Influence of N-Methylation and Conformation on Almiramide Anti-Leishmanial Activity

The almiramide N-methylated lipopeptides exhibit promising activity against trypanosomatid parasites. A structure–activity relationship study has been performed to examine the influences of N-methylation and conformation on activity against various strains of leishmaniasis protozoan and on cytotoxicity. The synthesis and biological analysis of twenty-five analogs demonstrated that derivatives with a single methyl group on either the first or fifth residue amide nitrogen exhibited greater activity than the permethylated peptides and relatively high potency against resistant strains. Replacement of amino amide residues in the peptide, by turn inducing α amino γ lactam (Agl) and N-aminoimidazalone (Nai) counterparts, reduced typically anti-parasitic activity; however, peptide amides possessing Agl residues at the second residue retained significant potency in the unmethylated and permethylated series. Systematic study of the effects of methylation and turn geometry on anti-parasitic activity indicated the relevance of an extended conformer about the central residues, and conformational mobility by tertiary amide isomerization and turn geometry at the extremities of the active peptides.


Introduction
Neglected tropical diseases caused by trypanosomatid protozoan infections, such as human African trypanosomiasis, Chagas disease and leishmaniasis, impact significantly on public health, especially in tropical countries, where their influences are compounded due to poverty, environment, and drug resistance [1][2][3]. Leishmaniasis is caused by an intracellular protozoan belonging to the genus Leishmania (L.) and transmitted by the phlebotomine sandflies [4]. Upon mammal host infection by way of a sand fly bite, the parasites enter the blood cells in the so-called promastigote stage, then begin to multiply in the amastigote stage spreading to other cells and tissues. Depending on the species, the parasites may disperse via blood and lymph fluid to other body sites, such as the skin and major organs.
With over 1 million new cases occurring annually, around 12 to 15 million people worldwide are infected with leishmaniasis [5,6], which presents as four main forms: cutaneous, mucocutaneous, visceral and post Kala-azar dermal leishmaniasis (CL, MCL, VL and PKDL) [7]. Leishmaniasis causes enlargement of the spleen and liver, anemia [8], skin lesions [9], and destruction of mucous membranes [10]. In addition, controlling co-infection with HIV and visceral leishmaniasis has become a serious challenge [11]. Preliminary structure-activity relationship studies on the almiramides have indicated the importance of the unsaturated lipophilic terminus. Almiramides B and C (2 and 3, Figure 1) with 2-methyl-7-octynoyl and -7-octenoyl tails were reported to exhibit low μM activity against Leishmania donovani, but their 2-methyl-7-oxooctanoyl counterpart almiramide A (1) was inactive [25]. Employment of a 6-heptynyl lipophilic terminus and N-methy1 valine 3 in acid 12 and dimethyl amide 13 gave peptides with μM IC50 values against L. donovani, and improved therapeutic indices with relatively lower cytotoxicity (CC50) against mammalian Vero kidney cells: selectivity index (CC50/EC50) 12 = 13 (50.2) > 2 (21.8) > 3 (17.4) [25]. In this series, the significance of the peptide C-terminal was illustrated by the methyl ester counterpart (e.g., 14) of peptide acid 12, which exhibited a significant loss of anti-parasitic activity and a relative gain in Vero cell cytotoxicity [25]. In addition, peptide 2 and novel almiramides D-H (4-8) possessing the common 2-methyl-   The intriguing potential of the almiramides for the development of anti-parasitic agents has prompted a more detailed study to ascertain the features responsible for their potency and selectivity. In spite earlier efforts, limited knowledge exists of the relevance of N-methylation and conformation for almiramide activity against parasite and host cells. Among the reported almiramide analogs, peptides having only two N-methyl residues [e.g., 5, Val(Me) 1 and Ile(Me) 4 ] to completely permethylated amides (e.g., acid 12) have been respectively isolated and synthesized. To the best of our knowledge, the activity of unmethylated and singly methylated almiramide peptides has yet to be reported nor has the effectiveness of almiramide analogs against resistant strains of Leishmania been examined. A systematic study of almiramide peptides has now been performed to shine light on the importance of tertiary amides and turn conformers for activity against Leishmania and resistant strains.

