Beyond the Active Site: First‐In‐Class Leishmania Infantum Trypanothione Reductase Dimerization Disruptors as Antileishmanial Leads

Abstract

Leishmania infantum trypanothione reductase (Li‐TryR) is an essential enzyme for parasite redox homeostasis and a validated target for antileishmanial therapy. Classical inhibitors acting at the active site have shown limited success due to poor selectivity, suboptimal pharmacological properties, and lack of in‐cell efficacy. This work presents a comprehensive strategy to develop first‐in‐class Li‐TryR dimerization disruptors, targeting the protein–protein interface required for enzymatic activity. Through molecular modeling, mutagenesis, and structural analysis, a key α‐helical hotspot (around residue E436) was identified as a druggable region. Interface‐derived peptides, helix‐stabilized stapled analogs, α/β‐foldamers, and cell‐penetrating conjugates demonstrated that dimer disruption effectively inhibits LiTryR and can translate into antiparasitic activity. Building on these insights, several families of small‐molecule proteomimetics, including imidazole‐phenyl‐thiazoles, triazole analogs, and triazolium salts, were developed, achieving low‐micromolar to submicromolar inhibition and significant activity in parasite cultures and intracellular models. Overall, this work establishes dimer‐interface targeting as a novel therapeutic paradigm for leishmaniasis and provides multiple optimized chemical scaffolds for future drug development.

This review outlines a pioneering strategy to inhibit Leishmania infantum trypanothione reductase by disrupting its dimer interface. From peptides to small‐molecule proteomimetics, multiple chemotypes achieved potent enzymatic inhibition and notable cellular antileishmanial activity, highlighting a new therapeutic paradigm.

1. Introduction

Leishmaniasis is one of the oldest neglected tropical diseases, caused by protozoa of the genus Leishmania, that affects humanity. Leishmaniases are disseminated by the bites of infected female Phlebotomus sandflies [1]. Lesismania donovani and Leishmania infantum parasites are responsible of Visceral leishmaniasis (VL) or kala azar, the most severe and aggressive form of the disease, which is fatal if untreated [2, 3, 4]. Leishmaniasis represents, after malaria, the second most deadly parasitic disease, causing about 600,000—1  million new cases (only around 200,000 reported to WHO) and around 50,000–90,000 deaths each year, according to WHO (World Health Organization) [5]. The disease is endemic in more than 97 countries worldwide [6, 7, 8]. Current chemotherapeutic options for leishmaniasis are scarce and include pentavalent antimonials, paramomycin, amphotericin B, and Miltefosine (the only oral drug). The use of these drugs is limited by toxicity, suboptimal efficacy (poor bioavailability), high cost, and increasingly resistance‐prone parasites [9, 10, 11]. These drawbacks underscore the urgent need to explore nonclassical mechanisms of inhibition that may lead to more effective therapies acting over new targets and/or through new mechanisms of action.

Within the strategies commonly used for the design of new drugs, the rational design based on the target represents a promising alternative to the repositioning of drugs or to the phenotypic screening used so far in the field of antiparasitics. Trypanothione reductase (TryR) is an NADPH‐dependent flavoenzyme that is unique to trypanosomatid parasites and plays an essential role in maintaining intracellular redox homeostasis, being crucial for the antioxidant defenses of the parasites, and it is absent in humans [12, 13, 14, 15]. Its mammalian counterpart is glutathione reductase (GR), which, however, shows significant structural differences from TryR. TryR catalyses the reduction of trypanothione disulfide to the dithiol form [12, 13, 16], which makes it a highly attractive and validated target for antiparasitic drug development against leishmaniasis and other trypanosomatid‐caused diseases.

TryR maintains trypanothione in its reduced dithiol form, T[SH]₂, thereby preserving a balanced intracellular redox state that functionally replaces the glutathione/GR system found in mammals. Although TryR and GR are structurally and mechanistically related, their substrate specificities are mutually exclusive: TryR reduces trypanothione disulfide, whereas GR acts on glutathione disulfide, a difference that can be exploited to achieve parasite selectivity. In pathogenic trypanosomatids, trypanothione is the sole donor of reducing equivalents to several peroxidase families, and genetic studies have demonstrated that TryR is essential for parasite survival under oxidative stress, highlighting its central role in the antioxidant defense network [14, 17, 18].

Most early efforts to inhibit L. infantum TryR (Li‐TryR) focused on identifying small molecules that target the active site pocket that binds oxidized trypanothione (T[S]₂), but their success was very limited [19, 20]. This cavity is large and highly charged because the natural substrate consists of two glutathione units linked by a positively charged spermidine moiety, which means that binding relies strongly on electrostatic interactions and can complicate the design of potent, drug‐like, and selective inhibitors [20]. Various inhibitor families have been reported [21, 22, 23, 24, 25], including compounds derived from tricyclic antidepressants, polyamine conjugates, substrate analogs, mepacrine (quinacrine), and 3,4‐dihydroquinazolines [26, 27, 28, 29, 30, 31, 32, 33, 34], but in most cases, the precise binding modes and structure–activity relationships (SAR) at the active site remain only partially characterized.

Despite the wide variety of TryR inhibitors described in the literature that act at the active site of the enzyme, satisfactory results have not been achieved in vivo. In order to develop novel inhibitors that could be useful in the treatment of Leishmaniasis, our research group initiated a new line of research aimed at exploring a new mechanism of inhibition based on the disruption of the dimerization process of the enzyme.

Because the biologically functional form of TryR is a homodimer, we devised an alternative approach to target the enzyme, that is, by interacting at the dimer interface to disrupt monomer–monomer association [35] and thereby inhibit enzymatic activity by means of protein–protein interaction (PPI) inhibitors [36]. Although PPI interfaces are often considered challenging drug targets due to the absence of well‐defined small‐molecule binding pockets [19, 37], binding energy is typically concentrated in small “hot spot” regions [38], suggesting that high‐affinity ligands need only engage a limited subset of key interface residues.

We used the X‐ray crystal structure of Li‐TryR [39, 40] to pinpoint potential hotspots involved in stabilizing the dimer. As a first step, a computational analysis of the crystal coordinates was carried out to evaluate the loss of solvent‐accessible surface area (SASA) for each monomer upon dimerization and, in a second step, to quantify the electrostatic and van der Waals participations to the interunit interaction energy [41]. Thus, by using a combination of molecular modeling and site‐directed mutagenesis, we identified an α‐helical segment at the dimer interface of Li‐TryR, encompassing residues P435‐M447 that played the most relevant contribution to dimer stabilization [36]. In this α‐helix, glutamate (E436) was identified and validated as a key amino acid and as a structural and functional “hotspot” for the catalytic activity of Li‐TryR [36] and for the structural integrity of the dimer [36]. Thus, the region of the α‐helix (PEIIQSVGISMKM) that accommodates E436 was considered a “hotspot” for dimerization of the enzyme (Figure 1A) and was contemplated as a potential binding site for the design of PPIs disruptors [36].

FIGURE 1.

(A) α‐helix “hotspot” for dimerization of the enzyme; (B) structures of selected linear peptides 1–8.

