J. Med. Chem. 2009, 52, 2157–2160
2157
Novel Peptidomimetics Containing a Vinyl Ester Moiety as Highly Potent and Selective Falcipain-2 Inhibitors Roberta Ettari,*,† Nicola Micale,† Tanja Schirmeister,‡ Christoph Gelhaus,§ Matthias Leippe,§ Emanuela Nizi,| Maria Emilia Di Francesco,| Silvana Grasso,† and Maria Zappala`† Dipartimento Farmaco-Chimico, UniVersity of Messina, Viale Annunziata, 98168 Messina, Italy, Institute of Pharmacy and Food Chemistry, UniVersity of Wu¨rzburg, Am Hubland, D-97074 Wu¨rzburg, Germany, Institute of Zoology, UniVersity of Kiel, Olshausenstrasse 40, D-24098 Kiel, Germany, Department of Medicinal Chemistry IRBM, MRL Rome, Via Pontina Km30.600, 00040 Pomezia, Roma, Italy ReceiVed January 17, 2009
This paper describes the synthesis and biological evaluation of a new class of peptidomimetic falcipain-2 inhibitors based on a 1,4-benzodiazepine scaffold combined with various R,β-unsaturated electrophilic functions such as vinyl-ketone, -amide, -ester, and -nitrile. The profile of reactivity of this class of derivatives has been evaluated and 4c, containing a vinyl ester warhead, proved to be a highly potent and selective falcipain-2 inhibitor. Introduction Malaria is one of the most prevailing and fatal infectious diseases in the world. It is a public health problem in about 90 countries, and approximately 40% of the world population is currently at risk of contamination.1 It is estimated that there are some 350-500 million cases, resulting in 2-3 million deaths each year. It strikes hardest at some of the poorest nations, mainly in tropical and subtropical regions of Africa, South America, and Asia, where it is itself a significant cause and consequence of poverty.2 Malaria is caused by protozoan erythrocytic parasites of the Plasmodium genus, with Plasmodium falciparum being the most lethal and widespread diseasecausing species. Although numerous efforts have been made to control this disease, the incidence of this parasitic infection has been increasing at an alarming rate due to the emergence of parasites resistant to the most common antimalarial agents. Particularly alarming is the resistance to chloroquine, which has been used for decades as the drug of choice to combat or limit the spread of malaria. Among several other therapeutic options taken into consideration so far, artemisinins (natural and semisynthetic) were the most promising drug candidates due to their efficacy on chloroquine-resistant strains, rapid mode of action, broad range of activity, and low tendency to induce parasite resistance. However, therapeutic use of artemisinins is hampered by severe limitations such as their short half-life, thermal instability, and high cost of therapy.3 The spread of multiresistant strains of P. falciparum, combined with the lack of effective vaccines, therefore represents a serious concern for the world population and creates a great need for the development of new and effective antimalarial agents.4 Several proteases are involved in the complicated life cycle of malaria parasites. Among them, the papain-like cysteine protease falcipain-2 (FP-2) of P. falciparum emerged as a valid target for antimalarial drug design in view of its multifunctional role. Not only does it catabolize hemoglobin at the early trophozoite stage to provide nutrients for parasites’ growth and * To whom correspondence should be addressed. Phone: +39 090 6766466. Fax: +39 090 355613. E-mail:
[email protected]. † Dipartimento Farmaco-Chimico, University of Messina. ‡ Institute of Pharmacy and Food Chemistry, University of Wu¨rzburg. § Institute of Zoology, University of Kiel. | Department of Medicinal Chemistry IRBM, MRL Rome.