Chemistry
Commencing with the potent and selective lead peptide 12, a systematic study of the influence of N-methyl groups on each of the five amides was begun by the preparation of the corresponding unmethylated peptide acid 15 ( Figure 2). Considering that long chain secondary amide derivatives of 12 retained significant anti-parasitic activity and offered potential for the synthesis of conjugates to study mechanism of action [26], the corresponding 1,6-hexanediamine amides 16 and 17 were examined for comparison with permethylated and unmethylated acids 12 and 15. Subsequently, an N-methyl scan was performed on amide 16 to provide singly methylated analogs 18-22. For systematic solidphase syntheses of N-methyl peptides 18-22, the required N-methyl amino acids (Fmoc-Ala(Me)-OH, Fmoc-Phe(Me)-OH and Fmoc-Val(Me)-OH) were respectively prepared in solution, according to literature protocols featuring acid-catalyzed condensation of an Fmoc-protected amino acid with paraformaldehyde, to form an oxazolidinone followed by reduction with triethyl silane and trifluoroacetic acid [33].  In the study of biologically active peptides, α-amino γ-lactam residues (Agl residues, Figure 2) so-called Freidinger-Veber lactams have been commonly used to restrict backbone ω and ψ dihedral angles to favor β-turn conformers, in which the Agl residue situates at the i + 1 position [34,35]. The related N-amino-imidazol-2-one (Nai) residues offer similar backbone constraint with potential to add substituents at the heterocycle 4-and 5positions to mimic side chain function with constrained χ dihedral angle geometry [35,36]. A biologically active almiramide bent conformer has been proposed based on the activity of sugar-peptide hybrids 9-11 [32]. To probe this hypothesis further, Agl and (5-Me)Nai residues were used to replace systematically the first four residues in the sequences of peptides 12 and 15-17 ( Figure 3). For example, six Agl almiramide derivatives (e.g., 23-28) were respectively prepared by replacing valine (e.g., Val 1 , Val 2 , Val 3 ) with the heterocycle in acids 12 and 15. The first four residues of permethylated and unmethylated pep-

22
H H H H Me NH(CH 2 ) 6 NH 2 In the study of biologically active peptides, α-amino γ-lactam residues (Agl residues, Figure 2) so-called Freidinger-Veber lactams have been commonly used to restrict backbone ω and ψ dihedral angles to favor β-turn conformers, in which the Agl residue situates at the i + 1 position [34,35]. The related N-amino-imidazol-2-one (Nai) residues offer similar backbone constraint with potential to add substituents at the heterocycle 4-and 5-positions to mimic side chain function with constrained χ dihedral angle geometry [35,36]. A biologically active almiramide bent conformer has been proposed based on the activity of sugar-peptide hybrids 9-11 [32]. To probe this hypothesis further, Agl and (5-Me)Nai residues were used to replace systematically the first four residues in the sequences of peptides 12 and 15-17 ( Figure 3). For example, six Agl almiramide derivatives (e.g., 23-28) were respectively prepared by replacing valine (e.g., Val 1 , Val 2 , Val 3 ) with the heterocycle in acids 12 and 15. The first four residues of permethylated and unmethylated peptide amides 16 and 17 were also systematically replaced by Agl residues to provide peptides 29-35. Moreover, (5-Me)Nai residues were respectively used to replace Val 1 and Val 2 of peptide acid 15 in peptides 36 and 37. For the Agl scan, Fmoc-Agl dipeptides [Fmoc-Agl-Val-OH, Fmoc-Agl-Ala-OH and Fmoc-Agl-Phe-OH (38a-c), supporting information] were respectively synthesized by a modification of the original Freidinger-Veber protocol, as recently reported for the preparation of the valine dipeptide [34,37]. In brief, N-(Boc)methionyl dipeptide tert-butyl esters were treated with iodomethane to prepare the corresponding sulfonium ion intermediate, which on treatment with NaH underwent intramolecular N-alkylation to furnish N-(Boc)Agl dipeptide tert-butyl esters (e.g., 39a-c, supporting information). The carbamate and ester groups were removed using trifluoroacetic acid (TFA) in dichloromethane and the Fmoc group was installed using N-(9-fluorenylmethoxycarbonyloxy)succinimide (Fmoc-OSu) and sodium carbonate in aqueous acetone to provide Agl dipeptides 38. During the cyclization to form Boc-Agl-Phe-Ot-Bu (39c, supporting information), ester epimerization occurred providing an inseparable 2:1 mixture of diastereomers, which was used to prepare the separable mixture of Agl-Phe peptide 32 and Agl-D-Phe isomer R-32. The corresponding Nai-dipeptide ester [Fmoc-(5-Me)Nai-Val-Ot-Bu (40, supporting information)] was synthesized from Fmoc-azaGly-Val-Ot-Bu (41, supporting information), by a route featuring oxidation to the corresponding azopeptide using N-bromosuccinimide (NBS) and lutidine in dichloromethane, proline-catalyzed alkylation with propionaldehyde, and dehydration using p-toluenesulfonic acid in chloroform [38]. Ester solvolysis using TFA in dichloromethane gave the Fmoc-(5-Me)Nai-Val-OH which without purification was coupled onto resin.

