In Vitro Leishmanicidal Efficacy of Synthesized Arylidene Analogues of Glitazone

ABSTRACT

Diabetes is a fast‐growing health issue in low‐ and middle‐income countries, with ~80% of diabetics living in the tropics and sub‐tropics. It is a deadly condition claiming the lives of millions of individuals annually, with no therapeutic treatment available to date. The management of diabetes is thus limited to symptomatic relief by glycemic control. Furthermore, the geographical overlap of diabetes and neglected tropical diseases (NTDs) is of concern, as diabetes is known to increase infection susceptibility and severity. In contrast, diabetes‐infection comorbidity can negatively affect treatment responses. Leishmaniasis ranks among the top 10 NTDs. Its current therapeutic treatment relies on a handful of drugs that are marred with two main shortcomings: toxicity and reduced efficacy due to pathogenic resistance. Hence, there is a pressing need for new, effective antileishmanial therapeutics. There is evidence of rising cases of leishmaniasis‐diabetes co‐infection, which may require the use of dual‐active therapeutics to curb them. In search of new effective antileishmanial agents with potential for dual use, we evaluated in vitro the antileishmanial and antidiabetic activities of a series of arylidenes derived from hydantoin, glitazone, and rhodanine scaffolds using phenotypic assays, some of which had previously been investigated for antidiabetic potential. Additionally, the antitrypanosomal potential of these compounds was also considered due to the taxonomic relation between Leishmania and Trypanosoma spp. and reported concerns of Chagas disease and human African trypanosomiasis‐diabetes comorbidities. Three leishmanicidal early leads with submicromolar activity were uncovered, but no antitrypanosomal or dual leishmaniasis‐diabetes active hits were identified.

1. Introduction

Diabetes and infectious diseases rank among the most lethal ailments to humans worldwide. Diabetes, a chronic metabolic condition characterized by high blood glucose levels (hyperglycemia) resulting from insufficient insulin production in the pancreas, is a major public health concern worldwide and ranks as the seventh leading cause of death globally (Glovaci et al. 2019). According to the international diabetes federation (IDF), as of 2021, there were an estimated 537 million diabetics worldwide, with the majority (3 in 4 adults or 80% of the world population) living in low‐and middle‐income countries (Dunachie and Chamnan 2019; IDF 2021; Unnikrishnan and Mohan 2016). Diabetes cases have been increasing steadily over several decades and are projected to reach 643 million by 2030 and 783 million by 2045 (IDF 2021), and 1.5 million deaths worldwide have been directly attributed to diabetes each year (WHO 2023a). Two types of diabetes mellitus (DM) affect humans, i.e., type 1 diabetes (T1D) resulting from insufficient insulin secretion by the pancreas and type 2 diabetes mellitus (T2DM) due to insulin resistance development, with the latter being the most prevalent form accounting for 90% of all diabetes cases (IDF 2021).

The management of diabetes, mainly its symptomatic relief, relies solely on the use of drugs focusing on good glycemic control, especially for T2D. Glycemic control refers to the maintenance of blood glucose levels within a desirable range to prevent both hypoglycemia and hyperglycemia (Perlmuter et al. 2008). The management of T1D is daily and involves the subcutaneous injection of insulin (Sheleme et al. 2020) with possible side effects such as hypoglycemia, pain, itching, redness, bleeding, or bruising at the injection site, lipohypertrophy (fatty lumps under the skin) (Aleali et al. 2018; Thota and Akbar 2024). The symptomatic relief of T2D, on the other hand, relies on the use of clinically approved antidiabetic drugs. Metformin is the mainstay drug for T2D patients with a second line among oral antidiabetic agents (ADAs), including sulphonylureas, glinides, thiazolidinediones (TZDs), α‐glucosidase inhibitors (AGIs), and dipeptidyl peptidase‐4 inhibitors (Chan et al. 2018). These drugs improve insulin resistance and glycemic control (Sharma and Patial 2022) by inter alia inhibiting α‐glucosidase and α‐amylase enzymes, which are responsible for breaking down carbohydrates into simple sugars while also inhibiting oxidative stress. However, they present various side effects, including gastrointestinal issues such as diarrhea, constipation, bloating, gas (flatulence), stomach upset, and nausea (Chaudhury et al. 2017).

Diabetes presents a significant risk factor for all kinds of infections that appear to be related to deficits in the immune system, particularly changes in innate immunity, which increases the morbimortality of individuals with concomitant conditions (Casqueiro et al. 2012). In most of these concomitant cases, the infection is the main cause of death and morbidity of the diabetics (Holt et al. 2024). The well‐established coexistence of diabetes with infection was spotlighted during the Covid‐19 pandemic, where a twofold higher mortality rate was observed among people with diabetes (Holt et al. 2024). With regard to neglected tropical diseases (NTDs), data on the relationship between diabetes and infection is scarce; however, indications of increased infection morbidity risk and protection against diabetes have been reported. Furthermore, cases of NTD treatments interfering with glycemic control and promoting diabetes development have also been observed (van Crevel et al. 2017).

Leishmaniasis is a zoonotic NTDs caused by more than 20 species of the genus Leishmania in mammals, including humans. The infection is transmitted by infected female phlebotomine sandflies of over 90 species, and the disease is endemic in 98 countries worldwide (WHO 2024a). Leishmaniasis manifests in three clinical forms in humans, namely cutaneous (CL), mucocutaneous (MCL), and visceral leishmaniasis (VL), with a cumulative 900,000 to 1.2 million new cases and approximately 65,000 deaths occurring annually (Cortes et al. 2020; WHO 2024a). Clinical drugs currently available to treat leishmaniasis include pentavalent antimonials, liposomal amphotericin, miltefosine, and paromomycin. These drugs are marred with several drawbacks, such as poor safety profiles manifesting in harsh side effects, high costs, the need for cold chain storage (Ferreira et al. 2022), the requirement of nosocomial conditions for administration, frequent relapses, and often occurrence of treatment failure due to drugs' reduced efficacy and pathogenic resistance (DNDi 2023; Santi and Murta 2022). An additional hurdle is the long treatment course, which leads to patients' non‐compliance.

