Three Decades of Targeting Falcipains to Develop Antiplasmodial Agents: What have we Learned and What can be Done Next?

Malaria is a devastating infectious disease that affects large swathes of human populations across the planet's tropical regions. It is caused by parasites of the genus Plasmodium, with Plasmodium falciparum being responsible for the most lethal form of the disease. During the intraerythrocytic stage in the human hosts, malaria parasites multiply and degrade hemoglobin (Hb) using a battery of proteases, which include two cysteine proteases, falcipains 2 and 3 (FP-2 and FP-3). Due to their role as major hemoglobinases, FP-2 and FP-3 have been targeted in studies aiming to discover new antimalarials and numerous inhibitors with activity against these enzymes, and parasites in culture have been identified. Nonetheless, cross-inhibition of human cysteine cathepsins remains a serious hurdle to overcome for these compounds to be used clinically. In this article, we have reviewed key functional and structural properties of FP-2/3 and described different compound series reported as inhibitors of these proteases during decades of active research in the field. Special attention is also paid to the wide range of computer-aided drug design (CADD) techniques successfully applied to discover new active compounds. Finally, we provide guidelines that, in our understanding, will help advance the rational discovery of new FP-2/3 inhibitors.

Enhancement of Inhibitory Activity by Combining Allosteric Inhibitors Putatively Binding to Different Allosteric Sites on Cathepsin K

BACKGROUND

Cathepsin K, which is involved in bone resorption, is a good target for treating osteoporosis, but no clinically approved medicine has been developed. Recently, allosteric inhibitors with high specificity and few side effects have been attracting attention for use in new medicines.

METHODS

Cathepsin K inhibitors were isolated from the methanol extract of Chamaecrista nomame (Leguminosae) using cathepsin K inhibition activity-assisted multi-step chromatography. Standard kinetic analysis was employed to examine the mechanism of cathepsin K inhibition when an isolated inhibitor and its derivative were used. The allosteric binding of these cathepsin K inhibitors was supported by a docking study using AutoDock vina. Combinations of allosteric cathepsin K inhibitors expected to bind to different allosteric sites were examined by means of cathepsin K inhibition assay.

RESULTS

Two types of cathepsin K inhibitors were identified in the methanol extract of Chamaecrista nomame. One type consisted of cassiaoccidentalin B and torachrysone 8-β-gentiobioside, and inhibited both cathepsin K and B with similar inhibitory potential, while the other type of inhibitor consisted of pheophytin a, and inhibited cathepsin K but not cathepsin B, suggesting that pheophytin a binds to an allosteric site of cathepsin K. Kinetic analysis of inhibitory activity suggested that pheophytin a and its derivative, pheophorbide b, bind allosterically to cathepsin K. This possibility was supported by a docking study on cathepsin K. The cathepsin K inhibitory activity of pheophytin a and pheophorbide b was enhanced by combining them with the allosteric inhibitors NSC 13345 and NSC94914, which bind to other allosteric sites on cathepsin K.

CONCLUSIONS

Different allosteric inhibitors that bind to different sites in combination, as shown in this study, may be useful for designing new allosteric inhibitory drugs with high specificity and few side effects.

Recent applications of computational methods to allosteric drug discovery

Interest in exploiting allosteric sites for the development of new therapeutics has grown considerably over the last two decades. The chief driving force behind the interest in allostery for drug discovery stems from the fact that in comparison to orthosteric sites, allosteric sites are less conserved across a protein family, thereby offering greater opportunity for selectivity and ultimately tolerability. While there is significant overlap between structure-based drug design for orthosteric and allosteric sites, allosteric sites offer additional challenges mostly involving the need to better understand protein flexibility and its relationship to protein function. Here we examine the extent to which structure-based drug design is impacting allosteric drug design by highlighting several targets across a variety of target classes.

