You are on page 1of 14

Article pubs.acs.

org/jmc

Carbonylhydrazide-Based Molecular Tongs Inhibit Wild-Type and Mutated HIV‑1 Protease Dimerization
Laure Dufau,‡ Ana Sofia Marques Ressurreica ̧o ̃ ,† Roberto Fanelli,† Nadjib Kihal,† Anamaria Vidu,† † † Thierry Milcent, Jean-Louis Soulier, Jordi Rodrigo,† Audrey Desvergne,‡ Karine Leblanc,† Guillaume Bernadat,† Benoit Crousse,† Michel̀ e Reboud-Ravaux,*,‡ and Sandrine Ongeri*,†

UMR−CNRS 8076, Molec ́ ules Fluoree ́ s et Chimie Med ́ icinale, LabEx LERMIT, Faculte ́ de Pharmacie, Universite ́ Paris-Sud 11, 5 rue J. B. Clem ́ ent, 92296 Chat ̂ enay-Malabry Cedex, France ‡ UPMC−UR4, Enzymologie Molec ́ ulaire et Fonctionnelle, Case 256, 7 Quai St. Bernard, 75251 Paris Cedex 05, France
S Supporting Information *

ABSTRACT: We have designed and synthesized new molecular tongs based on a rigid naphthalene scaffold and evaluated their antidimer activity on HIV-1 protease (PR). We inserted carbonylhydrazide and oligohydrazide (azatide) fragments into their peptidomimetic arms to reduce hydrophobicity and increase metabolic stability. These fragments are designed to disrupt the protein−protein interactions by reproducing the hydrogen bond pattern found in the antiparallel β-sheet formed between the Nand C-ends of the two monomers in the native PR. Kinetic analyses and fluorescent probe binding studies showed that several molecular tongs can inhibit PR dimerization. The best nonpeptidic molecular tongs to date were obtained with an inhibition constant Kid of 50 nM for PR and 80 nM for the multimutated protease ANAM-11. The PR inhibition was selective, the aspartic proteases renin and pepsin were not inhibited.

The post-translational processing of virus polyproteins and the subsequent generation of the structural and functional proteins by human immunodeficiency virus type 1 (HIV-1) protease (PR) are essential steps in the maturation of HIV.1 Inhibiting PR results in the production of an uninfectious virus.2 The PR inhibitors (PIs) used in highly active antiretroviral therapy (HAART) target the PR active site where the two catalytic aspartic residues are located. The emergence of resistance to PIs due to several mutations within or outside the active site3,4 has led to a search for alternative strategies. One of these is to prevent the essential dimerization process of PR because PR is a homodimeric aspartyl protease (99 residues per monomer) and each monomer contributes one catalytic aspartic residue. The dimeric enzyme is active, whereas the monomer is inactive. The PR homodimer is mainly stabilized by a four-stranded antiparallel β-sheet that involves the N-termini (H-Pro(1)Gln(2)-Ile(3)-Thr(4)) and C-termini (Cys(95)-Thr(96)Leu(97)-Asn(98)-Phe(99)-OH) of both monomers. These regions provide over 50% of the hydrogen bonds along the dimer interface5 and contribute close to 75% of the total Gibbs
© 2012 American Chemical Society

INTRODUCTION

free energy of PR dimerization.6 Moreover, these areas appear to be free of mutations.7 Dimerization inhibitors designed to target the PR termini β-sheet interface can block the formation of the homodimer or even disrupt it.8,9 The dimerization inhibitors that target this β-sheet region are C-terminal and Nterminal mimetic peptides,10−12 lipopeptides,13−15 bicyclic guanidinium derivatives,16 and cross-linked peptides with flexible17 or semirigid spacers.18 The recently approved PIs, tipranavir and darunavir, are active against HIV-1 PR variants that are resistant to many PIs and reportedly act as conventional protease inhibitors. They also block protease dimerization in cells but do not dissociate dimerized cellular protease.19 Our strategy for inhibiting PR dimerization is to design molecular tongs using a rigid spacer as this provides an entropy benefit.20−23 We previously described the design of organized molecular tongs based on a naphthalene or a quinoline scaffold in which two peptide strands were attached through a carboxypropyloxy linker to enable them to form an
Received: February 9, 2012 Published: July 17, 2012
6762
dx.doi.org/10.1021/jm300181j | J. Med. Chem. 2012, 55, 6762−6775

were anchored through a carboxypropyloxy linker.23 However. (B) putative complexes between HIV-1 protease monomer and dihydrazido peptidomimetic molecular tongs.24. Med. The azatide moieties were first directly attached to the carboxypropyloxy link (in molecules 3−6. these molecules were less efficient inhibitors of PR and mutated proteases (Kid up to 220 nM compared to 60 nM for the molecular tong with one tripeptide and one peptidomimetic strands23). We made several changes to the structures of the strands. while nonidentical peptides gave asymmetrical antidimers. These motifs were chosen because they are isosteric elements that can replace amino acid residues and introduce resistance to cleavage by peptidases. we then introduced two hydrazide moieties via an azatide (a sequence of aza amino acids) by inserting two azaglycines into the peptidomimetic arms (Figure 2). Figure 3). One drawback of our previous peptide or peptidomimetic molecular tongs was that they were completely insoluble in water. antiparallel β-sheet with the N. As our goal was to decrease the peptide character and increase the water solubility of the molecular tongs. 6763 Figure 2. According to the number of fragments introduced. Identical peptide strands led to efficient symmetrical antidimers.and C-ends of monomers and so disrupt dimers (Figure 1B). 55. thus inverting the sense of the peptide arm (Figure 2). 2012. (A) Putative complexes between HIV-1 protease monomer and P1 peptide molecular tong.23 Inserting a 5-acetamido-2-methoxybenzohydrazide group into one strand (while the other strand remained a tripeptide) increased both the metabolic stability of the molecular tongs and their ability to inhibit wild-type (PR) and mutated HIV-1 proteases in vitro (Kid = 60 nM for PR and 100 nM for the multimutated protease ANAM-11). the linear hydrogenbonding properties of carbonylhydrazide and oligohydrazide peptidomimetics have not yet. 6762−6775 . been exploited in the design of protein ligands and protein−protein interactions inhibitors.20. to our knowledge. Figure 3).26−28 However.23 We also inserted the 5-acetamido-2-methoxybenzydrazide group into both strands to produce the first nonpeptide molecular tongs. mimicking a tripeptidic sequence (except for molecule 6). This report describes the design and synthesis of new molecular tongs containing the original carbonylhydrazide (CONHNHCO) peptidomimetic fragments and their ability to inhibit PR and two multimutated mutated HIV-1 proteases (Figure 2). respectively.22. respectively.doi.1021/jm300181j | J.Journal of Medicinal Chemistry Article Figure 1. All the molecular tongs were based on a naphthalene scaffold and the new strands. We first inserted one carbonylhydrazide motif by linking two Val residues through a hydrazine bridge. This new peptidomimetic was introduced into one or two strands (molecules 1 and 2.org/10.22. This report describes the most potent nonpeptide molecular tongs obtained to date as dimerization inhibitors of HIV-1 PR.25 Oligohydrazide derivatives have also been used in the self-assembly of hydrogen bond-mediated supramolecular systems to produce well-established structures. We postulated that these peptidomimetic elements would operate mainly by binding to the antiparallel β-sheet formed between the N. Chem. General structure of motifs inserted into one or two arms of molecular tongs. the azatide moieties were dx. they may also modulate the balance between hydrophilicity and hydrophobicity and make peptides resistant to proteolytic cleavage of the newly designed molecular tongs.21 The next step was to replace one or two peptide fragments by peptidomimetic ones with keeping the same array of hydrogen bonds. Second. We are now concerned with obtaining more efficient and more water-soluble nonpeptide molecular tongs.and C-termini of one HIV-1 PR monomer (molecular tong P1 in Figure 1A). A Val and a β-Ala amino acid residue were coupled at the C-terminus of the arm in molecules 3 and 4 and in molecule 5.

Figure 3). Med. Acidic cleavage of the tert-butyl carbamate of 11 and its coupling to N-Boc-Val-OH gave compound 13. as reported for molecular tongs having the 5-acetamido-2-methoxybenzohydrazide derivative unit. attached to the carboxypropyloxy link through an amino acid residue at the N-terminus of the arm (in molecules 7−9. The presence of the β-Ala residue (8) or ε-amino-linked Lys residue (9) introduced flexibility in the arm (Figure 3). Hydrogenolysis of the benzyl carbamate of 11 gave 12 in good yield (96%).org/10. using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDCI) and N-hydroxybenzotriazole (HOBt) as coupling agents in good yield (88%) (Scheme 1). 55.1021/jm300181j | J.Journal of Medicinal Chemistry Article Figure 3.doi.23 We also designed a shorter dihydrazide-based peptidomimetic unit in which the third terminal amino acid residue was replaced by a methylamine group (molecule 6) (Figure 3).22. Chem. which possesses only one peptidomimetic arm (Figure 3) to inhibit PR dimerization. Structures of molecular tongs 1−10. We also evaluated the capacity of molecule 10. 6764 ■ CHEMISTRY The synthesis of the peptidomimetic 14 (to be introduced in molecular tongs 1. and the ε-amino group of an α-NH2 protected Lys residue in molecule 9. 2. 6762−6775 . to confirm that only molecular tongs with two arms attached to on the naphthalene scaffold are efficient inhibitors. and 10) is described is Scheme 1 and started with the preparation of the intermediate Z-ValNHNHBoc 11. The azatide moiety was coupled to Val in molecule 7. 2012. 11 was obtained by coupling Z-Val-OH to tert-butyl carbazate. dx. β-Ala in molecule 8. Benzyl carbamate hydrogenolysis of 13 gave the deprotected peptidomimetic strand 14 with a satisfactory yield (71%) (Scheme 1).