Bioactivity
The biological activity of peptides 15-37 was assessed against the L. infantum wildtype stain (WT), as well as mutants resistant to the common anti-leishmanial agents, such as antimony (Sb III ), amphotericin B (AmB) and miltefosine (MF): Sb2000.1, AmB1000.1 and MF200.5. Activity against Leishmania promastigotes was determined by monitoring the replication of parasites after 72 h of incubation at 25 • C in the presence of increasing concentrations of the different peptides 12-37 and reported as EC 50 values (Table 2). In general, almiramide peptides 12-37 exhibited activity against wild type and resistant strains in the range of 5-300 µM. The greatest potencies were observed using analogs against the amphotericin B resistant strain (e.g., AmB1000.1).
Examining peptide potency (EC 50 ) against wild type L. infantum, the 1,6-diaminohexane amides (e.g., 16 and 17) were respectively more potent than the acid counterparts (e.g., 15 and 12). In the Agl series, amides 29-31 and 33-35 were similarly more active than their acid counterparts 23-25 and 26-28. Moreover, the unmethylated analogs (e.g., 15, 16, 29 and 31) were typically more potent than the respective permethylated counterparts (e.g., Many of the structure-activity relationships, that were observed in the wild-type strain, were also found in the resistant strains. For example, the amides were more potent than their acid counterparts. The nonmethylated analogs were also typically more potent than their permethylated analogs; however, in the case of the amphotericin B resistant strain (e.g., AmB1000.1), permethylated analogs 17 and 35 were respectively 4-and 1.4-fold more active than nonmethylated counterparts 16 and 31, and in the case of the miltefosine resistant strain (e.g., MF200.5), 35 was 1.5-fold more active than 31. Although the order of potency for nonmethylated 16, MeVal 1 18 and MePhe 5 22 was contingent on the tested resistant strain, analogs with no tertiary amide and methylation at the Cand N-terminal residues were consistently more active than peptides 19-21 with N-methylation in the central residues. The potency of MePhe 5 22 was greater than Agl-Phe 5 32 for all strains tested. The Agl constraint at other positions tended to slightly favor activity compared to the tertiary amide counterpart in the antimony resistant strain (e.g., Sb2000.1) as was observed in wild type; however, the lactam analogs were significantly less active than the respective N-methyl counterparts against the miltefosine and amphotericin B strains. In addition, preliminary investigations of peptides (e.g., 12-22), which had notable potency against Leishmania (EC 50 = 5-90 µM), detected no activity against T. brucei and T. cruzi (data not shown).
The host cytotoxicity was evaluated by monitoring the survival rates of murine LM-1 macrophage at different concentrations of peptides 15-37 (0.0001 to 100 mM) over

Discussion
The structure-activity relationship studies obtained in wild type and resistant strains of L. infantum reflect likely a combination of ability to engage the target and pharmacokinetic properties that may influence peptide availability. The molecular targets of almiramide C have been studied using a combination of photo-affinity and fluorescent probes in T. brucei, and suggested to include integral membrane proteins found in glycosomes (e.g., GIM5 and PEX11), which are specific to kinetoplastid parasites [26]. Responsible for the first seven steps of glycolysis, the glycosome is a peroxisome-related organelle essential for parasite survival in the bloodstream stage [41]. Notably, translocation across the glycosomal membrane implicates transporter and pore-forming proteins [42], which may differ contingent on species and be modified in resistant strains.
