Rational Design of Triazololipopeptides Analogs of Kisspeptin

26 Mar 2015 - of the endogenous neuropeptide kisspeptin 10 (KP10). Improved resistance to ... Kisspeptin (KP) is the most potent secretagogue of the...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/jmc

Rational Design of Triazololipopeptides Analogs of Kisspeptin Inducing a Long-Lasting Increase of Gonadotropins Massimiliano Beltramo,*,† Vincent Robert,† Mathieu Galibert,‡ Jean-Baptiste Madinier,‡ Philippe Marceau,‡ Hugues Dardente,† Caroline Decourt,† Nicolas De Roux,§ Didier Lomet,† Agnès F. Delmas,‡ Alain Caraty,† and Vincent Aucagne*,‡ †

UMR Physiologie de la Reproduction et des Comportements (INRA, UMR85; CNRS, UMR7247; Université François Rabelais Tours; IFCE), F-37380 Nouzilly, France ‡ Centre de Biophysique Moléculaire (CNRS UPR4301), Rue Charles Sadron, F-45071 Orléans Cedex 2, France § INSERM U690, Hôpital Robert Debré, 75019 Paris, France S Supporting Information *

ABSTRACT: New potent and selective KISS1R agonists were designed using a combination of rational chemical modifications of the endogenous neuropeptide kisspeptin 10 (KP10). Improved resistance to degradation and presumably reduced renal clearance were obtained by introducing a 1,4-disubstituted 1,2,3-triazole as a proteolysis-resistant amide mimic and a serum albumin-binding motif, respectively. These triazololipopeptides are highly potent full agonists of KISS1R and are >100 selective over the closely related NPFF1R. When injected in ewes with a quiescent reproductive system, the best compound of our series induced a much prolonged increase of luteinizing hormone release compared to KP10 and increased follicle-stimulating hormone plasma concentration. Hence, this KISS1R agonist is a new valuable pharmacological tool to explore the potential of KP system in reproduction control. Furthermore, it represents the first step to develop drugs treating reproductive system disorders due to a reduced activity of the hypothalamo−pituitary−gonadal axis such as delayed puberty, hypothalamic amenorrhea, and hypogonadotropic hypogonadism.

■

INTRODUCTION Kisspeptin (KP) is the most potent secretagogue of the gonadotropin-releasing hormone (GnRH), a hypothalamic peptide controlling synthesis and secretion of gonadotropins: luteinizing hormone (LH) and follicle-stimulating hormone (FSH). The reduced activity of this system, known as the gonadotropic axis, is the cause of severe reproductive disorders leading to partial or complete infertility. Genetic defects in Kp receptor (KISS1R, also called GPR54) have been shown to cause isolated gonadotropic deficiency in human.1,2 Corroborating this finding, injection of the endogenous 10 amino acidlong form of KP (KP10) produces a rapid, albeit transient, increase of LH and FSH in several mammals including human.3−5 This makes the Kp system an appealing target for treating human reproduction disorders as well as for coping with problems in reproduction management of livestock. Seasonal breeders such as sheep represent valuable animal models to develop new approaches to treat human reproductive disorders. Seasonal breeders undergo a cyclic arrest of © 2015 American Chemical Society

reproductive activity, called anoestrus. During anoestrus, the activity of the gonadotropic axis is remarkably downregulated with both low LH and FSH pulse frequency and plasma levels that result in gonadal quiescence. In this respect, gonadotropic deficiency observed in some human pathologies resembles the anoestrus state. We previously showed that continuous administration of KP10 can induce ovulation in seasonal anoestrus ewe.4 This result was a breakthrough unveiling essential mechanisms controlling reproduction. However, the short lasting effect of KP injection blunts its usefulness to treat a deficient reproductive system. Despite the major importance of the KP system as a pharmacological target in reproduction, surprisingly few synthetic KISS1R agonists have been disclosed so far. Takeda Pharmaceuticals generated KISS1R agonists by designing KP10 pseudopeptide analogs chiefly to treat hormone-dependent Received: December 19, 2014 Published: March 26, 2015 3459

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Figure 1. Half maximal effective concentration and ewe serum stability of analogs 2−6 compared to native KP10 1. nd: not detectable. Ac: acetyl.

cancers.6,7 The design of these analogs was aimed at preventing protease-mediated degradation8,9 that together with fast renal excretion is the main factor accounting for KP timely restricted action. However, to our knowledge, no KP analogs have been evaluated in vivo for their capacity to restore reproductive system activity in human or in seasonally anoestrus ruminants. We herein report on the development of new potent and selective pseudopeptide agonists of KISS1R based on KP10 scaffold. The best analog of our series showed a much longer in vivo duration of action compared to natural KP10 due to both reduced proteolysis and, presumably, slower renal excretion. Notably, our agonists are capable of reactivating the dormant reproductive system in anoestrus ewe, inducing a prolonged increase of blood LH and FSH concentration after a single injection. This represents a major step forward in the development of a therapy to treat reproductive disorders and to master livestock reproduction.

N-Terminal Acetylation of KP10 Increases Both in Vitro Potency and Resistance to Serum Proteases. Previous report of analysis of KP10 degradation in mouse serum showed rapid hydrolysis of N-terminal Tyr1-Asn2 bond (t1/2 < 1 min), presumably by aminopeptidases.8 To prevent this specific degradation, we masked the Tyr1 N-terminal amine group as an N-acetylamide (compound 2), a classical approach in peptide medicinal chemistry. This resulted in a large improvement in resistance to proteolysis in ewe serum, as a fraction of the acetylated analog 2 was still present after a 6 h incubation (Figure 1). Conversely, the amount of the endogenous form of the peptide (compound 1) was below detection limits after only 1 h of incubation in serum. Surprisingly, acetylation also significantly increased compound potency (Figure 1). Introduction of Triazole as an Amide Mimic Overtly Increases Resistance to Serum Proteases. Another major degradation site of KP10 is the Gly7-Leu8 bond that is cleaved by matrix metalloproteinases leading to rapid peptide inactivation.8,10 On the basis of our previous work11,12 exploiting 1,4-disubstituted 1,2,3-triazole as mimics of amides,13 we investigated the introduction on the KP10 scaffold of a triazole as a peptide bond mimic. The heterocycle was formed in the course of solid phase peptide synthesis (SPPS) through copper-catalyzed alkyne/azide cycloaddition14,15 (CuAAC) of a solid-supported azidopeptide and an N-Fmoc-protected α-aminoalkyne. An inexpensive catalytic system was specifically optimized for this process: different sources of copper(I) were screened, tested in the presence of excess diisopropylethylamine in NMP and voluntarily excluding the use of costly copper(I) ligands such as TBTA. Copper(I) thiophene-2carboxylate (CuTC), [(CH3CN)4Cu]PF6, and (CuCF3SO3)2· C6H6 led to moderate yields, but the cheap CuBr·Me2S complex gave quantitative yields and reproducible reactions when used in excess (see Supporting Information Table S1 for detailed experiments). We synthesized three different triazole-containing pseudopeptides (3−5, Figure 1) as well as an azaglycine-containing

■

RESULTS To improve KP10’s poor pharmacokinetics and pharmacodynamics, a rational design approach based on stepwise targeted modifications of the endogenous peptide was used. First, its fast degradation by serum proteases was addressed through the introduction of chemical modifications at two main proteolysis sites. Second, it was conjugated with serum albumin interacting motifs known to reduce renal excretion rate. To assess the potency of analogs, we used a calcium mobilization assay in an HEK-293 cell line stably transfected with human KISS1R. Ewe KP10 (YNWNSFGLRY-NH2, 1) was used as a template for targeted modifications and as a standard to compare potency and efficacy of new compounds. Of note, ewe KP10 amino acid sequence is 100% identical to that of most mammals and differs from the human sequence only in the C-terminal amino acid that is Tyr instead of Phe. As expected, compound 1 produced a concentration-dependent intracellular increase of intracellular calcium, showing an EC50 comparable with literature data (Figure 1). 3460

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Figure 2. Effect of compounds 1, 2, and 3 on LH plasma concentration. Basal concentration of LH was recorded every 10 min for 30 min before intravenous injection (time 0) of 5 nmol/ewe of compound 1 (A), compound 2 (B), or compound 3 (C). Elevation in LH plasma concentration was significantly longer after analog injection compared to compound 1 (KP10). The total amount of LH secreted (area under the curve, AUC, panel D) was not different between compound 1 and compound 2, whereas compound 3 induced a significant increase of LH AUC. Values are the mean ± SEM (N = 5−6 per group). The effect on duration was evaluated using one way ANOVA multiple comparisons test for repeated measurement followed by uncorrected Fisher’s LSD test: (∗) P < 0.05. AUCs were analyzed by one way ANOVA followed by Dunn’s multiple comparison post hoc test: (∗∗) P < 0.01.