Tetracycline Derivatives Inhibit Plasmodial Cysteine Protease Falcipain-2 through Binding to a Distal Allosteric Site

Allosteric inhibitors regulate enzyme activity from remote and usually specific pockets. As they promise an avenue for less toxic and safer drugs, the identification and characterization of allosteric inhibitors has gained great academic and biomedical interest in recent years. Research on falcipain-2 (FP-2), the major papain-like cysteine hemoglobinase of Plasmodium falciparum, might benefit from this strategy to overcome the low selectivity against human cathepsins shown by active site-directed inhibitors. Encouraged by our previous finding that methacycline inhibits FP-2 noncompetitively, here we assessed other five tetracycline derivatives against this target and characterized their inhibition mechanism. As previously shown for methacycline, tetracycline derivatives inhibited FP-2 in a noncompetitive fashion, with K_i values ranging from 121 to 190 μM. A possible binding to the S' side of the FP-2 active site, similar to that described by X-ray crystallography (PDB: 6SSZ) for the noncompetitive inhibitor E-chalcone 48 (EC48), was experimentally discarded by kinetic analysis using a large peptidyl substrate spanning the whole active site. By combining lengthy molecular dynamics (MD) simulations that allowed methacycline to diffuse from solution to different FP-2 surface regions and free energy calculations, we predicted the most likely binding mode of the ligand. Of note, the proposed binding pose explains the low differences in K_i values observed for the tested tetracycline derivatives and the calculated binding free energies match the experimental values. Overall, this study has implications for the design of novel allosteric inhibitors against FP-2 and sets the basis for further optimization of the tetracycline scaffold to produce more potent and selective inhibitors.

In silico identification of noncompetitive inhibitors targeting an uncharacterized allosteric site of falcipain-2

Falcipain-2 (FP-2) is a Plasmodium falciparum hemoglobinase widely targeted in the search for antimalarials. FP-2 can be allosterically modulated by various noncompetitive inhibitors that have been serendipitously identified. Moreover, the crystal structures of two inhibitors bound to an allosteric site, termed site 6, of the homolog enzyme human cathepsin K (hCatK) suggest that the equivalent region in FP-2 might play a similar role. Here, we conduct the rational identification of FP-2 inhibitors through virtual screenings (VS) of compounds into several pocket-like conformations of site 6, sampled during molecular dynamics (MD) simulations of the free enzyme. Two noncompetitive inhibitors, ZINC03225317 and ZINC72290660, were confirmed using in vitro enzymatic assays and their poses into site 6 led to calculated binding free energies matching the experimental ones. Our results provide strong evidence about the allosteric inhibition of FP-2 through binding of small molecules to site 6, thus opening the way toward the discovery of new inhibitors against this enzyme.

Synthesis and kinetic characterization of hyperbolic inhibitors of human cathepsins K and S based on a succinimide scaffold

Cathepsins K and S are closely related papain-like cysteine peptidases and potential therapeutic targets for metabolic and inflammatory diseases such as osteoporosis and arthritis. Here we describe the reduction of a previously characterized succinimide (2,5-dioxopyrrolidine)-containing hyperbolic inhibitor of cathepsin K (methyl (RS)-N-[1-(4-methoxyphenyl)-2,5-dioxopyrrolidin-3-yl]glycinate), to obtain a better and more selective compound (compound 4a - methyl (2,5-dioxopyrrolidin-3-yl)glycinate), which acted as a hyperbolic mixed inhibitor/activator similar to already known allosteric effectors of cathepsin K. We then investigated the potential of the succinimide scaffold as inhibitors of cathepsins K and/or S and synthesized a library of such compounds by 1,4-addition of α-amino acid esters and related compounds to N-substituted maleimides. From the generated library, we identified the first small molecule hyperbolic inhibitors of cathepsin S (methyl ((R)-2,5-dioxopyrrolidin-3-yl)-l-threoninate (compound R-4c) and 3-{[(1S,2R,3'S)-2-hydroxycyclohexyl]amino}pyrrolidine-2,5-dione (compound (1S,2R,3'S-10)). The former acted via a similar mechanism to compound 4a, while the latter was a hyperbolic specific inhibitor of cathepsin S. Given the versatility of the scaffold, the identified compounds will be used as the basis for the development of high-affinity hyperbolic inhibitors of the individual peptidases and to explore the potential of hyperbolic inhibitors for the inhibition of cysteine cathepsins in in vitro models.