Resistant strains of Leishmania emerge by different mechanisms which contingent on the drug act commonly to reduce the active concentration inside the parasite by either decreasing uptake, increasing efflux or inhibiting activity [43]. Leishmania antimonial resistance is associated with thiol metabolism to prevent reduction of the Sb V to more active Sb III , and to sequester antimony in thiolate complexes amenable for efflux [44]. Miltefosine resistance is commonly associated with mutations in the Leishmania miltefosine transporter (LMT), a P-type ATPase responsible for the translocation of phospholipids, as well as overexpression of ABC transporters [44]. Amphotericin B resistant L. donovani promastigotes have been shown to feature substitution of ergosterol for another sterol, which alters the fluidity and AmB binding affinity in the cell membrane [45].
Although the conformational preferences of the almiramides alone and target bound have yet to be described, information gleaned from related peptides and their N-methyl and lactam counterparts offers a lens through which to interpret the structure-activity relationships. For example, the circular dichroism spectrum of the (Val-Val-Val-Ala) n oligomer in a mixture of hexafluoro-2-propanol and trifluoro ethanol indicated a curve shape typical of an extended β-sheet structure [46]. The corresponding Val-Val-Val-Ala region in nonmethylated almiramide analogs 15 and 16, as well as MePhe analog 22 may likely adopt an extended β-strand conformer. Introduction of N-methyl groups causes significant consequences on peptide conformation, due in part to creation of a tertiary amide which loses a potential NH hydrogen-bond donor and may adopt energetically similar cis and trans isomers [47]. Computational analysis of the N'-methyl amides of N-acetyl-N-methyland N-acetyl-alanine indicated that repulsive interactions between the N-methyl and carbonyl oxygen moieties of the former abolished the low-energy minimum β-conformer adopted by the latter [48]. In cyclic peptides, N-methyl residues have also induced backbone f and ψ dihedral angles consistent with βand γ-turn conformers [49], as well as altered side chain χ geometry [50]. N-Alkylation of the central amide of the hairpin inducing D-Pro-Aib dipeptide has also been shown by variable temperature CD spectroscopy to reinforce the central turn conformer and enhance the stability of the folded β-sheet peptide [51]. Introduction of N-methyl residues within the peptide chain causes likely a shift from an extended sequence to a dynamic series of cisand transamide conformers exhibiting a preference for turn geometry, which in peptides 19-21 reduces activity. In MeVal 1 analog 18 and permethylated analogs 12 and 17, methylation at the N-terminal enables a cis-amide conformer in which the lipid tail may fold in the direction of the peptide. Such a geometry may improve membrane transport by hiding hydrophilic NH moieties and may account for the significantly improved activity of 17 and 18 against the amphotericin B resistant strains. To further examine the influence of N-methylation on conformation, the 1 H NMR spectra of nonmethylated peptide 16 and N-methyl counterpart 18 were examined in DMSO-d 6 (supporting information). In contrast to peptide 16, which exhibited a narrow distribution of amide protons signals between 7.55 and 7.95 ppm characteristic of a linear conformer, the corresponding peaks were downfield shifted, dispersed between 7.65 and 8.35 ppm, and existed in isomeric pairs for MeVal 1 peptide 18, likely due the tertiary amides favoring cisand trans-amide isomers of similar energy and local turn conformers with intramolecular hydrogen bonds.
The propensity for Agl bearing peptides to adopt type II and II' β-turns contingent on α-carbon stereochemistry has been demonstrated using spectroscopic and computational methods, as well as X-ray diffraction, which has also characterized crystals of lactam analog in extended conformers [52]. Relatively diminished activities of Agl analogs 29-35 compared to nonmethylated peptide 16 may again be due to the favored turn geometry. The slightly better activity exhibited by Agl analogs 29-31 in comparison to N-methyl counterparts 19-21 in wild type L. infantum may be attributed to the capacity of the amino lactam to stabilize trans-amide isomers. On the other hand, the notably better activity of MePhe analog 22 relative to Agl counterpart 32 indicates the attributes of greater conformational flexibility at the C-terminal. Similarly, greater conformational dynamics of N-methyl analogs compared to the Agl counterparts appear to be important for the relatively better activity of the former in resistant Leishmania strains.