monitored over time. Pharmacodynamics analysis of compound 2 showed that Tyr1 N-acetylation only modestly affected maximal amplitude (basal 0.6 ± 0.5 ng·mL−1 vs KP10 3.3 ± 0.9 ng·mL−1 ; and basal 0.6 ± 0.5 ng·mL−1 vs compound 2 3.9 ± 1.1 ng·mL−1) as well as the duration of LH plasma concentration increase (Figure 2) compared to unmodified KP10. Conversely, combination of the two modifications, Tyr1 N-acetylation and triazole at position Gly7-Leu8 (compound 3), resulted in a clear increase of the maximal effect (basal 0.4 ± 0.2 ng·mL−1 vs compound 3 5.4 ± 2.3 ng·mL−1). Importantly, the duration of LH rise was more than doubled, which resulted in a significant increase of the total amount secreted (Figure 2). We further evaluated the activity of compound 3 by performing a dose response using 1, 5, and 15 nmol/ewe doses. A gradual increase in the maximal amplitude, duration of action, and total amount of LH secretion was observed. However, a clearcut dose response could not be delineated and the overall difference between doses remained modest (Supporting Information Figure S2). Similar results were also obtained for FSH (Supporting Information Figure S3). Design of Analogs Incorporating an Albumin-Binding Motif. Considering the relatively small size and hydrophilicity of compound 3, fast renal clearance might explain the absence of a marked in vivo dose response effect. Two main approaches exist to slow renal clearance of peptides and small proteins: covalent linking to polyethylene glycol polymers (PEGylation) or insertion of an albumin-binding motif. Introduction of such groups can be conveniently achieved through N-acylation of a lysine side chain’s amine group. To identify the most convenient site for N-acylated Lys introduction, without compromising analog potency, we performed a Lys(Ac) scan

derivative (6) for benchmark comparison, as that modification is a key feature in analogs previously developed by Takeda.6 We quantified their residual amount after a 6 h incubation in ewe serum. Incorporation of a triazole at position Gly7-Leu8 (3), Phe6-Gly7 (4), or both sites (5) led to a considerably improved stability compared to compound 2 (typically 15- to 30-fold). Remarkably, the beneficial effect was much more pronounced than for the azaGly-pseudopeptide 6 (typically 3- to 4-fold) (Figure 1). In the latter case, metabolites identified by LC−MS analysis clearly showed bond cleavage at both C- and N-termini of the azaGly bond (Supporting Information Figure S5). The triazolopeptides were then evaluated for their potency to activate KISS1R signaling pathways. Substitution of the Phe6Gly7 peptide bond by a triazole (compound 4) was detrimental to agonist potency, and introduction of two consecutive triazoles (compound 5) resulted in an even more dramatic loss of potency. Contrastingly, modification of the Gly7-Leu8 bond (compound 3) was perfectly tolerated. Hence, analog 3 showed about 30-fold increase in potency (EC50 = 0.07 ± 0.06 nM) compared to endogenous KP10 while preserving full efficacy (Supporting Information Figure S1). We therefore selected triazolopeptide 3 for further in vivo evaluation and kept this modification for all the compounds further tested in vivo in this work. Proteases Resistant Analogs Have a Longer-Lasting Effect on LH Plasma Concentration compared to KP10. To confirm that increased blood half-life was mirrored by prolongation of the physiological effect, compounds 2 and 3 were compared to compound 1 for their capacity to increase LH plasma concentration in anoestrus ewe. Test compounds were injected intravenously, and LH plasma concentration were 3461

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Figure 3. Lys(Ac) scan on peptide 2 to determine the best suited positions to introduce chemical modifications aimed at reducing renal excretion.

Figure 4. Half maximal effective concentration of analogs 13−18.

on compound 2 from Tyr1 to Ser5 (Figure 3). These five N-terminal residues are known to be relatively tolerant to mutation, whereas the C-terminus is crucial for bioactivity.16,17 Substitution of Asn4 (compound 10) and Ser5 (11) decreased potency, whereas substitution of Asn2 (8) and Trp3 (9) had no significant impact. We thus selected the two latter positions for conjugation. We started with PEGylation, a classical modification used to improve small protein half-life.18 However, preliminary studies using a KP analog conjugated with a 25 kDa PEG (compound 12) did not show prolongation of in vivo effect (Supporting Information Figure S4), even if the molecule was active in vitro (EC50 = 9.0 ± 4.2 nM). We therefore concentrated our efforts on the second strategy, which benefits from the long half-life of serum albumin present in high concentration in blood. Reversible binding of peptides to albumin leads to improved pharmacodynamics by increasing their apparent size, which in turn reduces renal excretion. We first opted for the introduction of a simple and well-established N-hexadecanoyl group19,20 and synthesized the corresponding conjugates in which the hydrophobic moiety was introduced by N-acylation of either Lys2 (13) or Lys3 side chain amine group. (14), following a validated solid-phase procedure.21 Disappointingly these modifications resulted in more than 1000fold decrease in potency (Figure 4). To try to overcome this dramatic potency loss, we introduced an L-glutamate spacer between the Lys and the hexadecanoyl groups, a modification also supposed to increase affinity binding to serum albumin.21,22 The γ-(N-hexadecanoyl-L-glutamyl) modifications

at both position 2 (15) and 3 (16) were quite effective in restoring potency, with EC50 of 2.9 ± 1.9 and 24.4 ± 19.6 nM, respectively (Figure 4). Finally, when a triazole was introduced at position Gly7-Leu8 (compound 17), no significant change in potency was observed compared to 15, whereas the same modification introduced in analog 16 significantly decreased potency (compound 18). Analog 17 was therefore chosen for further in vivo studies. Introduction of an Albumin Binding Motif Dramatically Increases Duration and Extent of LH and FSH Secretion. To assess the capacity of compound 17 to increase gonadotropins plasma concentration, anoestrus ewes were injected intravenously with three different doses (1, 5, and 15 nmol/ewe). A clear dose response was observed in duration of plasma LH rise, LH peak concentration, and total amount of LH secreted (Figure 5). At 1 nmol the effect, if any, was minimal and peak LH values were 0.3 ± 0.09 ng mL−1 for basal and 1.1 ± 1.4 ng mL−1 for treated animals. However, LH plasma concentration rose from 0.2 ± 0.09 ng·mL−1 to 5.3 ± 3.6 ng·mL−1 (more than 25-fold increase over basal) for the 5 nmol/ewe dose and from 0.3 ± 0.15 ng·mL−1 to 13.3 ± 5.3 ng·mL−1 (more than 40-fold increase over basal) for the 15 nmol/ewe dose. In the latter group mean LH concentration was still 3-fold higher than the basal level 9 h after administration. Also, the total amount of LH secreted increased dramatically (more than 8-fold, from 274 ± 200 ng·min·mL−1 to 2405 ± 768 ng·min·mL−1) when the dose was increased from 1 to 15 nmol/ewe. 3462

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Figure 5. Dose−response effect of compound 17 on LH plasma concentration. The time course of LH plasma concentration was analyzed after intravenous injection of three different doses of compound 17: 1 nmol/ewe (A), 5 nmol/ewe (B), or 15 nmol/ewe (C). Basal concentration of LH was recorded every 10 min for 30 min before injection (time 0). (D) Dose response AUC. Value are the mean ± SEM (N = 4−6 per group). The effect on duration was evaluated using one way ANOVA multiple comparisons test for repeated measurement followed by uncorrected Fisher’s LSD test: (∗) P < 0.05. AUCs were analyzed by one way ANOVA followed by Dunn’s multiple comparison post-test: (∗∗) P < 0.01, 15 nmol/ewe vs 5 nmol/ewe, (∗∗∗∗) P < 0.001, 15 nmol/ewe vs 1 nmol/ewe.

groups have substituted it with various amide mimics.8,31 Some of these isosteres, e.g., thioamide, azaglycine or the nonhydrolyzable (E)-alkene and hydroxyethylene were shown to be well tolerated by KISS1R. Several azaglycine-containing molecules modified at multiple other sites were further developed by Takeda Pharmaceutical8,32,33 as metabolically resistant KISS1R agonists. The intrinsic planarity and conformational restriction of azaglycine residue compared to an α-amino acid residue are supposed to prevent its degradation by proteases at both azaGly7-Leu8 and Phe6azaGly7 sites. In this work, we explored the effect of the introduction of a 1,4-disubstituted-1,2,3-triazole heterocycle as nonhydrolyzable peptide bond mimic. To our knowledge such a triazole-to-amide substitution approach aimed at improving the resistance to proteolysis of therapeutic peptide candidates has been reported only once.34 This can be viewed as extremely surprising, considering the great popularity of the CuAAC reaction and the well-known trans-amide-mimicking nature of 1,4-disubstituted triazoles. We optimized a simple and inexpensive protocol for the solid-phase CuAAC, using the commercially available copper(I) bromide−dimethyl sulfide complex that does not require costly CuI ligands such as TBTA.34 This represents a clear advantage in the perspective of developing similar type of molecules for clinical applications. Triazoles bonds were incorporated at either Gly7-Leu8 or Phe6Gly7 sites (compounds 3 and 4, respectively) or at both sites (5), and an azaGly-containing pseudopeptide (6) was also synthesized for comparison. Remarkably, all three triazolocontaining peptides 3−5 had a considerably increased resistance to degradation compared to the more classical azaGly residue (Figure 1). This suggests that Gly-to-azaGly mutation only slows enzymatic hydrolysis of azaGly7-Leu8 and/or Phe6-azaGly7 bonds rather than completely preventing it, as was previously observed for other aza-amino acidcontaining analogues of nonrelated peptides.35−37 Fitting with this hypothesis, traces of peptide fragments corresponding to these two reactions were detected in the LC−MS analysis of