2. Peptide Inhibitors

2.1. Interface‐Derived Linear Peptides

As a proof‐of‐concept of the novel approach, and on the basis of this hot‐spot region, a small library of 13‐mer linear peptides was initially designed and tested, with sequences directly derived from the Li‐TryR homodimerization α‐helix containing E436 (Figure 1) and spanning residues P435 to M447, with some variants incorporating isosteric Ser (as in peptides 1–4) and Nle (as in peptides 2 and 4) to replace oxidizable Cys and Met residues, respectively to prevent potential oxidation. Also, Glu at position 2 was replaced by Lysine (peptides 3 and 4) to study the importance of a possible interaction with E436´ of the other dimer subunit. Finally, peptides bearing randomly permuted amino acids were also prepared (those resulted inactive). The effects on dimer stability of these peptides were tested using an in‐house‐developed ELISA test (very versatile and highly sensitive) [36]. Also, enzymatic activity was determined within this series of peptides. Among these peptides, two 13‐mers, PKIIQSVGISMKM (3) and PKIIQSVGIS‐Nle‐K‐Nle (4), emerged as promising Li‐TryR dimerization disruptors with micromolar potency in enzyme activity assays (IC₅₀ = 1.6 ± 0.1 and 1.4 ± 0.2 μM, respectively) and disrupt TryR dimerization (IC₅₀ ^dim = 17.9 and 15.7 μM, respectively), providing a proof‐of‐concept for interface‐mimicking peptide inhibitors [36].

To better understand the binding interaction and the role of specific peptide residues, we ran explicit‐solvent MD simulations of the Li‐TryR monomer in complex with 3. The peptide 3 was modeled using the corresponding helical segment from the other monomer as a structural guide, with residue E at position 2 substituted by K. The results of the simulation showed that the positively charged side chain of the K2 residue hydrogen bonds with the carboxylate of E436’. K at position 2 seems to effectively anchor the N‐terminus of the peptide within the monomer's binding groove, disrupting dimer formation and inhibiting enzymatic activity [36].

Next, we explored the importance of each amino acid for activity, performing alanine scanning (Figure 1). Replacement of the native amino acid at position 5 (Q) with alanine (peptide 6) produced the most pronounced negative effect, suggesting that the side chain of Q at this position plays a major role in the binding affinity. Substituting K at position 2 with A (Figure 1, peptide 5) caused only a slight reduction in inhibition, indicating a less critical contribution to binding. In contrast, replacing S with A at position 6 (peptide 7) resulted in an increase in inhibitory activity [36]. Also, the minimal peptide length for activity was determined by synthesizing truncated peptides in which, to limit the conformational freedom of the N‐terminus, a proline was introduced at position 1. Our screening showed that nine amino acids (peptide 8) are the minimal length required for activity (IC₅₀ < 4 μM). Further deletion at the C‐terminus resulted in a complete loss of activity. The last four amino acids seem not to be essential to maintain significant binding affinity [36]. Replacing glutamic acid with lysine at position 2 enhanced inhibition of both TryR dimerization and oxidoreductase activity, while substituting glutamine at position 5 with alanine reduced inhibitory potency. Molecular dynamics simulations helped explain these observations. They showed that the peptide‐induced disruption of the dimer occurs slowly when the enzyme is in an inactive state, but the process becomes significantly faster once catalysis begins [36].

Encouraged by the previous in vitro findings, two of these peptides (4 and 8) were tested on L. infantum axenic amastigotes and promastigotes. However, none of the peptides showed activity against this parasite stage. This lack of effect may be, most likely, due to poor cellular uptake (inability to cross the cell membrane) or to the need for stronger enzyme inhibition to achieve significative leishmanicidal action [36].

To the best of our knowledge, this study is the first to uncover a crucial functional role for residue E436 in Li‐TryR. It also highlights a previously unrecognized potential binding site that could be exploited for designing small‐molecule inhibitors that disrupt PPIs.

2.2. Kinetic Analysis of Peptide 4

In 2020, an in‐depth kinetic analysis of peptide 4 was conducted after initial steady‐state analyses produced atypical results. Classical analyses failed to provide an accurate inhibition constant because reaction rates declined continuously over time, indicating a behavior inconsistent with standard reversible inhibition. This observation prompted a detailed kinetic investigation that ultimately revealed peptide 4 as a clear example of slow‐binding pseudoirreversible inhibition.

Progress‐curve experiments showed that the enzyme exhibited an initial velocity followed by a gradual decline, demonstrating that binding of 4 occurs through a time‐dependent process rather than reaching instantaneous equilibrium [42]. Mechanistically, the inhibition follows a two‐step pathway (Figure 2). The inhibitor first forms a rapid, reversible EI complex. This is followed by a much slower isomerization step that generates an inactive E*I species. The kinetic constants highlight the slow‐binding nature of this transition: the forward isomerization rate is approximately k ₅ ≈ 2.27 × 10⁻⁴ s⁻¹, while the reverse rate is extremely small (k ₆ ≤ 10⁻⁵ s⁻¹). Such a low dissociation rate effectively traps the inhibitor in its bound state. Experimentally, this pseudoirreversibility becomes evident when prolonged incubation of the enzyme–inhibitor complex is followed by dilution by several orders of magnitude without the recovery of enzymatic activity, confirming negligible dissociation within the experimental timeframe [42]. Interestingly, the inhibition was determined to be strictly noncompetitive with respect to oxidized trypanothione (TS₂), with an α value of approximately 1. This indicates that peptide 4 binds with a similar affinity to the free enzyme and to the enzyme–substrate complex. Physiologically, this is a crucial property: intracellular trypanothione levels fluctuate significantly under oxidative stress, which can weaken competitive inhibitors. In contrast, peptide 4 maintains its potency regardless of the substrate concentration, supporting its potential effectiveness inside the parasite [42].

FIGURE 2.

Time‐dependant two‐step mechanism of inhibition of Li‐TryR.

The potency of peptide 4 is further illustrated by the comparison of its inhibition constants. The rapid‐equilibrium step yields an initial Ki of around 1.34 μM. When the slow isomerization is incorporated into the full kinetic model, the refined Ki decreases to 0.46 μM. Most notably, the global Ki*, which accounts for the isomerization to the tightly bound E*I state, drops below 0.02 μM. This exceptionally low apparent inhibition constant reflects the long residence time of the inhibitor caused by the extremely slow dissociation rate, a feature increasingly recognized as a determinant for pharmacological efficacy and selectivity [42]. Despite the formation of a largely inactive E*I complex, peptide 4 does not completely eliminate catalytic turnover. The ternary ESI complex retains approximately 50% of maximal catalytic activity, suggesting that the peptide does not distort the active site itself but instead perturbs the enzyme structure or dynamics in a way that modulates, rather than entirely blocks, functionality. Peptide 4 represents one of the clearest demonstrations of how targeting oligomerization, rather than the catalytic site, can give rise to unconventional kinetic behavior and highly potent inhibitory profiles.

2.3. Interface‐Derived Stapled Peptides

Short peptides typically lack a defined structure in aqueous environments, while their biological targets often require a specific conformation for binding. Covalent cyclization, commonly through side chain–to–side chain links, can reinforce the secondary structure and protect peptide analogs from proteolysis [43, 44, 45]. To stabilize α‐helices, strategies such as the incorporation of lactam bridges [46, 47, 48, 49, 50, 51] and short hydrocarbon chains [52, 53, 54, 55, 56] as staples [57, 58, 59] have proven very effective in enhancing helicity and reducing enzymatic degradation by proteolysis. Hydrocarbon staples have also been associated with improved oral bioavailability [60] and enhanced cellular uptake by cell permeability [61, 62, 63, 64].