LM-1 Macrophages and Cytotoxicity Determination
The LM1-macrophages were grown in DMEM supplemented with 10% heat inactivated FBS. Cells at the concentration of 100,000 cells/mL were cultivated for 24 h in a 96-well plate (37 • C, 5% CO 2 ). The culture medium was removed and fresh medium containing the appropriate drug and concentration was added to the cells, which were incubated for 24 h (37 • C, 5% CO 2 ). Seven different concentrations were tested (100, 10, 1, 0.1, 0.01, 0.001 and 0.0001 mM) as well as controls (without drugs). After 24 h, the culture medium was removed and replaced for fresh medium containing 10% Alamar Blue (Thermo Fisher Scientific, Waltham, MA, USA) and incubated for 2 h (37 • C, 5% CO 2 ). Readings at 570 and 600 nm (Asys UVM340 Plate reader, Biochrom, Cambridge, UK) were taken and analyzed according to the manufacturer protocol. Survival rates at different drug concentrations and CC 50 values were calculated using the Excel software.

Materials
Anhydrous solvents (THF, DMF, CH 2 Cl 2 , and NMP) were obtained by passage through solvent filtration systems (GlassContour, Irvine, CA, USA). Unless specified otherwise, all reagents from commercial sources were used as received. amino acids, such as Fmoc-Phe-OH, Fmoc-Ala-OH, Fmoc-Val-OH, and 6-heptynoic acid were purchased from GL Biochem, ChemImpex and Combi-blocks; solvents were obtained from Fisher. Sodium hydride (60% dispersion in mineral oil) was washed with hexane three times to remove oil prior to use. The N-methyl amino acids, Fmoc-Phe(Me)-OH, Fmoc-Ala(Me)-OH and FmocVal(Me)-OH, were prepared by according to the literature procedure and exhibited 1 H NMR spectra data identical to that previously reported [33]. The Agl dipeptide, Fmoc-Agl-Val-OH was prepared according to the literature procedure and exhibited a 1 H NMR spectrum identical to that previously reported [37].
Chromatography was on 230−400 mesh silica gel. Analytical thin-layer chromatography (TLC) was performed on glass-backed silica gel plates (Merck 60 F254). Visualization of the developed chromatogram was performed by UV absorbance or staining with ninhydrin. 1  Almiramide peptide analog synthesis was performed using Fmoc-based solid-phase synthesis in an automated shaker commencing with 2-chlorotrityl chloride resin. All final peptides were purified on a preparative column (C18 Gemini column) using a gradient from pure water (0.1% FA) to mixtures with MeOH (0.1% FA) at a flow rate of 10 mL/min. Purity of peptides (>95%) was evaluated using analytical LC−MS on a 5 µM 50 mm × 4.6 mm C 18 Phenomenex Gemini column in two different solvent systems: water (0.1% FA) with CH 3 CN (0.1% FA) and water (0.1% FA) with MeOH (0.1% FA) at a flow rate of 0.5 mL/min using the appropriate linear gradient.
Fmoc-Phe(Me)-OH, Fmoc-Ala(Me)-OH and FmocVal(Me)-OH, were prepa ing to the literature procedure and exhibited 1 H NMR spectra data identica ously reported [33]. The Agl dipeptide, Fmoc-Agl-Val-OH was prepared ac literature procedure and exhibited a 1 H NMR spectrum identical to that ported [37].
Chromatography was on 230−400 mesh silica gel. Analytical thin-lay raphy (TLC) was performed on glass-backed silica gel plates (Merck 60 F2 tion of the developed chromatogram was performed by UV absorbance or ninhydrin. 1 H and 13 C NMR spectra were measured respectively in CDCl3 a 126 MHz, and referenced to CDCl3 Almiramide peptide analog synthesis was performed using Fmoc-bas synthesis in an automated shaker commencing with 2-chlorotrityl chloride peptides were purified on a preparative column (C18 Gemini column) us from pure water (0.1% FA) to mixtures with MeOH (0.1% FA) at a flow rate Purity of peptides (>95%) was evaluated using analytical LC−MS on a 5 μM mm C18 Phenomenex Gemini column in two different solvent systems: w with CH3CN (0.1% FA) and water (0.1% FA) with MeOH (0.1% FA) at a f mL/min using the appropriate linear gradient.