Identification of Plasmodium dipeptidyl aminopeptidase allosteric inhibitors by high throughput screening

Dipeptidyl aminopeptidases (DPAPs) are cysteine proteases that cleave dipeptides from the N-terminus of protein substrates and have been shown to play important roles in many pathologies including parasitic diseases such as malaria, toxoplasmosis and Chagas's disease. Inhibitors of the mammalian homologue cathepsin C have been used in clinical trials as potential drugs to treat chronic inflammatory disorders, thus proving that these enzymes are druggable. In Plasmodium species, DPAPs play important functions at different stages of parasite development, thus making them potential antimalarial targets. Most DPAP inhibitors developed to date are peptide-based or peptidomimetic competitive inhibitors. Here, we used a high throughput screening approach to identify novel inhibitor scaffolds that block the activity of Plasmodium falciparum DPAP1. Most of the hits identified in this screen also inhibit Plasmodium falciparum DPAP3, cathepsin C, and to a lesser extent other malarial clan CA proteases, indicating that these might be general DPAP inhibitors. Interestingly, our mechanism of inhibition studies indicate that most hits are allosteric inhibitors, which opens a completely new strategy to inhibit these enzymes, study their biological function, and potentially develop new inhibitors as starting points for drug development.

Rat cathepsin K: Enzymatic specificity and regulation of its collagenolytic activity

Human cathepsin K (hCatK), which is highly expressed in osteoclasts, has the noteworthy ability to cleave type I and II collagens in their helical domain. Its collagenase potency depends strictly on the formation of an oligomeric complex with chondroitin 4-sulfate (C4-S). Accordingly, hCatK is a pivotal protease involved in bone resorption and is an attractive target for the treatment of osteoporosis. As rat is a common animal model for the evaluation of hCatK inhibitors, we conducted a comparative analysis of rat CatK (rCatK) and hCatK, which share a high degree of identity (88%) and similarity (93%). The pH activity profile of both enzymes displayed a similar bell-shaped curve (optimal pH: 6.4). Presence of Ser134 and Val160 in the S2 pocket of rCatK instead of Ala and Leu residues, respectively, in hCatK, led to a weaker peptidase activity, as observed for mouse CatK. Also, regardless of the presence of C4-S, rCatK cleaved in the nonhelical telopeptide regions of both type I (tail) and type II (articular joint) rat collagens. Structure-based computational analyses (electrostatic potential, molecular docking, molecular dynamics, free energy calculations) sustained that the C4-S mediated collagenolytic activity of rCatK obeys distinct molecular interactions from those of hCatK. Additionally, T-kininogen (a.k.a. thiostatin), a unique rat serum acute phase molecule, acted as a tight-binding inhibitor of hCatK (Ki = 0.11 ± 0.05 nM). Taken into account the increase of T-Kininogen level in inflamed rat sera, this may raise the question of the appropriateness to evaluate pharmacological hCatK inhibitors in this peculiar animal model.

Dissecting a novel allosteric mechanism of cruzain: A computer-aided approach

Trypanosoma cruzi is the causative agent of Chagas disease, a neglected infection affecting millions of people in tropical regions. There are several chemotherapeutic agents for the treatment of this disease, but most of them are highly toxic and generate resistance. Currently, the development of allosteric inhibitors constitutes a promising research field, since it can improve the accessibility to more selective and less toxic medicines. To date, the allosteric drugs prediction is a state-of-the-art topic in rational structure-based computational design. In this work, a simulation strategy was developed for computational discovery of allosteric inhibitors, and it was applied to cruzain, a promising target and the major cysteine protease of T. cruzi. Molecular dynamics simulations, binding free energy calculations and network-based modelling of residue interactions were combined to characterize and compare molecular distinctive features of the apo form and the cruzain-allosteric inhibitor complexes. By using geometry-based criteria on trajectory snapshots, we predicted two main allosteric sites suitable for drug targeting. The results suggest dissimilar mechanisms exerted by the same allosteric site when binding different potential allosteric inhibitors. Finally, we identified the residues involved in suboptimal paths linking the identified site and the orthosteric site. The present study constitutes the first approximation to the design of cruzain allosteric inhibitors and may serve for future pharmacological intervention. Here, no major effects on active site structure were observed due to compound binding (modification of distance and angles between catalytic residues), which indicates that allosteric regulation in cruzain might be mediated via alterations of its dynamical properties similarly to allosteric inhibition of human cathepsin K (HCatK). The current findings are particularly relevant for the design of allosteric modulators of papain-like cysteine proteases.

