Antioxidant Activity, Phytochemical Profiling by UPLC-QTOF-MS, and In Silico Targeting of the RNA-Binding Protein DRBD18 of Trypanosoma brucei by Bioactive Compounds from Tetradenia riparia and Tetracera poggei

1. Introduction

African Trypanosomiasis is a neglected tropical disease caused by protozoan flagellates of the genus Trypanosoma, transmitted by tsetse flies (Glossina spp.) across sub-Saharan Africa [1]. Human African trypanosomiasis (HAT) is caused by Trypanosoma brucei gambiense in West and Central Africa and by T. b. rhodesiense in East and Southern Africa, while Animal African trypanosomiasis (AAT) is caused by T. b. brucei, T. congolense, and T. vivax [1]. AAT alone threatens more than 55 million cattle and imposes critical barriers to food security and economic development [2]. The Democratic Republic of the Congo (DRC) remains one of the most affected endemic countries [3].

A well-established but underappreciated aspect of trypanosomiasis pathogenesis is the induction of oxidative stress in the mammalian host. T. brucei infection triggers a strong inflammatory response associated with excessive production of reactive oxygen species (ROS), including superoxide anions, hydrogen peroxide, and hydroxyl radicals, by activated macrophages and neutrophils at the site of infection [4]. This oxidative reaction overwhelms host antioxidant defenses, leading to lipid peroxidation, DNA damage, and tissue injury, contributing to the characteristic neurological deterioration observed in late-stage HAT [5]. Conversely, the parasite itself relies on a thiol-based antioxidant system, centered on trypanothione to neutralize ROS generated both by its own metabolism and by the host immune response. Disruption of this system is lethal to the parasite, reinforcing the relevance of redox biology as a therapeutic target [6]. Plant-derived antioxidants therefore hold dual promise; they may attenuate host oxidative damage while simultaneously interfering with parasite antioxidant defense or other essential biological functions.

Currently available antitrypanosomal drugs including pentamidine, suramin, melarsoprol, eflornithine, and fexinidazole are associated with significant limitations such as complex parenteral administration, severe toxic side effects, emerging drug resistance, and high costs inaccessible to the most vulnerable populations [7,8]. These constraints highlight the urgent need for novel, safe, and affordable therapeutic agents derived from renewable natural sources.

At the molecular level, trypanosomes regulate gene expression almost exclusively by post-transcriptional mechanisms. RNA-binding proteins (RBPs) are central regulators of mRNA stability, trans-splicing, polyadenylation, and translation [9,10]. DRBD18 (Double RRM-containing protein 18; UniProt ID: Q57XR9) is an essential T. brucei RBP that regulates mRNA trans-splicing and polyadenylation of transcripts containing polypyrimidine tracts in their 3’-UTR [11]. DRBD18 has also been identified as a transcript-specific RNA editing auxiliary factor [12]. Its depletion by RNA interference severely impairs parasite growth, making it a compelling drug target. Polyphenolic compounds, known for their strong antioxidant properties, have been proposed as candidates capable of disrupting RRM domain–RNA interactions through hydrogen bonding and aromatic stacking [13,14]

In this study, two Central African medicinal plants were selected, Tetradenia riparia (Hochst.) Codd (Lamiaceae), widely used for its insecticidal, antiparasitic, antimalarial, and anti-inflammatory properties [15,16] and Tetracera poggei Gilg (Dilleniaceae), a lesser-studied specie whose leaves are traditionally used for infectious diseases in Central Africa, with no prior UPLC-MS or computational study reported [17]. Previous work from our group identified rosmarinic acid as a major compound of T. riparia [18,19].

The present study aims to (i) characterize the phytochemical composition of T. riparia and T. poggei by TLC screening, quantitative spectrophotometry, and UPLC-QTOF-MS; (ii) evaluate the antioxidant activity of decocted and percolated extracts by DPPH and ABTS assays; and (iii) perform in silico molecular docking of the major identified compounds against DRBD18 RRM1 and RRM2 domains, to identify candidate inhibitors with dual antioxidant and antiparasitic potential.