HCC(CH 2 ) 4 CO-Val(Me)-Val(Me)-Val(Me)-Ala(Me)-Phe(Me)-OH (12)
Under argon, peptide 15 (200 mg) was dissolved in THF (40 mL), cooled to 0 • C, treated with NaH (600 mg, 80 eq.), stirred for 5 min, and treated dropwise with iodomethane (0.6 mL, 9 mmol, 30 eq.). The cooling bath was removed. The reaction mixture warmed to room temperature. After 2 h, more iodomethane (0.6 mL, 30 eq.) was added to the reaction mixture, which was stirred for 20 h. The suspension was quenched with water, concentrated to a reduced volume, and acidified to pH = 1 using 10% aqueous HCl. The acidified mixture was extracted 3 times with EtOAc. The organic layers were combined, dried over Na 2 SO 4 , filtered and evaporated to 174 mg of residue, from which part (87 mg) was used to make peptide 17 as described below, and the remainder was purified by HPLC on a C 18 column using a gradient of 30% to 90% MeOH in H 2 O to obtain peptide 12 ( was extracted 3 times with EtOAc. The organic layers were combined, dried over Na2S filtered and evaporated to 174 mg of residue, from which part (87 mg) was used to ma peptide 17 as described below, and the remainder was purified by HPLC on a C18 colu using a gradient of 30% to 90% MeOH in H2O to obtain peptide 12 (16 mg, 13%), wh was shown to be >99% pure by LC-MS analysis [30−95% MeOH (0.1% FA) in H2O (0. FA), 14 min, RT 6.83 min].

HCC(CH2)4CO-Val-Val-Val-Ala-Phe-OH (15)
Fmoc-Phe-OH (848 mg, 1.5 eq.) a DIEA (0.8 mL, 3 eq.) were added to a suspension of 2-chlorotrityl chloride resin (1 g, 20 400 mesh, 1% DBV) swollen in CH2Cl2 (25 mL). The mixture was shaken for 18 h, filter and washed sequentially with CH2Cl2 (3 times for 1 min/wash) and DMF (3 times fo min/wash). The Fmoc group was removed upon treatment twice for 20 min with a 2 solution of piperidine in DMF (20 mL/g resin). Subsequently, Fmoc-Ala-OH (3 eq.) w coupled to the resin swollen in NMP (25 mL/g resin) using DIC (3 eq.) and HOBt (3 e After shaking for 16 h, the coupling mixture was filtered, and the resin was washed described above. Subsequent Fmoc group removal and coupling of Fmoc protected am acid residues were performed using the above protocols. Complete coupling reactio were confirmed by LC-MS monitoring. After coupling of the last residue, the 6-heptyn acid (3 eq.), DIC (3 eq.) and HOBt (3 eq.) were added to the resin. The mixture was shak for 18 h, filtered and washed as previously described. Then, the linear peptide was cleav from the solid support using a solution of TFA/TES/H2O (95/2.5/2.5). The volatiles w evaporated. The reduced volume was treated with diethyl ether. The precipitate was c lected by centrifugation (1200 rpm), washed with ether and recollected by centrifugat (3 × 10 min). Removal of diethyl ether afforded a colorless solid [2], which gave 387 mg residue, from which 200 mg was used to make peptide 12 as described below, and 60 purified by HPLC on C18 column using a gradient of 30% to 90% MeOH in H2O to obt peptide 15 (24 mg, 12%) , which was shown to be >99% pure by LC-MS analysis [30−9 MeOH (0.1% FA) in H2O (0.1% FA), 14 min, RT 8.33 min].