Similar results were obtained with FSH: 1 and 5 nmol/ewe doses had little effect, whereas the 15 nmol/ewe dose consistently resulted in a long-lasting effect (Figure 6). KP10 Analogs Are Selective. KP10 has been reported to be an agonist of NPFF1R.23−25 The ligand−receptor pair RFRP3-NPFF1R has been shown, pending on species, sex, and physiological conditions, to enhance or suppress reproductive system activity.26−30 Therefore, it is important to establish the selectivity of KP analogs toward NPFF1R. Potency of selected analogs was measured by evaluating variation in cAMP intracellular concentration in a cell line stably expressing hNPFF1R and compared to potency at KISS1R. Compounds 2 and 3 EC50 was >5 μM, and therefore, the NPPF1R/KISS1R ratio was superior to 10 000. Compound 17 showed an EC50 at NPFF1R of 222 nM and as a result an NPPF1R/KISS1R ratio of >100. Hence, the three compounds tested in vivo are all highly selective for KISS1R.

■

DISCUSSION AND CONCLUSIONS The present work discloses the design of new KP10 analogs based on original chemical modifications leading to improved pharmacokinetics and pharmacodynamics. These achievements are propaedeutic to therapeutic and reproduction management applications based on KISS1R modulation. All KP analogs reported so far as KISS1R agonists were designed with a focus on improving blood stability by modifying the sites of enzymatic cleavage by proteases. We herein report on the synthesis of a series of molecules dedicated at improving KP10 half-life through reduced susceptibility to proteolysis and, presumably, reduced renal clearance. The free amine N-terminus of KP10 makes it susceptible to rapid catabolism by aminopeptidases. Simple capping with an acetyl group (compound 2) leads to an improved in vitro stability in ewe serum, together with a significant increase of potency at KISS1R. Another major site of degradation is the Gly7-Leu8 bond. To prevent hydrolysis of this specific peptide bond, other 3463

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Figure 6. Dose−response effect of compound 17 on FSH plasma concentration. The time course of LH plasma concentration was analyzed after intravenous injection of three different doses of compound 17: 1 nmol/ewe (A), 5 nmol/ewe (B), or 15 nmol/ewe (C). Basal concentration of FSH was recorded every 10 min for 30 min before injection (time 0). (D) Dose response AUC. Values are the mean ± SEM (N = 4−6 per group). The effect on duration was evaluated using one way ANOVA multiple comparisons test for repeated measurement followed by uncorrected Fisher’s LSD test: (∗) P < 0.05. AUCs were analyzed by one way ANOVA followed by Dunn’s multiple comparison post-test: (∗) P < 0.05, 15 nmol/ewe vs 1 nmol/ewe dose.

a glutamate spacer between the Lys and the hexadecanoyl group (compounds 15 and 16, respectively, Figure 3). The beneficial effect of this modification has already been observed.21 A reasonable explanation of this effect could be a change of the physicochemical properties of the lipopeptide, e.g., reduction of hydrophobic-mediated aggregation, formation of micelles or related structures. Accordingly, compound 17, combining these modifications together with a Gly7-Leu8 triazole, showed an in vitro potency comparable to that of native KP10 and was selected for further in vivo evaluation. Analog 17 stimulated LH increase of larger magnitude and considerably longer duration compared to KP10 or compounds 2 and 3. Remarkably, the increase of LH plasma concentration following administration of analog 17 (15 nmol/ewe) lasted more than 9 h probably because of reduced renal excretion. However, another factor could eventually contribute to improved pharmacodynamics response of analog 17. The introduction of an hexadecanoyl group likely makes compound 17 more hydrophobic than compound 3, suggesting that the former could perhaps more easily cross the blood−brain barrier. On the basis of this hypothesis, compound 3 would mainly activate GnRH release from GnRH terminal outside the blood−brain barrier whereas compound 17 may perhaps activate also GnRH neurons in the brain. Further studies will be necessary to confirm or invalidate this hypothesis. KP effect on FSH release has been less thoroughly investigated than that on LH. However, studies in primate,38 rodents,39 and sheep40 have shown that FSH is generally less

compound 6 after incubation in blood serum. Compound 3, the sole triazole-containing compound to exhibit potency comparable to triazole-free compound 2, was selected for further in vivo evaluation. When injected in anoestrus ewes, both compounds 2 and 3 led to a prolonged increase of LH plasma concentration compared to native KP10 (1, Figure 2). However, considering that these two analogs are at least 35-fold more potent that KP10 and their in vitro serum half-life is considerably improved, the magnitude of both prolongation and increased amplitude of LH plasma level was lower than anticipated. From the analysis of compound 3 dose response (Supporting Information Figures S2 and S3), where no clear difference was observed between doses, it could be surmised that renal clearance is the limiting factor. Previous attempts to create KP analogs with prolonged effect only focused on improving resistance to degradation, not on slowing renal excretion. We thus decided to address this issue and engineered KP10 analogs that include motifs known to reduce renal clearance. To this goal we adopted a strategy taking benefits from the long circulating half-life (∼20 days) of serum albumin. Tethering peptides by reversible noncovalent binding to this protein has been shown to improve their pharmacokinetics.19−21 Initial modification implying derivatization with an albumin-binding N-hexadecanoyl group on the side chain of a Lys residue substituting either Asn2 or Trp3 led to considerable loss of potency of the resulting analogs (13 and 14). However, potency could be rescued by the introduction of 3464

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

compound represents an essential step toward the generation of new treatments to heal diseases related to reduced activity of the gonadotropic axis as well as to manage livestock reproduction.

sensitive to KP stimulation than LH. Nevertheless, since KP infusion induced ovulation in sheep,4,40 it is obvious that the increase of FSH, even though comparatively smaller, is sufficient. Consistent with previous data, our analogs (compounds 3 and 17) preferentially stimulated LH over FSH. The reason for this differential effect of KP and its analogs on gonadotropins could be due to the different GnRH pulse frequency required to elicit LH or FSH secretion. An essential achievement in the design of our analogs is their selectivity toward KISS1R versus NPFF1R, the receptor for RFRP3, a neuropeptide of the RFamide family to which KP also belongs.41 In vitro data on different recombinant systems showed that KP binds and activates NPFF1R even though with modest affinity and potency.23−25 Nevertheless, this is an important issue considering the potential role for RFRP3 in reproduction.42−45 RFRP3 impact on reproduction has been studied in several species with effects that range from inhibition to stimulation pending on species, sex, and reproductive status.26−28 In ewe, an inhibitory role of RFRP3 on LH secretion29,30 has been reported. However, our own experiments in ewe did not replicate this effect of RFRP3 on LH plasma level.46 Regardless of these conflicting results, the availability of a selective molecule is important to clearly separate KISS1R-mediated from potential NPFF1R-mediated effects. Unexpectedly, information on selectivity of previously reported synthetic KP agonists versus NPPF1R is not available. The analogs we tested in vivo (compounds 2, 3, and 17) are selective, with a NPFF1R/KISS1R ratio of >100. In our in vivo experiments the highest dose would lead to a blood concentration of analogs that does not exceed 3 nM. Considering that this concentration is at least 70 times lower than the EC50 of those analogs on NPFF1R, it seems reasonable to assume that effects mediated by compounds 2, 3, and 17 are exclusively accounted for by KISS1R activation. Hence, these compounds could be used to selectively activate KP system without triggering NPFF1R activation. It has been reported that other peptides belonging to the RFamide family (26RFa and its N-terminal extended form 43RFa, also named QRFP43), both binding to the receptor GPR103, elicit an increase of plasma LH concentration in rat.47,48 At present no evidence exists that members of the kisspeptin family could bind to this receptor. Nevertheless, future studies aimed at establishing whether this receptor takes part in the action of our analogs would help to further refine the pharmacological profiles of analogs and shed light on modulation of gonadotropins release. The present work focused on possible application of KP analogs in the area of reproduction. However, KP analogs could also be of interest to treat cancer. For example they could act in the case of sexual hormones-dependent cancer by an indirect action through reduction of GnRH secretion by agonistinduced desensitization or in other cases by a direct antimetastatic action on target tissues expressing KISS1R. Nevertheless, these applications will require the design of agonists with disease-specific characteristics such as the capacity to induce fast receptor desensitization to reduce GnRH secretion or, on the contrary, prolonged receptor activation for direct antimetastatic effect. In conclusion using original modifications dealing with degradation and renal clearance, we have designed new potent and selective KP analogs with improved pharmacokinetics and pharmacodynamics. One of these analogs (17) triggers a prolonged increase of both LH and FSH plasma concentration in the ewe during the anoestrus season. Therefore, this

