Synthesis and evaluation of dipeptide amides containing N omega-nitroarginine and D-2,4-diaminobutyric acids as inhibitors of neuronal nitric oxide synthase.

Selective inhibition of the isoforms of nitric oxide synthase (NOS) could be beneficial in the treatment of certain disease states arising from the overproduction of nitric oxide by NOS. Recently, we reported dipeptide amides containing a basic amine side chain as potent and selective inhibitors of neuronal NOS (Huang H. et al. (1999) J. Med. Chem., 42, 3147). The most potent nNOS inhibitor among these compounds is L-ArgNO2-L-Dbu-NH2 (1) (Ki = 130 nM), which also exhibits the highest selectivity over eNOS (> 1500-fold). The D,D-dipeptide, D-Lys-D-ArgNO2-NH2 (3) also shows high potency and selectivity. Here the dipeptide amides containing ArgNO2 and D-Dbu (9-12) were synthesized and evaluated. They are all modest inhibitors of nNOS, but poor inhibitors of eNOS and iNOS. D-Dbu-D-ArgNO2-NH2 (12) exhibits decreased inhibitory potency as compared with 3. A hypothesis regarding the binding at the active site of nNOS is proposed to explain the potency differences between the L- and D-form dipeptide amides.

Short, highly efficient syntheses of protected 3-azido- and 4-azidoproline and their precursors.

[structure: see text] An improved synthesis of protected cis- and trans-3-azido-L-proline and cis- and trans-4-azido-L- and -D-proline is reported. These compounds have been synthesized from the corresponding hydroxyproline precursors using diphenylphosphoryl azide under Mitsunobu conditions. Short, highly efficient syntheses of these precursors are described, based on a new lactone-opening reaction and p-nitrobenzoate hydrolysis under very mild conditions.

Mild and selective sodium azide mediated cleavage of p-nitrobenzoic esters.

[reaction: see text] A mild and selective cleavage of p-nitrobenzoic esters by sodium azide in methanol is reported. This new methodology is mild enough for use with acid- or base-sensitive compounds. No elimination byproducts are formed. Fmoc- and trifluoroacetyl-amino protecting groups, benzyl esters, and ethyl esters remain unaffected. Less reactive compounds are discussed in terms of steric factors, and yields are increased by altering the azide solvation.

Reduced amide bond peptidomimetics. (4S)-N-(4-amino-5-[aminoakyl]aminopentyl)-N'-nitroguanidines, potent and highly selective inhibitors of neuronal nitric oxide synthase.

Selective inhibition of the isoforms of nitric oxide synthase (NOS) could be therapeutically useful in the treatment of certain disease states arising from the overproduction of nitric oxide. Recently, we reported nitroarginine-containing dipeptide amides (Huang, H; Martasek, P.; Roman, L. J.; Masters, B. S. S.; Silverman, R. B. J. Med. Chem. 1999, 42, 3147.) and some peptidomimetic analogues (Huang, H; Martasek, P.; Roman, L. J.; Silverman, R.B. J. Med Chem. 2000, 43, 2938.) as potent and selective inhibitors of neuronal NOS (nNOS). Here, reduced amide bond pseudodipeptide analogues are synthesized and evaluated for their activity. The deletion of the carbonyl group from the amide bond either preserves or improves the potency for nNOS. Significantly, the selectivities for nNOS over eNOS (endothelial NOS), and iNOS (inducible NOS) are greatly increased in these series. The most potent nNOS inhibitor among these compounds is (4S)-N-(4-amino-5-[aminoethyl]aminopentyl)-N'-nitroguanidine (7) (K(i) = 120 nM), which also shows the highest selectivity over eNOS (greater than 2500-fold) and 320-fold selectivity over iNOS. The reduced amide bond is an excellent surrogate of the amide bond, and it will facilitate the design of new potent and selective inhibitors of nNOS.

Inactivation of C30A trimethylamine dehydrogenase by N-cyclopropyl-alpha-methylbenzylamine, 1-phenylcyclopropylamine, and phenylhydrazine.