Cases of diabetes‐leishmaniasis co‐infection in humans have been reported (Izri et al. 2021; Schwetz et al. 2018). DM is an aggravating factor for the parasitic infection by impairing the cellular immune response, mainly of macrophages against the infection through the decrease of chemotaxis (Roy et al. 2022), phagocytosis (Zhou et al. 2024), respiratory burst (Plunkett et al. 2024), and the antimicrobial ability of neutrophils (Dowey et al. 2021). Neutrophils and monocytes are the first line of defense against the protozoan Leishmania spp. (Díaz‐Varela et al. 2024; Fowler et al. 2024). Accordingly, cases of altered lesion presentations, increased lesion formation, infection relapse/recrudescence, and reduced drug responses have been reported to occur in diabetes patients (Cortes et al. 2020; Mostafavi et al. 2021; van Crevel et al. 2017). Co‐infection can further lead to diagnostic errors and delays, and influence the effectiveness and safety of treatment. As such, it must be dealt with seriously.

Other protozoan species of the Kinetoplastida class that Leishmania is part of (Kaufer et al. 2017) have presented similar comorbidity implications with diabetes. Chagas disease (American trypanosomiasis) is caused by infection with Trypanosoma cruzi via contact with the urine or feces of infected triatomine bugs. Chronic infection can result in significant cardiac, digestive, and neurological disorders in a third of infected patients (van Crevel et al. 2017; WHO 2024b). A single study observed that patients with Chagas‐related cardiomyopathy are more commonly presented with diabetes and hyperglycemia than healthy individuals and Chagas patients without cardiomyopathy (van Crevel et al. 2017). T. cruzi infection has been shown to alter sympathetic and parasympathetic activities, and it was accordingly deduced that this can contribute to T. cruzi‐induced intrapancreatic denervation changes. Hyperglycemia has, in turn, been associated with significantly lower control of experimental T. cruzi parasitemia and increased mortality in vivo. Human African trypanosomiasis (HAT or sleeping sickness) is caused by infection with Trypanosoma brucei gambiense and T.b. rhodesiense via the bite of infected tsetse flies (WHO 2023b). HAT is endemic in sub‐Saharan Africa and presents common infection symptoms during the initial (hemolymphatic) phase, which advances to significant neurological symptoms and behavioral changes as the infection spreads to the central nervous system (WHO 2024b). There are currently no clinical reports available on HAT and diabetes comorbidity; however, experimental murine infection with T.b. rhodesiense has been shown to protect against inducible diabetes (van Crevel et al. 2017).

There are currently no official WHO statistics on diabetes‐leishmaniasis/trypanosomiasis co‐infections. However, the significant overlaps in the geographical distribution of diabetes and these protozoan diseases increase the risk for comorbidity, affecting treatment efficacy, co‐treatment compatibility, and disease progression. The existing shortcomings of diabetes and leishmaniasis or trypanosomiasis treatments also exacerbate these concerns. Therefore, there is a pressing need for new therapeutics for both individual diseases and co‐infections.

In search of such agents, we investigated the antileishmanial and antitrypanosomal potential of a series of previously synthesized arylidene derivatives of glitazone and related heterocycles, rhodanine, and hydantoin, some of which possessed antidiabetic properties inhibiting α‐glucosidase (Tshiluka et al. 2023). Antidiabetic agents glitazone (a class of TZDs) and rhodanine (a chemical family derived from thiazolidine and structurally similar to glitazone) have been shown to inhibit carbonic anhydrases (Mueller et al. 2021; Naeem et al. 2024). This family of zinc metalloenzymes catalyzes important processes such as respiration, carbon dioxide homeostasis, pH, and tumor survival. Carbonic anhydrases are therefore deemed druggable targets in NTD treatment, and their inhibition has been shown to have in vitro and in vivo antileishmanial and antitrypanosomal effects (Renzi et al. 2025). Furthermore, rhodanine and hydantoin, which inhibit sodium glucose co‐transporter 2 (SGLT2) and stimulate insulin secretion in vitro (Cho et al. 2019; Sergent et al. 2008), are actively used by our research group (Badenhorst et al. 2025; Sechoaro et al. 2024; Zuma et al. 2023) due to reports of their antiprotozoal activities (Cho et al. 2019; Havrylyuk et al. 2014; Schadich et al. 2020). Hence, we herewith report the discovery of potential hits and early lead compounds from phenotypic screening.

2. Materials AND Methods

2.1. Materials

Previously synthesized arylidene derivatives of hydantoin 1–3 (Tshiluka et al. 2022; Tshiluka et al. 2025), glitazone 4–10 (Tshiluka et al. 2023; Tshiluka et al. 2021) and rhodanine 11 and 12 (Tshiluka et al. 2023) as shown in Figure 1, were subjected to phenotypic assays to determine their cytotoxicity and antitrypanosomatid activity profiles followed by the identification of potential hits/leads for future in vivo investigation. Derivatives 1–3, 5, and 9–12 were previously evaluated for cytotoxicity and α‐glucosidase inhibition. Compounds 5, 9, and 10 were found to be nontoxic on the human colorectal adenocarcinoma (CaCO₂) cells, while 8 and 12 reduced cell viability compared to the other tested compounds. At the highest experimental concentration of 200 µM, most derivatives displayed poor inhibition of α‐glucosidase, apart from 9 that showed moderate activity with 51% inhibition (Tshiluka et al. 2023).

Figure 1.

Structures of investigated arylidene derivatives.

Of note, compound 8 is new, and its synthesis and characterization data are reported in the supporting material file.