■

EXPERIMENTAL SECTION

General Information. Unless stated otherwise, all reagents and anhydrous solvents were used without further purification. Protected amino acids, aminomethyl polystyrene, ChemMatrix resin, Rink linker, and HCTU were purchased from Merck Biosciences (Nottingham, U.K.). Peptide synthesis grade DMF was obtained from Applied Biosystems (Courtaboeuf, France). PEG25000-NHS was obtained from Iris Biotech GmbH (Marktredwitz, Germany). Ultrapure water was prepared using a Milli-Q water system from Millipore (Molsheim, France). All other chemicals were purchased from Sigma-Aldrich (St-Quentin-Fallavier, France) and solvents from SDS-Carlo Erba (Val de Reuil, France) and were used without any further purification. Polypropylene syringes fitted with polypropylene frits were obtained from Torviq (Niles, MI, USA) and were equipped with PTFE stopcocks from Chromoptic (Courtaboeuf, France). General Procedure for Peptide Characterization and Purification. HPLC analyses were carried out on a LaChrom Elite system consisting of a Hitachi L-2130 pump, a Hitachi L-2455 diode array detector, and a Hitachi L-2200 autosampler. The machines were equipped with Nucleosil C18 reversed-phase columns, 300 Å, 5 μm, 250 mm × 4.6 mm for analysis (flow rate, 1 mL/min), except when stated otherwise, and 300 Å, 5 μm, 250 mm × 10 mm for purification (flow rate, 3 mL/min). Solvents A and B were 0.1% TFA in H2O and 0.1% TFA in MeCN, respectively. After purification, fractions were pooled, evaporated, and lyophilized to give pure peptides as a white powder. The purity of all peptides tested in cellulo was >95%, and it was >98% for those tested in vivo (C18 RP-HPLC analysis at 280 nm). MS analyses were performed on an Ultraflex MALDI-TOF instrument (Bruker Daltonics, Bremen, Germany) equipped with a 337 nm nitrogen laser and a gridless delayed extraction ion source. External calibration was accomplished using Flex-Control software (Bruker). The sample was cocrystallized with a solution of α-cyano-4hydroxycinnamic acid (HCCA) as a matrix. The observed m/z corresponds to the monoisotopic ions unless stated otherwise. General Procedure for Solid-Phase Peptide Synthesis (SPPS). SPPS was run on an automated synthesizer 433A from Applied Biosystem using Fmoc/tBu chemistry at a 0.1 mmol scale with HCTU as coupling reagents and Rink amide ChemMatrix as solid support. The elongation was carried out automatically using a 10-fold excess of protected amino acids (or acetic acid for N-terminal acetylation), 9.5-fold excess of coupling reagents, and 20-fold excess of iPr2NEt in NMP. Standard side chain protecting groups were used: Arg(Pbf), Asn(Trt), Ser(tBu), Trp(Boc), Tyr(tBu), Thr(tBu), Lys(Boc) unless otherwise mentioned. Fmoc deprotection was performed using a 20% piperidine solution in NMP. The 0.1 mmol scale program purchased from the manufacturer was used, with a single coupling followed by capping with acetic anhydride. The crude peptide was released from the resin with TFA/H2O/iPr3SiH/phenol, 87.5/5/2.5/5 for 2 h, and the peptide was precipitated by dilution into an ice-cold 1:1 diethyl ether/petroleum ether mixture, recovered by centrifugation, and washed 3 times with diethyl ether. Synthesis of the Nonconventional SPPS Building Blocks. The specific building blocks, (S)-2-azido-4-methylpentanoic acid, Fmochydrazine, N-Fmoc-propargylamine, (S)-1-phenylbut-3-yn-2-amine, were prepared as described (see Supporting Information data). The synthesis and full characterization of N-Fmoc-(S)-1-phenylbut-3-yn-2amine are reported in the Supporting Information. General Procedure for Triazole Incorporation through Solid-Supported CuAAC. N-Fmoc-α-aminoalkyne (0.4 mmol, 4 equiv) and CuBr·Me2S (82 mg, 0.4 mmol, 4 equiv) were dissolved in NMP (10 mL) under argon. After addition of iPr2NEt (70 μL, 0.4 mmol, 4 equiv), the mixture was transferred into a syringe fitted with a frit containing azidopeptide resin (0.1 mmol) swollen in NMP. 3465

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

Ac-Lys(Ac)-Asn-Trp-Asn-Ser-Phe-Gly-Leu-Arg-Tyr-NH2 (7). MALDI-TOF MS (m/z): [M + H]+ calcd for C64H91N18O16, 1367.68; found, 1367.70. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 17.5 min. Ac-Tyr-Lys(Ac)-Trp-Asn-Ser-Phe-Gly-Leu-Arg-Tyr-NH2 (8). MALDI-TOF MS (m/z): [M + H]+ calcd for C69H94N17O16: 1416.70, found: 1416.70. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 21.4 min. Ac-Tyr-Asn-Lys(Ac)-Asn-Ser-Phe-Gly-Leu-Arg-Tyr-NH2 (9). MALDI-TOF MS (m/z): [M + H]+ calcd for C62H90N17O17, 1344.67; found, 1344.66. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 12.6 min. Ac-Tyr-Asn-Trp-Lys(Ac)-Ser-Phe-Gly-Leu-Arg-Tyr-NH2 (10). MALDI-TOF MS (m/z): [M + H]+ calcd for C69H94N17O16: 1416.70, found: 1416.70. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 20.3 min. Ac-Tyr-Asn-Trp-Asn-Lys(Ac)-Phe-Gly-Leu-Arg-Tyr-NH2 (11). MALDI-TOF MS (m/z): [M + H]+ calcd for C70H95N18O16: 1443.71, found: 1443.70. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 19.5 min. Ac-Tyr-Lys(PEG25000)-Trp-Asn-Ser-Phe-GlyΨ[Tz]Leu-ArgTyr-NH2 (12). The synthesis of the intermediate Ac-Tyr-Lys-Trp-AsnSer-Phe-GlyΨ[Tz]Leu-Arg-Tyr-NH2 was conducted as for peptide (3). MALDI-TOF MS (m/z): [M + H]+ calcd for C68H92N19O14, 1398.71; found, 1398.71. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 18.0 min. Ac-Tyr-Lys-Trp-Asn-Ser-PheGlyΨ[Tz]Leu-Arg-Tyr-NH2 (2 mM final concentration) was then dissolved in a mixture of 100 mM HEPES buffer, pH 8/MeCN (2:1). Then, PEG25000-NHS (0.7 equiv) dissolved in the same mixture was added to this solution. The reaction mixture was stirred for 6 h at room temperature, and 12 was purified by semipreparative HPLC. MALDI-TOF MS (m/z): [M + H]+ calcd for C77H107N20O18 + 464 × (C2H4O)* (*, n = 464 corresponds to the most intense peak in the MS spectrum of the polydisperse commercially available conjugate), 22041.11; found, 22043.04 (average mass, not monoisotopic). HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 16.4 min.. Ac-Tyr-Lys(hexadecanoyl)-Trp-Asn-Ser-Phe-Gly-Leu-ArgTyr-NH2 (13). The elongation of the peptide was performed by standard automated solid phase synthesis using Fmoc-Lys(Dde)-OH as a nonstandard building block. After elongation of the peptide, the resin was treated with 2% hydrazine monohydrate in NMP (2 × 10 mL) for 5 min. Then, hexadecanoic acid (10 equiv) was coupled on lysine side chain using HCTU (9.5 equiv) and iPr2NEt (20 equiv) in NMP/CH2Cl2 (1:4). MALDI-TOF MS (m/z): [M + H]+ calcd for C83H122N17O16, 1612.92; found, 1612.91. HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 24.6 min. Ac-Tyr-Asn-Lys(hexadecanoyl)-Asn-Ser-Phe-Gly-Leu-ArgTyr-NH2 (14). The synthesis of 14 was conducted as described for peptide 13. MALDI-TOF MS (m/z): [M + H]+ calcd for C76H118N17O17, 1540.89; found, 1540.94. HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 19.2 min. Ac-Tyr-Lys(γ-(N-hexadecanoyl-Glu-OH))-Trp-Asn-Ser-PheGly-Leu-Arg-Tyr-NH2 (15). The elongation of the peptide was performed by standard automated solid phase synthesis using FmocLys(Dde)-OH as a nonstandard building block. After elongation of the peptide, the resin was treated with 2% hydrazine monohydrate in NMP (2 × 10 mL) for 5 min. Then, Fmoc-Glu-OtBu (10 equiv) was coupled on amine-containing peptide resin by standard automated SPPS protocol followed by standard Fmoc deprotection. Then, hexadecanoic acid (10 equiv) was coupled using HCTU (9.5 equiv) and iPr2NEt (20 equiv) in NMP/CH2Cl2 (1:4). MALDI-TOF MS (m/z): [M + H]+ calcd for C88H129N18O19, 1741.97; found, 1742.00. HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 20.1 min. Ac-Tyr-Asn-Lys(γ-(N-hexadecanoyl-Glu-OH))-Asn-Ser-PheGly-Leu-Arg-Tyr-NH2 (16). The synthesis of 16 was conducted as described for peptide 15. MALDI-TOF MS (m/z): [M + H]+ calcd for C81H125N18O20, 1669.93; found, 1669.96. HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 16.9 min.