Trimethylamine dehydrogenase (TMADH) from the bacterium Methylophilus methylotrophus (sp. W(3)A(1)) and its C30A mutant were inactivated with three known inactivators of monoamine oxidase, namely, phenylhydrazine, N-cyclopropyl-alpha-methylbenzylamine, and 1-phenylcyclopropylamine. All three compounds irreversibly inactivated both the wild-type and C30A mutant enzymes, although phenylhydrazine was 10 times more potent than N-cyclopropyl-alpha-methylbenzylamine, which was much more potent than 1-phenylcyclopropylamine. The change in the UV--visible absorption spectra upon modification indicated that the flavin was modified. In the case of the C30A mutant, the absence of a covalent attachment of the flavin to the polypeptide has permitted LC-electrospray mass spectrometry of the reaction product to be undertaken, demonstrating new mass peaks corresponding to various chemically modified forms of the flavin cofactor. In the case of N-cyclopropyl-alpha-methylbenzylamine, masses corresponding to hydroxy-FMN and hydroxyriboflavin were detected. 1-Phenylcyclopropylamine inactivation of the C30A mutant produced three modified flavins, as evidenced by the electrospray mass spectrum: hydroxy-FMN, FMN plus C(6)H(5)COCH(2)CH(2), and hydroxy-FMN plus C(6)H(5)COCH(2)CH(2). Phenylhydrazine inactivation of the C30A mutant gave at least seven different modified flavins: hydroxyriboflavin, hydroxy-FMN, two apparently isomeric compounds corresponding to hydroxy-FMN plus one phenyl group, two apparently isomeric compounds corresponding to FMN plus one phenyl group, and FMN plus two phenyl groups. Covalent flavin adduct formation appears to be the only modification because dialysis of the inactive enzyme followed by reconstitution with FMN restores the enzyme activity to that of a noninactivated control.

Inactivation of monoamine oxidase B by 1-phenylcyclopropylamine: mass spectral evidence for the flavin adduct.

Incubation of 1-phenylcyclopropylamine with bovine liver MAO (MAO B), followed by complete enzymatic digestion to single amino acid residues and subsequent analysis by on-line liquid chromatography-electrospray ionization mass spectrometry, was used to investigate the resulting flavin adduct structure.

Spectrometric evidence for the flavin-1-phenylcyclopropylamine inactivator adduct with monoamine oxidase N.

1-Phenylcyclopropylamine (1-PCPA) is shown to be an inactivator of the fungal flavoenzyme monoamine oxidase (MAO) N. Inactivation results in an increase in absorbance at 410 nm and is accompanied by the concomitant loss of the flavin absorption band at 458 nm. The spectral properties of the covalent adduct formed between the flavin cofactor of MAO N and 1-PCPA are similar to those reported for the irreversible inactivation product formed with 1-PCPA and mammalian mitochondrial monoamine oxidase B [Silverman, R. B., and Zieske, P. A. (1985) Biochemistry 24, 2128-2138]. There is a hypsochromic shift of the 410 nm band upon lowering the pH to 2, indicating that an N(5)-flavin adduct formed upon inactivation. Use of the fungal enzyme, MAO N, which lacks the covalent attachment to the flavin adenine dinucleotide (FAD) cofactor present in the mammalian forms MAO A and MAO B, has allowed for the isolation and further structural identification of the flavin-inactivator adduct. The incorporation of two (13)C labels into the inactivator, [2,3-(13)C(2)]-1-PCPA, followed by analysis using on-line liquid chromatography/electrospray ionization mass spectrometry and nuclear magnetic resonance spectroscopy, provided a means to explore the structure of the flavin-inactivator adduct of MAO N. The spectral evidence supports covalent attachment of the 1-PCPA inactivator to the cofactor as N(5)-3-oxo-3-phenylpropyl-FAD.

Rapid, high-yield, solid-phase synthesis of the antitumor antibiotic sansalvamide A using a side-chain-tethered phenylalanine building block.

A 10-step solid-phase synthesis of the cytotoxic depsipeptide sansalvamide A (1) has been accomplished in an overall yield of 67% with >95% purity employing polymer-bound phenylalanine building block 2. Both the N- and C-termini of 2 are extended followed by on-resin head-to-tail macrocyclization of the linear peptide in a high yield. This should be a general stategy for the synthesis of diverse libraries of cyclic peptides and depsipeptides that contain exclusively phenylalanine and other hydrophobic side chains.