To stabilize the α‐helical character of the linear peptide prototype 13‐mer 2, we carried out initial studies aimed to systematically compare two helix‐stabilization chemistries by covalent side‐chain–to–side‐chain “stapling” strategies [64, 65], hydrocarbon stapling [52, 53, 54, 55, 56] versus Glu‐Lys lactam cyclization [49, 50, 51], generating analogs with either hydrocarbon staples bridging residues at positions i, i+4 (one turn) or i, i+7 (two turns) (peptides 9 and 10) or a lactam bridge (peptide 11, Figure 3). To preserve the interactions with Li‐TryR, the staples were located at the opposite face required for binding to the enzyme monomer. These modifications were systematically evaluated with respect to three key parameters: (i) α‐helical content, (ii) resistance to proteolysis, and (iii) ability to disrupt Li‐TryR dimerization [66].

FIGURE 3.

Structures of selected hydrocarbon and lactam stapled peptides 9–17.

Both hydrocarbon‐stapled (i, i+4 and i, i+7) and lactam‐bridged cyclic peptides exhibited significantly enhanced α‐helical content relative to the linear prototype (2). NMR spectroscopy in TFE/H₂O mixtures confirmed that both strategies produce highly ordered helical conformations (≈70%–90% helicity) [66].

Regarding the disruption of Li‐TryR dimerization, despite their high helicity, hydrocarbon‐stapled peptides (9 and 10) were inactive as dimerization inhibitors in a highly sensitive in‐house ELISA‐based Li‐TryR monomer displacement assay [36] (IC₅₀ > 75 μM). In contrast, the lactam‐bridged cyclic peptide (11) retained micromolar inhibitory potency, showing IC₅₀ values comparable to those of the active linear prototype counterpart (≈25 μM). These findings revealed that the chemical nature of the covalent constraint critically determines biological activity, even when the stapling positions are identical. On the other hand, MD simulations of peptide–protein complexes (with 10) revealed that hydrocarbon stapling disrupts key binding interactions, particularly introducing unfavorable hydrophobic constraints that alter the orientation of the critical hotspot residue (Gln5), previously identified as essential for Li‐TryR inhibition [66]. In contrast, lactam cyclization (11) preserved the native‐like orientation of this residue and maintained key intermolecular contacts with the enzyme surface [66].

Proteolitic stability studies showed that the lactam‐bridged peptide (11) exhibited a dramatic improvement in proteolytic stability, with a ≈14‐fold increase in half‐life against proteinase K (a promiscuous nonspecific serine protease) compared to the linear peptide 2 [66]. This improvement was achieved without the loss of functional activity, identifying the lactam‐cyclized peptide (11) as a validated lead combining efficacy and enhanced pharmacokinetic robustness. Hydrocarbon‐stapled peptides (9, 10) also showed enhanced stability; however, this did not translate into functional inhibition [66]. The results allowed us to conclude that lactam‐based cyclization achieves a superior balance between structural stabilization, functional binding, and resistance to proteolytic degradation, resulting in a peptide (analog 11) that retains activity while displaying markedly improved proteolytic stability relative to the linear peptide 2 (14‐fold) and better drug‐like properties [66].

A similar strategy was followed with linear peptide prototypes 13‐mer 4 and 9‐mer 8, which were the most potent as Li‐TryR dimerization disruptors, and were further optimized by introducing side‐chain–to–side‐chain lactam bridges, that were separated by one helical turn (i, i+4), to stabilize the helical conformation, improve proteolytic resistance, and preserve key interaction surfaces. Systematically scanning stapling positions (covalent amide cyclization) was explored in both the 13‐mer and truncated 9‐mer peptides (Figure 3, cyclic peptides 12–17), all positioned on the alpha‐helix face opposite to that of binding to the Li‐TryR monomer (recognition face). Both 13‐mer and 9‐mer cyclic peptides were synthesized using solid‐phase peptide synthesis and orthogonal protection strategies [67]. Conformational analysis by circular dichroism and molecular dynamics simulations showed that lactam cyclization consistently increased α‐helicity relative to the linear precursors, particularly in the 13‐mer series (12–15). Reached 38%–65% helicity versus 16% for the linear 13‐mer 4. MD simulations (explicit solvent) supported the experimental data and highlighted that the staple position dictates helix continuity and flexibility, with 13‐mer 14 (7, 11 link) locked in a rigid, nonbioactive conformation, while 12 (3, 7 link), 13 (4, 8 link), and 15 (8, 12 link) retained conformations compatible with activity [67].

Biological evaluation showed that the position of the staple was decisive for activity in the 13‐mers. Several 13‐mer cyclic peptides (12, 13, and 15) inhibited Li‐TryR oxidoreductase activity, with IC₅₀ values in the low micromolar range (0.9–5.3 μM) and effectively disrupted enzyme dimerization (IC_50dim = 9.9–14.4 μM), while peptide 14 was inactive in both assays (>75 μM) [67]. In contrast, shorter 9‐mer cyclic peptides (16 and 17) were inactive, highlighting the importance of peptide length and conformational flexibility for productive interaction with the dimer interface. In addition to improved potency. The most active cyclic peptides showed a two to threefold increase in resistance to proteolytic degradation compared with their linear counterparts, supporting the stabilizing effect of lactam stapling when the staple is correctly positioned [67]. The location of the lactam bridge was critical: certain constraints preserved inhibitory activity (as 13‐mers 12, 13, and 15), whereas others rigidified the peptides into nonbioactive conformations (as 13‐mer 14 and 9‐mers 16 and 17) [67]. However, despite their favorable enzymatic profiles, neither linear nor cyclic peptides were intrinsically leishmanicidal up to 25 μM (in vitro against L. infantum promastigotes and axenic amastigotes), which was attributed to poor membrane permeability, as assessed by fluorescence microscopy using fluorescein‐labeled peptide conjugates linear 18 and cyclic 19 (Figure 4) (of the prototype 4 and the lactam‐bridged analog 12, respectively) [67].

FIGURE 4.

Structures of selected peptide conjugates 18–20.

To overcome this limitation and to increase cell permeability, selected TryR‐inhibitory peptides were conjugated to cationic cell‐penetrating peptides (CPPs) at their N‐ or C‐terminal ends [67]. Polyarginine (R₉) and Tat sequences were used as CPPs due to their low toxicity and efficient uptake by Leishmania parasites and macrophages [68, 69, 70]. These conjugates exhibited potent in vitro activity against both promastigote and axenic amastigote forms of L. infantum, with IC₅₀ values (3.5–4.6 and 1.5–2.8 μM, respectively) comparable to those of the reference drugs miltefosine and edelfosine [67]. Fluorescence microscopy studies further demonstrated efficient parasite uptake of a fluorescein‐tagged R₉ peptide conjugate 20 (Figure 4), bearing a fluorescein moiety linked through a polymethylene spacer to the N‐terminal position of prototype 4 and an R₉ chain conjugated at the C‐terminal position. The compound was localized initially at the flagellar pocket and later followed by broader intracellular distribution, directly visualizing parasite entry, which supports a mechanistic link between cellular entry and antiparasitic activity [67].

On the whole, this study provided the first evidence that disruption of TryR dimerization can translate into antileishmanial activity at the cellular level. It establishes helix‐stabilized CPP‐conjugated peptides as a viable platform for targeting PPIs in parasitic enzymes and opens new avenues for the development of nonclassical TryR inhibitors with therapeutic potential.