HCC(CH 2 ) 4 CO-Val-Val-Val-Ala-Phe-OH (15)
Fmoc-Phe-OH (848 mg, 1.5 eq.) and DIEA (0.8 mL, 3 eq.) were added to a suspension of 2-chlorotrityl chloride resin (1 g, 200-400 mesh, 1% DBV) swollen in CH 2 Cl 2 (25 mL). The mixture was shaken for 18 h, filtered and washed sequentially with CH 2 Cl 2 (3 times for 1 min/wash) and DMF (3 times for 1 min/wash). The Fmoc group was removed upon treatment twice for 20 min with a 20% solution of piperidine in DMF (20 mL/g resin). Subsequently, Fmoc-Ala-OH (3 eq.) was coupled to the resin swollen in NMP (25 mL/g resin) using DIC (3 eq.) and HOBt (3 eq.). After shaking for 16 h, the coupling mixture was filtered, and the resin was washed as described above. Subsequent Fmoc group removal and coupling of Fmoc protected amino acid residues were performed using the above protocols. Complete coupling reactions were confirmed by LC-MS monitoring. After coupling of the last residue, the 6-heptynoic acid (3 eq.), DIC (3 eq.) and HOBt (3 eq.) were added to the resin. The mixture was shaken for 18 h, filtered and washed as previously described. Then, the linear peptide was cleaved from the solid support using a solution of TFA/TES/H 2 O (95/2.5/2.5). The volatiles were evaporated. The reduced volume was treated with diethyl ether. The precipitate was collected by centrifugation (1200 rpm), washed with ether and recollected by centrifugation (3 × 10 min). Removal of diethyl ether afforded a colorless solid [2], which gave 387 mg of residue, from which 200 mg was used to make peptide 12 as described below, and 60 mg purified by HPLC on C 18 column using a gradient of 30% to 90% MeOH in H 2 O to obtain peptide 15 (24 mg, 12%), which was shown to be >99% pure by LC-MS analysis [ was extracted 3 times with EtOAc. The organic layers were combined, dried over Na2SO4, filtered and evaporated to 174 mg of residue, from which part (87 mg) was used to make peptide 17 as described below, and the remainder was purified by HPLC on a C18 column using a gradient of 30% to 90% MeOH in H2O to obtain peptide 12 (16 mg, 13%), which was shown to be >99% pure by LC-MS analysis [30−95% MeOH (0.1% FA) in H2O (0.1% FA), 14 min, RT 6.83 min].

HCC(CH2)4CO-Val-Val-Val-Ala-Phe-OH (15)
Fmoc-Phe-OH (848 mg, 1.5 eq.) and DIEA (0.8 mL, 3 eq.) were added to a suspension of 2-chlorotrityl chloride resin (1 g, 200-400 mesh, 1% DBV) swollen in CH2Cl2 (25 mL). The mixture was shaken for 18 h, filtered and washed sequentially with CH2Cl2 (3 times for 1 min/wash) and DMF (3 times for 1 min/wash). The Fmoc group was removed upon treatment twice for 20 min with a 20% solution of piperidine in DMF (20 mL/g resin). Subsequently, Fmoc-Ala-OH (3 eq.) was coupled to the resin swollen in NMP (25 mL/g resin) using DIC (3 eq.) and HOBt (3 eq.). After shaking for 16 h, the coupling mixture was filtered, and the resin was washed as described above. Subsequent Fmoc group removal and coupling of Fmoc protected amino acid residues were performed using the above protocols. Complete coupling reactions were confirmed by LC-MS monitoring. After coupling of the last residue, the 6-heptynoic acid (3 eq.), DIC (3 eq.) and HOBt (3 eq.) were added to the resin. The mixture was shaken for 18 h, filtered and washed as previously described. Then, the linear peptide was cleaved from the solid support using a solution of TFA/TES/H2O (95/2.5/2.5). The volatiles were evaporated. The reduced volume was treated with diethyl ether. The precipitate was collected by centrifugation (1200 rpm), washed with ether and recollected by centrifugation (3 × 10 min). Removal of diethyl ether afforded a colorless solid [2], which gave 387 mg of residue, from which 200 mg was used to make peptide 12 as described below, and 60 mg purified by HPLC on C18 column using a gradient of 30% to 90% MeOH in H2O to obtain peptide 15 (24 mg, 12%) , which was shown to be >99% pure by LC-MS analysis [30−95% MeOH (0.1% FA) in H2O (0.1% FA), 14 min, RT 8.33 min].