55. or Nα-Boc-Nε-Z-Lys-OH. (d) N-Boc-Val-OH. Acid cleavage of the tert-butyl carbamate gave the common intermediate 25. (e) H2. The molecular tongs 1 was synthesized in three steps from 4[7-(3-carboxy-propoxy)naphthalene-2-yloxy]butyryl-Val-LeuVal-OMe 3221 (Scheme 4).1021/jm300181j | J. MeOH. Acid cleavage of the tert-butyl carbamates gave the new peptidomimetic arms 21 and 23 (Scheme 2). CH2Cl2. room temp. (d) HCl·H-Val-NH2. CH2Cl2. room temp. The molecular tongs 3 was synthesized directly by condensing 4-[7-(3-carboxy-propoxy)naphthalene-2-yloxy]dx. AcCN. 10% Pd/C. 12%. (g) HCl·H-β-Ala-OMe. Compound 17 was then coupled to HCl·H-Val-NH2 to give the Boc-protected peptidomimetic arm 18 in good yield (93%). Thus. EDCI. (b) H2NNH2·H2O. AcCN. Synthesis of Peptidomimetics Containing Strands 19. a Scheme 2. (e) TFA/CH2Cl2 (1:1. 2012. Et3N. DIPEA.N′.29 Treatment of 15 with hydrazine hydrate gave the intermediate 16 in 72% yield. The acid hydrolysis of 26 and 28 gave the deprotected peptidomimetic strands 27 and 29. Et3N. and benzyl carbamate hydrogenolysis of 30 gave the peptidomimetic strand 31 (Scheme 3). Chem.org/10. room temp. room temp. 21. MeOH. Classical acid hydrolysis of 18 gave the new peptidomimetic arm 19 (Scheme 2). v/v). which was coupled to N-Boc-Val6765 OH. These modest yields may be partly due to the hydrophilic compounds 26. (f) HCl·NH2CH3. 10% Pd/C. dioxane. room temp. The procedure designed to obtain the new peptidomimetic arms involved the synthesis of a dihydrazide motif 16 that was common to all azatide-based arms (molecules 3−9). 32 was first condensed with ValNHNHBoc 12 to form the intermediate 33 in 76% yield. Et3N. DMF. Compound 17 was also coupled to methyl amine to give 20 (yield 70%) and to HCl·H-β-Ala-OMe to give 22 (yield 67%). DIPEA. EDCI. room temp. compound 15 was prepared in 97% yield by reacting the commercially available compounds tert-butyl carbazate and phenyl chloroformate in dichloromethane in the presence of pyridine (Scheme 2). and 30 remaining in the aqueous phase during the workup of the coupling reaction and consequently being very difficult to extract with organic solvents. to give 17 in good yield (91%). HOBt. 6762−6775 . room temp. 28. room temp.N′-tetramethyl-uronium-hexafluoro-phosphate (HBTU) and HOBt as coupling agents to give compounds 26.N. room temp. room temp. using Obenzotriazole-N. Synthesis of Peptidomimetics Containing Strands 12 and 14a Article (a) BocNHNH2.29 Compound 16 was then treated with phenyl chloroformate in THF. (c) ClCO2Ph. Med. and 23a a (a) Pyridine. Compound 16 was acetylated to give 24 in good yield (96%) (Scheme 3). or 30 in modest yields (46%. room temp. HOBt.Journal of Medicinal Chemistry Scheme 1. (c) HCl/dioxane 4 M. N-Boc-β-Ala-OH. AcCN. THF. or 30%). 28. MeOH. tertButyl carbamate cleavage of 33 was then followed by coupling to N-Boc-Val-OH to give molecular tongs 1 in satisfactory yield (55%).doi. pyridine. in the presence of pyridine. (b) H2.

HBTU. room temp. room temp.org/10. (d) N-Boc-βAla-OH.1021/jm300181j | J. (b) HCl/dioxane 4M. DMF. Molecular tongs 2 was synthesized from 4-((7-(3-carboxypropoxy)-2-naphthyl)oxy)butanoic acid 3420 by a standard coupling protocol (using HBTU and HOBt) to peptidomimetics 14 in 39% yield (Scheme 5). MeOH. DMF.7) and 30 °C. HOBt. (b) HBTU. room temp. The mutant MDR-HM30 contains six amino acid mutations located within and outside the active site of the enzyme (Leu10Ile/Met46Ile/Ile54Val/ Val82Ala/Ile84Val/Leu90Met). which has only one peptidomimetic arm. (c) N-Boc-Val-OH. MDR-HM and ANAM-11. room temp. in 9% yield. room temp. 2012.21 with the peptidomimetic 19 in satisfactory yield (49%) (Scheme 4). room temp. reflux. and 31a Article (a) Ac2O. in agreement with dimerization inhibition (Figure 4). HOBt. (c) N-Boc-Val-OH. In contrast. Et3N. a Scheme 4. Molecule 10. 5 days. in satisfactory yields (40− 73%) (Scheme 5). 6766 RESULTS AND DISCUSSION We used a fluorimetric assay to test the capacity of compounds 1−10 to inhibit recombinant wild-type PR (PR) at the optimum pH (4. DIPEA. HOBt.doi. 6762−6775 ■ protocol (using HBTU and HOBt) to peptidomimetics 19. Synthesis of Molecular Tongs 2 and 4−10a a (a) HBTU. HBTU. compounds 1−4 efficiently inhibited both PR and the two multimutated proteases. DIPEA. (b) HCl/EtOH. 8. HBTU. 19 or 21 or 23 or 27 or 29 or 31. room temp. respectively.Journal of Medicinal Chemistry Scheme 3. . 27. Med. DMF. room temp. where vi were the initial rates. room temp. DIPEA. Plots of [E]0/√vi versus √vi were constructed. DIPEA. HBTU. DMF. butyryl-Val-Leu-Val-OMe 32. 14. HOBt. We used Zhang−Poorman kinetic analysis to identify the mechanism of inhibition. while ANAM-1131 has 11 mutations: (Leu10Ile/Met36Ile/Ser37Asp/Met46Ile/ Arg57Lys/Leu63Pro/Ala71Val/Gly73Ser/Ile84Val/Leu90Met/Ile93Leu). DIPEA. room temp. whereas compounds 6 and 7 inhibited the enzyme poorly (13 and 21% at 28 μM. dioxane. DMF. was obtained as a byproduct. and 31. Synthesis of Peptidomimetics Containing Strands 27. DIPEA. (d) 19. Chem. 29. DMF. HOBt. DIPEA. DMF. (f) H2. respectively). HBTU. and 9 were not inhibitors at the highest tested concentration of 28 μM. DMF. HBTU. 21. resulting from incomplete coupling. 10% Pd/C. 23. The corresponding inhibition constants Kid are dx. Synthesis of Molecular Tongs 1 and 3a a (a) 12. THF. HOBt.20−22 Compounds 5. In all cases. 55. 29.18. room temp. parallel lines were obtained. (e) Nα-Boc-Nε-Z-Lys-OH. The molecular tongs 4−9 were synthesized from 4-((7-(3-carboxypropoxy)-2naphthyl)oxy)butanoic acid 3420 by a standard coupling Scheme 5.

5 μM 4 (■) and 5. 11 μM 4 (▲). Med. 55. 6762−6775 . (A−C) Zhang kinetic analyses of PR.Journal of Medicinal Chemistry Article Figure 4. 2 μM 2 (▲). (C) ANAM-11 by 4. Enzyme activity was also always measured without inhibitor (●). shown in Table 1. The hydrophobic 1-anilino-8-naphthalene sulfonate (ANS) is 6767 preferentially bound to the hydrophobic surfaces of proteins. (D) PR inhibition by 17 μM 4 (■). (A) PR inhibition by 2. The mechanism of inhibition was confirmed for the four compounds using fluorescent probe binding.1021/jm300181j | J.32 We observed important enhancements of the fluorescence due to ANS when compounds 1−3 and P1 were added to the dx. 2012.8 μM 2 (■).doi. (B) MDR-HM inhibition by 2.6 μM 4 (▲).8 μM 2 (■) and 1.8 μM 2 (▲). MDR-HM. Chem.9 μM 2 (▲). and ANAM-11 inhibition by compounds 2 and 4 at pH 4. (F) ANAM-11 by 22 μM 4 (■) and 17 μM 4 (▲).2 μM 2 (■) and 2. (E) MDR-HM inhibition by 8.7 and 30 °C.org/10.

3 (▲). 55.1 5. Remarkably. 4 (◆).doi. ANS emission fluorescence (470 nm) measured at pH 4.2 and 11.14.7 and 25 °C. the inhibitory activity was only decreased by a factor of 1. or amprenavir were Figure 5.20. For the most soluble compound.95 for cLogP). hydrophobic surface area. Med.19. The inhibition of the two mutated proteases by compounds 1−3 remained essentially identical for a given mutated protease (for example. clogP.7)a Kid (nM) compd P1 1 2 3 4 compd 5 6 7 8 9 10 a Article clogP 7.3 4. This is in agreement with the expected exposure of hydrophobic interface area when the dimer is destabilized or dissociated. Similarly. The final DMSO concentration was always 0.6 6. enzyme (less marked for 4) (Figure 5) but not when the activesite inhibitors acetylpepstatin. An important decrease of cLogP (factor of 78) and hydrophobic surface (factor of 2.org/10. Inhibitors 1−4 and the parent peptidic molecule P1 were bound to the unmasked hydrophobic surfaces of PR. We used molecular modeling to estimate three physical properties of the molecules. Compound dx. Fluorescence intensity was corrected for the fluorescence of the same concentrations of ANS and inhibitor without enzyme. 2 (●).Journal of Medicinal Chemistry Table 1. inhibition was not affected by inserting an azatide moiety (diazaglycine) into one arm directly linked to the carboxypropyloxy group while the other arm was a tripeptide (Val-Leu-Val) (Kid of 150 nM for 3 and 90 nM for 1). Inhibition of PR and Mutated HIV-1 Proteases MDR-HM and ANAM-11 by Compounds 1−10 (30 °C and pH 4. active-site inhibitors changed the conformation from semiclosed (dimer alone) to closed (active-site inhibitor/ dimer complex). 6762−6775 . inhibitory activity was decreased when azatide moieties were introduced in both arms and directly linked to the carboxypropyloxy group (Kid of 700 nM for 4 and 50 nM for 2). and compounds 1 (■). The experimental results summarized in Table 1 indicate that the introduction of pseudopeptidic arms containing two Val residues linked by an hydrazine bridge dramatically increased the inhibitory potency (Kids of 90 nM for 1 and 50 nM for 2) over those previously reported for the peptidic arm Val-Leu-Val (P1. These data show that both activity and solubility can be increased (compound 3). 560 nM)20 and to the Val-(5-acetamido-2-methoxybenzo6768 hydrazide) arm (∼200 nM).2 1. The increased flexibility at the N-termini destroyed all inhibitory activity in contrast to our previous observation with 5-acetamido-2-methoxybenzohydrazide and 3amino-6-methylpyridine(1H)-one strands. and hydrophilic surface area (Table 1). saquinavir. Both inhibitory activity and cLogP were improved for compound 3 (factor of 3.2 against PR). nd: not determined.7 for PR inhibition and 1.4 2. Lastly. in agreement with our previous results showing that there must be two peptide strands on the molecular tongs to inhibit dimerization. The excitation wavelength was 370 nm. whereas clogP was poorly increased (factors of 1.0 0. which are analogous to the proteases found in multiresistant viruses (Table 1). further increasing the flexibility by introducing a β-Ala residue or a Lys residue linked through its ε-amino group completely suppressed inhibitory activity (8 and 9 relative to 4).5 9.8 8.2% (v/v).4 3. the four new molecular tongs 1−4 behaved as antidimers against the two multimutated proteases ANAM-11 and MDR-HM. 4). in order to further correlate the inhibitory activity with the hydropathic properties of the molecules. These two last observations are in accordance with our previous results showing that a tripeptide sequence is better than the shorter dipeptide or longer tetrapeptide sequences. and P1 (▼) prior to adding ANS. Inserting an azatide moiety at the C-terminus via a Val residue made the symmetric molecule 7 almost completely devoid of inhibitory activity (21% inhibition at 28 μM). A favorable influence of hydrophobicity on inhibitory activity was observed as previously reported for some molecular tongs. These data highlight the difficulty but the possibility to find a compromise between hydrophilicity and inhibitory activity.1 clogP 3.20 Third. added. shortening both arms by removing the C-terminal valinamide gave the poorly active compound 6. 2012.1021/jm300181j | J.3 mg/mL). Lengthening both arms with a β-Ala residue linked to the C-terminal side (compound 5) gave a poorer inhibitor.15 leading to a smaller decrease in the ANS fluorescence than that observed with the PR alone.23 The azatide peptidomimetic unit is thus best directly attached to the carboxypropyloxy linker (3. saquinavir (○).21 The decreased inhibitory activity of molecular tongs 5 and 6 can be also ascribed to the lack of hydrophobic valine side chain.22 They also show that the one-arm molecule 10 is not an inhibitor. However. respectively).23 The presence in the molecular tongs arms of carbonylhydrazide (CONHNHCO) peptidomimetic fragments that have particular hydrogen bonding properties led to a noticeable increase of the inhibitory activity for compounds 1 and 2 compared to P1 (factors of 6. Chem.15 In contrast. clogP and solvent accessible surface areas are indicated.08 and 1.4 hydrophobic surface (Å2) 1467 1443 1391 1079 673 hydrophobic surface (Å2) 797 639 911 794 1304 924 hydrophilic surface (Å2) 232 281 326 398 555 PR MDRHM ANAM11 56020 90 50 150 700 hydrophilic surface (Å2) 570 465 468 548 595 270 nd nd 160 100 160 80 270 170 1600 800 % inhibition at 28 μM 0 13 21 0 0 0 Standard errors of initial rates are less than 5%. Kid of 2: 120 nM for MDR-HM and 80 nM for ANAM-11).25 whereas cLogP was improved by a factor of 78.39) did not comprise the inhibitory efficiency (compound 4). while compound 4 has valine residues. 4 (≈ 0.21. PR (275 nM) was preincubated in the absence of inhibitor (△) or in the presence of acetylpepstatin (×).17) and increase in hydrophilic surface (2.22.