Based on our previous studies on linear and lactam‐bridged peptide inhibitors, the next challenge was to explore the synthesis of all‐hydrocarbon stapled peptides 21, 22 (Figure 5), analogs of the most potent linear peptide 13‐mer 4. These analogs were designed to stabilize the bioactive α‐helical conformation and enhance protease resistance (proteolytic stability) with respect to the previous linear and lactam‐bridged peptides [71]. Hydrocarbon staples were introduced via i, i+4 side‐chain cross‐linking between the amino acid side chains separated by one helical turn. Hydrocarbon staples were placed at positions previously shown to tolerate cyclizations (3, 7 and 4, 8 of the peptide sequence) and located on the nonrecognition face of the helix to preserve key interaction residues. This strategy aimed to improve structural preorganization while maintaining inhibitory activity against the target enzyme. Thus, all‐hydrocarbon stapled peptides 21 and 22 (3/7 and 4/8 cyclized, respectively) were pursued. Moreover, fluorescein‐labeled peptide 23 and polyarginine‐conjugated 24 (of the most proteolytically stable hydrocarbon‐stapled peptide 21) were also prepared (Figure 5) [71].

FIGURE 5.

Structures of selected all‐hydrocarbon stapled peptides 21–24.

The all‐hydrocarbon stapled peptides were synthesized by Fmoc‐based solid‐phase peptide synthesis, followed by on‐resin ring‐closing metathesis using a second‐generation Grubbs catalyst under microwave irradiation [71]. Incorporation of α‐methyl‐alkenyl amino acids enabled the formation of the hydrocarbon cross‐link. Circular dichroism analysis revealed that hydrocarbon stapling markedly increased α‐helical content compared to the linear prototype, with helicity values reaching ~64% for the most stable stapled analog 21 [71]. Importantly, stapling conferred a high improvement in proteolytic stability, with up to an 18‐fold increase in the half‐life against proteinase K relative to the linear peptide and a substantial enhancement over previously reported lactam‐bridged analogs [67, 71].

In biochemical assays, the stapled peptides retained micromolar inhibitory activity against Li‐TryR oxidoreductase function, although with somewhat reduced potency compared to the most active linear and lactam‐bridged peptides [71]. Unexpectedly, evaluation of their effects on enzyme oligomerization revealed a profound change in the mechanism of action. Unlike earlier peptides that disrupted TryR dimerization, the all‐hydrocarbon‐stapled analogs abolished dimer disruption and instead acted as stabilizers of the homodimer [71]. This paradoxical behavior suggests that inhibition of TryR catalytic activity can arise not only from dimer destabilization but also from excessive stabilization of the dimer, potentially restricting the conformational dynamics required for catalysis. Notably, the all‐hydrocarbon‐stapled peptides activated, rather than inhibited, human GR, pointing to mechanistic divergence between the parasite and host enzymes [71].

Despite their enhanced stability and helicity, the all‐hydrocarbon‐stapled peptides displayed negligible intrinsic antileishmanial activity, which correlated with their poor ability to cross parasite membranes, as shown by fluorescence microscopy with fluorescein‐labeled peptide 23 (prepared by fluorescein‐labeling the most proteolytically stable peptide 21). 23 was not able to cross the cell membrane. To overcome this limitation and to enhance intracellular delivery, peptide 21 was conjugated to a nona‐arginine (R₉) cell‐penetrating sequence to give 24 that exhibited potent in vitro activity against both promastigote and axenic amastigote forms of L. infantum. This compound showed EC₅₀ values (1.95 ± 0.14 and 2.14 ± 0.5 μM, respectively), which were better than those of clinically used drugs such as miltefosine and edelfosine [71]. Cytotoxicity assays indicated an acceptable selectivity profile [71]. The results of this study demonstrated that all‐hydrocarbon stapling is a powerful tool to enhance the conformational stability and protease resistance of peptide‐based TryR inhibitors, but it also alters their mode of interaction with the enzyme, shifting from dimer disruption to dimer stabilization. However, while stapling alone did not improve cell permeability, combination with a cell‐penetrating moiety successfully translated biochemical potency into cellular antileishmanial activity.

2.4. α/β‐Peptide Foldamers

Continuing with our objective to develop dimerization disruptors of Li‐TryR proteolytically stable and to overcome the poor cellular uptake of α‐helical peptides, we carried out a backbone modification strategy. We designed α‐helical backbone‐modified peptide flexible foldamers [72, 73, 74, 75], composed of both natural α‐amino acids and unnatural β³‐amino acid residues (mixed α/β‐peptides) following the strategy described by Gellman [73, 74, 75]. The design strategy involved periodic replacement of α‐amino acids with β³‐amino acids that conserve the native side chains of the previously identified α‐helical linear prototype peptide 4, following established α/β³ substitution patterns intended to mimic α‐helical conformations. Earlier studies have shown that such mixed α/β³‐peptide can successfully mimic helical conformations and markedly reduce vulnerability to proteolytic enzymes. Additionally, because β³‐amino acids include an extra CH₂ group, they may enhance protease resistance and permit residual flexibility in the peptide backbone compatible with protein–protein recognition [76, 77, 78, 79]. Multiple lengths (9–13 residues) and different α/β substitution patterns were explored to map the interplay between preorganization, flexibility, and binding at the dimer interface [80].

A small library of 9‐ to 13‐residue α/β³‐peptide foldamers (Figure 6) with different α/β³ substitution patterns was synthesized using solid‐phase synthesis under microwave‐assisted conditions on Rink amide resin using HCTU/DIEA couplings for both α‐ and β³‐amino acids [80]. The most stable analog was further derivatized with 5‐FITC (5‐fluorescein 5‐isothiocyanate) via a PEG linker for uptake visualization and covalently coupled to nona‐arginine (R₉) for delivery. This library design retained the side‐chain sequences of the parent 13‐mer/11‐mer/9‐mer peptides while modulating backbone composition in defined α/β patterns (Figure 6).

FIGURE 6.

Structures of selected foldamers 25–31.

Several foldamers displayed a high improvement in proteolytic stability compared with both linear and cyclic α‐peptide prototypes, with half‐lives increased by more than an order of magnitude. Thus, 13‐mer foldamer 27 reached t _{1 /2} ≈ 517 min, >22‐fold longer than the matched linear α‐peptide 4 (t _1/2 ≈ 23 min) [80] and about 8–10 times longer than prior lactam‐stapled α‐analogs 12–15 (t _{1 /2} ≈ 52–64 min) [67]. Not all patterns improved stability, underscoring the importance of positional control in α → β³ replacement. However, as with free peptides, α/β foldamers showed no intrinsic antileishmanial activity up to 25 µM, which is consistent with limited permeability [66, 67, 80].

Enzymatic inhibition and dimer disruption evaluation showed a strong length dependence for enzyme inhibition. 11‐mer and 9‐mer foldamers showed lower activity than that of the parent α peptides, while most of the 13‐mer α/β‐foldamers with the ααβαααβ pattern (25, 26, 28, and 29) retained low‐micromolar inhibitory activity (IC₅₀ = 1.65 ± 0.43–3.11 ± 0.17 µM). This activity was similar to that of the parent α‐prototype 4 against Li‐TryR oxidoreductase function and was the best dimer disruptor, while 1‐mer and 9‐mer analogs lost activity [80]. MD‐guided energy decomposition rationalized this: truncation weakened interactions with residues such as Asp453 and reduced the number of significant protein contacts, lowering computed binding energies (13‐mer > 11‐mer > 9‐mer) [80]. As free peptides, α/β foldamers showed no intrinsic antileishmanial activity, which again indicated limited cell permeability. To overcome this limitation, the most stable foldamer (27) was further functionalized with a fluorescent probe (FITC via a PEG linker) for uptake studies (Figure 6, compound 30) and covalently coupled to an oligoarginine (R₉) cell‐penetrating peptide (compound 31) to enhance cellular delivery. Conjugation yielded potent in vitro activity against L. infantum promastigotes and axenic amastigotes (IC₅₀ ≈ 0.98  ± 0.02 µM and 1.9 ± 0.5 µM, respectively) and reduced THP‐1 cytotoxicity (LC₅₀ ≈ 5.8 ± 1.4 µM) relative to earlier R₉‐conjugated α‐linear and lactam‐stapled peptides, thus, improving the selectivity index while maintaining efficacy. Control (R₉ alone) was inactive, confirming that activity depends on the TryR interface sequence delivered by the cell‐penetrating peptide [80].