HCC(CH 2 ) 4 CO-Val-Val-Val-Ala-Phe-NH(CH 2 ) 6 NH 2 (16)
A suspension of 2-chlorotrityl chloride resin (250 mg, 200-400 mesh, 1% DBV) swollen in CH 2 Cl 2 (25 mL) was treated with 1,6-diaminohexane (51 mg, 1.2 eq.) and DIEA (0.2 mL, 3 eq.), shaken for 24 h, filtered and washed sequentially with CH 2 Cl 2 (3 times for 1 min/wash) and DMF (3 times for 1 min/wash). Subsequently, Fmoc-Phe-OH (3 eq.) was coupled to the resin swollen in NMP (25 mL/g resin) using DIC (3 eq.) and HOBt (3 eq.). After shaking for 16 h, the coupling mixture was filtered, and the resin was washed as described above. Subsequent Fmoc group removals, couplings of Fmoc protected amino acid residues, and acylation with 6-heptynoic acid, all were performed using the protocols described for the synthesis of peptide 15. The linear peptide was cleaved from the solid support using a solution of TFA/TES/H 2 O (95/2.5/2.5). Removal of the volatiles gave a colorless oil, which was purified by HPLC on a C 18  with 6-heptynoic acid, all were performed using the protocols described for the synthesis of peptide 15. The linear peptide was cleaved from the solid support using a solution of TFA/TES/H2O (95/2.5/2.5). Removal of the volatiles gave a colorless oil, which was purified by HPLC on a C18 column using a gradient of 30% to 90% MeOH in H2O to obtain peptide 16 as a white solid (8.8 mg, 3%), which was shown to be >99% pure by LC-MS analysis [ (25 mL). The resulting amine resin was swollen in NMP (25 mL/g resin), treated with Fmoc-Phe-OH (3 eq.), DIC (3 eq.) and HOBt (3 eq.), shaken for 16 h, filtered, and washed as described above. The Fmoc group was removed upon treatment twice for 20 min with a 20% solution of piperidine in DMF (20 mL/g resin). After Fmoc group removal, the peptide was elongated, coupled to 6-heptynoic acid using HATU (3 eq.) in 0.4 NMM in DMF (25 mL/g resin), cleaved and purified as described for the synthesis of peptide 16. Evaporation of the collected fractions gave peptide 18 (8.2 mg, 3%), which was prepared and shown to be >99% pure by with 6-heptynoic acid, all were performed using the protocols described for the synthesis of peptide 15. The linear peptide was cleaved from the solid support using a solution of TFA/TES/H2O (95/2.5/2.5). Removal of the volatiles gave a colorless oil, which was purified by HPLC on a C18 column using a gradient of 30% to 90% MeOH in H2O to obtain peptide 16 as a white solid (8.8 mg, 3%), which was shown to be >99% pure by LC-MS analysis [30−95% MeOH (0.1% FA) in H2O (0.1% FA), 14 min, RT 7.33 min].

Conclusions
The unsaturated lipid tail and the C-terminal acid and carboxamide functions of almiramide peptides have previously been shown to have relevance for activity against trypanosomatid parasites [24,25]. Moreover, active almiramide analogs possessing sugar amino acids were suggested to adopt bent structures [31]. The influences on anti-parasite activity of amide N-methylation and turn-inducing Agl and Nai residues within almiramide peptides have now been investigated in wild type and resistant strains of L. infatum and compared with macrophage cytotoxicity. Peptide amides exhibited consistently better activity than their C-terminal acid counterparts. Within a set of almiramide 1,6-diaminohexane amides, more potent peptides with relatively high selectivity were typically obtained without methylation (e.g., 16) and with a single methyl group in MePhe 5 almiramide 22 than with analogs possessing a methyl amide at other positions and with permethylated analog 17. On the other hand, permethylated and MeVal 1 peptide amides 17 and 18 exhibited µM inhibitory activity against the amphotericin B resistant strain with high therapeutic indices. Replacement of amino acid residues by turn-inducing counterparts caused typically losses of activity against the L. infatum strains; however, permethylated Agl 2 amide had similar activity and better selectivity against wild type L. infatum compared to permethylated analog 17. Although studies of the consequences of such structural modifications on metabolism and mechanism of action are in progress, conformers extended about the central residues and mobile at the extremities of the peptide may favor almiramide activity.
Supplementary Materials: The following are available online, synthesis procedure of Agl and Nai dipeptides; NMR Spectra; Ascertainment of purity by HPLC; Dose-response curves of Leishmania strains and LM-1 mac-rophage survival rate in the presence of peptides 12-37.