT inhibitors reported to date.

Molecular recognition of sub-micromolar inhibitors by the epinephrine-synthesizing enzyme phenylethanolamine N-methyltransferase.

The crystal structures of human phenylethanolamine N-methyltransferase in complex with S-adenosyl-l-homocysteine (7, AdoHcy) and either 7-iodo-1,2,3,4-tetrahydroisoquinoline (2) or 8,9-dichloro-2,3,4,5-tetrahydro-1H-2-benzazepine (3, LY134046) were determined and compared with the structure of the enzyme complex with 7 and 7-aminosulfonyl-1,2,3,4-tetrahydroisoquinoline (1, SK&F 29661). The enzyme is able to accommodate a variety of chemically disparate functional groups on the aromatic ring of the inhibitors through adaptation of the binding pocket for this substituent and by subtle adjustments of the orientation of the inhibitors within the relatively planar binding site. In addition, the interactions formed by the amine nitrogen of all three inhibitors reinforce the hypothesis that this functional group mimics the beta-hydroxyl of norepinephrine rather than the amine. These studies provide further clues for the development of improved inhibitors for use as pharmacological probes.

The NH---FC dipole orientation effect for pendant exocyclic CH(2)F.

[structure: see text] DFT and MMFF force field calculations for 2 (R = H) predict that two conformers dominate in water (>/=95%) and both sustain a geometry in which the C-F and H-N dipoles align oppositely in a near-planar arrangement. The (1)H NMR spectra (D(2)O and DMSO-d(6)) and X-ray structure for 2 (R = SO(2)NHEt) confirm the predictions in all essentials. A novel single-conformer system is proposed.

Exploring the active site of phenylethanolamine N-methyltransferase with 1,2,3,4-tetrahydrobenz[h]isoquinoline inhibitors.

1,2,3,4-Tetrahydrobenz[h]isoquinoline (THBQ, 11) is a potent, inhibitor of phenylethanolamine N-methyltransferase (PNMT). Docking studies indicated that the enhanced PNMT inhibitory potency of 11 (hPNMT K(i)=0.49microM) versus 1,2,3,4-tetrahydroisoquinoline (5, hPNMT K(i)=5.8microM) was likely due to hydrophobic interactions with Val53, Met258, Val272, and Val269 in the PNMT active site. These studies also suggested that the addition of substituents to the 7-position of 11 that are capable of forming hydrogen bonds to the enzyme could lead to compounds (14-18) having enhanced PNMT inhibitory potency. However, these compounds are in fact less potent at PNMT than 11. Furthermore, 7-bromo-THBQ (19, hPNMT K(i)=0.22mM), which has a lipophilic 7-substituent that cannot hydrogen bond to the enzyme, is twice as potent at PNMT than 11. This once again illustrates the limitations of docking studies for lead optimization.

Inhibitors of phenylethanolamine N-methyltransferase devoid of alpha2-adrenoceptor affinity.

A series of 3-trifluoromethyl-1,2,3,4-tetrahydroisoquinolines was synthesized and evaluated for their phenylethanolamine N-methyltransferase (PNMT) inhibitory potency and affinity for the alpha(2)-adrenoceptor. Although their PNMT inhibitory potency decreased compared with corresponding 3-methyl-, 3-hydroxymethyl- or 3-unsubstituted-THIQs, some of them showed good selectivity due to their extremely low alpha(2)-adrenoceptor affinity.

Mode of binding of methyl acceptor substrates to the adrenaline-synthesizing enzyme phenylethanolamine N-methyltransferase: implications for catalysis.

Here we report three crystal structure complexes of human phenylethanolamine N-methyltransferase (PNMT), one bound with a substrate that incorporates a flexible ethanolamine side chain (p-octopamine), a second bound with a semirigid analogue substrate [cis-(1R,2S)-2-amino-1-tetralol, cis-(1R,2S)-AT], and a third with trans-(1S,2S)-2-amino-1-tetralol [trans-(1S,2S)-AT] that acts as an inhibitor of PNMT rather than a substrate. A water-mediated interaction between the critical beta-hydroxyl of the flexible ethanolamine group of p-octopamine and an acidic residue, Asp267, is likely to play a key role in positioning the side chain correctly for methylation to occur at the amine. A second interaction with Glu219 may play a lesser role. Catalysis likely occurs via deprotonation of the amine through the action of Glu185; mutation of this residue significantly reduced the kcat without affecting the Km. The mode of binding of cis-(1R,2S)-AT supports the notion that this substrate is a conformationally restrained analogue of flexible PNMT substrates, in that it forms interactions with the enzyme similar to those observed for p-octopamine. By contrast, trans-(1S,2S)-AT, an inhibitor rather than a substrate, binds in an orientation that is flipped by 180 degrees compared with cis-(1R,2S)-AT. A consequence of this flipped binding mode is that the interactions between the hydroxyl and Asp267 and Glu219 are lost. However, the amines of inhibitor trans-(1S,2S)-AT and substrate cis-(1R,2S)-AT are both within methyl transfer distance of the cofactor. These results suggest that PNMT catalyzes transfer of methyl to ligand amines only when "anchor" interactions, such as those identified for the beta-hydroxyls of p-octopamine and cis-AT, are present.

Exploring the active site of phenylethanolamine N-methyltransferase: 3-alkyl-7-substituted-1,2,3,4-tetrahydroisoquinoline inhibitors.

A series of 3-alkyl-7-substituted-1,2,3,4-tetrahydroisoquinolines was synthesized and these compounds were evaluated for their PNMT inhibitory potency and affinity for the alpha2-adrenoceptor. 7-Nitro-, 7-bromo-, 7-aminosulfonyl-, or 7-N-2,2,2-trifluoroethylaminosulfonyl-THIQs that possess a 3-alkyl substituent that is longer than a methyl group showed decreased PNMT inhibitory potency, except for 3-propyl-7-aminosulfonyl-THIQ, which displayed excellent PNMT inhibitory potency. The rank order for selectivity (PNMT vs the alpha2-adrenoceptor) is 3-alkyl-7-aminosulfonyl-THIQs congruent with 3-alkyl-7-N-2,2,2-trifluoroethylaminosulfonyl-THIQs>3-alkyl-7-nitro-THIQs>3-alkyl-7-bromo-THIQs.