respectively). monomeric renin and pepsin. there were still 13 hydrogen bonds in the complex with molecular tongs 4 (Figure 6). However. However. molecules 1−4 interacted more strongly with the monomer than did molecules 7−9. These results suggest that they are more resistant to hydrolysis than the previously described peptide molecules and about the same as those with 5-acetamido-2-methoxybenzohydrazide peptidomimetics (no significant degradation after 33 h). The carbonylhydrazide peptidomimetic arms still had an extended conformation. 6762−6775 ■ .org/10. We used molecular modeling to build several complexes that might be formed between our new molecular tongs 1−9 and the PR monomer and used them to correlate inhibitory activity with structural characteristics. 55. The half-lives of compounds 1 and 3 were about 24 h. The designed carbonylhydrazide peptidomimetics had an extended conformation compatible with the formation of a β-sheet. For example. we tested the ability of these molecules to disrupt the PR termini β-sheet interface in order to inhibit PR. Their original hydrogen bonding properties made these peptidomimetics good modulators of protein−protein interactions involving β-sheet structures. The same holds if we compare compound 4 (only one valine side chain per arm) with compound 1 (two valine and one leucine side chains in one arm. The PR− molecular tongs complexes had an extensive network of hydrogen bonds. Valine side chain(s) may reinforce the binding of molecular tongs to the monomer ends and explain the improved inhibitory efficiency of molecular tongs 1−4. we determined the stability of the symmetrical molecular tongs 2 and 4 and asymmetrical molecular tongs 1 and 3 in RPMI culture medium containing 20% fetal calf serum (Figure 4S in Supporting Information). and two valine side chains in the other) and with compound 2 (two valine side chains per arm). Med.22 The metabolic stability of the molecular tongs with two peptidomimetic arms was also better than that of the asymmetrical molecules bearing one tripeptidic arm. Conversely. Last. As a proof-of-concept. compounds 2 and 4 were not significantly degraded after incubation for 48 h. The biological results are in agreement with the proposed complexes designed by molecular modeling. We used one monomer from the crystallographic data for the dimeric PR and built molecular tongs using the C. Most of them were with the C-terminal end (nine for molecular tongs 4 and only five with the N-terminal end). the inhibitory potencies varied greatly according to the molecule structures.1021/jm300181j | J. (B) the C. This minimization did not disrupt the postulated β-sheet structure formed between the N. The absence of inhibitory activity of 5 and 6 compared to 4 may be explained by the lack of hydrophobic amino acid side chains in 5 and 6 because hydrophobic interactions are beneficial for stabilizing β-sheet structures.doi.and N-terminal ends of the removed second monomer of PR as templates. as indicated above. 4 was the poorest inhibitor of both PR and the mutated proteases (Table 1). Finally.and C-termini of the remaining PR monomer and molecular tongs 1−6.and C-termini of the remaining PR 6769 monomer was higher for 4 than for 1 and 2 (13 against eight and nine for compounds 1 and 2. Putative complexes between PR monomer and the peptidomimetic molecular tong 4 (color by atom type): (A) one PR-monomer (ribbon tube representation). Conversely. Chem. there was no stable β-sheet structure with molecular tongs 7−9 after minimization (see Supporting Information).Journal of Medicinal Chemistry Article Figure 6. Neither renin nor pepsin was inhibited by 10 μ M concentrations of compounds 1−4. We observed that the number of hydrogen bonds formed with the inhibitor and the N. CONCLUSION We have designed and synthesized a new series of naphthalenebased molecular tongs containing carbonylhydrazide (CONHNHCO) and oligohydrazide (azatide) peptidomimetic side arms. hydrophobic interactions are missing for compound 4 to further stabilize the β-sheet structure. The azatide-based scaffolds also increased water solubility. 2012. we evaluated the selectivity of molecules 1−4 toward two other aspartic proteases. indicating that there must be appropriate balances between dx. which is very similar to the 28 h half-life of the P1 peptide molecular tong bearing ValLeu-Val21 (Figure 1). The energy of the monomer− molecular tongs complexes was minimized. The correct array of alternating 10-member and 14-member rings involving two consecutive hydrogen bonds was maintained after minimization (see Figure 6 and Supporting Information). Their scaffold gave the molecular tongs some rigidity and forced the arms of the tongs to have the extended structure essential for mimicking a βstrand. The best of our nonpeptide molecular tongs inhibited PR dimerization with an inhibition constant Kid of 50 nM. which is the best inhibition of HIV-1 PR dimerization reported to date for molecular tongs.and N-termini of one PR-monomer (green backbone).