2. Materials and Methods

2.1. Plant Collection and Preparation

Leaves of Tetradenia riparia were collected between March and April 2025 in the Ngansele district, Mont-Ngafula commune, Kinshasa, DRC (S 4°26’17.197”; E 15°17’32.812”; altitude 0 m). Leaves of Tetracera poggei (local names: Bojo, Monji) were collected in the Kimwenza area, Mont-Ngafula commune (S 4°27’4.251”; E 15°17’37.693”). Plant collection adhered to all ethical guidelines; permission was obtained from the Faculty of Sciences and Technologies, University of Kinshasa. Voucher specimens were deposited at the Herbarium of Kinshasa (IUK), INERA-RDC. Collected plant material was oven-dried (Raypa) at 40 °C for 4 days and ground using an electric grinder (Blinder Butterfly B-592). Geographic collection sites are shown in Figure 1.

2.2. Preparation of Extracts

Extracts were prepared by decoction and percolation following the protocol of Ngolo et al. (2025) [19] with minor modifications. For decoction, 10 g of powder were mixed with 100 mL of distilled water, heated at 100 °C for 10 min, cooled, and filtered through Whatman filter paper. For percolation, 10 g of powder were macerated cold in 50 mL of dichloromethane/methanol (1:1, v/v) for 48 h and the percolate collected dropwise. Both extracts were evaporated and stored as dry extracts at 4 °C. Extraction yield (R) was calculated as:

R (%) = (Mex / Mmv) × 100, where Mex is the dry residue mass and Mmv is the initial dry plant material mass.

2.3. Phytochemical Screening

Qualitative phytochemical screening was performed on the aqueous and organic extracts following the methods of Hamilton-Amachree et al. (2024) [20], Shaikh (2020) [21], and Tourabi et al. (2025) [22], detecting phenols, flavonoids, alkaloids, saponins, quinones, terpenoids, anthraquinones, tannins, anthocyanins, leucoanthocyanins, and steroids. Total polyphenol content was determined by the Folin–Ciocalteu method with gallic acid as standard [23]. Total flavonoid content was quantified by the AlCl3 colorimetric method [23]. All measurements were performed in triplicate, and results are expressed as mean ± SD.

2.4. UPLC-QTOF-MS Analysis

Compound identification was carried out following the protocol of Ngolo et al. (2025) [19]. Dry extract (1 mg) was re-dissolved in 1 mL MeOH/water (1:1) and filtered through a 0.22 µm nylon syringe filter. Analysis was performed on a Waters Acquity UPLC system (Waters BEH C18 column, 2.1 × 100 mm, 1.7 µm) coupled to a Waters Synapt G2 QTOF mass spectrometer in negative ESI mode, as described previously [19]. Chromatographic separation used a Waters BEH C18 column (2.1 mm × 100 mm, 1.7 µm) with a gradient of solvent A (water + 0.1% formic acid) and solvent B (methanol + 0.1% formic acid), applied as follow: 0 min, 3% B; 0.10 min, 3% B; 14 min, 100% B; 16 min, 100% B; 16.5 min, 3% B; 20 min, 3% B. The flow rate was set at 0.3 mL/min. The injection volume was 5 µL. The following ESI source parameters were used for negative mode: source temperature 120 °C; sampling cone 20 V; extraction cone 4.0 V; desolvation temperature 300 °C; cone gas 10.0 L/h; desolvation gas 600 L/h; capillary 2.6 kV; trap collision energy 28 V. Molecular formulae were generated with MassLynx V4, and tentative identifications were made by comparing fragmentation patterns with the KEGG and PubChem databases via MetFrag (https://msbi.ipb-halle.de/MetFrag/).

2.5. Antioxidant Activity

Radical scavenging activity was evaluated by DPPH (2,2-diphenyl-1-picrylhydrazyl) and ABTS (2,2’-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid)) assays as described by Bukatuka et al. [24] with minor modifications. For DPPH, 1980 µL of 0.1 mM DPPH in methanol was mixed with 20 µL of extract at varying concentrations, incubated in the dark for 30 min, and absorbance read at 517 nm. For ABTS, the radical cation was generated by reacting 20 mM ABTS with 10 mM K2S2O8 (1:1) for 16 h in the dark; the working solution was diluted to an absorbance of 0.8–1.0, mixed with extract, and read at 734 nm after 30 min. IC50 values were calculated from inhibition–concentration curves using GraphPad Prism 9.5.1.733. All measurements were performed in triplicate.