The suspension was stirred by syringe rotation for 2 h at room temperature, and the resin was flow-washed successively with NMP (3 × 2 min), CH2Cl2 (2 × 2 min), 1 M pyridine hydrochloride in CH2Cl2/MeOH 95:5 (2 × 2 min), CH2Cl2 (2 × 2 min), and DMF (2 × 2 min). Elongation of the peptide was completed by standard automated solid phase synthesis. Synthesis of the KISS1R Agonists. H-Tyr-Asn-Trp-Asn-SerPhe-Gly-Leu-Arg-Tyr-NH2 (1) (KP10). The elongation of the peptide was performed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C63H84N17O15, 1318.6; found, 1318.6. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 18.4 min. Ac-Tyr-Asn-Trp-Asn-Ser-Phe-Gly-Leu-Arg-Tyr-NH2 (2). The elongation of the peptide was performed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C65H86N17O16, 1360.6; found, 1360.6. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 21.9 min. Ac-Tyr-Asn-Trp-Asn-Ser-Phe-GlyΨ[Tz]Leu-Arg-Tyr-NH2 (3). The elongation of the peptide was performed by standard automated solid phase synthesis up to Arg9. The coupling of (2S)-2-azido-4methylpentanoic acid was also performed by the automated procedure. Triazole formation was performed by reaction with Fmoc-propargylamine following the general solid supported CuAAC protocol. Elongation of the peptide was completed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C66H86N19O15, 1384.7; found, 1384.6. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 20.5 min. Ac-Tyr-Asn-Trp-Asn-Ser-PheΨ[Tz]Gly-Leu-Arg-Tyr-NH2 (4). The elongation of the peptide was performed by standard automated solid phase synthesis up to Gly7. After Fmoc deprotection, the peptide resin was stirred for 1 h with the diazo transfer reagent 1H-imidazole1-sulfonyl azide·H2SO446 (5 equiv) and K2CO3 (10 equiv) dissolved in DMF/H2O (3/7, 3 mL). Caution: gas evolution, need venting of the syringe. The resin was washed with DMF, and Fmoc-(S)-1-phenylbut3-yn-2-amine was coupled using solid supported CuAAC protocol. Elongation of the peptide was completed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C66H86N19O15, 1384.65; found, 1384.65. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 20.8 min. Ac-Tyr-Asn-Trp-Asn-Ser-PheΨ[Tz]GlyΨ[Tz]Leu-Arg-Tyr-NH2 (5). The elongation of the peptide was performed by standard automated solid phase synthesis up to Arg9. The coupling of (2S)-2azido-4-methylpentanoic acid was also performed by the automated procedure. Triazole formation was performed by reaction with Fmocpropargylamine following the general solid supported CuAAC protocol. After Fmoc deprotection, the peptide resin was stirred for 1 h with the diazo transfer reagent: 1H-imidazole-1-sulfonyl azide· H2SO449 (5 equiv) and K2CO3 (10 equiv) dissolved in DMF/H2O (3/7, 3 mL). Caution: gas evolution, need venting of the syringe. The resin was washed with DMF, and Fmoc-(S)-1-phenylbut-3-yn-2amine was coupled using solid supported CuAAC protocol. Elongation of the peptide was completed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C67H86N21O14: 1408.66, found: 1408.66. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 21.9 min. Ac-Tyr-Asn-Trp-Asn-Ser-Phe-azaGly-Leu-Arg-Tyr-NH2 (6). The elongation of the peptide was performed by standard automated solid phase synthesis up to Leu8. Separately, Fmoc-NHNH2 (10 equiv) was suspended in DMF (1 mL), and under ice cooling a solution of CDI (9.5 equiv) in dry THF (10 mL) was added. Subsequently, iPrNEt2 (20 equiv) was added dropwise to the mixture, followed by stirring at room temperature for 1h. The resulting solution was added to the peptide resin washed with dry THF, followed by stirring at room temperature for 16 h. Elongation of the peptide was completed by standard automated solid phase synthesis. MALDI-TOF MS (m/z): [M + H]+ calcd for C64H85N18O16, 1361.63; found, 1361.63. HPLC: analytical gradient, 20−40% B/A over 30 min; retention time, 19.1 min. Lys(Ac) Scan (Compounds 7−11). The elongation of the peptide was performed by standard automated solid phase synthesis using Fmoc-Lys(Ac)-OH as a nonstandard building block. 3466

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

peptide being obtained by mixing 420 μL of serum with 5 μL of a commercially available protease inhibitor cocktail (Sigma-Aldrich, St-Quentin-Fallavier, France, ref P8340, solution in DMSO) and 25 μL of solution of phenylalaninol, followed by 50 μL of the KP10 analog and then immediately treating the resulting solution using the protocol described above. Evaluation of Analog Effects on LH and FSH Plasma Concentration. Experiments were conducted using intact acyclic anestrous ewes (March-June) of Ile-de-France breed (3−5 years old, weighing 55−70 kg) maintained under normal husbandry at the INRA research center (Unité Expérimentale PAO No. 1297 (EU0028) of INRA, Nouzilly, France). Experiments were carried out in accordance with national and international regulations (Directive 2010/63/UE, Authorization No. A 38801 and E37-175-2 of the French Ministry of Agriculture), and all procedures were approved by the local Animal Ethics Committee (Comité d’Ethique en Expérimentation Animale Val de Loire) (Authorization No 2012-03-7). The day before the experiment ewes were placed in small contiguous pens and a catheter (i.d. of 1.0, L = 52 mm, Intraflon 2, Vygon, France) was inserted in the jugular vein. On the day of experiment test compound was injected intravenously at the desired dose (ranging from 1 to 15 nmol) and blood samples (2 mL) were taken every 10 min for the first 2 h, then every 20 min for an additional 4 h and finally every hour until the end of the experiment. Data are expressed as the mean ± SEM (N = 4−6). Hormone Assays. Plasma LH concentrations were measured using a previously described radioimmunoassay,51,52 with all samples from one experiment included in a single assay. The assay standard was 1051-CY-LH (equivalent to 0.31 NIH-LH-S1). Intraassay and interassay coefficient of variation averaged 9% and 15%, respectively; assay sensitivity was 0.1 ± 0.05 ng/mL (four assays). Plasma FSH levels were measured using the reagents supplied by Tucker Endocrine Research Institute (Atlanta, GA, USA). Intraassay and interassay coefficient of variation averaged 7.6% and 9%, respectively; assay sensitivity was 0.1 ng/mL relative to the standard (Tuenere oFSHstd.1 equiv to 1.0 NIH-FSH-S1). The cross reactivity with ovine LH was 0.03%. cAMP Assay on hNPFF1R Cell Line. CHO cell line stably transfected with hNPFF1R was a kind gift from Dr. F. Simonin (ESBS, Illkirch, France). hNPFF1R is a Gαi coupled receptor. To monitor receptor activation, the cell line expressing the receptor was transfected with GloSensor (Promega, Charbonnières-les-Bains, France) following the manufacturer’s instructions. The assay uses a genetically encoded biosensor variant with cAMP binding domain fused to a mutant form of Photinus pyralis luciferase and allows for kinetic evaluation of receptor response. Cell were grown on DMEM/F12 (50/50) containing 10% fetal calf serum, 1% penicillin, 1% streptomycin, 600 μg/mL Geneticin, and 100 μg/mL hygromycin B. Cells were seeded at 30 000 cells/well in 96-well μclear F-bottom, white plate (Dutscher, Brumath, France) 24 h before the experiment. On the day of experiment medium was removed and plate washed with phosphate buffered saline twice before incubating in GloSensor cAMP reagent for 2 h. Then test compounds, prepared as described above, were added, luminescence was recorded with a plate reader (PolarStar Optima, BMG Labtech) for 15 min followed by addition of a solution containing 300 nM forskolin and 0.25 mM IBMX and luminescence recorded for additional 120 min. Data were analyzed as described in the calcium mobilization assay section.