The compound was characterized by NMR spectroscopy, where the ¹H NMR spectrum was categorized by the absence of aldehydic proton peaks at ~9 ppm, confirming the consumption of the aldehyde. In addition, the spectrum was characterized by the appearance of an aryl proton peak as a singlet at 7.90 ppm. The ¹³C NMR spectrum of the compounds was characterized by the lack of an aldehydic carbon peak at ~190 ppm with the appearance of a new aryl carbon signal at 134 ppm, confirming that the condensation reaction was successful.

2.2. In Silico Prediction of Pharmacokinetic Properties

SwissADME (http://www.swissadme.ch), a free web tool, was used to compute the physicochemical descriptors and predict the absorption, distribution, metabolism, and excretion (ADME) parameters, pharmacokinetic properties, drug‐like nature, and medicinal chemistry friendliness of the hydantoin arylidene analogues 1–3, glitazone arylidene analogues 4–10, and rhodanine arylidene analogues 11–12 from their molecular structures using the most relevant computational methods (Daina et al. 2017).

2.3. Cytotoxicity Assays

2.3.1. General Cytotoxicity

The resazurin assay was used to assess the cytotoxicity of the compounds. The assay includes the irreversible enzymatic reduction of oxidized blue resazurin dye to pink, highly fluorescent resorufin by viable cells (Czekanska 2011). This nontoxic reagent is useful for evaluating cell proliferation and drug toxicity. Emetine was used as the negative control.

The assay was performed using Vero cells as reported in Zuma et al. (Zuma et al. 2022). Briefly, 96‐well plates containing cell suspension (60,000 cells/mL, 100 μL/well) were incubated for 24 h and subsequently treated with a two‐fold dilution range of 100 μM compound solution. After incubating the treated plates for 72 h, resazurin was added to all wells and incubated for 2 h. The absorbances, data analysis, and CC₅₀ determinations were carried out as described in Zuma et al. (Zuma et al. 2022).

2.3.2. Animal Cytotoxicity

The compounds' toxicity was also assessed in Madin‐Darby bovine kidney (MDBK) cell line to assess their potential safety profile in livestock in relation to the animal trypanosomiasis‐causing Trypanosoma species screened.

The CCK‐8 assay was used as reported in Kannigadu et al. (Kannigadu et al. 2025). Briefly, 96‐well plates were prepared with two‐fold serial dilutions of the compound and 10 000 cells/mL MDBK cell suspension. The plates were incubated at 37°C and 5% CO₂ in a humidified atmosphere for 72 h. Cell Counting Kit‐8 (CCK8) reagent (10 µL/well; Dojindo, Kumamoto, Japan) was subsequently added and the plates incubated for a further 4 h. The absorbances, data analysis, and CC₅₀ determinations were carried out as described in Kannigadu et al. (Kannigadu et al. 2025).

2.3.3. Host Cell Cytotoxicity

The compounds' toxicity was also assessed in the host macrophage cell line used in the Leishmania intramacrophage assay, as detailed in Zuma et al. (Zuma et al. 2022). Briefly, infected phorbol 12‐myristate 13‐acetate (PMA)‐differentiated THP1 cells (500,000 cells/mL) were treated with seven two‐fold dilution concentrations of 100 μM compound and incubated for 72 h before adding resazurin and further incubation for 2 h. The absorbances, data analysis, and CC₅₀ determinations were carried out as described in Zuma et al. (Zuma et al. 2022).

2.4. Antileishmanial Assays

2.4.1. Antipromastigote Assay

The antipromastigote activity of the derivatives was evaluated according to Mangwegape et al. (Mangwegape et al. 2021). The two Leishmania strains used were Leishmania donovani (strain 9515 (MHOM/IN/95/9515)) and L. major (strain NIH S (MHOM/SN/74/Seidman)) promastigotes. The promastigotes were cultured in Media 199 (M199) with Hank's salts, 0.68 mM l‐glutamine, 4.2 mM sodium bicarbonate, 0.0001% biotin, 0.0005% hemin, 25 mM Hepes, 0.1 mM adenine (Sigma Aldrich), 10% fetal bovine serum (FBS) and 50 U/mL Penicillin/Streptomycin (Pen/Strep) solution (Sigma Aldrich). The pH was adjusted to be between 7.3 and 7.4, and the promastigotes were maintained at 26°C. Briefly, 96‐well plates were prepared with 1.25 × 106 cells/mL (50 μL/well) logarithmic phase promastigotes and 50 μL of (i) 10 μM of compound for the screening of activity, or (ii) seven two‐fold dilution concentrations of compounds for the determination of IC₅₀ values. Amphotericin B (AmB,10 μM) served as the clinically available antileishmanial (standard) drug, and the blank consisted of growth medium without parasites. After incubation at 26°C in a humidified atmosphere for 48 h, resazurin solution (50 μL of 0.01% in phosphate‐buffered saline (PBS)) was added, and the plates were further incubated in the dark for 24 h. The absorbances, data analysis, and IC₅₀ determinations were carried out as described in Mangwegape et al. (Mangwegape et al. 2021).

2.4.2. Intramacrophage Antileishmanial Assay

The resazurin assay and two Leishmania strains were also used to assess the efficacy of the derivatives against intramacrophage amastigotes. A literature‐modified technique (Jain et al. 2012; Njanpa et al. 2021) was applied as detailed in Zuma et al. (Zuma et al. 2022).

Briefly, 96‐well plates of adherent THP1 macrophage cells were infected with Leishmania donovani (strain 9515 (MHOM/IN/95/9515)) or L. major (strain NIH S (MHOM/SN/74/Seidman)) stationary phase promastigotes with a MOI of 30:1 or 10:1, respectively, and then treated with either seven two‐fold dilution concentrations of 100 µM compounds for IC₅₀ determination for 72 h. Upon the completion of the treatment, the host cells were lysed with sodium dodecyl sulphate for 30 s, and promastigote growth medium containing 10% FBS was added to terminate the lysis. To determine the antileishmanial activity, resazurin was added, and the plates were incubated in the dark for 24 and 72 h for the IC₅₀ determination to assess anti‐amastigote activity and potential recovery to proliferative promastigotes, respectively. The absorbances, data analysis, and IC₅₀ determinations were carried out as described in Zuma et al. (Zuma et al. 2022).