This study positions mixed α/β foldamers as a generalizable platform to target enzyme dimerization interfaces in kinetoplastids: it couples durable protease resistance with retained PPI recognition and, when paired with a CPP module, delivers potent in‐cell antiparasitic effects while improving selectivity versus host cells. The study complements stapling approaches by showing that backbone engineering can tune the critical balance between preorganization and conformational freedom required at dynamic PPIs like the TryR dimer interface.

2.5. Small‐Molecule Peptide Conjugates (SMPCs)

To advance our development of peptide‐based Li‐TryR dimerization disruptors, we sought to enhance the binding affinity of our previously reported peptide inhibitors by conjugating them to selected hydrophobic groups (Figure 7). The design of these SMPC was guided by computational modeling of prototype peptide 8 (PKIIQSVGI) bound to a Li‐TryR monomer [81]. The C‐terminally truncated 9‐mer peptide 8 was chosen as the prototype since it was identified as the minimal sequence that inhibited Li‐TryR oxidoreductase activity while also interfering with the dimerization of the enzyme [36]. The computational study revealed nearby regions that could be exploited for additional contacts. Our structural analyses identified two large, spatially contiguous hydrophobic cavities lined by residues Phe396, Pro398, and Leu399, positioned approximately 5 Å from the N‐terminal region of peptide 8 [81]. Based on these structural features, we designed, synthesized, and evaluated a series of SMPCs (Figure 7), incorporating bulky nonpolar hydrophobic groups (phenyl, naphthyl, biphenyl, acridine, adamantyl, tricyclic naphtholactam) to the N‐terminus of 8, via linkers of different length and chemical nature (PEG1, PEG2, β‐Ala) to modulate reach, flexibility, and solubility. Moreover, a series of SNPCs with no spacers were also prepared. Two kinds of linkages (amide bonds or aminoalkyl linkages) were also explored to enable structural and electronic variation of the connection between the peptide and its appended groups. Hydrophobic groups were intended to interact with adjacent pockets involving Phe396, Pro398, and Leu399 to potentiate binding and promote both dimer‐disrupting capability and inhibition of enzymatic activity by ultimately shifting TryR from its functional dimeric form toward an inactive monomer [81].

FIGURE 7.

Selected peptide conjugates 32–38.

Biological testing showed that all conjugates outperformed peptide prototype 8 with significant increases in both enzymatic inhibition and dimer disruption. The nature of the hydrophobic group is the major determinant of the inhibitory potency. Among them, naftolactam‐based conjugates (32–35, Figure 7) were the most potent, with IC₅₀ values around 6.7–10.1 µM for enzymatic inhibition and 5.3–7.8 µM for dimer disruption. Phenyl groups were significantly less effective. Adamantyl and acridine moieties also enhanced potency (IC₅₀ = 7.6 and 8.1 µM, respectively), although the former suffered from reduced solubility and the latter showed preferential effects on dimerization (IC₅₀ = 5.5 µM). The linker length and type (PEG, β‐alanine, or none) have a minimal influence on activity. However, the kind of linkage is important for activity; aminoalkyl‐type linkers (36–38) improve oxidoreductase inhibition (IC₅₀ = 6.3–9.2 µM) but show a slight decrease in dimerization disruption (IC_{50 dim} = 12.6–21.2 µM). A positive charge at the N is detrimental for dimerization [81].

Docking and molecular dynamics simulations support a mechanism in which hydrophobic conjugation stabilizes the α‐helical fold of 8 within the interfacial cavity but ultimately promotes local structural rearrangements in the G448–S470 region. This destabilization progressively shifts the equilibrium toward the monomeric inactive form of TryR, which is consistent with the observed dual impact on activity and dimerization [81].

All these results position SNPCs, particularly those bearing amphipathic tricyclic lactams (naftolactams), as an effective approach for targeting Li‐TryR dimerization and as promising leads for antileishmanial drug development. It also provides a valuable framework for further optimization of peptide‐based dimerization disruptors and underscores the therapeutic potential of targeting oligomeric interfaces in trypanosomatid enzymes.

3. Nonpeptidic Dimerization Disruptors (Small‐Molecule Proteomimetics)

3.1. Pyrrolopyrimidines Versus 5‐6‐5 Imidazole‐Phenyl‐Thiazole

All the work mentioned up to this point allowed us to identify linear and cyclic peptides as well as α/β‐peptide foldamers, derived from an α‐helical region at the Li‐TryR dimer interface, that could effectively disrupt dimerization and inhibit enzymatic activity. However, their therapeutic potential was limited by poor cell permeability and the need for conjugation to cell‐penetrating sequences to achieve intracellular delivery and exert antiparasitic effects in cultured Leishmania parasites.

To address these shortcomings, our subsequent efforts shifted toward the design of nonpeptidic Li‐TRyR dimerization disruptors based on small‐molecule proteomimetics (scaffolds) capable of mimicking the key molecular interactions of the active prototype 4. Two structural families were explored: pyrrolopyrimidines [82, 83, 84, 85] and 5‐6‐5 imidazole‐phenyl‐thiazole [85, 86] scaffolds, reported in the literature to mimic an α‐helix [86, 87], each decorated with functional groups mimicking the three key side chains of residues (Lys‐2, Gln‐5, and Ile‐9) located at the i, i+3, and i+7 positions of the original α‐helix (Figure 8). These side chains were suggested by alanine scanning and MD simulations of a complex of a monomer of LiTryR with peptide prototype 4 to have a large contribution to the inhibitory potency [82]. The objective was to obtain drug‐like, cell‐permeable small molecules able to disrupt Li‐TryR dimerization and/or inhibit its enzymatic activity. Although the synthetic optimization of these scaffolds required substantial methodological refinement, a diverse library of 15 compounds was successfully synthesized [82]. The pyrrolopyrimidine series required extensive optimization of both solid‐phase and solution‐phase methods. The imidazole‐phenyl‐thiazole series was synthesized via nucleophilic aromatic substitution, Ullmann coupling, and Hantzsch thiazole assembly [82].

FIGURE 8.

(A) Lysine, glutamine and isoleucine residues at positions i, i+3 and i+7 in the prototype peptide 4. (B) Selected pyrrolopyrimidine proteomimetics 39–41. (C) Selected imidazole‐phenyl‐thiazole proteomimetics 42–43.