2.6. Molecular Docking Studies

2.6.1. Protein Preparation

T. brucei DRBD18 (UniProt: Q57XR9; gene: Tb927.11.14090) was selected as the target protein [12]. The AlphaFold-predicted model (AF-Q57XR9-F1, version 6) was retrieved from the AlphaFold Protein Structure Database [25] and visualized in PyMOL (v2.5) and NGL Viewer. The model comprises 480 residues with two RRM domains: RRM1 (residues 90–170) and RRM2 (residues 260–340). Low-confidence terminal regions (low pLDDT scores) were excluded; water molecules and non-standard atoms were removed. Binding pocket coordinates were as follows: RRM1 (x = 3, y = 12, z = −25) and RRM2 (x = −34, y = 3, z = 24).

2.6.2. Ligand Selection and Docking

Ligands were selected from UPLC-MS profiles based on peak abundance and structural diversity. For T. riparia, rosmarinic acid, astragalin, luteolin, acacetin, asiatic acid, diterpenoid SP-II, abieta-8,11,13-triene-7,15,18-triol, and trihydroxy octadecenoic acid were used. For T. poggei, epi-gallocatechin 3-O-gallate, catechin 3-O-gallate, (−)-epicatechin-3-(3-O-methyl)-gallate, isorhamnetin, 7-O-methylquercetin, 7-O-methylkaempferol, galloyl-hexose-sulfate, and 1-O-palmitoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-glycerol were considered. Three-dimensional structures were retrieved from PubChem and optimized with Open Babel. Molecular docking was performed with AutoDock Vina in PyRx 0.9.8 [26]. The lowest binding energy pose was retained. Ligand efficiency (LE), Fit Quality (FQ), Binding Efficiency Index (BEI), and estimated Ki values were calculated. Interactions were visualized using BIOVIA Discovery Studio Visualizer.

3. Results

3.1. Qualitative Phytochemical Screening

Table 1 summarizes the qualitative phytochemical composition of the aqueous and organic extracts of T. riparia and T. poggei. Polyphenols, flavonoids, anthocyanins, tannins, bound quinones, alkaloids, leucoanthocyanins, and saponins were detected in the aqueous extracts of both species. In the organic fractions, steroids were present in both species, while triterpenoids were identified exclusively in T. riparia and free quinones exclusively in T. poggei. Tannins were absent from the aqueous extract of T. poggei.

3.2. Total Phenolic and Flavonoid Content

T. poggei exhibited significantly higher polyphenol and flavonoid contents (polyphenols: 510.227 ± 74.608 mgGAE/gDW; flavonoids: 855.394 ± 5.292 mgQE/gDW) than T. riparia (polyphenols: 331.56 ± 89.22 mgGAE/gDW; flavonoids: 445.727 ± 1.8 mgQE/gDW) (Table 2).

3.3. UPLC-QTOF-MS Metabolite Profiling

UPLC-QTOF-MS in negative ESI mode tentatively identified eleven compounds in T. riparia (Figure 2, Table 3) and thirteen compounds in T. poggei (Figure 3, Table 4).

In Tetradenia riparia methanolic extracts (Figure 2), rosmarinic acid (m/z 359.0772, RT 6.85 min, error 1.4 ppm) was the dominant polar constituent, consistent with Ngolo et al. (2025) [19]. Caffeic acid (m/z 179.0337) and its biosynthetic precursor were also detected. Flavonoids were represented by astragalin (m/z 447.0926), luteolin (m/z 285.0400), acacetin (m/z 283.0617), and nepetoidin A or isomer (m/z 313.0707). Terpenoids included asiatic acid (m/z 487.3438), diterpenoid SP-II (m/z 335.2220), and abieta-8,11,13-triene-7,15,18-triol (m/z 317.2119). Trihydroxy octadecenoic acid (m/z 327.2168) and dibutyl succinate (m/z 229.1449) were also identified. This metabolic diversity is consistent with earlier reports for T. riparia [15,16]. The compounds tentatively identified in the extracts from Tetradenia riparia are listed in Table 3.

In Tetracera poggei (Figure 3), the profile is dominated by galloylated catechin derivatives including epi-gallocatechin 3-O-gallate (m/z 457.0765, error −1.3 ppm), catechin 3-O-gallate (m/z 441.0818), and (−)-epicatechin-3-(3-O-methyl)-gallate (m/z 455.0967), alongside simpler phenolic acids such as gallic acid (m/z 169.0134), protocatechuic acid (m/z 153.0184), and methyl 6-O-galloyl-β-D-glucopyranoside (m/z 345.0818).