2.5. Antitrypanosomal Assays

2.5.1. Antitrypanosomal Assay

The antitrypanosomal activity of the derivatives was evaluated according to Kannigadu et al (Kannigadu et al. 2025). The five Trypanosoma strains used were Trypanosoma brucei brucei GUTat3.1, T.b. gambiense IL1922, T.b. rhodesiense IL1501, T. equiperdum IVM‐t1, and T. congolense IL3000 blood‐stages. Briefly, all compounds were initially screened at 0.25 µg/mL and 25 µg/mL for potential activity. Compounds with >50% growth inhibition at 25 µg/mL qualified for further IC₅₀ determination. Assay plates were prepared with 0.25 and 25 µg/mL compound concentrations for initial testing, or two‐fold serial dilutions for IC₅₀ determination, and parasite suspension (2500 cells/mL for T. brucei spp., 10,000 cells/mL for T. equiperdum, and 100,000 cells/mL T. congolense). The plates were then incubated for 72 h. Cell‐TiterGlo reagent (Promega, USA) was subsequently added to the wells (25 µL/well) and the plate placed on an orbital shaker for 2 min. The plates were then incubated at room temperature for 10 min to promote cell lysis. Bioluminescence, data analysis, and IC₅₀ determinations were carried out as described in Kannigadu et al. (Kannigadu et al. 2025).

2.5.2. Intracellular Anti‐T. Cruzi Assay

The activity of the compounds against the intracellular amastigotes of T. cruzi strain Tulahuen Clone C4 ( + lacZ). A literature‐modified method of Martinez‐Peinado et al. (Martinez‐Peinado et al. 2021) was applied as detailed in N'Da et al. (Seetsi et al. 2024). As with the intramacrophage antileishmanial assay, all compounds were tested for growth inhibition, and only those with growth inhibition of more than 60% were considered for further IC₅₀ determination.

Briefly, 96‐well plates were prepared with 100 µL of a 1:1 MOI mixture of Vero cells and trypomastigotes (400,000 cells/mL; ≥ 80% trypomastigote forms). After the plates were incubated for 24 h to promote infection of the Vero cells, the plates were treated with 200 μL of 10 μM of compound for single‐point screening. The plates were incubated for 72 h in a humidified atmosphere at 37°C and 5% CO₂. After treatment, 400 μM ONPG and 0.3% Triton X‐100 in PBS were added, and the plate was incubated for 24 h at room temperature. Absorbance, data analysis, and IC₅₀ determinations were carried out as described in N'Da et al. (Seetsi et al. 2024).

2.6. α‐Amylase Inhibition Assay

The α‐amylase inhibition assay was performed using a colorimetric method using 3,5‐dinitrosalicylic acid (DNSA) reagent, adapted from Parmar and Rupasinghe (2015) and NOOR MOHAMAD ZIN et al. (2022). The enzymatic hydrolysis of starch to reducing sugars is measured by the formation of a colored complex with DNSA (Deshavath et al. 2020), the intensity of which is directly proportional to the concentration of reducing sugars and inversely proportional to the test compound's α‐amylase inhibitory activity.

All reagents were purchased from Sigma‐Aldrich (South Africa). A 20 mM sodium phosphate buffer containing 6 mM NaCl was prepared and adjusted to pH 6.9 using NaOH. A 1 U/mL α‐amylase solution was freshly prepared using sodium phosphate buffer, kept at room temperature, and vortexed before use. A 0.5% (w/v) starch solution was freshly prepared by dispersing the starch in sodium phosphate buffer, vortexing, and heating at 85°C in a water bath for 15 min. The solution was cooled to room temperature and vortexed again before use. The DNSA reagent was prepared according to a protocol by NOOR MOHAMAD ZIN et al. (2022) and Nyambe‐Silavwe et al. (2015) by dissolving 0.8 g of NaOH in 10 mL of distilled water and heating it at 70°C to yield a 2 M solution. Thereafter, 12 g of potassium sodium tartrate tetrahydrate was dissolved in 8 mL of the hot NaOH solution and heated until fully dissolved (to give a 5.3 M solution). Separately, 0.438 g of DNSA was dissolved in 20 mL of distilled water at 70°C (to provide a 96 mM solution). Then, 12 mL of hot distilled water, the DNSA solution, and the potassium sodium tartrate tetrahydrate solution were combined, vortexed, and stored in a light‐protected container at room temperature.

The assay was conducted in a 96‐well plate format. The test compound solutions were prepared at eight concentrations by two‐fold serial dilution (to give a final highest concentration of 100 µM, lowest 0.78125 µM). Each well contained 20 µL of α‐amylase solution (1 U/mL) and 20 µL of test compound solution (or buffer for negative control), followed by pre‐incubation at 37°C with 5% CO₂ for 15 min. Next, 20 µL of the 0.5% starch solution was added to each well. The plate was mixed on a microplate shaker at 1250 rpm for 1 min and then incubated at 37°C with 5% CO₂ for 15 min. The enzymatic reaction was terminated by adding 100 µL of DNSA reagent to each well. The microplate was sealed with Parafilm, placed in a heat‐resistant plastic bag, and incubated in a water bath at 85°C for 15 min. After cooling to room temperature, 90 µL of distilled water was added to each well. The absorbance was measured at 540 nm using a Multiskan Go microplate reader (SkanIt Software 4.1 Research Edition, Thermo Scientific). Samples were appropriately diluted and re‐measured if the optical density (OD) exceeded 1.5.