Deuterium isotope effects and product studies for the oxidation of N(omega)-allyl-L-arginine and N(omega)-allyl-N(omega)-hydroxy-L-arginine by neuronal nitric oxide synthase.

The nitric oxide synthases (NOS), which require heme, tetrahydrobiopterin, FMN, FAD, and NADPH, catalyze the O2-dependent conversion of L-arginine to L-citrulline and nitric oxide. N(omega)-Allyl-L-arginine, a mechanism-based inactivator of neuronal NOS, also is a substrate, producing L-arginine, acrolein, and H2O (Zhang, H. Q.; Dixon, R. P., Marletta, M. A.; Nikolic, D.; Van Breemen, R.; Silverman, R. B. J. Am. Chem. Soc. 1997, 119, 10888). Two possible mechanisms for this turnover are proposed, one initiated by allyl C-H bond cleavage and the other by guanidino N H cleavage, and these mechanisms are investigated with the use of N(omega)-allyl-L-arginine (1), N(omega)-[1,1-(2)H2]allyl-L-arginine (7), N(omega)-allyl-N(omega)-hydroxy-L-arginine (2) and N(omega)-[1,1-(2)H2]allyl-N(omega)-hydroxy-L-arginine (8) as substrates. Significant isotope effects on the two kinetic parameters, kcat and kcat/Km, are observed in case of 1 and 7 during turnover, but not with 2 and 8. No kinetic isotope effects are observed for either compound in their role as inactivators. These results support a mechanism involving initial C-H bond cleavage of N(omega)-allyl-L-arginine followed by hydroxylation and breakdown to products.

Monoamine oxidase-catalyzed oxidation of endo,endo-2-amino-6-[(Z)-2'-phenyl]ethenylbicyclo[2.2.1]heptane, a potential probe for a radical cation intermediate.

An 11-step synthesis of endo,endo-2-amino-6-[(E)-2'-phenyl]ethenylbicyclo[2.2.1]heptane (6) and the corresponding (Z)-isomer (7) was carried out in an attempt to make a compound that could trap the purported amine radical cation intermediate during monoamine oxidase (MAO)-catalyzed oxidation of amines. The E-isomer was not a substrate for MAO, and the Z-isomer was a very poor substrate. No trapping product was observed. Possible explanations for the inability of these compounds to trap a potential radical cation intermediate are discussed.

Synthesis and evaluation of peptidomimetics as selective inhibitors and active site probes of nitric oxide synthases.

Nitric oxide synthase (NOS) catalyzes the conversion of L-arginine to L-citrulline and nitric oxide (NO). Selective inhibition of the isoforms of NOS could have great therapeutic potential in the treatment of certain disease states arising from pathologically elevated synthesis of NO. Recently, we reported dipeptide amides containing a basic amine side chain as potent and selective inhibitors of neuronal NOS (Huang, H.; Martasek, P.; Roman, L. J.; Masters, B. S. S.; Silverman, R. B. J. Med. Chem. 1999, 42, 3147). The most potent nNOS inhibitor among these compounds is L-ArgNO2-L-Dbu-NH2 (1) (Ki = 130 nM), which also exhibits the highest selectivity over eNOS (>1,500-fold) with excellent selectivity over iNOS (190-fold). Here we describe the design and synthesis of a series of peptidomimetic analogues of this dipeptide as potential selective inhibitors of nNOS. The biochemical evaluation of these compounds also revealed the binding requirements of the dipeptide inhibitors with NOS. Incorporation of protecting groups at the N-terminus of the dipeptide amide 1 (compounds 4 and 5) resulted in dramatic decreases in the inhibitory potency of nNOS. Masking the NH group of the peptide bond (peptoids 6-8 and N-methylated compounds 9-11) also gave much poorer nNOS inhibitors than 1. Both of the results demonstrate the importance of the alpha-amine of the dipeptide and the NH moiety of the peptide bond for binding at the active site. Modifications at the C-terminus of the peptide included converting the amide to the methyl ester (12), tert-butyl ester (13), and carboxylic acid (14) and also descarboxamide analogues (15-17), which revealed less restricted binding requirements for the C-terminus of the dipeptide. Further optimization should be possible when we learn more about the binding requirements at the active sites of NOSs.