Comparative biological evaluation revealed that pyrrolopyrimidine derivatives (39–41) were significantly less potent than imidazole‐phenyl‐thiazole derivatives (42 and 43). The pyrrolopyrimidines showed moderate inhibitory activity against Li‐TryR, and this was strongly dependent on the hydrophobic R₃ substituent (Ile mimetic). Alkyl or benzylamino substituents gave inactive compounds. However, the presence of a phenylethyl, a phenylpropyl, or polycyclic aromatic groups as naphthyl or biphenyl (39–41), gave the best analogs of this series (IC₅₀ = 39–52 µM). The compounds showed no significant capacity to disrupt the dimeric assembly [82]. In contrast, the imidazole‐phenyl‐thiazole derivatives were significantly more potent, with compounds 42 and 43 bearing bulky hydrophobic substituents, achieving low‐micromolar IC₅₀ values in activity (5.1 and 8.6 µM, respectively) against Li‐TryR and inhibiting dimerization (IC₅₀ = 26 and 32 µM, respectively). Their enhanced performance correlates with the capability of their extended aromatic R₃ groups to occupy a large hydrophobic pocket at the dimer interface, as supported by molecular dynamics simulations [82]. These analyses further revealed that compound 43 can reproduce critical binding interactions seen in the prototype peptide inhibitor, thereby stabilizing conformations incompatible with dimer maintenance. Regarding their ability to disrupt dimerization, 42 and 43 behave as moderate disruptors with IC₅₀ values of 26 and 32 µM, respectively [82]. Notably, compounds 42 and 43 possessed high aqueous solubility and intrinsic fluorescence, which facilitates the direct visualization of their uptake by parasites [82].

X‐ray crystallography of compound 39 bound to Li‐TryR revealed that pyrrolopyrimidines bind in the active site (mepacrine site), overlapping with the natural polyamine‐binding region interacting with Trp21 and blocking the access of the natural substrate but without triggering functional dimer dissociation [82]. Attempts to co‐crystallize the most potent inhibitors 42 and 43 with the enzyme failed due to precipitation of the protein. MD simulations of imidazole‐phenyl‐thiazole 43 bound at the dimer interface indicated that 43 binds at the dimer interface, with the biphenyl group interacting in a large hydrophobic pocket. The results suggest that the presence of a bulky group at R₃ is very important for the binding and highlights the importance of a bulky aromatic group for dimer disruption.

The biological evaluation underlines the therapeutic promise of this scaffold. While most analogs lacked antiparasitic activity, compounds 42 and 43 showed potent leishmanicidal effects against both promastigotes (EC₅₀ ^{promastigotes} = 12.8 and 5.3 µM, respectively) and intracellular amastigotes (EC₅₀ ^amastigotes = 12.8 and 5.3 µM, respectively). Their activities were comparable to those of miltefosine and the previously optimized peptide inhibitors, but importantly, they did not require any delivery‐enhancing conjugation. Both compounds exhibited good cell permeability, as confirmed by fluorescence microscopy. Both molecules have the ability to penetrate macrophages and reach phagolysosomal compartments, leading to substantial parasite clearance with acceptable cytotoxicity profiles on human THP‐1 cells (EC₅₀ ^TPH−1 = 26 and 14,2 µM, respectively) [82].

Collectively, these findings position imidazole‐phenyl‐thiazole proteomimetics as the first nonpeptidic Li‐TryR dissociative inhibitors with demonstrable antileishmanial activity in cellular models. The strong correlation between Li‐TryR inhibition, partial dimer disruption, and antiparasitic efficacy supports the enzyme as a biologically relevant target for these compounds. This work not only validates the proteomimetic strategy for modulating PPIs at enzymatic interfaces but also opens a new avenue for the development of drug‐like, cell‐permeable small molecules aimed at treating leishmaniasis.

3.2. 1,2,3‐Triazole‐Phenyl‐Thiazole‐Based Dimerization Disruptors

As a further step in our interests toward the development of drug‐like small‐molecule dimerization disruptors of the Li‐TryR homodimerization interface, we shifted, using a scaffold hopping approach, to a novel 1,2,3‐triazole‐phenyl‐thiazole scaffold (Figure 9) designed as a bioisostere of the previously studied imidazole‐phenyl‐thiazole analogs [82, 88].

FIGURE 9.

Selected triazole‐phenyl‐thiazole proteomimetics 44–50, and symmetrical analog 51.

A chemically diverse library of 31 triazole‐based compounds (Figure 9) was synthesized with variations at four positions (R1–R4) of the new scaffold to explore SAR [88]. Several compounds were potent inhibitors of enzymatic activity and significantly improved the ability to destabilize the LiTryR homodimer when compared with the reference imidazoles. The screening revealed that analogs bearing bulky hydrophobic substituents at the R₃ position, such as naphthyl, biphenyl groups, or dibenzofuranyl groups (44, 45, and 47), showed similar or stronger inhibitory effects on both enzymatic activity and dimer disruption (IC₅₀ ^act = 5.9–10.9 µM; IC₅₀ ^dim = 4,8–8.8 µM) with respect to imidazole‐containing counterparts. Disubstituted analogs at the thiazole ring (48–51) also showed IC₅₀ values in the low micromolar range for enzyme inhibition (approximately 4–19 μM) and sub‐10 μM potencies in the dimer disruption assay. The best compound of this series was disubstituted thiazole analog 48 (IC₅₀ ^act = 4.3 µM; IC₅₀ ^dim = 7.1 µM). As will be discussed below, this compound behaves as a potent slow‐binding inhibitor, with an overall Ki* of ≈0.5 μM, and as a disruptor of dimer dissociation. Further SAR studies confirmed the essential structural features required for activity: a positively charged butylamine at R₁ is indispensable for activity, its removal or replacement dramatically reduces activity; an imidazolidinone moiety at R₂ optimizes both potency and cytotoxicity profiles; and a hydrophobic poly(hetero∫)aromatic substituent at R₃ is critical for dimer disruption, with naphthyl, biphenyl, and dibenzofuranyl substituents providing the best complementarity to the interfacial cavity (with key interfacial subpockets of the enzyme) [88]. Compounds lacking these features, such as truncated scaffolds or analogs with aliphatic R₃ groups, were essentially inactive [88].

Mechanistic enzymology revealed that the most active molecules, particularly 44, 45, and 48, and polyaromatic analogs such as 47, behave as potent noncompetitive slow‐binding inhibitors via a two‐step isomerization (EI → E*I). After rapid formation of an initial enzyme‐inhibitor (EI) complex, a slower isomerization event yields a high–affinity complex (E*I), consistent with progressive loss of enzyme activity during the reaction. Dimer disruption correlates with the slow isomerization step and catalytic inactivation. This kinetic behavior parallels that of earlier peptide disruptors [42], confirming that small molecules can effectively mimic the conformational destabilization induced by larger proteomimetic inhibitors [88].

In parallel, computational modeling provided compelling structural support for these findings. Since crystallographic attempts failed to produce enzyme–ligand complexes (due to large precipitation of the enzyme), we used FTMap, CAVER, and AutoDock analyses to examine druggable regions within the Li‐TryR homodimer. These methods converged on a previously unexplored hydrophobic cavity at the center of the dimer interface, located near the interfacial α‐helices implicated in dimer stabilization. This region is not only accessible through solvent‐connected tunnels but is also ideally suited to accommodate the polyaromatic fragments of the new triazole series. Docking and molecular dynamics simulations confirmed that the most active triazole derivatives (44, 45, and 48) insert their aromatic R₃ substituents deep into hydrophobic pockets shaped by residues such as Pro435′, Phe367′, Leu72/Leu72′, and their R₂ imidazolidinone toward hydrogen‐bond donors onThr65 or Ser433, while the terminal ammonium at R₁ forms stabilizing interactions with Glu466/Glu466′. The disruption of key inter‐monomer hydrogen bonds, particularly those involving Glu436, emerged as a plausible structural rationale for the observed dimer destabilization [88].