2H). TR = 24. 55.1021/jm300181j | J.97 (m.0. Elemental analyses (C. HPLC purity: method 1. 13C.4. 28. 6. J = 6. 1163.37 (s. 7. J = 8.0.8.97−4.97 (d.20 (bs. 2H).2 Hz. 7. 66. 19. DMSO-d6): δ 11. 7H). 4.06 (t. 831.25 (t. 2H). HPLC purity: method 1. 1 equiv). 27.82 (s. 1516. Compounds 19 (88 mg.0.9. 4. TLC was performed on 60F-250 silica gel (0.05 (s.1. 827.77−0.91−8. 2H). 1H NMR (400 MHz. 1. 1H). 6770 ■ EXPERIMENTAL SECTION TR = 14. 911 [M + 39]+. 12H).0. in order to obtain powerful inhibitors.3 Hz. 155. 106. 5 μm.26 mm thick plates. CH2Cl2.0. which was washed successively with CH2Cl2. 1210 cm−1. 0.20 were prepared according to published methods. 1.72 (s. 1. 2. J = 8.8. 2. Product 10: mp °C. Elemental analysis data are included in the Supporting Information.4.1. 31. HPLC purity: method 1. J = 8.85 (d. 30.9. 18. 57.2.66 (m.0. J = 6.31−2. 56%).0.1 Hz. 7. DMSO-d6): δ 9. 75 MHz) or Bruker AVANCE 400 (1H. 10% aqueous K2CO3 (10 mL). 66. 30.2. Synthesis of Molecular Tongs (3). 3. 30. 170.org/10. 128. 171. IR (neat): νmax 3221.4 Hz.2. N) were performed on a Perkin-Elmer CHN Analyzer 2400 at the Microanalyses Service of the Faculty of Pharmacy at Chat ̂ enay-Malabry (France). and broad singlet (bs). J = 8. 6H). 2H). (C33H48N4O9) C. ESI+ MS m/z 895 [M + Na]+.1. mixture A/B from 40/60 to 0/100 in 30 min. and NBoc-Val-OH (20 mg. 7. brine. 1169. 3H). 0. 24. 9%) as white solids.6.2 Hz.5. and again with CH2Cl2 to yield 1 (49 mg. 49%).01 (m. 18. and HOBt (52 mg. Method 1: column SUNFIRE. 8. doublet triplet (dt). 2H). 58. 39%) and 10 (26 mg. 19. 7. aqueous 0.9. APCI+ MS m/z 970 [M + H]+.9. 9.99 mmol. 67. mixture A/B from 60/40 to 0/100 in 20 min. 155. 56. Dimethylformamide (DMF) was distilled over CaSO4.1 min. doublet doublet (dd).08 (t.92 (s.45 mmol.43 (m.2 Hz. Med. a mixture of (A) water (0.7. 157.2. 2.86 (d.9. 2H). 2963. 1545.09 (d. (C50H79N7O12·6H2O) C.1. CDCl3 was used as solvent unless otherwise stated.90 (m. 1209. method 2. 18.2.40 (m.9. 19. 7. 135.7.7. 1169. 1 equiv) were added and the mixture stirred again at room temperature for 5 days. mixture A/B from 60/40 to 0/100 in 20 min. 158. 169. and 2 mL of 9 M HCl in ethanol was added and the mixture stirred for 4 h. 3H). 115. N.7. Chemical shifts (δ) are in ppm. 7. 1H).63 (s. 24H). 2036. 1H NMR (400 MHz. Product 2: mp 220−223 °C. mp 213-216 °C. 3. Trituration in EtOAc yielded a beige solid that was collected on a filter and washed successively with EtOAc.70 (d. 22. 9. DMSO-d6): δ 171. 3. Spots were visualized with UV light (254 nm).69 (s. 2.90 (bs. DMSO-d6): δ 9. 30.66 (d.9. 2H).2 equiv).Journal of Medicinal Chemistry rigidity and flexibility.1. 1743. UV detector PDA 2996. 92 μmol.8 Hz. 6. 123. 129. 4H).1 Hz.92 (s. J = 7. 2. All solvents were purchased from commercial sources. 57. 1.42 mmol. 19. 3.37 (m. 4-[7-(3-Carboxy-propoxy)naphthalene-2-yloxy]butyryl-ValLeu-Val-OMe 3221 and 4-((7-(3-carboxypropoxy)-2-naphthyl)oxy)butanoic acid 34.99 mmol. 40. 57. TR = 24. 18.71 (d.8. Liquid chromatography was performed on Merck 60 silica gels (230−400 mesh).6. 0. J = 6.2 Hz. The purity of molecular tongs was determined by HPLC using a WATERS modular system (quaternary pump + controller E600. and Et2O to give 3 as a white powder (150 mg. The example of the inhibitor 4 shows that water solubility does not compromise inhibitory efficiency. 6.1 M HCl. 66. 1H).9 min. J = 8. 2. 2012.N′. The solvent was removed under reduced pressure to give a beige oil.38 (s.61 (s.8.9. tetrahydrofuran (THF) was distilled over sodium/benzophenone. 9H).9 Hz.3.20 (s.92−7.60 (s.40 (m. 2H). H.68 (d.8. 56. 1H). 106. J = 8. Article Chemistry. 17. 100 MHz) spectrometers. Synthesis of Molecular Tongs (1). 7H).38 mmol. 2H).6 Hz. 1. DMF was removed under reduced pressure.1. 19. 2H).7 Hz. 31. J = 8.85 (m. 4H). 2967. 1.9.17 (s. mixture A/B from 60/40 to 0/100 in 20 min. 77. 57. C18. 1H NMR (300 MHz. J = 7. C18.95 (s. 106. 6. 123.N. 51. mixture A/B from 40/60 to 0/100 in 20 min. Product 33 (80 mg.6.0.3.6 Hz.78 (s.5. 1209. Synthesis of Molecular Tongs (4). DMSO-d6): δ 9. 25.4 Hz. H2O. 1.86 (d. 1. 123. 96%. 1 equiv) was dissolved in DMF (4 mL).2. 135. 1605. 12H).doi.36−4. 171. 7. 29.9. aqueous 10% NaHCO3. 1H). 0. HPLC conditions were as follows. 21.1. 2H).0. (C42H65N9O11·1.24 (m.2 Hz. Anal. 1. 110 μmol.08 (m. CH3OH. J = 8. 66. Product 4 was obtained as a white powder (174 mg.2 equiv) was dissolved in DMF (5 mL) at 0 °C was mixed with compound 34 (150 mg. DMSO-d6): δ 172.9.2. J = 8. 0.2. 7. 40. 1H NMR (400 MHz.05% TFA). and the mixture was stirred for 24 h at room temperature under an argon atmosphere.8. water (10 mL). 1.79 (s. H. 158. 1H). 2H).9.6 mm.6. Chem. 7. J = 9. 115. 7.05% TFA).7.1. 13C NMR (100 MHz. J = 8. 9. triplet (t). 0. 9. 3H). 171.20 (br s.77 (s. 24.3. 1478.6. 170.2. 4. 0. 96%. 6H).3. 1632. 2H). 0.60 (bs. 1. 128. 135. J = 8. 0. 158. 1021. 4. 652 cm−1. 66.98−2. 17. 1H NMR (400 MHz. 115.4. 6.7.9. room temperature.7. IR (neat): νmax 3252.5H2O) C. and the mixture was stirred for 48 h at room temperature. 4H).90−2. 3. 1.71 (d. 19.5 Hz. 1H). 24. to which were added DIPEA (207 μL. HOBt (134 mg. 158.05% TFA) and (B) MeOH (0. 6H).70 (d. 2H). 6H).5. IR (neat): νmax 3221. 28.7 equiv). 172. 7. 13C NMR (75 MHz. and acetonitrile was distilled over CaCl2.05 (s. 7. or with a solution of ninhydrin in ethanol.37 (q.8. DMSO-d6): δ 171.1. and 1hydroxybenzotriazol (HOBT) were purchased from commercial sources. 18.5 min.5. 157. 1. 30.7. 6. 150 mm × 4. 7. 155. 4H). HBTU (42 mg.7. 31. 31.6. mobile phase a mixture of (A) water (0.98 (bs. 8H).5. Purification by HPLC gave products 2 (168 mg. 55%) as a light-brown solid. 0.1 min.17 (m.8. and CH2Cl2 was added to the crude product to precipitate a solid.4. Compound 14 (327 mg.54−1.0.6. N.1. 7. 29.1 min. 19.1 equiv). 30 H). 10% aqueous 10% citric acid (10 mL). 1211 cm−1. 157.1 Hz. 1605. 24. quadruplet (q).6. DMSO-d6) δ 9.2.8 Hz. 7.34 (m.5. J = 6. J = 6.03 (m.7. 13C NMR (100 MHz.3.12− 4. IR (neat): νmax 3278. 7.3.6 Hz. J = 6. 4H). 18.9. 128.18 (s. 0.1.8. 3. 115.2 equiv).81 (t. 29. 2H).06 (m. 1H).33−2. 4. 4H). 2H).4. 1H).7 Hz. 2H). Anal.3. 55. 829. 0.9. and between hydrophilicity and hydrophobicity.5% phosphomolybdic acid in ethanol. DIPEA (145 mg. 3.08 (d.33−2. 4H).1. N.1. 24. 2H). 4.3. 1H).1 min.6. 24. and HBTU (426 mg.6. NMR spectra were obtained on Ultrafield AVANCE 300 (1H. 1H). 51. 7.7. 171. Method 2: column SUNFIRE. 135.63 (s. DMSO-d6): δ 173. flow rate 1 mL/min. HBTU (159 mg. 1H). multiplet (m).8 Hz. 300 MHz .22 (t. J = 6. 123.5. 169.2.7.5%.4 Hz. O-benzotriazol-1-ylN. 2.99 (t. 6H).99 mmol. (C48H76N8O12·5H2O) C. 50. 2H). method 2. Then.99−4. 9.9.7. 2H).96 (m.6. 18.9.43 (m.99 mmol. 50.5 equiv).1. 77. 129. 2. 170. 2H). 135. 2H). 1386. The ability of these proteolysis-resistant molecules to inhibit the multimutated ANAM-11 highlights their potential to successfully overcome the resistance to classical PI protease inhibitors presently encountered. 106. 5 μm. detection at 254 nm. 106. 156. 29.1. 0.0. 24. TR = 19. 1H). Anal. All molecular tongs tested in biological assays were 95% pure.3.7 Hz. room temperature. 1657.2 mmol. H.7. 8. 1.7. 6762−6775 . Anal. 799 [M + dx. 2H). 2H). Mass spectra (APCI and ESI) were obtained using a Bruker Esquire-LC apparatus at the SAMM (Faculty of Pharmacy at Chat ̂ enay-Malabry).9. 1H).8 μmol. 2H). mixture A/B from 40/60 to 0/100 in 30 min.71 (d. and again with CH2Cl2 (5 mL). 30. 248 μmol.6. 1H). 4H). 1209. 78.91−2.97 (d. 4H).6. 31. 91. mp 218−220 °C. mobile phase.4. 1.1. J = 6. 18.7. 2H). 651 cm−1. 23. 13C NMR (100 MHz.81−0.8. The DMF was removed under reduced pressure.5. 30. 2.8. 1H).N′-tetramethyluronium hexafluorophosphate (HBTU). 4. 58.5 Hz. 18H). detection at 235 nm. 7. 13C NMR (100 MHz. J = 8. Same procedure as for 3 from 34 and 19.17 (s. 1633. DMSOd6): δ 173.1. mixture A/B from 60/40 to 0/100 in 20 min. 2H). 150 mm × 4. 21. 6.35−4.97 (d. TR = 8.49 (m. 1H).9.38 mmol) and 32 (230 mg. and the following abbreviations are used: singlet (s). Et3N (35 μL. 18. 57. TR = 11. 99%. 57.81 (t.91 (d.9. 2 equiv). J = 8.6 mm. 1.5.7. J = 6. H. ESI− MS m/z 643 [M − H]−. 24. 13C.18 (s. 1478.32 (bs. J = 6. 123. J = 6. 1H). 30. 6. 3H).1. 0. 170.2 Hz. J = 8. 2H).5. 2H). 1.35 mmol) were dissolved in DMF. 66. 2963. 25. 3.0.3. 7. 115.2 equiv).82 (s. Protected amino acids. 2H). 8.98 (d. H.1. 2036. 157.0. TR = 8.97 (m. 0.7 Hz.2. J = 8. 6H). 668 cm−1.37 (s. ESI+ MS m/z 783 [M + Na]+. 4H). doublet (d).94 (m. 400 MHz.05% TFA) and (B) AcCN (0. 6.03 (m.9. 1.6 Hz. 1H). 1537. 0. autosampler 717). J = 8. 0. 6H). 97%.82 (d.8. mp 181−185 °C. HPLC purity: method 1. Synthesis of Molecular Tongs (2) and Molecule (10). 100%.32 (bs. Melting points were determined on a Kofler melting point apparatus.44−1.5.92 (d.97 (d. and the resulting solid was washed successively with CH2Cl2 (5 mL). 97.2 equiv). 4. 169.2 Hz.9 Hz. 30. 170. method 2. ESI+ MS m/z 979 [M + Na]+. IR (neat): νmax 3277.44 (bs. 1H). N.2.97 (d. 97%.5 min. 4H). flow rate 1 mL/min.06 (t. 9 H).1 Hz.