Allosteric Site Inhibitor Disrupting Auto-Processing of Malarial Cysteine Proteases

Falcipains are major haemoglobinases of Plasmodium falciparum required for parasite growth and development. They consist of pro- and mature domains that interact via 'hot-spot' interactions and maintain the structural integrity of enzyme in zymogen state. Upon sensing the acidic environment, these interactions dissociate and active enzyme is released. For inhibiting falcipains, several active site inhibitors exist, however, compounds that target via allosteric mechanism remains uncharacterized. Therefore, we designed and synthesized six azapeptide compounds, among which, NA-01 & NA-03 arrested parasite growth by specifically blocking the auto-processing of falcipains. Inhibitors showed high affinity for enzymes in presence of the prodomain without affecting the secondary structure. Binding of NA-03 at the interface induced rigidity in the prodomain preventing structural reorganization. We further reported a histidine-dependent activation of falcipain. Collectively, for the first time we provide a framework for blocking the allosteric site of crucial haemoglobinases of the human malaria parasite. Targeting the allosteric site could provide high selectivity and less vulnerable to drug resistance.

N-tert-butyloxycarbonyl-Phe-Leu-Phe-Leu-Phe (BOC2) inhibits the angiogenic activity of heparin-binding growth factors

The peptides N-tert-butyloxycarbonyl-Phe-Leu-Phe-Leu-Phe (BOC2) and BOC-Met-Leu-Phe (BOC1) are widely used antagonists of formyl peptide receptors (FPRs), BOC2 acting as an FPR1/FPR2 antagonist whereas BOC1 inhibits FPR1 only. Extensive investigations have been performed by using these FPR antagonists as a tool to assess the role of FPRs in physiological and pathological conditions. Based on previous observations from our laboratory, we assessed the possibility that BOC2 may exert also a direct inhibitory effect on the angiogenic activity of vascular endothelial growth factor-A (VEGF-A). Our data demonstrate that BOC2, but not BOC1, inhibits the angiogenic activity of heparin-binding VEGF-A₁₆₅ with no effect on the activity of the non-heparin-binding VEGF-A₁₂₁ isoform. Endothelial cell-based bioassays, surface plasmon resonance analysis, and computer modeling indicate that BOC2 may interact with the heparin-binding domain of VEGF-A₁₆₅, thus competing for heparin interaction and preventing the binding of VEGF-A₁₆₅ to tyrosine kinase receptor VEGFR2, its phosphorylation and downstream signaling. In addition, BOC2 inhibits the interaction of a variety of heparin-binding angiogenic growth factors with heparin, including fibroblast growth factor 2 (FGF2) whose angiogenic activity is blocked by the compound. Accordingly, BOC2 suppresses the angiogenic potential of human tumor cell lines that co-express VEGF-A and FGF2. Thus, BOC2 appears to act as a novel multi-heparin-binding growth factor antagonist. These findings caution about the interpretation of FPR-focusing experimental data obtained with this compound and set the basis for the design of novel BOC2-derived, FPR independent multi-target angiogenesis inhibitors.

Computational investigation of conformational variability and allostery in cathepsin K and other related peptidases

Allosteric targeting is progressively gaining ground as a strategy in drug design. Its success, however, depends on our knowledge of the investigated system. In the case of the papain-like cysteine peptidase cathepsin K, a major obstacle in our understanding of allostery is represented by the lack of observable conformational change at the active site. This makes it difficult to understand how binding of effectors at known allosteric sites translates into modified enzyme activity. Herein, we address this issue by a computational approach based on experimental data. We analyze the conformational space of the papain-like family and the positioning of cathepsin K within it using principal component analysis and molecular dynamics simulations. We show that human cathepsin L-like endopeptidases (cathepsins L, K, S and V) adopt similar conformations which are distinct from their non-animal counterparts and other related peptidases. Molecular dynamics simulations show that the conformation of cathepsin K is influenced by known allosteric effectors, chondroitin sulfate and the small molecules NSC13345 and NSC94914. Importantly, all effectors affect the geometry of the active site around sites S1 and S2 that represent the narrowest part of the active site cleft and the major specificity determinant in papain-like endopeptidases. The effectors act by stabilizing pre-existing conformational states according to a two-state model and thereby facilitate or hinder the binding of substrate into the active site, as shown by molecular docking simulations. Comparison with other related enzymes shows that similar conformational variability and, by implication, allostery also exist in other papain-like endopeptidases.