maturation, but it also prevents the premature red blood cell lysis that might occur where parasite growth is not compensated for by reduced host-cell volume.5 During the late trophozoite and schizont stages, FP-2 is also involved in the degradation of erythrocyte membrane skeletal proteins including ankyrin and the band 4.1 protein.6 This latter activity is thought to contribute to destabilization of the erythrocyte membrane, leading to host cell rupture and release of the mature merozoites. Most of the earlier attempts to design FP-2 inhibitors have been focused on peptide-based structures.7 These inhibitors possess a suitable recognition motif for the enzyme and a C-terminal electrophilic moiety, which tends to form covalent bonds with the thiolate of the catalytic cysteine. However, their susceptibility to protease degradation and their poor absorption through cell membranes limit their utility as therapeutic agents. Moreover, their poor selectivity for parasitic cysteine proteases over the human cysteine proteases (i.e., cathepsins) remains an issue of significant concern and an intriguing challenge for medicinal chemists. Efforts to alleviate these problems have been undertaken,8 ranging from the simple replacement of natural with unnatural amino acids9,10 to the synthesis of conformationally constrained surrogates devoid of any peptide bond mimicking the secondary structures of proteins.11 Our research in this field began with the design and synthesis of aldehyde-based peptidomimetics featuring a 1,4-BZD scaffold embedded within the peptide sequence as a D-Ser-Gly replacement, which were found to be reversible inhibitors of recombinant FP-2 at a micromolar level.12 These preliminary results prompted us to focus our efforts on irreversible inhibitors with the aim of improving the pharmacological profile of this novel class of peptidomimetics: in particular, an irreversible inhibition of the enzyme would be advantageous in the control and elimination of the parasite.13 Previous reports supported the choice of the vinyl sulfone warhead at P1 site as an electrophilic moiety able to covalently trap in an irreversible manner the active site thiol function of the target enzyme. They also supported the introduction of the homoPhe residue at P1 site to improve the potency of FP-2 inhibitors.14 As a matter of fact, all new vinylsulfones (i.e., 1a and 2a, Chart 1) proved to be potent FP-2 inhibitors with high second-order rate constants (in the range 161-634 × 103 M-1 min-1) against the target enzyme, and the homoPhe series (i.e., 1a) turned out to be the most active.15 Notably, these derivatives
10.1021/jm900047j CCC: $40.75 2009 American Chemical Society Published on Web 03/18/2009
2158 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 7
Brief Articles
Chart 1. Structure of the FP-2 Irreversible Inhibitors 1-2
This synthetic approach allowed us to recover the CMproducts as E-isomers in excellent yields, particularly for the highly reactive vinyl esters and vinyl ketones. The reaction was carried out by employing the Hoveyda-Grubbs second generation catalyst, a homogeneous ruthenium carbene complex easy to handle in air and tolerant toward a large variety of functional groups.17 Furthermore, the use of microwave irradiation to promote the CM reaction resulted in reduced reaction times (from 24 h to 1 h) and byproduct formation. Biological Activity. All compounds 3a-3c and 4a-4d were tested on recombinant FP-2 using Cbz-Phe-Arg-AMC as fluorogenic substrate.18 First, a preliminary screening with inhibitor concentrations of 50 µM was performed. An equivalent volume of DMSO was used as negative control, and the irreversible standard inhibitor of clan CA, family C1 cysteine proteases (papain-family), namely E-64,19 was used as positive control. The screening showed all compounds to abolish enzyme activity. Continuous assays (progress curve method)20 were then performed to determine the first-order rate constants of inhibition kinac (min-1), the dissociation constants Kinac (µM), and the second-order rate constants of inhibition ksecond (M-1 min-1), as ksecond ) kinac/Kinac (Table 1). In contrast to the SAR described for vinyl sulfones,15 derivatives 4a-4c, all bearing a glycine residue at P1 site, proved to be more active than the homoPhe analogues 3a-3c, with potency in the order ester > amide > ketone. In particular, a surprising inhibitory potency was observed for the vinylester 4c (Table 1), which showed the highest value of second-order rate constant of inhibition (ksecond 3571000 M-1 min-1), a potency two times higher than that of the standard E-64 (ksecond 1586000 M-1 min-1) and the highest enzymatic affinity (Ki ) 17 nM) among all the described inhibitors. A promising profile was also obtained for the amide 4b (ksecond 1091000 M-1 min-1 and Ki ) 76 nM), which proved to be nearly as potent as the standard E-64. In contrast, a strong reduction in potency was observed when the electron-withdrawing group (EWG) is a nitrile (i.e., 4d with ksecond 387000 M-1 min-1) or a ketone (i.e., 4a with ksecond 264000 M-1 min-1). In these latter cases, the inhibitory potency proved to be almost comparable to that shown by the reported vinyl sulfones.15 All the Michael acceptors 3 and 4 were also tested against papain family human cysteine proteases such as cathepsins B and L in order to check their selectivity against the target enzyme. As shown in Table 2, all the synthesized compounds 3-4 generally proved to be weak inhibitors of cathepsins B and L. In most cases, the compounds inhibit the enzymes in a irreversible manner, with ksecond values 1-3 orders of magnitude lower than that reported for the inhibition of FP-2. Conversely, the inhibition transpired to be non time-dependent for compounds 3b, 3c, and 4c against cathepsin L and 4d against cathepsin B, with Ki values 1-2 orders of magnitude higher than that shown toward the parasitic protease. Vinyl ester 4c, the most potent inhibitor of this class, also displayed an excellent selectivity toward FP-2 (ksecond 3571000 M-1 min-1 and Ki 17 nM) with respect to the inhibition of cathepsin B (ksecond 6000 M-1 min-1 and Ki 7.3 µM) and L (Ki 8.4 µM). Compounds 4a-d were also tested against P. falciparum strain FCBR (Table 1). The dose-dependent effects of the compounds on parasite development were quantified using the Malstat assay, which measures the activity of the Plasmodiumspecific enzyme lactate dehydrogenase.21
Scheme 1a
a
Reagents and conditions: (a) Hoveyda catalyst, CH2Cl2, 100 °C, MW,
1 h.