0. 123. 123.3. 2.5. 2. 24. 100%.00 (bs. 4H).97 mmol. 2. Anal.3 mmol. 7. 29. 60. 3H). Method 2. 1H). 26. J = 8. DMSO-d6): δ 171.1 equiv). TR = 11. Compound 11 (1.1 equiv) in CH2Cl2 (10 mL) was added dropwise over 30 min at 0 °C.9 mmol.74 g. The residue was dissolved in EtOAc (400 mL) and washed successively with 10% aqueous citric acid (50 mL). 19.19 g. 3.8. 2H). 8. 6. Anal. 1.1 g. 6.0 Hz.75 g.63−7.93 (s. 66. 157.org/10. 33. 1642. 28.9. 81. 18. mixture A/B from 90/10 to 0/100. 2H).3. 158.3. 4H). 173. 13C NMR (75 MHz. 2H). 4. The resulting crude product was purified by flash chromatography on silica gel and eluted with EtOAc/cyclohexane (20/80).9.8. 156.2. Chem. method 2. TR = 5. IR (neat): νmax 3224. 1664. 1515. 2H). 4H).6 min.7. 1H). N. Anal.36 (t. J = 8.2. 3 equiv). 1H).5. N. 12H). mixture A/B from 70/30 to 0/ 100 in 20 min. 4H).12 (s.0 Hz. 97%. 127. 169.8. 137. J = 7.1 equiv). 1. and concentrated under reduced pressure to yield 13 as a white solid (1. 1433. 4H). 4H). 30. 6.02 (br.1 Hz. 3. DIPEA (1. IR (neat): νmax 3275.1. 1. 2. 66. 7.Journal of Medicinal Chemistry K]+.7. It was then dried over MgSO4. 2967.7.6. Anal.13 (m. 4H). 155. CDCl3): δ = 7. 10% aqueous K2CO3 (50 mL). 28.9.6. 8. 2 Hz.3 Hz.8. 100%. 125. ESI+ MS m/z 1040 [M + Na]+ ]+. 24. 106. 3. 0. tert-butyl carbazate (1. 4H). 170. 2. J = 7. 159. 2H).10 (m.59 (s. 59. 31.5.3 mmol. 1210 cm−1. Same procedure as for 3 from 34 and 21. 157.1021/jm300181j | J.2. 159. 2H). 2. The organic phase was dried over MgSO4.70−7. Anal.3.8. 2256. 135. 1514. 2H). 1167. 1 H NMR (300 MHz. H.97 (m. ESI+ MS m/z 613 [M + Na]+.48 (s.4.96 (d.19 (s.92 (m. 6762−6775 . 4H).1.08 mL. N. 155. 20. 3.4. mixture A/B from 40/60 to 0/100 in 30 min. 3. and brine (5 mL). 9.9. 67. Same procedure as for 3 from 34 and 27. 31. 129. 3H). 1. 1.36 (m. 3H). 157. tert-Butyl N-[(1S)-1-[[[(2S)-2-(Benzyloxycarbonylamino)-3-methyl-butanoyl]amino]carbamoyl]-2-methyl-propyl]carbamate (13). (C24H34N10O8·2H2O) C.41 mmol) was dissolved in MeOH (5 mL). 123.7.0. mixture A/B from 70/30 to 0/100 in 20 min. and EDCI (1. 1212 cm−1.00 (m. H. 115. 2H). 127. 4. J = 8. for 24 h at room temperature. 1.7 min. 2H).4. and water (50 mL). 2912. 2.48 (s.1.42 (m. 40.7. tert-Butyl N-[[(2S)-2-(Benzyloxycarbonylamino)-3-methylbutanoyl]amino]carbamate (11).5. 2H). 1607. 1H NMR (400 MHz. and the residue was dissolved in EtOAc (50 mL). 2H).4 min. mp 141−143 °C. DMSO-d6): δ 9. 155.1. 96%). IR (neat): νmax 3277. IR (neat): νmax 3280.9. and 10% Pd/C (15 mg.4.2.9. 2H).94 (dd.2 equiv) were dissolved. 1. 1248. 4H). DMSO-d6): δ 170.9. 135. 6H).57 (s. 18H). 158. 97%). 4H). 4H). 1H).1 min. 6H).9. 7. HPLC purity: method 1. 1H NMR (300 MHz.38 (s. 115. DMSO-d6): δ 9. 51. 1630.7.53 (bs. 100%. 2H).1 equiv) were successively added and the reaction mixture was stirred at room temperature for 24 h.9 Hz.60 (s.85 (m.83 (d. 2H). Synthesis of Molecular Tongs (7). (C15H30N4O4·2H2O) C.1 Hz.5. mixture A/B from 80/20 to 0/100 in 30 min. DMSO-d6): δ 171. Phenyl N-(tert-Butoxycarbonylamino)carbamate (15). 2H). mp 225−227 °C.1 min.3. TR = 4.08 (m.2 Hz. 1663.08 (br. 29. 2. 7. 26.0. ESI+ MS m/z 725 [M + Na]+. method 2.5. 1H NMR (300 MHz. DMSO-d6): δ 7. TR = 9. and concentrated under reduced pressure to give 11 as a white powder (2. HPLC purity: method 1. 1089 cm −1.24 (m. 741 [M + K]+ ]+.2. 13C NMR (75 MHz.23 (br.1 min. ESI+ MS m/z 487 [M + Na]+. 57. ESI+ MS m/z 388 [M + Na]+. 2H). HPLC purity: method 1. 40%).3. 54.05 (s.1 equiv). 3.33−2. CD3OD): δ 7.73 (s. 8. The resulting organic phase was washed successively with 10% aqueous citric acid (10 mL).0. 58. Product 5 was obtained as a pale-gray powder (174 mg. TR = 3. 137. 1. 1275. 158. and CH2Cl2 was evaporated off under reduced pressure. CD3OD): δ 7. filtered. (C32H48N12O10·1. 115. 1. filtered. under a hydrogen atmosphere. This mixture was stirred 6771 Article overnight at room temperature and then diluted with EtOAc (50 mL). 1H). 1531. filtered.9.8. 6H). 2H). 1290.2. IR (neat): νmax 3276. 58. TR = 24.5H2O) C. 2.5.1. 7. mp 132−134 °C. N. 1039. 77. 1H NMR (400 MHz. 9. 2H). 130. DIPEA (3.9. N. 169. J = 7.7. CD3OD): δ = 173. J = 9 Hz.8. N. mixture A/B 60/40.20−4. 1214 cm −1. TR = 4. 4H). Anal. method 2.14−7.70 (s.01 (br.80 (s.9 Hz. mixture A/B from 70/30 to 0/100 in 20 min. 117. J = 7.6. and concentrated.96 (dd. 30.97 (s. dx. 1685.2. C.97 (d. 0. 12H).3. J = 7. mp 156−158 °C.2. 1H). 4H). 1. CDCl3): δ 173. 9.7. mp 208−210 °C. DMSO-d6): δ 171.35 (bs.9. 9H).1 mmol. Anal. 1H).84 (s. tert-Butyl N-[[(2S)-2-Amino-3-methyl-butanoyl]amino]carbamate (12). 69%).77 mmol.5. 1.9.1 equiv). 4. in CH2Cl2 (10 mL) and phenyl chloroformate (2. water (10 mL).1 equiv).14 g. 1208 cm−1.51 (m. 1629. 157. 28. 135. 1631. (C18H27N3O5) C.72 (m. 2. 2H). IR (neat): νmax 3215.3 Hz.68 g. 8.42 (t. 66. ESI+ MS m/z 387 [M + Na]+. 1H NMR (300 MHz. 2369. 58. H.18 (s. to give 15 as a white powder (3.2. 20. mp 157−159 °C. 6.2. 66. 13C NMR (100 MHz.9. 0.28 (m.9. 30.21 (d. 1. 95%.6. Same procedure as for 12 from 13. 2.3 Hz. ESI+ MS m/z 757 [M + Na]+.3.89 (m. 1542.9.4. J = 7. 73%). H. 1514.75H2O).01 (m. 1186. 1. 10% aqueous K2CO3 (10 mL). Synthesis of Molecular Tongs (9). 3.04 (m. J = 7.98 (t.1. 4H). 4H). Then. 158.15 (m.00−1. 4H). 4H). 55%).8. 68. 1. 2.8 Hz.77 mmol. 6. Product 14 was obtained as a white solid (691 mg. 12H). 4 equiv). 30. DMSO-d6): δ 9. 2H). Synthesis of Molecular Tongs (6).9.16 g. 8. 6.27 (m. 1H NMR (400 MHz. 7. 2. 16. 1. 135.77 mmol. 1552. 3. 2012.70 (d.89 (m. 3. 155. 55. 128.8 min. N-Boc-Val-OH (0. Anal. (C46H72N12O14·4H2O) C.0.3 mmol.0.1. 34. The mixture was stirred overnight at room temperature. C10H21N3O3 (%) C.50 (bs.90 (bs.1.6.6. 1H). The organic phase was washed with 10% aqueous citric acid (5 mL).7. 100%. 7. 5H).10 (m.63 (d.3. 1 equiv) was dissolved in CH2Cl2 (30 mL) at 0 °C and to it were added successively. TR = 15. Anal.9.doi. N-Z-Val-OH (2 g. 8.09 (br. J = 8. 13 C NMR (100 MHz. 158. 7. 18.27 (m. with stirring.96 (s. 80.49 (s. 1H NMR (400 MHz. 2947. H. water (10 mL).80 (s.5.88 min. 4H). 4H).5. 1043 cm−1. H.64 (m. mp 138−142 °C. 32. 175. HOBt (0. 20.14 (bs. Benzyl N-[(1S)-1-[[[(2S)-2-Amino-3-methyl-butanoyl]amino]carbamoyl]-2-methyl-propyl] Carbamate (14). N.0. DMSO-d6): δ 170. 1694.15 (m.61 (s. 81. TR = 19. J = 8. 7. N. tert-Butyl carbazate (2 g. 2.03 (d.38 (m. DMSO-d6): δ 9.7. APCI + MS m/z 782 [M + Na]+. 7. 6H).4.7.1. 4. 15.3. 128. The mixture was then filtered through a pad of Celite. 7. 19. H. 9H). 4H).08 (br. The organic phase was dried over Na2SO4.9. mp 177−187 °C. N.7.43 (m. 3. 65.05 (m.7.35. J = 8. 33.90−0. 1. 2H).6.70 (d.8. 28. 71%).. 56.2. 1 equiv) was dissolved in 4 M HCl in dioxane (10 mL) at 0 °C and stirred at room temperature until completely dissolved.71 mL. 0. 28.19 (s.47 (s. C23H36N4O6 (%) C. 1162. 2. 1.29 mmol.12 (s. The solvent was evaporated off. 12. 2H). 13C NMR (75 MHz.7. 115. 2H).1.4. 30. 10% aqueous K2CO3 (5 mL). 698 cm−1. 170.1 Hz.18 (s. IR (neat): νmax 3291. 2H).72 (m. Same procedure as for 3 from 34 and 31. J = 8. 32. 7. 0.99 (d.32 (t. 24. 129. 1. 8H).75 (m. 13C NMR (100 MHz. 7. 1H). 16.4 Hz.0. 4H). 7.5 g. 77.1 g. 9H).9. 19. mixture A/B from 70/30 to 0/100.59 (s. 23. Same procedure as for 3 from 34 and 23. 13C NMR (100 MHz. 2H).97 (d. DMSO-d6): δ 174. Product 6 was obtained as a white powder (98 mg. 106. 107.9. TR = 10. 7. CD3OD): δ 175. 1H). HPLC purity: method 1. 6. Anal. 19.0. 4. 2H). J = 8. 19. water (100 mL).5. 6.30 (s. 168. 0.65 g. 1 H NMR (300 MHz.3. H.24 (bs. 1160 cm−1. 128. 158. 8.14 (bs. and the cake was washed with methanol (10 mL). 18.0. 11 (150 mg. 1H NMR (300 MHz. 106. 35.7.7. 35. 9H).1. 0. DMF was removed under reduced pressure.68 g. 1H). 31. 168.7 Hz. IR (neat): νmax 3205.43−2. mixture A/B from 50/50 to 20/80 in 20 min.47 (m. 5. 5.9. 19. 1H).43 (s. 8. method 2. (C30H42N10O12·2H2O) C.42 (m. and the residue was dissolved in DMF (6 mL). 98%. HOBt (1. 1. H. The reaction flask was flushed three times with hydrogen and stirred. 4. 2.2. 1H). 2H).9.19 (m.82 (s. (C30H42N10O10·3H2O) C. 24. 106. 6H). 4. 4H). 1604. 2H). 1H). 7. 1677. 82% over 2 steps). 128. 2H). 742. 6H). 2. 10% mass) was added. 6.64 mmol.7.82 (bs.9. 13C NMR (75 MHz.46 g. 2H).08−4. Product 7 was obtained as a brown solid (140 mg. 3H). Med. 1.91 (s. 24. 88%). mp 206−210 °C. 6. (C34H50N10O10·0.13 mmol) and pyridine (2. 2H). ESI+ MS m/z 254 [M + Na]+.01 mmol. 5H).6. 157. J = 8. 1. 1687. 3.97 (dd.1.1.8.69 (s. 6. 100%. 97%.4 Hz. HPLC purity: method 1.2. mixture A/B from 70/30 to 0/100 in 20 min.3. 138. 129. and EDCI (0.4. 156. mixture A/B from 50/50 to 0/100 in 30 min. Same procedure as for 3 from 34 and 29. The filtrate was evaporated under reduced pressure to give 12 as a white solid (91 mg. 2H).2. 123. Product 9 was obtained as a light-yellow solid (120 mg. 4H). TR = 13 min.22−3.9.8. 2.5 min. H. N.3.8. 2947. 30.9. Synthesis of Molecular Tongs (8).0. 2H). 3H). 128. DMSO-d6): δ 9.5. 60%). DMSO-d6): δ 9. Synthesis of Molecular Tongs (5). mp 172−175 °C. 1248. 4H). Product 8 was obtained as a light-brown solid (53 mg.6. HPLC purity: method 1.8. and brine (5 mL). 1. 33.1. 2H). 96%.04 (s.92 (m. 100%. 13C NMR (75 MHz.93 (m. IR (neat): νmax 3304.2 min. 1209 cm−1.