The following equation was used to calculate the percentage α‐amylase inhibition:

$$\%\mathrm{\alpha }-\text{Amylase}\text{Inhibition}=((\text{Absorbance}\text{control}–\text{Absorbance}\text{sample})/\text{Absorbance}\text{control})\mathrm{x}100$$

The half maximum inhibitory concentration (IC₅₀, µM) of the test compounds was determined for each biological replicate using nonlinear regression in GraphPad Prism version 5 (GraphPad Software, San Diego, CA, USA). The results are presented as the mean ± standard deviation (SD) from at least three independent biological replicates.

3. Results AND Discussion

3.1. Predicted Pharmacokinetic Properties

For a compound to exert a biological effect, either desirable or undesirable, there are two key factors: (i) the intrinsic activity of the compound and (ii) its potential to reach the site of action in a sufficient concentration (Madden 2010). The current section is devoted to the predicted pharmacokinetic properties, including bioavailability, of the hydantoin arylidene analogues 1‐3, glitazone arylidene analogues 4‐10, and rhodanine arylidene analogues 11‐12. Tables S1–3 and Figure S1, 2 (Supporting Material) summarize the computed physicochemical descriptors and predicted ADME parameters, pharmacokinetic properties, drug‐like nature, and medicinal chemistry friendliness of the said analogues.

Lipinski's rule‐based filter states that a compound is more likely to show absorption problems; that is, poor absorption or permeability if MW > 500, logP_o/w >5, N or O > 10, and NH or OH > 5 (Lipinski et al. 1997a). Analogues 1‐12 did not violate any of Lipinski's rules and were thus considered drug‐like. However, Zhang and Wilkinson warn that although Lipinski's “rule of five” is a valuable guideline for drug discovery, it should not be overemphasized (Zhang and Wilkinson 2007). Indeed, adherence to these rules does not guarantee oral bioavailability.

The WLOGP‐versus‐TPSA referential, also known as the BOILED‐Egg model (Fig. S2), was also used to predict passive human gastrointestinal absorption (HIA) and blood‐brain barrier (BBB) permeation (Daina et al. 2017; Daina and Zoete 2016). All test compounds, except analogues 6, 7, 11, and 12, were expected to have high gastrointestinal absorption upon oral administration. Absorption alone does not determine systemic availability; once a compound enters systemic circulation, it is distributed throughout the body to reach its intended site of action (Madden 2010). According to the BOILED‐Egg model, no analogues were predicted to cross the BBB. The TPSA of a compound that is transported by the transcellular route impacts its oral absorption (TPSA ≤ 120 Ų) and brain penetration (TPSA 60 − 70 Ų) (Kelder et al. 1999). Uptake into tissues is also influenced by influx and efflux transporters; for example, the permeability protein (Pgp) is an efflux transporter that can modulate biological response by decreasing cellular entry (Madden 2010). Fortunately, only analogues 2‐3 were predicted substrates for Pgp.

Lead‐likeness describes promising lead compounds with physicochemical features that allow for molecular optimization. Since lead compounds are subject to chemical modifications that increase their size and lipophilicity, lead‐like compounds must be smaller and less hydrophobic. None of the analogues was considered lead‐like regarding chemical optimization due to their molecular weight (MW > 350), number of rotatable bonds (rotors > 7), and/or lipophilicity (XLOGP > 3.5) (Teague et al. 1999).

3.2. Cytotoxicity

All derivatives (except 1) possess low basal and organ‐specific toxicity, as can be seen from their CC₅₀ > 100 µM (Adewusi et al. 2013), which is indicative of their potential intrinsic activity. Compound 1, on the other hand, displayed mild to moderate toxicity (10 μM < CC₅₀ < 50 μM) (Finiuk et al. 2017). As such, its antileishmanial activity may not be intrinsic.

3.3. In Vitro Antileishmanial Activity

The Leishmania parasite has two developmental forms: promastigotes (in the insect vector) and amastigotes (in mammal hosts). The latter is responsible for the onset of symptoms and progression of the infection to a disease state. As such, amastigotes bear clinical relevance and hence have been the logical targeted forms in antileishmanial drug development (Katsuno et al. 2015).

Furthermore, the antileishmanial activity of compounds varies with the species involved, and the geographical location of the parasite ultimately contributes to different clinical outcomes (Rama et al. 2015; Stuart et al. 2008; Zulfiqar and Avery 2022). Thus, two species of Leishmania were selected to evaluate the activity of the compounds. These were L. donovani 9515 causing visceral leishmaniasis in India [WHO reference strain MHOM/IN/95/9515], and Leishmania major NIH S [WHO reference strain MHOM/SN/74/Seidman also referred to as NIH SD (MHOM/SN/74/SD)] that causes cutaneous leishmaniasis in Senegal, West Africa.

All the compounds were screened for Leishmania antipromastigote activity against antimony‐resistant L. donovani strain 9515 (Lira et al. 1999; Potvin et al. 2021) and L. major strain NIH, using a maximum concentration of 100 μM for promoting structure–activity relationship analysis. The compounds were then screened at 100 μM for potential anti‐amastigote activity. Amphotericin B (AmB) was used as a reference drug. Compounds with anti‐amastigote growth inhibition > 90% potentially translate to IC₅₀ < 10 μM, deemed therapeutically relevant (Katsuno et al. 2015), and were selected for further IC₅₀ determination. A modified method of Jain et al. (2012)'s parasite rescue and transformation assay was utilized (Jain et al. 2012). The anti‐amastigote activity was measured 24 and 72 h after lysis of the host cells to measure anti‐amastigote activity and potential recovery to proliferative promastigotes, respectively. Anti‐amastigote activity 24 h post‐lysis may be either static or cidal in nature. In contrast, the retainment of the activity up to 72 h post‐lysis hints towards cidal activity as the parasites could not recover to dissipate the activity. The resulting IC₅₀ values are summarized in Tables 1 and 2.

Table 1.

Antileishmanial activity (μM ± Standard deviation) of compounds against L. donovani strain 9515.