showed a good selectivity toward human cysteine proteases cathepsin B and L. Conversely, our attempt to further optimize the pharmacophore portion by replacement of the vinyl sulfone moiety with a vinyl phosphonate one (i.e., 1b, 2b, Chart 1) generated unsatisfactory results with a decrease in both potency and selectivity.16 The aim of the present work is to explore the strength and efficacy of different R,β-unsaturated electrophilic functions (Michael acceptors) in the interaction with the active site thiol group of the target enzyme. Vinyl-ketones 3a, 4a, -amides 3b, 4b, -esters 3c, 4c, and -nitrile 4d were evaluated in order to definitively establish an optimized pharmacophore portion for this class of inhibitors. Results and Discussion Chemistry. The key step in the synthesis of the Michael acceptors 3a-3c and 4a-4d (Scheme 1) was the olefin crossmetathesis (CM) reaction, an innovative and efficient method selected to functionalize the pseudotripeptides 5a-5b. These intermediates, synthesized as previously described by our group,15 were reacted with the CM partners 6a-6d containing the appropriate electron-withdrawing group.
Brief Articles
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 7 2159
Table 1. Inhibition of Falcipain-2 and Antiplasmodial Activity of Compounds 3 and 4
compd
R
EWG
ksecond (M-1 min-1)
kinac (min-1)
Ki (µM)
3a 3b 3c 4a 4b 4c 4d E64
CH2CH2Ph CH2CH2Ph CH2CH2Ph H H H H
COMe CONMe2 COOMe COMe CONMe2 COOMe CN
4000 ( 1700 78000 ( 15600 21000 ( 2000 264000 ( 31500 1091000 ( 218000 3571000 ( 400000 387000 ( 96000 1586000 ( 400000b
a 0.074 ( 0.011 a a 0.08 ( 0.017 0.06 ( 0.009 0.074 ( 0.012 0.46 ( 0.07
a 0.95 ( 0.7 a a 0.076 ( 0.023 0.017 ( 0.004 0.19 ( 0.12 0.29 ( 0.09
P. falciparum IC50 (µM)
9.3 ( 1.4 >100 12.0 ( 2.1 36.1 ( 2.7
a Because the kobs vs [I] diagrams were restricted to the linear range, only the second-order rate constant could be determined as kobs/[I]. b kassociation ) kinac/Kiapp ) 11.31 ((2.70) × 103 M-1 s-1; literature value19 with 25 µM Cbz-Leu-Arg-AMC as substrate: kassociation ) 12.28 ((1.40) × 103 M-1 s-1. All results include standard deviations from two independent measurements, each performed in duplicate.
Table 2. Inhibition of Cathepsins B and L of Compounds 3 and 4 cathepsin B
cathepsin L
compd
ksecond (M-1 min-1)
kinac (min-1)
Ki (µM)
ksecond (M-1 min-1)
kinac (min-1)
Ki (µM)
3a 3b 3c 4a 4b 4c 4d
17000 ( 400 8000 ( 400 5000 ( 2000 3000 ( 700 22000 ( 3000 6000 ( 1000 a
0.15 ( 0.03 0.051 ( 0.001 0.071 ( 0.012 0.02 ( 0.002 0.02 ( 0.002 0.044 ( 0.004 a
8.4 ( 2.0 6.4 ( 2.3 13.7 ( 2.5 6 ( 0.3 0.85 ( 0.001 7.3 ( 3.2 9.9 ( 0.5
12000 ( 7000 a a 23000 ( 140 10000 ( 2000 a 12000 ( 360
0.09 ( 0.02 a a 0.12 ( 0.02 0.02 ( 0.004 a 0.04 ( 0.003
7.5 ( 2.2 7.4 ( 0.5 2.7 ( 0.1 5.9 ( 3.4 1.9 ( 0.1 8.4 ( 0.8 3.2 ( 0.37
a
Non-time dependent inhibition. All results include standard deviations from two independent measurements, each performed in duplicate. Table 4. Cytotoxicity of 4c towards Different Cell Lines
Table 3. Membrane Permeability in Caco2 Cells compd
Papp (nm/s)a mean ( SD
mass balancebmean %
4c caffeine vinblastine cimetidine A f B cimetidine B f A
38.2 ( 0.6 267.8 ( 69.5 5.1 ( 0.9 5.6 ( 1.3 15 ( 2.3
45 75 120 98 105
Absorption classification: Papp > 50: high; Papp 10-50: medium; Papp < 10: low. The results include standard deviations from two independent measurements. b Mass balance