Selective inhibition of monoamine oxidase B by aminoethyl substituted benzyl ethers.

Aminoethyl 3-chlorobenzyl ether was shown previously (Ding, C.Z. and Silverman, R.B. (1993). Bioorg. Med. Chem. Lett., 3, 2077-2078) to be a potent and selective time-dependent, but reversible inhibitor of monoamine oxidase B (MAO B). Based on this result, a series of novel aminoethyl substituted benzyl ethers was synthesized and the compounds were examined as potential inhibitors of both isozymic forms of MAO. Each compound in the series inhibits both MAO A and MAO B competitively, and IC50 values for each compound were determined. In general, the B isozyme is much more sensitive to these inhibitors than the A isozyme (except for the o- and p-substituted nitro analogues), in some cases by more than two orders of magnitude. The selectivity in favor of MAO B inhibition is relatively high for all of the meta-substituted analogues and quite low for all of the ortho-substituted analogues. Having the substituent at the ortho-position is most favorable for MAO A inhibition. With MAO B the meta-analogues were, in general, more potent than the corresponding ortho- and para-analogues with respect to their reversible binding constants. The meta-iodo analogue is the most potent analogue.

Traceless solid-phase synthesis of chiral 3-aryl beta-amino acid containing peptides using a side-chain-tethered beta-amino acid building block.

[reaction: see text] A general method for the attachment of a chiral aromatic side-chain-containing beta-amino acid to a polymer support using a traceless silyl linkage strategy has been developed. Using this building block, solid-phase synthesis was carried out to obtain tripeptide analogues with the aromatic ring either unsubstituted or halogenated (Br, I) at the position of the silyl group. The building blocks could generate libraries of peptidomimetics or cyclic peptides containing beta-amino acids with nonpolar side chains.

A new class of conformationally rigid analogues of 4-amino-5-halopentanoic acids, potent inactivators of gamma-aminobutyric acid aminotransferase.

Recently, we found (Qiu, J.; Pingsterhaus, J. M.; Silverman, R. B. J. Med. Chem. 1999, 42, 4725-4728) that conformationally rigid analogues of the GABA aminotransferase (GABA-AT) inactivator vigabatrin were not inactivators of GABA-AT. To determine if this is a general phenomenon of GABA-AT inactivators, several mono- and di-halogen-substituted conformationally rigid analogues (7-15) of other GABA-AT inactivators, 4-amino-5-halopentanoic acids, were synthesized as potential inactivators of GABA-AT. Four of them, (+)-7, (-)-9, (+)-10, and (+)-15, were inactivators, although not as potent as the corresponding open-chain analogues. The maximal inactivation rate constants, k(inact), for the fluoro- and bromo-substituted analogues were comparable, indicating that cleavage of the C-X bond is not rate determining. Consistent with that observation is the finding that [3-(2)H]-10 exhibits a deuterium isotope effect on inactivation of 3.3, suggesting that C-H bond cleavage is the rate-determining step. The rate of inactivation of GABA-AT by the fluorinated analogue 7 is 1/15 that of inactivation by the corresponding open-chain analogue, 4-amino-5-fluoropentanoic acid (3a). Whereas inactivation by 3a releases only one fluoride ion, inactivation by 7 releases 148 fluoride ions, accounting for the less efficient inactivation rate. Inactivation leads to covalent attachment of 2 equiv of inactivator after gel filtration; upon urea denaturation, 1 equiv of radioactivity remains bound to the enzyme. This suggests that, unlike the open-chain anlogue, the conformationally rigid analogue becomes, at least partially, attached to an active-site residue. It appears that the conformational constraint has a larger effect on inactivators that inactivate by a Michael addition mechanism than by an enamine mechanism.

The multiple active enzyme species of gamma-aminobutyric acid aminotransferase are not isozymes.