Compd.	General cytotoxicity	Host cell cytotoxicity	Antipromastigote activity		Static anti‐amastigote activity			Cidal anti‐amastigote activity
	Vero CC50 (μM)	THP1 CC50 (μM)	IC50 (μM)	SI1 a	IC50 (µM)	SI2 b	SpI1 c	IC50 (µM)	SI3 d	SpI2 e
1	47.28 ± 5.09	40.91 ± 8.96	>10	—	4.87 ± 0.01	10	2.1	7.68 ± 0.65	6	1.3
2	>100	—	>100	—	>10	—	—	>10	—	—
3	>100	—	>100	—	>10	—	—	>10	—	—
4	>100	—	>100	—	>10	—	—	>10	—	—
5	>100	>100	9.98 ± 0.03	10	0.54 ± 0.07	185	19	0.57 ± 0.00	175	18
6	>100	—	>100	—	>10	—	—	>10	—	—
7	>100	>100	>100	—	0.88 ± 0.00	114	114	0.31 ± 0.00	323	323
8	>100	—	85.86 ± 0.39	1	>10	—	—	>10	—	—
9	>100	—	91.37 ± 4.60	1	>10	—	—	>10	—	—
10	>100	—	>100	—	>10	—	—	>10	—	—
11	>100	—	59.63 ± 2.35	2	>10	—	—	>10	—	—
12	98.58 ± 2.01	>100	>100	—	0.62 ± 0.18	161	161	0.31 ± 0.01	318	323
Em	0.08 ± 0.009	—	—	—	—	—	—	—	—	—
AMB	57.80 ± 3.20	14.86 ± 0.09	0.02 ± 0.009	2890	0.05 ± 0.00	1156	0.40	0.45 ± 0.05	128	0.04

Vero: African green monkey kidney epithelial cells; THP‐1, human acute monocytic leukemia; Red, compounds qualifying as early antileishmanial leads (Katsuno et al. 2015); Blue, compounds qualifying as antileishmanial hits (Katsuno et al. 2015). All data reported in the tables were significant at p < 0.05.

^a Selectivity Index of L. donovani: SI₁ = CC₅₀ Vero/IC₅₀ promastigote.

^b Selectivity Index of L. donovani: SI₂ = CC₅₀ Vero/static IC₅₀ amastigote

^c Specificity index of L. donovani SpI₁ = IC₅₀ promastigote/static IC₅₀ amastigote.

^d Selectivity Index of L. donovani: SI₃ = CC₅₀ Vero/cidal IC₅₀ amastigote. Specificity index (SpI) < 0.4 indicates more antipromastigote activity, 0.4 < SpI < 2.0 indicates activity against both forms, SpI > 2.0 indicates more anti‐amastigote activity.

^e Specificity index of L. donovani: SpI₂ = IC₅₀ promastigote/cidal IC₅₀ amastigote.

Table 2.

Antileishmanial activity (μM ± Standard deviation) of compounds against L. major strain NIH S.

Compd.	Antipromastigote activity		Static anti‐amastigote activity			Cidal anti‐amastigote activity
	IC50 (μM	SI4 a	IC50 (µM)	SI5 b	SpI3 c	IC50 (µM)	SI6 d	SpI4 e
1	> 100	—	7.95 ± 2.90	6	1.3	7.45 ± 3.60	6	1.3
2	> 100	—	> 10	—	—	> 10	—	—
3	> 100	—	> 10	—	—	> 10	—	—
4	> 100	—	> 10	—	—	> 10	—	—
5	> 100	—	5.29 ± 0.13	19	1.9	5.14 ± 0.14	19	1.9
6	> 100	—	> 10	—	—	> 10	—	—
7	> 100	—	9.99 ± 0.02	10	1.0	9.97 ± 0.04	10	1.0
8	> 100	—	> 10	—	—	> 10	—	—
9	> 100	—	> 10	—	—	> 10	—	—
10	> 100	—	> 10	—	—	> 10	—	—
11	> 100	—	> 10	—	—	> 10	—	—
12	> 100	—	> 10	—	—	> 10	—	—
Em	—	—	—	—	—	—	—	—
AMB	0.11 ± 0.01	525	0.76 ± 0.11	76	0.14	0.75 ± 0.01	77	0.15

Vero: African green monkey kidney epithelial cells; THP‐1, human acute monocytic leukemia; Red, compounds qualifying as early antileishmanial leads (Katsuno et al. 2015); Blue, compounds qualifying as antileishmanial hits (Katsuno et al. 2015). All data reported in the tables were significant at p < 0.05.

^a Selectivity Index of L. major: SI₄ = CC₅₀ Vero/IC₅₀ promastigote.

^b Selectivity Index of L. major: SI₅ = CC₅₀ Vero/static IC₅₀ amastigote.

^c Selectivity Index of L. major: SpI₃ = IC₅₀ promastigote/static IC₅₀ amastigote.

^d Specificity index of L. major: SI₆ = CC₅₀ Vero/cidal IC₅₀ amastigote.

^e Specificity index of L. major: SpI₄ = IC₅₀ promastigote/cidal IC₅₀ amastigote.

With a focus on the anti‐amastigote activity due to its clinical relevance, it must also be observed that in this arylidene series, most compounds, including 2‐4, 6, and 8‐10, presented with no activity against antimony‐resistant L. donovani 9515 and L. major NIH S strains whereas 1, 5, 7 and 12 displayed potent activity against L. donovani (Figure 2). Of these, 1 possessed micromolar cidal activity while 5, 7 and 12 displayed comparable equipotent submicromolar anti‐amastigote activities which were deemed cidal after 72 h. Against L. major strain, however, only 1, 5, and 7 showed comparable micromolar cidal activity (Figure 3). Analogue 12, active against L. donovani, lost its potency against L. major.

Figure 2.

Variation of anti‐amastigote activity against L. donovani strain 9515.

Figure 3.

Variation of anti‐amastigote activity against L. major strain NIH S.