Based on these computational insights, we carried out the rational design of the symmetric analog 51 (Figure 9), intended to exploit the C₂ symmetry of the Li‐TryR homodimer and to bind in the so far unexplored hydrophobic central interfacial cavity as a putative binding site. The compound was designed to simultaneously and symmetrically occupy both entrance tunnels (solvent‐connected) leading into the central interfacial cavity. Docking simulations predicted optimal alignment of the two aromatic arms with the hydrophobic subpockets and reinforced interactions with acidic residues on both monomers [88]. Compound 51 showed a remarkable increase in potency, reaching submicromolar inhibition of LiTryR (IC₅₀ of 0.4 μM) and the strongest dimer disruption (IC₅₀ of 8 μM) among all synthesized analogs [88]. Thus, compound 51 validated this design rationale, emerging as the most potent inhibitor of the entire series and possessing robust dimer‐destabilizing capacity.

The biological relevance of these findings was confirmed through a broad panel of antiparasitic assays. Several triazole derivatives demonstrated micromolar antileishmanial activity against extracellular promastigotes and axenic amastigotes, with compounds such as 45, 46, and 51 being among the most active (EC₅₀ ^{promastigotes} = 3.8–5.8 μM) and (EC₅₀ ^amastigotes = 4.9–11.6 μM). Compound 46, which has a shorter R₁ linker and a biphenyl R₃, displayed particularly favorable selectivity indices (SI^pro = 10; SI^am = 11.6), while the symmetric analog 51 ranked as one of the most potent and selective inhibitors for promastigotes (EC₅₀ = 3.8 μM) and axenic amastigotes (EC₅₀ = 4.9 μM). Notably, both 46 and 51 also exhibited low cytotoxicity toward HepG2 cells, improving that of prior imidazole‐based candidates [88]. On the other hand, some of these molecules also showed activity against intracellular amastigotes within THP‐1‐derived macrophages (TPH‐1 infection model), an essential benchmark for therapeutic relevance. Although intracellular activity was generally decreased compared with axenic assays, compound 51 again stood out, achieving an EC₅₀ of ~22 μM, making it the first small‐molecule dimerization disruptor to show meaningful intracellular efficacy. Moreover, compound 51 also showed good activity against other trypanosomal TryR enzymes from T. brucei (TbTryR) and Trypanosoma congolense (TcoTryR), causing sleeping sickness and nagana, respectively. IC_50 values (LiTryR = 0.4 μM; TbTryR = 1.0 μM; TcoTryR = 0.5 μM) [88].

Fluorescence microscopy further revealed that triazole derivatives (that were fluorescent) can enter Leishmania parasites efficiently, suggesting that insufficient permeability is no longer the main limitation of this chemical class of inhibitors [88].

Taken together, our studies allowed us to identify triazole‐phenyl‐thiazoles as the most potent small‐molecule Li‐TryR dimerization disruptors reported to date. Their combined enzymatic, structural, and cellular behaviors highlight the relevance of the central dimer interface as a newly validated druggable site. In particular, the discovery and validation of the C₂‐symmetric analog 51 provide a strong proof‐of‐concept of a new druggable region on Li‐TryR for ligand binding within this interfacial region and offer a strong foundation for next‐generation antileishmanial drug discovery.

3.3. From 1,2,3‐Triazoles to 1,2,3‐Triazolium Salt‐Based Dimerization Disruptors

Continuing with our efforts to the development of improved inhibitors, starting from the 1,2,3‐triazole‐phenyl‐thiazole derivatives (noncompetitive, slow‐binding Li‐TryR dimer disruptors), N‐alkylation (quaternization) of the triazole ring was explored as a strategic approach to generate 1,2,3‐triazolium salts with an additional formal positive charge and modified H‐bonding capacity [89], aiming at to enhance Li‐TryR inhibition, to improve water solubility, and to increase antileishmanial potency and selectivity, especially against intracellular amastigotes.

The synthesis of the triazolium series of compounds (Figure 10) was carried out through an optimized alkylation of the corresponding Cbz‐protected triazoles using methyl triflate under mild conditions, followed by hydrogenolytic deprotection to afford the final compounds 52–56 [89]. An additional functionalization step at the R₂ position allowed the introduction of piperidine‐containing fragments, yielding derivatives 57 and 58 (the design of these compounds was guided by crystallographic studies). Across the series, N‐alkylation consistently increased water solubility by one to two orders of magnitude relative to the parent neutral triazoles (i.e., 53: 23.5 mg/mL vs. 2.1 mg/mL; 54: 12.4 mg/mL vs. 0.2 mg/mL) [89]. Biological evaluation showed that several compounds were potent inhibitors of enzymatic activity with IC₅₀ values in the low micromolar range (2.7–9.4 μM).

FIGURE 10.

Selected triazolium salts‐phenyl‐thiazole chemotypes 52–58.

From a mechanistic point of view, quaternization profoundly reshaped the inhibition mechanisms. Many triazolium salts lost their ability to act as dimerization disruptors but instead behaved as classical competitive inhibitors, exemplified by compound 52, which showed kinetic data consistent with a competitive inhibition pattern with a K _i of 7.3 ± 0.9 μM, and binds directly within the polyamine‐binding region of the active site of the enzyme, as confirmed by X‐ray crystallography [89]. The crystal structures of Li‐TryR in complex with 52 (2.5 Å) and with the piperidinyl analog 57 (3.0 Å) demonstrated a conserved active‐site binding mode stabilized by electrostatic interactions with Glu18 and Asp116, hydrophobic stacking with Trp21, Met113, and Tyr110, and orientation of the aminobutyl R₁ chain toward negatively charged residues at the entrance of the cavity [89]. These structural insights guided further refinement, such as the design of piperidinium‐containing derivatives (e.g., 57) and rationalized the enhanced potency of the piperidine variants, which exhibit stronger interactions with key active‐site residues [89].

Despite this general mechanistic shift, several triazolium salts, particularly those incorporating a bulky, highly hydrophobic biphenylethyl substituent at R₃ (54, 55, 56), retained the ability to disrupt the Li‐TryR dimer, with IC₅₀ values (IC₅₀ ^dim = 10–57 μM). Among them, compound 54 was the most potent dimerization disruptor (IC₅₀ ^dim = 10.3 μM) and displayed a characteristic slow‐binding time‐dependent inhibition consistent with a two‐step induced fit inhibition mechanism, which ultimately perturbs the delicate equilibrium between the active homodimer and its inactive monomeric form [89]. Kinetic analysis revealed the formation of a low‐affinity EI complex followed by conversion into a high‐affinity E*I state with submicromolar overall dissociation constants. Molecular dynamics simulations supported a model in which 54 initially binds to the enzyme at the active site, as observed for 52, but subsequently accesses the interfacial cavity between the two subunits, where its bulky hydrophobic R₃ moiety penetrates deeply into the cleft of a hydrophobic cavity at the subunit interface and destabilizes the dimer interface. In contrast, active‐site binders such as 52 fail to reach energetically favorable minima within this interfacial region, which is in agreement with their inability to trigger dimer disruption [89]. A key outcome of this dual mechanistic landscape is that the triazolium scaffold becomes inherently bifunctional, supporting both active‐site targeting and dimer–interface interaction depending on peripheral substituents. This work demonstrates how subtle changes to a heterocycle can switch the behavior of an inhibitor from allosteric slow‐binding to classical competitive inhibition.