35.9. and the mixture was stirred until a homogeneous solution was obtained. 1.3.34 (s. 1H). 1H).1. 2936.1021/jm300181j | J.38 (s. 1H). 3.9.1 equiv) were added successively. 5. The solvent was removed.3. J = 6. and 10% aqueous K2CO3 (2 × 20 mL).5 equiv) was added to a suspension of compound 16 (625 mg. under an argon atmosphere. 1.39 (s. 3H). 121.02 (m. 1H). 8. 1H). The corresponding TFA salt of 21 was used in the next step without further purification. 158.89−0.1. IR (neat): νmax 3270.0. and the mixture was stirred for 15 min at room temperature.1752. 1. 1474. IR (neat): νmax 3272. 158. 1H). Anal. 2H). 1723. 1.1. The residue was purified by flash chromatography on silica gel (EtOAc/cyclohexane.80 (s. 9H). 12%). 168. 1H).1 Hz. 1H). 2936. HRMS calcd for C13H25N5O5 + Na (354. 2.74 (m.05 (d. Compound 24 (88 mg. 78.68 (s. 0. 91%). 2.77−6.00 (dd. 6. 1H). 9H).1 Hz. DMSO-d6): δ 8. 6.3.5 g. 125.46 (s. IR (neat): νmax 3223. 1H NMR (300 MHz.20 (br. 150. 3H). Methyl 3-[[(tert-Butoxycarbonylamino)carbamoylamino]carbamoylamino]propanoate (20). 129.07 mmol. 5H). 13C NMR (75 MHz. 15. 28. 0.0. 2H). 10/1) to give product 26 as a white solid (150 mg. H.4. mp 232−234 °C. 1H). 1. DMSO-d6): δ 172. 155. ESI+ MS m/z 355 [M + 23]+.2. 1. the solvent was removed under reduced pressure. 8.49 (s. The product 20 was obtained as a white solid (274 mg. 7. 9H).05 (bs. HBTU (440 mg. 80/20) to give product 17 as a white solid (923 mg. 35. 1758. The precipitate obtained was washed with petroleum ether to yield compound 24 as a white solid (3.95−2. 1H). 1H). 13C NMR (75 MHz.54 (bs. 20. 1-Acetamido-3-[[(2S)-2-amino-3-methyl-butanoyl]amino]urea (27).4. 51. (C8H16N4O4·0. 9H). DMSO-d6): δ 9. DMSO-d6): δ 8. 1161 cm−1.28−2. 7.4. 9H).6. mp 173−174 °C. 13C NMR (75 MHz. 1158 cm−1. N. 157. 1651. 1241. H. 3H). and the HCl salt of 25 was obtained as a white solid (64 mg. N. and methanol was added and immediately evaporated off. 1H). N. The resulting oily residue was triturated in petroleum ether.4. IR (neat): νmax 3259. 28.1 equiv) was added. 28.0.1. J = 8. and DIPEA (1. and then ether was added and again immediately evaporated off to give the corresponding TFA salt of 19.32 (br.6 Hz.8.05 (bs. 3. 125.1. 1.0. Pyridine (678 μL. 3H). and the resulting residue was purified by flash chromatography on silica gel (EtOAc/CH3OH.13 (s. 2.56.03 (s.5. 3H). 1H). Anal. 1. 1. 167. 1H). Anal. 1159 cm−1. 9H).3.2. DMSO-d6): δ 173. N. 1229. and triethylamine (741 μL.97 mmol) and N-BocVal-OH were dissolved in DMF (15 mL) with stirring at 0 °C. 8. 78.74 (m. H. 2. tert-Butyl N-[3-[2-(Acetamidocarbamoyl)hydrazino]-3-oxopropyl]carbamate (28). DMSO-d6): δ 9. (C11H21N5O5) C. 20. 1489. 0. 82.6.9. ESI+ MS m/z 354 [M + Na]+. Anal. 1507. 2974. 1154 cm−1. H.1.07 (s. The aqueous phase was then concentrated and extracted twice with EtOAc because compound 26 is extremely soluble in water.21 (br. 1H NMR (300 MHz. 155. 1665. 1158 cm−1. 1. DMSO-d6): δ 171. 158. 46%).04 (bs. 6.7.41 (s.6. 1H). 9.97 mmol.2. The mixture was stirred. 1H NMR (300 MHz. 26.7. 7.8 mmol. 1.9. ESI− MS m/ z 302. mp 110−112 °C. 3H). 8. ESI+ MS m/z 255 [M + Na]+. 13C NMR (75 MHz. 10 equiv). DMSO-d6): δ 8.81 (bs. 18. 150. 78. 1H). 30. 1H).61 (bs. N.40 (s. 1. 13C NMR (75 MHz.13 M).15H2O) C.07 (bs.09 (m.1. 1668. 25 (163 mg. 0.37 (s. N. tert-Butyl N-[[[(1S)-1-Carbamoyl-2-methyl-propyl]carbamoylamino]carbamoylamino] carbamate (18).7. The organic phase was washed with water (2 × 50 mL) and concentrated under reduced pressure. 4. cooled to room temperature. (C12H16N2O4) C. 2012. 1H). Same procedure as for 19 from 20. mp 134−136 °C.5 Hz. DMSO-d6): δ 159.99 (bs. J = 8. 9H).0. 1H). The HCl salt of 29 was used in the next step without further purification. 1H NMR (300 MHz.41 (bs. 1-Acetamido-3-(3-aminopropanoylamino)urea (29).09 (s. for 3 days at room temperature. tert-Butyl N-(Hydrazinecarbonylamino)carbamate (16). 160.7 Hz. Same procedure as for 25 from 26.2.45 mL. 155.4.68 (bs.8. 0. 173. 13C NMR (75 MHz. 1H). 9H). 6. 82.2 equiv). an equal volume of TFA was added. phenyl chloroformate (458 μL. 25.8. J = 5. 13C NMR (75 MHz.77 (s.5 g. 96%).15H2O) C. 1519. 1H). 1. 1.82 mmol.5 h.5. 1511.73 (d. 1177. 1-Amino-3-(methylcarbamoylamino)urea (23). 7. 58. 156.7. 1H NMR (300 MHz. 3106. 13C NMR (75 MHz. 16 (3 g. 5. 121. 90/10) to give product 18 as a white solid (272 mg. Anal. Anal. 158. 1682. The solution was then cooled to 0 °C. 6.5. Same procedure as for 25 from 28.2. 7.43 (s. 72%). The product 22 was obtained as a white solid (84 mg. 3. 1730. N. 1.87 (m.8. IR (neat): νmax 3301.Journal of Medicinal Chemistry 1H).54 (s. DMSOd6): δ 8. 3H).5 h. H.61 (s.79−3. The reaction mixture was stirred under argon at room temperature for 24 h.16 mmol.80 (s. The HCl salt of 27 was used in the next step without further purification. 13C NMR (100 MHz. 6762−6775 .14 mL. The solvent was evaporated off under reduced pressure. 28. ESI+ MS m/z 275 [M + Na]+. 1. 1-Acetamido-3-amino-urea (25). 70%). 8.0. IR (neat): νmax 3252. The oily crude product was purified by flash chromatography on silica gel (EtOAc 100% to EtOAc/CH3OH. 1H).83 (m.6. 6H). 8.25 (dt.7. H. 1663.7.doi. 1226. 7.65 (br. 18 was dissolved in CH2Cl2 (0. 1658. 155.2. (2S)-2-[(Hydrazinecarbonylamino)carbamoylamino]-3-methylbutanamide (19). (C12H24N6O5·0. (C6H14N4O3) C.6 mmol. mp 118−120 °C.37 (bs. 3. 17.54 g. 1H). tert-Butyl N-(Acetamidocarbamoylamino)carbamate (24). 2H). 1. 79. 1213. and the residue was dissolved in EtOAc (200 mL). 8. 2H). IR (neat): νmax 3262.17− 3. J = 4. 1H).2.49 (s. 1164.1.1 equiv) were dissolved in acetonitrile (17 mL). 1712. and acetic anhydride (7. DMSO-d6): δ 9.2EtOAc) C. 155. 28. 1H). 79.77 mmol) and hydrazine monohydrate (8. 0.4 Hz. 28. 8.54 (s. 80. and the resulting residue was triturated in CH2Cl2. Compound 17 (273 mg.05.87 (d. N. 9H). 20.4. dried over Na2SO4. 1H).1. 1H NMR (300 MHz.09−7.3. ESI+ MS m/z 333 [M + Na]+.6. 1H). 1161 cm−1. 7. (C11H21N5O6··0. 1H). (C13H18N4O5) C. H.50 (s. 7.1. The solvent was removed under reduced pressure.0 mL. IR (neat): νmax = 3245. 1H). 2969. 2H).1. dx. 8.9. The DMF was removed under reduced pressure.1. 1692.org/10. 1238.3. 3. 9. 28. Chem. 0. 1. found 354. (C8H17N5O4) C. 1173 cm−1. 7. 37. and concentrated under reduced pressure. 93%). 129. 28.3. mp 122−124 °C. ESI+ MS m/z 213 [M + Na]+.82 (d. mp 154−156 °C.2. 6. 1H NMR (300 MHz.08 (s. DMSO-d6): δ 155.95−1.61 (s. 2H). DMSO-d6): δ 174. 158. tert-Butyl N-[(1S)-1-[(Acetamidocarbamoylamino)carbamoyl]-2methyl-propyl]carbamate (26).2 mmol.40 (s. 3238. 1H NMR (400 MHz. The mixture was heated under reflux for 1. DMSO-d6): δ 159.7.56 (d.6 Hz. The product 28 was obtained as a light-brown solid (95 mg. 79. and the solvent removed under reduced pressure. 3106. 1529. 155. CDCl3): δ 158.0. 2H). 57. 100%) and used in the next step without further purification. 1666. filtered.3 mmol.80 (s. 30. 1244. 157. ESI+ MS m/z 342 [M + 23]+. 1542. 1H). and the residue was dissolved in EtOAc (50 mL). 2H). 2H). The solvent was removed.31 (m.3 mmol) in THF (50 mL). DMSO-d6): δ 168.45 (t.7 mmol) was dissolved in dry THF (180 mL) with stirring.4. This was used in the next step without further purification.88 mmol) and HCl·H-Val-NH2 (151 mg.2. Compound 15 (6. 78. 4. J = 6. and the mixture was stirred at room temperature for 3 h. 55. IR (neat): νmax 3261.37 (s.6.8. 6. J = 8. 19. Same procedure as for 19 from 22. 6772 Article Methyl 3-[(Hydrazinecarbonylamino)carbamoylamino]propanoate (21). 1. The organic layer was washed sucessively with water (2 × 25 mL). Anal. Same procedure as for 18 from 17 and HCl·H-β-Ala-OCH3. 19. ESI+ MS m/z 270 [M + 23]+. 2969. The TFA salt of 23 was used in the next step without further purification.80 (s. H. 1. 155. 1519.51 (s. Anal.7. 5 equiv) was added dropwise. 1025 cm−1. 3. Same procedure as for 18 from 17 and HCl·NH2CH3.5 mmol. 1659. Same procedure as for 26 from 25 and Boc-β-Ala-OH. 5. J = 11. 67%). Med.38 mmol) was dissolved in 4 M HCl in dioxane (4 mL) and stirred at room temperature for 2 h. 28.7 Hz. 1521. 8. 5% aqueous citric acid (2 × 20 mL).22 (m. 1H). tert-Butyl N-[(Methylcarbamoylamino)carbamoylamino]carbamate ( 22 ).5 equiv) in methanol (100 mL) were heated under reflux for 1. Phenyl N-[(tert-Butoxycarbonylamino)carbamoylamino]carbamate (17). 5. 6 equiv) was added to the mixture. and HOBt (145 mg.1753). and the precipitate obtained was recrystallized from EtOAc/petroleum ether to give 16 as a white solid (3.6. mp 145−147 °C. 2H). CDCl3): δ 7. 34.