Furthermore, hit compounds are defined as active against the intracellular amastigotes with cellular potency IC₅₀ < 10 µM with tenfold higher selectivity over to the mammalian Vero cells (SI > 10), and early lead compounds are identified as those with antiparasitic cellular potency IC₅₀ < 1 µM and selectivity thereof with index, SI > 100 (Katsuno et al. 2015). Thus, glitazone arylidene analogues 5 and 7 qualified as hits against L. major; however, they were promoted to potential early leads against L. donovani, along with rhodanine arylidene 12. This indicates the selective antiparasitic activity of 5 and 7 towards the L. donovani sp. Furthermore, these compounds presented no significant antipromastigote activities in both species, indicating high selectivity towards the clinically relevant amastigote form. Hydantoin arylidene analogue 1 presented as a potential hit against L. donovani but did not qualify as a hit against L. major due to the low selectivity index (SI < 10) as a result of moderate cytotoxicity (IC₅₀ < 50 μM). Of note is that 1 presented with a minor loss in activity from 4.87 μM 24 h post‐host lysis to 7.68 μM 72 h post‐host lysis, which may indicate a tendency towards parasite recovery posttreatment. The remaining active arylidenes did not show significant recovery, and the reported activities were accordingly deemed cidal in nature.

Moreover, the sparsity and limited active compounds within this narrow series of arylidenes did not allow for the drawing of sensible structure–activity relationships.

3.4. In Vitro Antitrypanosomal Activity

The compounds were also screened for in vitro activity against HAT subspecies T.b. gambiense and T.b. rhodesiense, Chagas disease species T. cruzi, and animal trypanosomiasis subspecies T.b. brucei, T. equiperdum, and T. congolense. The antitrypanosomal drugs suramin and diminazene aceturate were used as reference drugs, and the selectivity of the compounds' activity was determined using MDBK CC₅₀ values. The results are summarized in Table 3.

Table 3.

Antitrypanosomal activities (μM ± Standard deviation) of compounds against five Trypanosoma subspecies.

Compd.	Cytotoxicity	Antitrypomastigote activity
	CC50 ± SD (µM)a	IC50 ± SD (µM)b (SI)c
	MDBK	Tbg	Tbr	Tbb	Tc	Teq
		IL1922	IL1501	GuTat3.1	IL3000	IVM‐t1
1	63.93 ± 11.91	> 64	> 64	> 64	65.95 ± 5.58	> 64
					(1)
2	145.48 ± 21.82	> 57	> 57	> 57	36.47 ± 5.90	> 57
					(4)
3	> 266	> 66	> 66	> 66	40.41 ± 6.43	> 66
					(2)
4	> 256	> 64	> 64	> 64	14.11 ± 0.92	> 64
					(18)
5	84.24 ± 6.69	> 59	> 59	> 59	> 59	> 59
6	X	> 66	> 66	> 66	> 66	> 66
7	> 256	> 64	> 64	> 64	37.29 ± 3.74	> 64
					(7)
8	97.13 ± 9.30	> 58	> 58	> 58	> 58	> 58
9	112.07 ± 18.86	> 63	> 63	> 63	36.16 ± 1.29	> 63
					(3)
10	> 230	> 58	> 58	> 58	> 58	> 58
12	> 276	> 69	> 69	> 69	> 69	> 69
Suramin	> 70	0.467 ± 0.08	0.150 ± 0.04	0.07 ± 0.02	8.46 ± 0.12	0.071 ± 0.01
		(150)	(467)	(1000)	(8)	(1401)
Diminazene aceturate	> 355	0.011 ± 0.00	0.039 ± 0.00	0.068 ± 0.00	0.09 ± 0.00	0.022 ± 0.002
		(33333)	(9091)	(5263)	(4000)	(16340)

MDBK: Madin‐Darby bovine kidney cells; Tbg IL1922: T. brucei gambiense strain IL1922; Tbr IL1501: T. brucei rhodesiense strain; Tbb GUTat3.1: T. brucei brucei strain GUTat3.1; Tc IL3000: T. congolense strain IL3000; Teq IVM‐t1: T. equiperdum strain IVM‐t1; X: not screened.

^a Half maximal cytotoxic concentration (CC₅₀, µM) represented as the mean ± standard deviation (SD), three biological replicates.

^b Half maximal inhibitory concentration (IC₅₀, µM) represented as the mean ± standard deviation (SD), three biological replicates.

^c Selectivity index (SI): CC₅₀ of MDBK/IC₅₀ of trypanosome.

The compounds presented with activity against only T. congolense, the causative species of animal trypanosomiasis, or wasting disease termed nagana. Compound 4 presented with the best activity (14.11 μM) that was deemed intrinsic towards the parasite (SI 18), whereas compounds 2, 3, 7, and 9 showed some activity (IC₅₀ 36 – 41 μM). The activity was not deemed significant if Katsuno et al. (Katsuno et al. 2015)'s guidelines are taken into consideration. As indicated in section 3.3, sufficient activities were not available for a clear structure–activity relationship determination.

The compounds were also screened for activity against the intracellular obligate amastigote form of T. cruzi Tulahuen Clone C4 ( + lacZ), a clone of the very pathogenic Chilean Tulahuen strain. A single point assay determined that the compounds presented with very low growth inhibition at 100 μM (0–9%, with the exception of compound 11 with 21%), thus the compounds were deemed inactive against this strain. The IC₅₀ values of the compounds were accordingly not determined.

3.5. Antidiabetic Activities

The α‐amylase inhibitory activity of compounds 1, 4, 5, 6, 7, 8, 10, and 12 was evaluated. At the highest concentration of 100 µM, most derivatives displayed no significant inhibition compared to the reference inhibitor, acarbose, which had an IC₅₀ value of 10.45 µM under the same conditions. Of the eight analogues subjected to α‐glucosidase inhibition testing, two (5 and 6) were inactive, while the remaining five, 1–3, 9 and 11, were found to be poor (1–3 and 11) to moderately (9) active against α‐glucosidase at a concentration range of 100–200 μM (Tshiluka et al. 2023) as reported in Table 4. Compound 7 was screened at 200 μg/mL (500 μM), resulting in a low 53.85 ± 1.37% inhibition (Tshiluka 2018).