The biological consequences of these distinct binding modes were reflected in the antiparasitic assays. Many triazolium salts (as 54 and 55) showed comparable or superior antileishmanial potency against axenic amastigotes (EC₅₀ = 4.8 and 11.6 μM, respectively) with respect to the parent neutral triazoles. More importantly, they showed much lower cytotoxicity toward HepG2 cells (CC₅₀ > 75 µM), leading to substantially improved selectivity indices (SI = 15.6 and 6.5, respectively). The most compelling findings emerged from testing against intracellular amastigotes (in macrophage infection assays). Here, the dimerization‐disrupting derivatives demonstrated the most notable improvements. Compounds 54 and 55 showed potent activity (EC₅₀ values below 10 µM; 8.7 and 7.3 μM, respectively) and low toxicity (CC₅₀ values > 75 µM), resulting in excellent selectivity indices (SI > 8.6–0.3.6), superior to those of the parent triazoles (whose intracellular efficacy was modest and whose toxicity frequently limited their therapeutic window). These results highlight a clear correlation between dimer disruption and enhanced intracellular potency [89].

To provide evidence of on‐target activity within the parasite, nonprotein thiol levels in L. infantum promastigotes (a readout directly connected directly to the function of Li‐TryR) were quantified using CMF‐based flow cytometry. Treatment with compound 54 resulted in a detectable reduction in thiol content, particularly under conditions of mild oxidative stress or blockade of glutathione biosynthesis via buthionine sulfoximine, compatible with inhibition of endogenous Li‐TryR and impaired regeneration of reduced trypanothione (in other words, providing direct functional evidence that the inhibitor perturbs trypanothione homeostasis within the parasite, supporting on‐target activity in living cells) [89]. Such demonstration is rare among reported TryR inhibitors and strenghthens the translational relevance of the triazolium series of inhibitors. Moreover, fluorescence microscopy additionally confirmed an efficient intracellular accumulation of 54. All mentioned results indicate that 54 is capable of interacting with Li‐TryR within intact parasites and perturbing the thiol redox balance [89].

In summary, this study makes several noteworthy contributions to antileishmanial drug discovery. It provides a new class of Li‐TryR inhibitors whose activity can be tuned between competitive and allosteric dimer‐disrupting mechanisms. It identifies highly promising lead compounds (54 and 55) with robust intracellular potency, excellent selectivity, and demonstrated involvement of trypanothione metabolism in intact parasites. It also provides a clear mechanistic rationale for how N‐alkylation reshapes Li‐TryR inhibition. Consequently, triazolium salts can be regarded as a versatile and compelling chemotype capable of addressing long‐standing challenges in targeting the redox machinery of kinetoplastid parasites and setting the stage for further optimization toward clinically relevant antileishmanial agents.

4. Summary and Outlook

The work presented in this manuscript represents a comprehensive journey toward developing first‐in‐class L. infantum trypanothione reductase (LiTryR) dimerization disruptors as a novel therapeutic strategy beyond classical active‐site inhibitors. By acting at a validated PPI hotspot within the dimer interface, these compounds provide an alternative mechanism capable of overcoming some of the intrinsic challenges associated with active‐site inhibition. Starting from rational target analysis, the dimer interface, particularly the α‐helix‐containing hotspot residue E436, was validated as a critical structural region for enzyme function and an exploitable site for inhibitor design. Interface‐derived linear peptides, stapled peptides, and α/β‐peptide foldamers demonstrated that disrupting Li‐TryR dimerization is feasible and leads to potent enzymatic inhibition. Although free peptides lacked cellular activity, conjugation to CPPs successfully restored antileishmanial efficacy.

Several generations of small‐molecule proteomimetics were designed to mimic key interfacial interactions. Imidazole‐phenyl‐thiazoles and later 1,2,3‐triazole‐phenyl‐thiazoles emerged as potent, drug‐like dimer disruptors with cellular activity, selectivity, and improved physicochemical properties. Structure–activity studies pointed to the importance of hydrophobic aromatic substituents for targeting an interfacial cavity and destabilizing the dimer. The most advanced derivatives, including symmetrical triazole analogs and triazolium salts, showed low‐micromolar to submicromolar Li‐TryR inhibition, effective dimer disruption, and good antiparasitic activity, including in intracellular amastigotes. However, important limitations remain, such as variable cellular activity among different chemotypes and the need for further optimization of pharmacokinetic properties and in vivo efficacy. Compared with classical active‐site inhibitors, dimerization disruptors offer several potential advantages, such as improved target selectivity, noncompetitive and slow‐binding inhibition mechanisms that are less sensitive to fluctuations in substrate concentration, access to previously unexplored druggable regions of the enzyme, and reduced competition with the natural substrate. These features make Li‐TryR dimerization disruption a promising and innovative therapeutic paradigm for the development of next‐generation antileishmanial agents.

Funding

This study was supported by Spanish MINECO, MICINN, and MICIU (SAF2009‐13914‐C02, SAF2012‐39760‐C02, SAF2015‐64629‐C2‐1R, PID2019‐104070RB‐C21, and PID2022‐136307OB‐C21), Comunidad de Madrid (PLATESA2‐CM S2018/BAA‐4370, BIPPED‐CM S‐BIO‐0214‐2006, and BIPPED‐2 S2010/BMD‐2457), and CSIC (PIE 201280E093 and PIE‐201980E028).

Conflicts of Interest

The authors declare no conflicts of interest.

Biographies

María‐José Camarasa Ph.D. in Chemistry (Complutense University of Madrid). Full Professor at IQM‐CSIC. She leads a renowed medicinal chemistry group focused on antiviral, antiparasitic, and anticancer drug discovery. Author of over 250 papers and inventor of 18 patents, she has coordinated major national, European, and industry projects. She has held key science management roles, serves on multiple editorial boards, and has received prestigious awards, including the Descartes Prize and recognition among the most cited CSIC scientists in the world's top 2%.

Miguel Maldonado received his B.S. in Chemistry from Universidad Complutense de Madrid and his Inter‐University M.S. in Drug Discovery (UCM, UAH, and CEU San Pablo). He is currently in the final year of his Ph.D. at the Institute of Medicinal Chemistry and Biomedicine (IQMB‐CSIC), supervised by Prof. María‐José Camarasa and Dr. Sonia de Castro. His doctoral research centers on medicinal chemistry strategies, specifically on the design and organic synthesis of novel antiparasitic agents and biomarker tools.

Alvaro de la Cruz‐Potenciano holds a Ph.D. in Medicinal Chemistry from the Universidad Complutense de Madrid. His doctoral research focused on the design and synthesis of antiviral compounds, with particular emphasis on therapies targeting influenza viruses and coronaviruses. In recognition of his contributions to this field, he received the Ramón Madroñero Award in 2025 from the Spanish Society of Therapeutic Chemistry. He is currently a postdoctoral researcher at the Institute of Medicinal Chemistry and Biomedicine (IQMB‐CSIC) and has gained international research experience through his stay at the Medicinal University of Graz.

Sonia de Castro obtained her Ph.D. in Chemistry from the Universidad Autónoma de Madrid. She conducted postdoctoral research at the National Institutes of Health (USA) as a Fulbright fellow and later joined the Instituto de Química Médica (IQM‐CSIC), where she became a Tenured Scientist in 2023. Over the last 14 years, her work has focused on the design, synthesis, and evaluation of heterocycles, peptides, peptidomimetics, and PNAs as pharmacological tools or drugs to treat potentially fatal diseases, such as AIDS, influenza, or coronavirus.

Dedicated to the memory of Dr. Sonsoles Velázquez.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.