31.24 (m. 1H). ESI+ MS m/z 361 [M + H]+.2. 170.9. EDANS. 3H). ANS emission spectra were measured using an excitation wavelength of 370 nm and an emission wavelength of 470 nm with bandwidths of 10 and 5 nm. 1. 3H).5 Hz.43 (m.24 (t. J = 9. 150 μL). Other reagents and solvents were purchased from commercial sources. The proteolytic activity of pepsin from porcine gastric mucosa was determined fluorometrically using the fluorogenic substrate Abz-ThrIle-Nle-(p-NO2-Phe)-Gln-Arg-NH2 (Km = 13.7. 8. 13H).0 Hz.37−7. acetylpeptsatin.93 (m.45 (m. 4. aqueous HCl 0. 2H). The proteolytic activities of PR and mutated proteases were determined fluorometrically using the fluorogenic substrate DABCYL-γ-abu-SQNYPIVQ-EDANS (λex = 6773 Article 340 nm.49 mmol. λem = 500 nm). (C45H70N6O11·3H2O) C. 1.83−1. 2H).4.6. 1H NMR (300 MHz. final concentrations: 10−60 μM) was then added. 2. (C22H34N6O7·1H2O) C.36 dx.3 Hz.40−1.5. 2H). 1740. 1210. Wild-Type and Mutated Proteases. The fluorogenic pepsin substrate Arg-Glu(EDANS)-Ile-His-Pro-Phe-HisLeu-Val-Ile-His-Thr-Lys(DABCYL)-Arg. The stability of inhibitors in RPMI (Roswell Park Memorial Institute medium) culture medium containing 20% fetal calf serum was assessed by studying their kinetics of breakdown at 37 °C for up to 2 days. in DMF (55 mL) and DIPEA (0. 155. Med. 1H). H.60 (s.7. 24. 13 C NMR (75 MHz.2 Hz. 1H NMR (300 MHz. 6762−6775 . 8.12−4.5 equiv). 22. 115. The experimental data were fitted to the appropriate equations according to ref 14.80 (s. 2H).0. 1.73 (s. 2H). 4-(4′-dimethylaminophenylazo)benzoyl.6. 1. 1H).00 (s. 1H NMR (300 MHz.47 (m. pH 8.7. the final DMSO concentration was 0. 4. 171. 1 mM EDTA.8. The substrate and the compound(s) were first dissolved in DMSO.794 mmol. pH 4. 66. 76%).46 (m. 173. Enzymatic Studies. 23. λem = 410 nm). respectively (Perkin-Elmer LS 50B spectrofluorometer) as described in ref 21.70 (s. respectively. 4H).66 (m.44 software33 and then energy minimized (by 2000 steps) using the Powell method as implemented in SYBYL 8. All experiments were performed at least in triplicate. mp 150−160 °C. 159. The mixture was poured at 4 °C and centrifuged (10000 rpm for 10 min). 1.69 (t.35 Solvent-accessible surface areas and clogP were calculated for these models with QikProp 3. 129. 78. 2012. 7H).3. PR (350 nM) was dissolved in the assay buffer (100 mM sodium acetate.80−0. The final DMSO concentration was kept at 3% (v/v). The proteolytic activity of renin was determined fluorometrically using the fluorogenic substrate ArgGlu(EDANS)-Ile-His-Pro-Phe-His-Leu-Val-Ile-His-Thr-Lys(DABCYL)-Arg (Km = 3.1. 1535.97 (d. H.3 Hz. 106. Same procedure as for 26 from 25 and Nα-Boc-Nε-Z-Lys-OH.org/10. 81. 4. 1H). Models of Molecular Tongs−PR Complexes. 55. 1H). compounds 1. and 1 M NaCl at pH 4. Hydrophobic ANS preferentially binds to the hydrophobic surfaces of proteins. In all cases.32−2. 31.7.8. 1246. 1.04 (s. The multimutated HIV-1 proteases ANAM-11 and MDR-HM contain 11 mutations (L10I/M36I/S37D/M46I/R57K/L63P/A71 V/G73S/ I84 V/L90M/I93L) (ref 31) and six mutations (L10I/M46I/I54 V/ V82A/I84 V/L90M) (ref 30). saquinavir. 19.6.9.5. 2H). 138. 1H). 123. mp 220−222 °C. 1. 100 mM NaCl. and compounds 1.1. J = 6. 2.doi. 32 (982 mg. 3H). 1. 2365. 1H). Chem. 51. 32. 3. tert-Butyl N-[(1S)-1-[(Acetamidocarbamoylamino)carbamoyl]-5amino-pentyl]carbamate (31). Q7K.0. 100%). was purchased from Bachem (Germany). 4. 1.3.1.1.9.56− 1. J = 6. Anal. Each fluorescence measurement was made five times and averaged. 30. 1 M NaCl. A distance-dependent (ε = 4r) dielectric function was used.6. 67. 128. 1522. 18.61−1.7) and incubated at 25 °C with or without inhibitor (5 μM.0 software.3. 1663. 1. 7. brine. 1665. N.1021/jm300181j | J. 6.9. The mixture was stirred at room temperature for 5 days. CD3OD): δ 174. 55. 5-[(2-aminoethyl)amino]naphthalene1-sulfonic acid)]. and ANAM-1122) used in this study were expressed and purified as described before.7 and 30 °C (final volume. Same procedure as for 12 from 30.1. and a substrate concentration of 20 μM (λex= 337 nm. with stirring. 7. 7.2 equiv) were added successively. Fluorescent Probe Binding. 1167 cm−1. and the DMF was removed under reduced pressure.5. Incubations were terminated by adding ethanol.70 (d.6. 50. Threedimensional coordinates for free compounds 1−10 and P1 generated with CORINA software were energy-minimized with OPLS_2005 force field as implemented in MacroModel 9. Assays were performed in 50 mM Tris−HCl. 7. 1162 cm−1. Synthesis of Molecule (33). Molecular Modeling. Analysis of the complex models involved the study of the hydrogen bond network and counting of favorable interactions. Inhibition of Renin and Pepsin. The HIV-1 proteases (PR. 156. A solid was precipitated from the oily crude with CH2Cl2. 1. 28.0. The product 30 was obtained as a white solid (88 mg. 2H).39 mL. the renin substrate Abz-ThrIle-Nle-(p-NO2-Phe)-Gln-Arg-NH2. 32. ANS (2 μL. 2. 80. 24. and a substrate concentration of 5 μM (λex= 340 nm.5. The half-life of breakdown was obtained by least-squares linear regression analysis of the semilogarithmic plot of remaining inhibitor concentration versus incubation time. 24 H).49 (br.32 (br. CD3OD): δ 4. 20. 1H).9.1. IR (neat): νmax 3289.05 (m. acetylpepstatin. 171. 2977. and P1).0. 13C NMR (75 MHz. 1508.3 μM).9.6. N.9. 173.8. 2H).6.92−3.9. 1-Anilino8-naphthalene sulfonate (ANS) was from Sigma-Aldrich. 2956. MDR-HM. 159. 32. Models of molecules 1−10 were built using CORINA 3. 55. DMSO-d6): δ 175.22 (m. ESI+ MS m/z 894 [M + Na]+. 2. ESI− MS m/z 493 [M − H]−. and saquinavir and amprenavir were from the NIH (USA). The protease domain (PR) has five protective mutations. 8. 41.0 Hz.6.8.05 kcal·mol−1·Å−1. 1. 3. mp 129−131 °C. 21.24 mmol. 5H). 128. 6. 1 mM EDTA.94−8. L63I to minimize the autoproteolysis of the protease and C67A and C95A to prevent cysteine−thiol oxidation. 24. DMSO-d6): δ 9. The product 31 was obtained as a white solid (164 mg.6. DMSO-d6): δ 172. 160. The intrinsic fluorescence of PR. J = 7. 1 mM EDTA. 5. Fluorescence intensities were measured in a BMG Fluostar microplate reader. 40. 170.7 Hz.99−2. Enzyme and Inhibition Assays. The minimizations were carried out by 1000 steps of steepest descent followed by a conjugate gradient minimization until the rms gradient of the potential was lower than 0.3 equiv) and HBTU (680 mg.32 (m.0.89−2. This was washed successively with CH2Cl2. 13H).6.9. 0. 158. 158. 1H). 41. Evaluation of Metabolic Stability. 135.2 μM) with at least six enzyme concentrations (5. 28. 383 [M + Na]+.53− 1.9.6. 1H). Aliquots of the clear supernatant were injected onto the HPLC column (Waters E600). and recombinant renin and pepsin were purchased from Sigma−Aldrich.28 (bs.0.41 (m.17 (s.8.41−1. aqueous NaHCO3.6. 29.00 (m. J = 9.61 (m.90 (m. 3H). 129.11 Kinetic experiments were carried out at a constant substrate concentration (5.9.65 (br. saquinavir.8. 1635. 27. 1H). J = 5. 9 H). 7.4. 12 (449 mg.0.7.4.01 (t.1.3 μM). 24. 2H). 1H).0. 1368. 1 equiv) was dissolved. The mechanism of inhibition and the corresponding kinetic constants Kid (dimerization inhibition) or Kic (competitive inhibition) were determined using Zhang−Poorman kinetic analysis. 57.35−4. 57. and P1 was negligible under the experimental conditions used. pH 3. IR (neat): νmax 3307.98 (s. 18. 1. 13C NMR (75 MHz. λem = 490 nm) in 100 mM sodium acetate. Anal. 4H). 55. 2. 1.94 mmol. The geometries of the resulting complexes were optimized by energy minimization using the AMBER7 FF02 force field as implemented in SYBYL to relax the structure and to remove steric bumps. γ-amino butyric acid. 30.5.38 (s. Absorbance measurements were made with a Perkin-Elmer LS 50B spectrophotometer.8.1 M.5. IR (neat): νmax 3279. 18. 3H).6 nM) and inhibitor concentrations from 0.9. 1. 8.2. 1H).87 (d.14 They were produced in Escherichia coli using the expression vector pET-9 and the host bacterium strain Rosetta (DE3)pLysS.34 Dihedral angles were then adjusted so that each molecule was superimposed over the backbone of chain B (subsequently removed) from the crystallographic structure of HIV protease 7HVP (PDB accession code).5. 30. 4. Product 33 was obtained as a white solid (986 mg. 1. L33I.Journal of Medicinal Chemistry tert-Butyl N-[(1S)-1-[(Acetamidocarbamoylamino)carbamoyl]-5(phenoxycarbonylamino) pentyl]carbamate (30). Estimation of Accessible Surface Areas and clogP. 2.0. γ-abu.5 to 28 μM. 3. and CH2Cl2. 3. 20.2% (v/v). DMSO-d6): δ 9.85 (m. 30%). 18. J = 6.33−18. 1162 cm −1.2.6. The fluorogenic substrate for PR and mutated proteases was DABCYL-γ-abu-Ser-Gln-Asn-Tyr-Pro-Ile-Val-GlnEDANS [DABCYL.05 (t. 9. Assays were performed in 10 mM NaHCOO.1.