Table 4.

Antidiabetic activities of compounds.

Compd.	α‐Glucosidase inhibition (%)		α‐Amylase inhibition IC50 (μM ± SD)b
	100 μMa	200 μMa
1	nd	39.32 ± 1.01.	> 100
2	nd	44.58 ± 0.99	nd
3	28.30 ± 17.30	57.41 ± 2.79	nd
4	nd	nd	> 100
5	0	nd	> 100
6	0	nd	> 100
7	nd	nd	> 100
8	nd	nd	> 100
9	51.32 ± 3.62	nd	nd
10	nd	nd	> 100
11	20.80 ± 13.90	nd	nd
12	30.00 ± 23.90	27.03 ± 2.06	> 100
EGCG	97.03 ± 0.41	97.03 ± 0.41	nd
Acarbose	nd	nd	10.45 nd ± 2.23

^a Tshiluka et al. (2023) and Tshiluka et al. (2025).

^b Half maximal inhibitory concentration; (IC₅₀, µM) represented as the mean ± standard deviation (SD), three biological replicates; nd, not determined; EGCG, epigallocatechin gallate.

This observed low activity towards α‐amylase and α‐glucosidase enzymes discourages the modulation of glucose metabolism as a potential antidiabetic working mechanism. However, TZD is known for promoting insulin sensitization via peroxisome proliferator‐activated receptor (PPAR) agonism (Kaminskyy et al. 2017; Mueller et al. 2021). TZD is a substructure of the antidiabetic glitazone drug class, and studies have indeed reported glitazone and rhodanine derivatives with PPAR binding and activation (Kaminskyy et al. 2017). Accordingly, glitazone analogues 4 to 10 and rhodanine derivatives 11 and 12 may still have antidiabetic potential as insulin sensitizers. Similarly, hydantoin derivatives have been shown to induce antidiabetic activity via alternative routes to glucose metabolism. Inhibition of SGLT2 (Cho et al. 2019) reduces blood glucose levels by reducing renal glucose reabsorption (Hsia et al. 2017). Additionally, hydantoin has been shown to have insulinotropic activity, with derivatives stimulating insulin secretion in rat pancreatic Langerhans islets (Sergent et al. 2008.

3.6. Identification of Hits for Diabetes‐Antiprotozoan Co‐Infection

Of the compounds presenting poor antidiabetic activity, analogue 1 was an antileishmanial hit against the L. donovani strain that could inhibit parasite proliferation, but did not qualify as hit against the L. major strain due to low selectivity (Table 2). Analogues 10 and 12 were antileishmanial inactive and cidal, respectively. On the other hand, 3 and 9 displayed moderate intrinsic antidiabetic activity as they were nontoxic on Caco‐2 cells (Tshiluka et al. 2023; Tshiluka et al. 2025), but these analogues were found to be inactive against both Leishmania spp. Furthermore, the compounds also did not present significant activity against HAT and Chagas disease strains. Hence, no hit candidate for potential diabetes‐antiprotozoan co‐infection treatment was uncovered during this study.

4. Conclusion

A series of arylidene analogues of hydantoin, glitazone, and rhodanine was investigated in vitro for antiprotozoan efficacy using two strains of Leishmania and six strains of Trypanosoma. The series did not present any significant activity against Trypanosoma spp., whereas micromolar active glitazone analogues 5 and 7 were identified as hits against L. major while also being uncovered as submicromolar active leishmanicidal in nature alongside rhodanine derivative 12 against L. donovani sp. These analogues qualified as early antileishmanial leads based on their cellular potency and selective antiparasitic action. However, since lead compounds are subject to chemical modifications that increase their size and lipophilicity, lead‐like compounds must be smaller and less hydrophobic. Thus, neither was considered lead‐like regarding chemical optimization due to their off‐range parameters, molecular weight, number of rotatable bonds, and/or lipophilicity. Nonetheless, these compounds that fulfilled the lead criteria with respect to both activity and selectivity could be good candidates for further antileishmanial in vivo profiling.

Furthermore, all analogues with inhibitory activity against glucose metabolism showed no antiprotozoan activity and vice versa; hence, no dual active compound with potential for diabetes‐antiprotozoan co‐infection treatment was unraveled. However, further investigation into the antidiabetic effects of the glitazone, rhodanine, and hydantoin analogues may be required to confirm this, as insulin sensitivity and secretion, as well as glucose excretion, are reported mechanisms of action for these three moieties.

Author Contributions

Conceptualization: [David D. N'Da and]; Methodology: [Janine Aucamp and Helena D. Janse van Rensburg]; Formal analysis and investigation: [Janine Aucamp, Helena D. Janse van Rensburg]; Writing‐original draft preparation: [David D. N'Da, Helena D. Janse van Rensburg and Simon S Mnyakeni‐Moleele]; Writing‐review and editing: [David D. N'Da, Helena D. Janse van Rensburg and Simon S Mnyakeni‐Moleele]; Funding acquisition: [David D. N'Da and Keisuke Suganuma]; Resources: [David D. N'Da and Keisuke Suganuma]; Supervision: [David D. N'Da].

Disclosure

Any opinions, findings and conclusions, or recommendations expressed in this material are those of the authors, and therefore, the NRF does not accept any liability in regard thereto.

Ethics Statement

The current study does not involve human samples. All in vitro procedures conducted were approved by the Human Research Ethics Committee of the North‐West University (NWU‐00223‐21‐A1), South Africa, and the Obihiro University of Agriculture and Veterinary Medicine, Japan.

Conflicts of Interest

There authors declare no conflicts of interest.

Supporting information

Supporting Information 1. Chemistry data of novel compound 8.

Data Availability Statement

The data supporting this study's findings are provided in the supporting information.