Ac-Tyr-Lys(γ-(N-hexadecanoyl-Glu-OH))-Trp-Asn-Ser-PheGlyΨ[Tz]Leu-Arg-Tyr-NH2 (17). The elongation of the peptide was performed by standard automated solid phase synthesis up to Arg9. The introduction of the triazole moiety was performed as described for the synthesis of peptide 3 and the introduction of γ-(N-hexadecanoylL-glutamyl) moiety as for the synthesis of peptide 15. MALDI-TOF MS (m/z): [M + H]+ calcd for C89H129N20O18, 1765.97; found, 1765.99. HPLC: analytical gradient, 45−75% B/A over 30 min; retention time, 20.1 min. Ac-Tyr-Asn-Lys(γ-(N-hexadecanoyl-Glu-OH))-Asn-Ser-PheGlyΨ[Tz]Leu-Arg-Tyr-NH2 (18). The elongation of the peptide was performed by standard automated solid phase synthesis up to Arg9. The introduction of the triazole moiety was performed as described for the synthesis of peptide 3 and the introduction of γ-(N-hexadecanoylL-glutamyl) moiety as for the synthesis of peptide 15. MALDI-TOF MS (m/z): [M + H]+ calcd for C82H125N20O19, 1693.94; found, 1693.97. HPLC: analytical gradient, 45−75% B/A over 6 min (column, Chromolith High Resolution RP-18e, 150 Å, 10 mm × 4.6 mm; flow rate, 3 mL/min); retention time, 3.02 min. Calcium Mobilization Assay. Wild-type human KISS1R receptor was cloned into the pcDNA3.1 expression vector (Invitrogen, Cergy Pontoise, France) in fusion at its 5′ end with a HA tag. The resulting construct was transfected in HEK293 cell line and selected for stable expression using Geneticin.50 HEK293 transfected cells were grown in DMEM (with GlutaMAX, high glucose, and without pyruvate), 10% fetal calf serum, 1% penicillin, 1% streptomycin, 200 μg/mL Geneticin, and HEPES (25 mM). KISS1R is a Gαq-coupled receptor, and to assess analogs potency and efficacy, the dynamics of intracellular Ca2+ mobilization induced by test compound application was monitored using Fluo4 NW Ca2+ assay kit. Cells were plated 48 h before the experiment into 96-well blackplate (Dutscher, Brumath, France) at a concentration of 40 000 cells/well. On the day of the experiment test compounds were diluted from stock solution in LoBind Tube (Eppendorf, Hamburg, Germany) or in nonbinding plate (VWR, Strasbourg, France) to 20× the final desired concentration (ranging from 1 pM to 1 μM). The medium was discharged, cell rinsed once with PBS, and incubated with the kit’s dye (95 μL/well) for 30 min at 37 °C and 30 min at rt. Basal fluorescence was measured 5 times at 7 s interval with a plate reader (PolarStar Optima, BMG Labtech). Immediately after basal reading an amount of 5 μL of test compound was added to each well to obtain the final test concentration. Intracellular Ca2+ dynamic was monitored for 7 min. To generate concentration activity curves, mean basal was subtracted to value obtained after stimulation. The area under the curve (AUC) was calculated and plotted against concentrations. Concentration activity data points were fitted to a sigmoid curve generated by GraphPad Prism 5, and EC50 was automatically calculated. To check for nonspecific signal, control experiments were performed in nontransfected HEK293 cells using the same method. Evaluation of KP10 Analogs Stability in Ewe Serum. Blood was collected from the jugular vein of Ile de France ewes, centrifuged and serum collected, stored at −20 °C, and thawed just before use. Five independent experiments were carried out for each analog. Peptides and L-phenylalaninol (Sigma-Aldrich, St-Quentin-Fallavier, France) were dissolved in Milli-Q water (0.5 and 6.6 mM, respectively). Serum, stock solutions of KP10 analogs, and phenylalaninol were preheated at 39 °C for 30 min. KP10 analog (50 μL, final concentration in serum of 50 μM) was mixed with 425 μL of serum and 25 μL of phenylalaninol and then incubated at 39 °C for 6 h. An amount of 75 μL of the solution was then diluted with 150 μL of MeCN to precipitate serum proteins. The suspension obtained was centrifuged at 14 000 rpm during 10 min at 4 °C. An amount of 100 μL of the supernatant was diluted in 900 μL of a 0.1% TFA solution in water, then analyzed by HPLC (column, Chromolith RP-18 end-capped High Resolution, 3 mL/min, gradient, 2−52% MeCN/H2O + 0.1% TFA over 5 min, UV detection at λ = 214 nm, and mass spectrometry). The amount of intact KP10 analog after incubation was determined by integrating the corresponding peak, using L-phenylalaninol as an internal calibrator. It is expressed in % of the peak area at time t = 0, the value corresponding to 100% of intact

■

ASSOCIATED CONTENT

S Supporting Information *

Synthetic protocols and analytical data for the nonconventional SPPS building blocks, detailed optimization of the catalytic system used for solid-supported CuAAC, HPLC chromatograms of peptides 1−18, and five figures showing the concentration−activity response of compounds 1−3 potency and efficacy at KISS1R, the dose−response effect of compound 3 on LH plasma concentration, the dose−response effect of compound 3 on FSH plasma concentration, the dose−response 3467

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry

M. Characterization of the potent luteinizing hormone-releasing activity of KISS-1 peptide, the natural ligand of GPR54. Endocrinology 2005, 146, 156−163. (4) Caraty, A.; Smith, J. T.; Lomet, D.; Ben Said, S.; Morrissey, A.; Cognie, J.; Doughton, B.; Baril, G.; Briant, C.; Clarke, I. J. Kisspeptin synchronizes preovulatory surges in cyclical ewes and causes ovulation in seasonally acyclic ewes. Endocrinology 2007, 148, 5258−5267. (5) Chan, Y. M.; Butler, J. P.; Pinnell, N. E.; Pralong, F. P.; Crowley, W. F., Jr.; Ren, C.; Chan, K. K.; Seminara, S. B. Kisspeptin resets the hypothalamic GnRH clock in men. J. Clin. Endocrinol. Metab. 2011, 96, E908−E915. (6) Matsui, H.; Tanaka, A.; Yokoyama, K.; Takatsu, Y.; Ishikawa, K.; Asami, T.; Nishizawa, N.; Suzuki, A.; Kumano, S.; Terada, M.; Kusaka, M.; Kitada, C.; Ohtaki, T. Chronic administration of the metastin/ kisspeptin analog KISS1-305 or the investigational agent TAK-448 suppresses hypothalamic pituitary gonadal function and depletes plasma testosterone in adult male rats. Endocrinology 2012, 153, 5297− 5308. (7) Scott, G.; Ahmad, I.; Howard, K.; MacLean, D.; Oliva, C.; Warrington, S.; Wilbraham, D.; Worthington, P. Double-blind, randomized, placebo-controlled study of safety, tolerability, pharmacokinetics and pharmacodynamics of TAK-683, an investigational metastin analogue in healthy men. Br. J. Clin. Pharmacol. 2013, 75, 381−391. (8) Asami, T.; Nishizawa, N.; Ishibashi, Y.; Nishibori, K.; Nakayama, M.; Horikoshi, Y.; Matsumoto, S.; Yamaguchi, M.; Matsumoto, H.; Tarui, N.; Ohtaki, T.; Kitada, C. Serum stability of selected decapeptide agonists of KISS1R using pseudopeptides. Bioorg. Med. Chem. Lett. 2012, 22, 6391−6396. (9) Curtis, A. E.; Cooke, J. H.; Baxter, J. E.; Parkinson, J. R.; Bataveljic, A.; Ghatei, M. A.; Bloom, S. R.; Murphy, K. G. A kisspeptin10 analog with greater in vivo bioactivity than kisspeptin-10. Am. J. Physiol.: Endocrinol. Metab. 2010, 298, E296−E303. (10) Takino, T.; Koshikawa, N.; Miyamori, H.; Tanaka, M.; Sasaki, T.; Okada, Y.; Seiki, M.; Sato, H. Cleavage of metastasis suppressor gene product KISS-1 protein/metastin by matrix metalloproteinases. Oncogene 2003, 22, 4617−4626. (11) Valverde, I. E.; Lecaille, F.; Lalmanach, G.; Aucagne, V.; Delmas, A. F. Synthesis of a biologically active triazole-containing analogue of cystatin A through successive peptidomimetic alkyne-azide ligations. Angew. Chem., Int. Ed. 2012, 51, 718−722. (12) Aucagne, V.; Valverde, I. E.; Marceau, P.; Galibert, M.; Dendane, N.; Delmas, A. F. Towards the simplification of protein synthesis: iterative solid-supported ligations with concomitant purifications. Angew. Chem., Int. Ed. 2012, 51, 11320−11324. (13) Ko, E.; Liu, J.; Perez, L. M.; Lu, G.; Schaefer, A.; Burgess, K. Universal peptidomimetics. J. Am. Chem. Soc. 2011, 133, 462−477. (14) Tornøe, C. W.; Christensen, C.; Meldal, M. Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)-catalyzed 1,3dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 2002, 67, 3057−3064. (15) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A stepwise Huisgen cycloaddition process: copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew. Chem., Int. Ed. 2002, 41, 2596−2599. (16) Orsini, M. J.; Klein, M. A.; Beavers, M. P.; Connolly, P. J.; Middleton, S. A.; Mayo, K. H. Metastin (KISS-1) mimetics identified from peptide structure-activity relationship-derived pharmacophores and directed small molecule database screening. J. Med. Chem. 2007, 50, 462−471. (17) Gutiérrez-Pascual, E.; Leprince, J.; Martinez-Fuentes, A. J.; Ségalas-Milazzo, I.; Pineda, R.; Roa, J.; Duran-Prado, M.; Guilhaudis, L.; Desperrois, E.; Lebreton, A.; Pinilla, L.; Tonon, M.-C.; Malagón, M. M.; Vaudry, H.; Tena-Sempere, M.; Castano, J. P. In vivo and in vitro structure-activity relationship and structural conformation of kisspeptin-10-related peptides. Mol. Pharmacol. 2009, 76, 58−67. (18) Harris, J. M.; Chess, R. B. Effect of pegylation on pharmaceuticals. Nat. Rev. Drug Discovery 2003, 2, 214−221.