F. A. C. The English text was edited by Dr Owen Parkes. Chem. R. Semo. 2003.R. A.. Yarchoan. Sicsic.. Y.. Pannecouque. PR inhibitors... Schramm. N.N.. 33(0)-146835743.V. de Rosny. 4686−4690. Van Baelinghem. Matsumi. 1998. J. Comparative analysis of the sequences and structures of HIV-1 and HIV-2 proteases.. Biol. Bishop. Design. Immunol. M. 957−962. Sicsic. D. Acad. 1321−1324. Baldridge.org. 6762−6775 . R. Schramm. Luftig.S. Reboud-Ravaux. Dimerization inhibitors of HIV-1 protease.S.: phone. (19) Koh. W. Jamart-Gregoire... T. Biochem. Ghirlando. Purity: HPLC chromatograms of molecular tongs 1−9 and analytical data of compounds.. (12) Franciskovich. C. 2004. W. (18) Ulysse. 28709−28720. HIV-1 reproduction is inhibited by peptides derived from the N.. A systematic evaluation of the inhibition of HIV-1 protease by its Cterminal and N-terminal peptides. 4841−4845.. HIV Med. J... 5196−5207. Chiemlevski. 475−488. MDR-HM. Chem. L.. Soc. J.. 2002. M. A. Loveday. Mitsuya. J. Biol. Biol. MD. M. Reboud-Ravaux. (2) Kohl. S. Chem. J. Res. J. E. H. R. A. Reboud-Ravaux. tert-butyloxycarbonyl. Am. Soc. γ-amino butyric acid. PIs.. 1991.doi. E.. 1999. AUTHOR INFORMATION Corresponding Author *For S.. A. J. Schramm. 1998. EDTA. Z. 379−387. M. A. Ongeri. 1988. A. P. Shafer. A. Demeter.: phone. (3) Carr. 3. Pethe. E. and N. F. 283. synthesis. Baltimore. G. J... interfacial peptide inhibitors of HIV-1 protease. γ-abu. (7) Gustchina. M. Conway.1021/jm300181j | J.. Weber. 282. O-benzotriazoleN. Proc.. Boggetto. M. 30−36.. (6) Todd.. Sci. L. Curr. root-mean-square ■ (1) Orozlan. fax.. Richman. L.N′-tetramethyl-uronium-hexafluoro-phosphate. (20) Bouras.. HBTU. Chem. Pharmacol.. J. Biol.. Toxicity of antiretroviral therapy and implications for drug development..K. M. J. E. S. R. S. Nature Rev. 65. ACKNOWLEDGMENTS We thank Claire Troufflard (UMR 8076. N. Boggetto. E-mail. New constrained “molecular tongs” designed to dissociate HIV-1 protease dimer. R. Y. Badet. Lett. Ongeri. NIH.. J. (14) Dumond.. 6774 ■ REFERENCES E. Microbiol. A. Merabet.. 380.. multimutated HIV-1 protease containing six mutations. 35. J. R. R. E. Ressad. Schwetizer. 276. 6392−6400. F. 2007. Heinrikson. 765−768. N.. (13) Schramm. 33(0)-144275140. 422.. 1997. Chem. for the generous gift of the plasmid encoding the ANAM-11 mutant. Faculty of Pharmacy at Chat ̂ enay-Malabry) for the NMR experiments.. Schramm. Folded monomer of HIV-1 protease. Reboud-Ravaux. R. (16) Breccia. 2004. Med. J. Dumond. Kinetic characterization of the inhibition of HIV-1 protease by its COOHterminal tetrapeptide.. Takahashi. (b) De Clercq. Collinet. EDANS. T. Kessler. H. S.. P. L.. M.. The structural stability of the HIV-1 protease. Biol. Reboud-Ravaux. K. Christine Philippe for their help with the syntheses.. R... 49110−49116. Molecular tongs containing amino acid mimetic fragments: new inhibitors of dx. Leschenko. 33(0)144275078. Chem. 21−25. 42. J. Biochemistry 2009. H. Collinet.. Thyroxine-derivatives of lipopeptides: bifunctional dimerization inhibitors of immunodeficiency virus-1 protease.. de Mendoza. S. 704−720. Z. Prados. Dufau. M. S. N. N. Chem.. Rene. T.. Ernesto Freire of the Department of Biology and Biocalorimetry Center. Yamamoto.. J. Commun.. L.. Chem.. J. P. M..A.O. DABCYL. Sigal. (17) Zutshi. Dimer disruption and monomer sequestration by alkyl tripeptides are successful strategies for inhibiting wild-type and multidrug-resistant mutated HIV-1 proteases. Email. C. L.. We thank Prof. 2002. Natl. Kuritzkes. U. Drug Discovery 2003. Chem. Louis.. Lett.. Dissociative inhibition of dimeric enzymes. Trans. J. Boggetto.F.. Boggetto. Benatalah. V. Dumond. 1990. 55. Torchia. A. Sicsic.. Wilson. J. Reboud-Ravaux. 10. Johns Hopkins University. N. 847−851.. 624−634. J. 15591−15594. M. 119. I. Das. 1anilino-8-naphthalene sulfonate. W.. 47. An alternative strategy for inhibiting multidrugresistant mutants of the dimeric HIV-1 protease by targeting the subunit interface.reboud@ upmc. K..fr. 8.and C-termini of HIV-1 protease. Reboud-Ravaux. Van Gool. and the Italian Research Ministery provided finances for R. A. H. Poorman. J. fax. ABBREVIATIONS USED PR... S. Biochem. multimutated HIV-1 protease containing 11 mutations. Scolnick. M.D.. 10. ANS. Reboud-Ravaux. Targeting the dimerization interface of HIV-1 protease: inhibition with cross-linked interfacial peptides. Davis. ANAM-11. sandrine. 3281−3286. N. (4) (a) D’Aquila. S. J. 4-(4 ′ dimethylaminophenylazo)benzoyl. 1-ethyl-3-(3dimethylaminopropyl) carbodiimide. 383. B. J. Chem.. Amano.. Kezdy. Antivirals and antiviral strategies. Heimbach. Slultz. 2012. B. Top. Med. EDCI. I. L. 325− 329. Maggiora. Retroviral proteinases. Microbiol. W. Perez-Fernandez. Drug Resistance Mutations in HIV-1... Lipopeptides as dimerization inhibitors of HIV-1 protease. T.. 2001.. M.. HOBt. Gronenborn. For M. (15) Bannwarth. and evaluation of conformationally constrained tongs. R.. de Rosny.org/10. S. 1999... M. 1993. D. M. 157.N′... 33(0)146835740. M. L.. L. Brun-Vezinet. R. (Marie Curie Early Stage training Fellowship of the European Community’s Sixth Framework Programme: contract MESTCT-2004-515968) was provided by the European Community. M. 1097−1102. 2003.. E. D. NIAID...... (22) Bannwarth. De Clercq. Chiemlevski. Rose. Tozser. H. Saquinavir was obtained through the NIH AIDS Research and Reference Reagent Program. Davis. B. S... (b) Banwarth. This material is available free of charge via the Internet at http://pubs. 179. L. Med. 46. (10) Schramm. Med.. (21) Merabet. Mueller. Houseman. J. R. 2. Emini.. Top. L. Chmielewski. 1991. D. and Lixian Qin for carrying some ANS experiments. Ghosh. Johnson. L.. A. 2. M. Nature Rev. HIV-1 protease.Journal of Medicinal Chemistry ■ ■ ■ Article ASSOCIATED CONTENT S Supporting Information * Putative complexes between the PR monomer and peptidomimetic molecular tongs 1−9. M. J. Dixon. J.. M. Biophys. Med. Shioda. R. M.. Dick. (5) Ishima. Li. A. Dimerization inhibitors of HIV-1 protease based on a bicyclic guanidinium subunit..acs.. rms.. B. Freire. A. B. B. Schleif. Active human immunodeficiency virus protease is required for viral infectivity. Boc. N-hydroxybenzotriazole. R.. Franciskovich. Division of AIDS. Proteins 1991. (8) (a) Boggetto. Biol. Bioorg. Grant. (9) Hamacher... Boggetto.. N. E. new inhibitors of HIV-1 protease dimerization.. 266. Restricting the flexibility of crosslinked. B. Potent inhibition of HIV-1 replication by novel nonpeptidyl small molecule inhibitors of protease dimerization. R. Schapiro. 153−185.. A. 551−554. Bioorg.. Takahashi. Med.. J. 593−596. J. Notes The authors declare no competing financial interest.M. M.. michele. D. Financial support for A. W. 5-[(2-aminoethyl)amino]naphthalene-1-sulfonic acid)}. Mol. N. J. D. Reboud-Ravaux. K Relating sequence evolution of HIV1-protease to its underlying molecular mechanics. Biochem. Gene 2008. 85.R-R. Chem. Nakashima. 2007. Chem. Clotet. Vanderesse. Buttner. ethylenediaminetetraacetic acid. B..ongeri@u-psud... (11) Zhang. H. N. 48...fr... The Minister ̀ e de la Recherche et des Technologies (MRT) provided financial support for L. N. Wakayama. Metabolic stability: HPLC chromatograms. J. A.

Inf. 1049− 1057. U. J. 1996... A. (b) Witt. Bannwarth. X. E. 2006.-T. A. Comparison of Automatic Three-Dimensional Model Builders Using 639 X-Ray Structures. J. J... Hydrogen bonded oligohydrazide foldamers and their recognition for saccharides.. T.. 2012. S. J.. J. Zhou. (34) SYBYL 8. Klebe. Li. H.. LLC: New York. Parker. Weije.. Azatides: solution and liquid phase synthesis of a new peptidomimetic. Kopecek. Freire. X. Z.. T. S.. Inhibition of cathepsin K with lysosomotropic macromolecular inhibitors. and supramolecular substitution.. D. J. J. (32) Stryer. 2011. The Interaction of a Naphthalene Dye with Apomyoglobin and Apohemoglobin. Pechar.. G.. A Fluorescent Probe of Nonpolar Binding Sites. J. J. L.. (31) Muzammil. Soc. Tripos International: 1699 South Hanley Rd. X.. (23) Vidu. H. Jiang. 118. 49. Chem. Q.. 2011. B. Janda. Med.-Q.1021/jm300181j | J. 739−743. Chem. Ireson. Chen. C. Self-assembly of hydrazide-based heterodimers driven by hydrogen bonding and donor−acceptor interaction. J. 1996.9. Davis.Journal of Medicinal Chemistry wild-type and mutated HIV-1 protease dimerization. assembling selectivity. 279.. (b) Wang. K. Schön. (30) Ohtaka. 13. Hruby. Wang. 2004. Ther. J. Hydrazide-based quadruply hydrogen-bonded heterodimers. 34. 2006. L. USA. M. J. 1200−1208. S. Chen. Structure. A. Biochemistry 2003. Chen. 2003. P. J. 8849−8859.. Comput. Missouri. Misicka. 1899−1906. D. Article 6775 dx.. P. Preparation and properties of some α-aza-amino-acid derivatives. Y. Soulier.0. 33. K. K. Germany. Chem. 13659−13666. 12386− 12394. 2006... Biol. 42. Louis. Li. Chem. Y. Tetrahedron 1977.doi. T.. Chin. E. A.. 63144. Toward the First Nonpeptide Molecular Tong Inhibitor of Wild-Type and Mutated HIV-1 Protease Dimerization. LLC: New York. 2539−2544. (28) Feng.... Chem. L. Lipkowski..4.. ChemMedChem. L. 4657−4664. J. CNS drug delivery: opioid peptides and the blood−brain barrier. V. Wang. G. (b) The 3D structure generator CORINA is available from Molecular Networks GmbH. X. Am. 15128−15139. D. 1965. X. Soc. Multidrug Resistance to HIV-1 Protease Inhibition Requires Cooperative Coupling between Distal Mutations. a unique double-enkephalin analog. their possible use in peptide synthesis. 125. Kopeckova. S. J.. (24) (a) Han. 1994.-J. A. and its halogenated derivatives. Li. AAPS J. Exp. 631−638. X. Z. Li. Chem. Pharmacol. 55. Dufau. 42. (35) MacroModel. Brömme.. Mol. (29) Gray. J. Reboud-Ravaux. Z. P. version 3. (36) QikProp.. 6762−6775 .. 126. C.org/10. A Major Role for a Set of Non-Active Site Mutations in the Development of HIV-1 Protease Drug Resistance.. Jiang. (33) (a) Sadowski. Blood-to-central nervous system entry and stability of biphalin. Am. R... St. version 9. Ross. Freire. D. T.. Williams. Schrödinger. L. Z. A. K. (27) Hou. Biochemistry 2002. Am. Gasteiger. J. J. X. G. Sci. L. Piarulli. Soc. 8 (Article 9). Biochemistry 2003. P. Ongeri. M. 24. Erlangen. Schrödinger. Med.. 1000−1008. 11. 2010. Chem. C. Shao. 482−495. 41.. (26) Zhao. W. (25) (a) Abbruscato. E76−E88.. Sicsic. J.