Purified gamma-aminobutyric acid aminotransferase (GABA-AT) from pig brain under certain conditions gave a single band on 12% NaDodSO(4)-PAGE, whereas two or three distinct bands were observed on 7.5% native PAGE. These multiple active species were isolated by 5% preparative gel electrophoresis and characterized by N-terminal sequencing and MALDI-TOF mass spectrometry. The results indicate that these active enzyme species are not GABA-AT isozymes in pig brain, but are the products of proteolysis of the N-terminus of GABA-AT, differing by 3, 7, and 12 residues from the full sequence (as deduced from the cDNA), respectively. Conditions for obtaining the nontruncated GABA-AT were found, and the potential cause for the proteolysis was determined. It was found that Na(2)EDTA inhibits the N-terminal cleavage during GABA-AT preparation from pig brain. The presence of Triton X-100 in the homogenization step is partially responsible for this proteolysis, and Mn(2+) strongly enhances the protease activity, suggesting the presence of a membrane-bound matrix metalloprotease that causes the N-terminal cleavage.

1H-pyrazole-1-carboxamidines: new inhibitors of nitric oxide synthase.

1H-Pyrazole-1-carboxamidines were prepared as potential inhibitors of the three isozymes of nitric oxide synthase. All of the compounds were found to be competitive inhibitors of all three isoforms. The most selective compound prepared was 1H-pyrazole-N-(3-aminomethylanilino)-1-carboxamidine (14), which is 100-fold selective for nNOS over eNOS with a Ki value of 2 microM.

Inhibition and substrate activity of conformationally rigid vigabatrin analogues with gamma-aminobutyric acid aminotransferase.

Several cyclopentene GABA analogues were synthesized as conformationally rigid analogues of the epilepsy drug vigabatrin and tested as inhibitors and substrates of gamma-aminobutyric acid aminotransferase (GABA-AT). None of these compounds produced time-dependent inhibition. (1R, 4S)-(+)-4-Amino-2-cyclopentene-1-carboxylic acid ((+)-3), (4R)-(-)-4-amino-1-cyclopentene-1-carboxylic acid ((-)-4), and d, l-3-amino-1-cyclopentene-1-carboxylic acid (6) are good substrates. The K(m) and k(cat) values for the latter two compounds are very similar to those of GABA, suggesting that they bind in an orientation similar to that of GABA. The K(m) value for (+)-3 is 24 times lower than that for GABA, although its k(cat) value is only one-fourth that for GABA; nonetheless, it is a better substrate for GABA-AT than is GABA. All of these compounds, as well as the enantiomers of 3 and 4 and d, l-trans-4-amino-2-cyclopentene-1-carboxylic acid (5), are competitive inhibitors of GABA-AT. These results demonstrate the effects of the carboxylate group orientation and the stereochemistry of the amino and carboxylate groups on the substrate activity and inhibitor activity, and this should be important to the future design of inhibitors of GABA-AT.

Aminomethyl-2,6-difluorophenols as a novel class of increased lipophilicity GABA(C) receptor antagonists.

3- and 4-(Aminomethyl)-2,6-difuorophenols were tested for activity against the three major classes of GABA receptors. 4-(Amninomethyl)-2,6difluorophenol was shown to be a competitive and somewhat selective antagonist at p1 GABA(C) receptors expressed in Xenopus oocytes (K(B) = 75.5 microM with a 95% Confidence Interval range of 75.2 microM to 75.8 microM). This is the first in a novel class of increased lipophilicity GABA(C) receptor antagonists with little activity at alpha1beta2gamma2 GABA(A) and GABA(B) receptors.

Imidazole-containing amino acids as selective inhibitors of nitric oxide synthases.

Two series of imidazole-containing amino acids (1a-e and 2a-c), all larger homologues and analogues of L-histidine, were prepared. Since imidazole and phenyl substituted imidazoles have been reported to be inhibitors of NOS and the mode of action of these compounds as heme ligands is a potential mechanism of inhibitory action, we designed imidazole-containing amino acids as combined inhibitors at both the amino acid as well as heme binding sites. To study the influence of the distance between the amino acid moiety and the imidazole moiety on inhibitory potency, the number of carbons between these two functional groups was varied from two to six. The structure-activity relationships of this class of inhibitors can be correlated with the distance between the heme and the amino acid binding sites of the enzyme. Two of the compounds (1b and 1d) with three and five methylenes between the imidazole and amino acid functional groups, respectively, were found to be potent and selective inhibitors for nNOS and iNOS over eNOS. When phenyl was substituted on the nitrogen of the imidazole, both the potency and isoform selectivity diminished.