effect of PEGylated compound 12 on LH plasma concentration, and the LC−MS evidence of cleavage of the serumincubated azaGly-containing peptide 6 at both sides of the azaGly residue. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*M.B.: phone, +33 (0)2 47 42 73 60; fax, + 33 (0)2 47 42 77; e-mail, [email protected]. *V.A.: phone, +33 (0)2 38 25 55 77; fax, +33 (0)2 38 63 15 17; e-mail, [email protected]. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The Région Centre council is greatly acknowledged for financial support (Reprokiss Project). We thank the CBM spectrometric platforms and in particular Dr. Guillaume Gabant for recording most of the MALDI-TOF spectra, Dr. Cyril Colas for LC−MS, and Hervé Meudal for the NMR spectra. CHO cell lines stably transfected with hNPFF1R was a kind gift from Dr. F. Simonin (ESBS, Illkirch, France). We thank Gwenhael Jegot for her precious assistance with the set up of the cAMP assay. We warmly thank CIRE personnel for assistance with jugular cannulation and the personnel of the Unité Expérimentale PAO No. 1297 (EU0028) of INRA, Centre Val de Loire for help with animal husbandry.

■

ABBREVIATIONS USED Ac, acetyl; AUC, area under the curve; cAMP, cyclic adenosine monophosphate; CDI, 1,1′-carbonyldiimidazole; CuAAC, copper-catalyzed alkyne/azide cycloaddition; CuTC, copper(I) thiophene-2-carboxylate; Dde, N-[l-(4,4-dimethyl-2,6dioxocyclohexylidene)ethyl]; DMEM, Dulbecco’s modified Eagle medium; DMF, dimethylformamide; FSH, folliclestimulating hormone; HCTU, 1-[bis(dimethylamino)methylene]-5-chlorobenzotriazolium 3-oxide hexafluorophosphate; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; KP, kisspeptin; LH, luteinizing hormone; NHS, Nhydroxysuccinimide ester; NMP, N-methyl-2-pyrrolidone; SPPS, solid-phase peptide synthesis; TBTA, tris[(1-benzyl1H-1,2,3-triazol-4-yl)methyl]amine; TFA, trifluoroacetic acid; Tz, 4-(1,2,3-triazol-1-yl)

■

REFERENCES

(1) de Roux, N.; Genin, E.; Carel, J. C.; Matsuda, F.; Chaussain, J. L.; Milgrom, E. Hypogonadotropic hypogonadism due to loss of function of the KISS1-derived peptide receptor GPR54. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10972−10976. (2) Seminara, S. B.; Messager, S.; Chatzidaki, E. E.; Thresher, R. R.; Acierno, J. S., Jr.; Shagoury, J. K.; Bo-Abbas, Y.; Kuohung, W.; Schwinof, K. M.; Hendrick, A. G.; Zahn, D.; Dixon, J.; Kaiser, U. B.; Slaugenhaupt, S. A.; Gusella, J. F.; O’Rahilly, S.; Carlton, M. B.; Crowley, W. F., Jr.; Aparicio, S. A.; Colledge, W. H. The GPR54 gene as a regulator of puberty. N. Engl. J. Med. 2003, 349, 1614−1627. (3) Navarro, V. M.; Castellano, J. M.; Fernandez-Fernandez, R.; Tovar, S.; Roa, J.; Mayen, A.; Nogueiras, R.; Vazquez, M. J.; Barreiro, M. L.; Magni, P.; Aguilar, E.; Dieguez, C.; Pinilla, L.; Tena-Sempere, 3468

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry (19) Kurtzhals, P.; Havelund, S.; Jonassen, I.; Kiehr, B.; Larsen, U. D.; Ribel, U.; Markussen, J. Albumin binding of insulins acylated with fatty acids: characterization of the ligand-protein interaction and correlation between binding affinity and timing of the insulin effect in vivo. Biochem. J. 1995, 312, 725−731. (20) Markussen, J.; Havelund, S.; Kurtzhals, P.; Andersen, A. S.; Halstrøm, J.; Hasselager, E.; Larsen, U. D.; Ribel, U.; Schäffer, L.; Vad, K.; Jonassen, I. Soluble, fatty acid acylated insulins bind to albumin and show protracted action in pigs. Diabetologia 1996, 39, 281−288. (21) Bellmann-Sickert, K.; Elling, C. E.; Madsen, A. N.; Little, P. B.; Lundgren, K.; Gerlach, L. O.; Bergmann, R.; Holst, B.; Schwartz, T. W.; Beck-Sickinger, A. G. Long-acting lipidated analogue of human pancreatic polypeptide is slowly released into circulation. J. Med. Chem. 2011, 54, 2658−2667. (22) Knudsen, L. B.; Nielsen, P. F.; Huusfeldt, P. O.; Johansen, N. L.; Madsen, K.; Pedersen, F. Z.; Thøgersen, H.; Wilken, M.; Agersø, H. Potent derivatives of glucagon-like peptide-1 with pharmacokinetic properties suitable for once daily administration. J. Med. Chem. 2000, 43, 1664−1669. (23) Lyubimov, Y.; Engstrom, M.; Wurster, S.; Savola, J. M.; Korpi, E. R.; Panula, P. Human kisspeptins activate neuropeptide FF2 receptor. Neuroscience 2010, 170, 117−122. (24) Oishi, S.; Misu, R.; Tomita, K.; Setsuda, S.; Masuda, R.; Ohno, H.; Naniwa, Y.; Ieda, N.; Inoue, N.; Ohkura, S.; Uenoyama, Y.; Tsukamura, H.; Maeda, K.; Hirasawa, A.; Tsujimoto, G.; Fujii, N. Activation of neuropeptide FF receptor by kisspeptin receptor ligand. ACS Med. Chem. Lett. 2011, 2, 53−57. (25) Elhabazi, K.; Humbert, J. P.; Bertin, I.; Schmitt, M.; Bihel, F.; Bourguignon, J. J.; Bucher, B.; Becker, J. A.; Sorg, T.; Meziane, H.; Petit-Demouliere, B.; Ilien, B.; Simonin, F. Endogenous mammalian RF-amide peptides, including PrRP, kisspeptin and 26RFa, modulate nociception and morphine analgesia via NPFF receptors. Neuropharmacology 2013, 75C, 164−171. (26) Kriegsfeld, L. J.; Mei, D. F.; Bentley, G. E.; Ubuka, T.; Mason, A. O.; Inoue, K.; Ukena, K.; Tsutsui, K.; Silver, R. Identification and characterization of a gonadotropin-inhibitory system in the brains of mammals. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 2410−2415. (27) Ancel, C.; Bentsen, A. H.; Sebert, M. E.; Tena-Sempere, M.; Mikkelsen, J. D.; Simonneaux, V. Stimulatory effect of RFRP-3 on the gonadotrophic axis in the male Syrian hamster: the exception proves the rule. Endocrinology 2012, 153, 1352−1363. (28) Ubuka, T.; Inoue, K.; Fukuda, Y.; Mizuno, T.; Ukena, K.; Kriegsfeld, L. J.; Tsutsui, K. Identification, expression, and physiological functions of siberian hamster gonadotropin-inhibitory hormone. Endocrinology 2012, 153, 373−385. (29) Clarke, I. J.; Sari, I. P.; Qi, Y.; Smith, J. T.; Parkington, H. C.; Ubuka, T.; Iqbal, J.; Li, Q.; Tilbrook, A.; Morgan, K.; Pawson, A. J.; Tsutsui, K.; Millar, R. P.; Bentley, G. E. Potent action of RFamiderelated peptide-3 on pituitary gonadotropes indicative of a hypophysiotropic role in the negative regulation of gonadotropin secretion. Endocrinology 2008, 149, 5811−5821. (30) Smith, J. T.; Coolen, L. M.; Kriegsfeld, L. J.; Sari, I. P.; Jaafarzadehshirazi, M. R.; Maltby, M.; Bateman, K.; Goodman, R. L.; Tilbrook, A. J.; Ubuka, T.; Bentley, G. E.; Clarke, I. J.; Lehman, M. N. Variation in kisspeptin and RFamide-related peptide (RFRP) expression and terminal connections to gonadotropin-releasing hormone neurons in the brain: a novel medium for seasonal breeding in the sheep. Endocrinology 2008, 149, 5770−5782. (31) Tomita, K.; Oishi, S.; Ohno, H.; Peiper, S. C.; Fujii, N. Development of novel G-protein-coupled receptor 54 agonists with resistance to degradation by matrix metalloproteinase. J. Med. Chem. 2008, 51, 7645−7649. (32) Asami, T.; Nishizawa, N.; Matsui, H.; Nishibori, K.; Ishibashi, Y.; Horikoshi, Y.; Nakayama, M.; Matsumoto, S.; Tarui, N.; Yamaguchi, M.; Matsumoto, H.; Ohtaki, T.; Kitada, C. Design, synthesis, and biological evaluation of novel investigational nonapeptide KISS1R agonists with testosterone-suppressive activity. J. Med. Chem. 2013, 56, 8298−8307.

(33) Asami, T.; Nishizawa, N.; Matsui, H.; Takatsu, Y.; Suzuki, A.; Kiba, A.; Terada, M.; Nishibori, K.; Nakayama, M.; Ban, J.; Matsumoto, S.; Tarui, N.; Ikeda, Y.; Yamaguchi, M.; Kusaka, M.; Ohtaki, T.; Kitada, C. Physicochemically and pharmacokinetically stable nonapeptide KISS1 receptor agonists with highly potent testosterone-suppressive activity. J. Med. Chem. 2014, 57, 6105−6115. (34) Valverde, I. E.; Bauman, A.; Kluba, C. A.; Vomstein, S.; Walter, M. A.; Mindt, T. L. 1,2,3-Triazoles as amide bond mimics: triazole scan yields protease-resistant peptidomimetics for tumor targeting. Angew. Chem., Int. Ed. 2013, 52, 8957−8960. (35) Dutta, A. S.; Giles, M. B. Polypeptides. Part XIV. A comparative study of the stability towards enzymes of model tripeptides containing alpha-aza-amino-acids, L-amino-acids, and D-amino-acids. J. Chem. Soc., Perkin Trans. 1 1976, 244−248. (36) Hoitink, M. A.; Beijnen, J. H.; Boschma, M. U. S.; Bult, A.; Hop, E.; Nijholt, J.; Versluis, C.; Wiese, G.; Underberg, W. J. M. Identification of the degradation products of gonadorelin and three analogues in aqueous solution. Anal. Chem. 1997, 69, 4972−4978. (37) Tal-Gan, Y.; Freeman, N. S.; Klein, S.; Levitzki, A.; Gilon, C. Metabolic stability of peptidomimetics: N-methyl and aza heptapeptide analogs of a PKB/Akt inhibitor. Chem. Biol. Drug. Des. 2011, 78, 887−892. (38) Plant, T. M.; Ramaswamy, S.; Dipietro, M. J. Repetitive activation of hypothalamic G protein-coupled receptor 54 with intravenous pulses of kisspeptin in the juvenile monkey (Macaca mulatta) elicits a sustained train of gonadotropin-releasing hormone discharges. Endocrinology 2006, 147, 1007−1013. (39) Navarro, V. M.; Castellano, J. M.; Fernandez-Fernandez, R.; Tovar, S.; Roa, J.; Mayen, A.; Barreiro, M. L.; Casanueva, F. F.; Aguilar, E.; Dieguez, C.; Pinilla, L.; Tena-Sempere, M. Effects of KISS-1 peptide, the natural ligand of GPR54, on follicle-stimulating hormone secretion in the rat. Endocrinology 2005, 146, 1689−1697. (40) Sebert, M. E.; Lomet, D.; Said, S. B.; Monget, P.; Briant, C.; Scaramuzzi, R. J.; Caraty, A. Insights into the mechanism by which kisspeptin stimulates a preovulatory LH surge and ovulation in seasonally acyclic ewes: potential role of estradiol. Domest. Anim. Endocrinol. 2010, 38, 289−298. (41) Tsutsui, K.; Bentley, G. E.; Kriegsfeld, L. J.; Osugi, T.; Seong, J. Y.; Vaudry, H. Discovery and evolutionary history of gonadotrophininhibitory hormone and kisspeptin: new key neuropeptides controlling reproduction. J. Neuroendocrinol. 2010, 22, 716−727. (42) Clarke, I. J.; Qi, Y.; Puspita Sari, I.; Smith, J. T. Evidence that RF-amide related peptides are inhibitors of reproduction in mammals. Front. Neuroendocrinol. 2009, 30, 371−378. (43) Kriegsfeld, L. J.; Gibson, E. M.; Williams, W. P., 3rd; Zhao, S.; Mason, A. O.; Bentley, G. E.; Tsutsui, K. The roles of RFamide-related peptide-3 in mammalian reproductive function and behaviour. J. Neuroendocrinol. 2010, 22, 692−700. (44) Tsutsui, K.; Bentley, G. E.; Bedecarrats, G.; Osugi, T.; Ubuka, T.; Kriegsfeld, L. J. Gonadotropin-inhibitory hormone (GnIH) and its control of central and peripheral reproductive function. Front. Neuroendocrinol. 2010, 31, 284−295. (45) Simonneaux, V.; Ancel, C.; Poirel, V. J.; Gauer, F. Kisspeptins and RFRP-3 act in concert to synchronize rodent reproduction with seasons. Front. Neurosci. 2013, 7, 22. (46) Caraty, A.; Blomenröhr, M.; Vogel, G. M. T.; Lomet, D.; Briant, C.; Beltramo, M. RF9 powerfully stimulates gonadotrophin secretion in the ewe: Evidence for a seasonal threshold of sensitivity. J. Neuroendocrinol. 2012, 24, 725−736. (47) Navarro, V. M.; Férnandez-Férnandez, R.; Nogueiras, R.; Vigo, E.; Tovar, S.; Chartrel, N.; Le Marec, O.; Leprince, J.; Aguilar, E.; Pinilla, L.; Dieguez, C.; Vaudry, H.; Tena-Sempere, M. Novel role of 26RFa, a hypothalamic RFamide orexigenic peptide, as putative regulator of the gonadotropic axis. J. Physiol. 2006, 573.1, 237−249. (48) Patel, S. R.; Murphy, K. G.; Thompson, E. L.; Patterson, M.; Curtis, A. E.; Ghatei, M. A.; Bloom, S. R. Pyroglutamylated RFamide peptide 43 stimulates the hypothalamic-pituitary-gonadal axis via gonadotropin-releasing hormone in rats. Endocrinology 2008, 149 (9), 4747−4754. 3469

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470

Article

Journal of Medicinal Chemistry (49) Fischer, N.; Goddard-Borger, E. D.; Greiner, R.; Klapötke, T. M.; Skelton, B. W.; Stierstorfer, J. Sensitivities of some imidazole-1sulfonyl azide salts. J. Org. Chem. 2012, 77, 1760−1764. (50) Chevrier, L.; de Brevern, A.; Hernandez, E.; Leprince, J.; Vaudry, H.; Guedj, A. M.; de Roux, N. PRR repeats in the intracellular domain of KISS1R are important for its export to cell membrane. Mol. Endocrinol. 2013, 27, 1004−1014. (51) Pelletier, J.; Kann, G.; Dolais, J.; Rosselin, G. Radioimmunologic determination of plasma lutenizing hormone in sheep. Comparison with the biologic determination of LH by the diminution of ovarian ascorbic acid, and example of the application to measure of the blood LH in ewes. C. R. Hebd. Seances Acad. Sci., Ser. D 1968, 266, 2352−2354. (52) Montgomery, G. W.; Martin, G. B.; Pelletier, J. Changes in pulsatile LH secretion after ovariectomy in Ile-de-France ewes in two seasons. J. Reprod. Fertil. 1985, 73, 173−183.

3470

DOI: 10.1021/jm5019675 J. Med. Chem. 2015, 58, 3459−3470