Methylated flavonols included isorhamnetin or isomer (m/z 315.0512), 7-O-methylquercetin or isomer (m/z 315.0498), and 7-O-methylkaempferol (m/z 299.0554). Sulphated compounds galloyl-hexose-sulfate (m/z 415.2162) and 1-O-palmitoyl-3-O-(6-sulfo-α-D-quinovopyranosyl)-glycerol (m/z 555.2838) were detected at later retention times. This chemical profile has not previously been reported for T. poggei and represents a significant expansion of the known phytochemistry of the genus Tetracera [17]. The compounds tentatively identified in the extracts from Tetradenia poggei are listed in Table 4.

3.4. Antioxidant Activity

The antioxidant activities of decocted and percolated extracts are summarized in Table 5. Percolated extracts consistently showed higher radical scavenging activity than decocted extracts in both species. For DPPH, T. poggei percolated extract showed the strongest activity (IC50 = 4.141 ± 0.175 µg/mL), followed by T. riparia percolated extract (7.211 ± 0.222 µg/mL). In the ABTS assay, T. riparia percolated extract showed higher activity (IC50 = 10.79 ± 0.56 µg/mL) than T. poggei (13.99 ± 0.15 µg/mL). All IC50 values were below 100 µg/mL.

3.5. Molecular Docking Against DRBD18

3.5.1. T. poggei Compounds

Table 6 shows the docking results for T. poggei compounds. Among the eight compounds evaluated, (−)-epicatechin-3-(3-O-methyl)-gallate showed the highest RRM1 affinity (ΔG = −7.3 kcal/mol; Ki = 4.424 µM), followed by epi-gallocatechin 3-O-gallate (−7.2 kcal/mol; Ki = 4.645 µM) and catechin 3-O-gallate (−7.1 kcal/mol; Ki = 6.202 µM). In the RRM2 domain, catechin 3-O-gallate showed the highest affinity (−6.7 kcal/mol; Ki = 12.187 µM). Methylated flavonols showed moderate affinities (−5.7 to −6.6 kcal/mol for RRM1). The sulphated lipid derivatives showed the weakest binding (−4.5 to −6.1 kcal/mol).

The interaction profiles of epi-gallocatechin-3-O-gallate (EGCG) at RRM1 and catechin-3-O-gallate (C3G) at RRM2 reveal distinct but well-defined binding characteristics (Figure S1). Within the RRM1 binding pocket, EGCG forms conventional hydrogen bonds with residues GLY186, TYR177, ARG180, THR182, and GLU185, along with a π-sigma interaction with GLU181, π-alkyl contacts with LEU179 and ARG178, and a T-shaped π–π interaction with PHE188 (Figure 4). This pattern highlights the combined role of hydrogen bonding and aromatic contacts in RRM1 binding.

In contrast, the C3G interaction network at RRM2 is more complex and diverse; the compound forms a carbon hydrogen bond with THR232, conventional hydrogen bonds with ALA230, GLN243, and GLU231, a π-cation interaction with ARG247, and π–π stacking with TRP233, supplemented by π-alkyl interactions with ARG244 and LEU240 (Figure 5). This combination of electrostatic, aromatic, and hydrophobic interactions suggests a more balanced and versatile binding mode at RRM2.

Interaction distances generally fall in the range of approximately 1.6–4.8 Å, indicating favorable spatial positioning of the ligands within both binding pockets.

3.5.2. T. riparia Compounds

Table 7 shows the docking results for T. riparia compounds. Diterpenoid SP-II showed the highest RRM1 binding affinity (ΔG = −7.3 kcal/mol; Ki = 4.424 µM), followed by asiatic acid (−7.2 kcal/mol; Ki = 4.645 µM), astragalin (−6.8 kcal/mol), and abieta-8,11,13-triene-7,15,18-triol (−6.8 kcal/mol). Phenylpropanoids and flavonoids including rosmarinic acid (−6.7 kcal/mol), luteolin (−6.6 kcal/mol) and acacetin (−6.5 kcal/mol) also showed moderate to strong RRM1 binding. In RRM2, diterpenoid SP-II (−6.1 kcal/mol) and acacetin (−6.0 kcal/mol) formed the most stable complexes. Trihydroxy octadecenoic acid showed the lowest affinities (RRM1: −3.5 kcal/mol; RRM2: −2.9 kcal/mol; Ki > 2,700 µM).

The interaction profile of diterpenoid SP-II with RRM1 and RRM2 reveals limited but distinct contacts with both domains (Figure S2). Within the RRM1 binding site, the compound forms an alkyl interaction with LEU179 and conventional hydrogen bonds involving GLY186 and THR182, reflecting structural complementarity between the diterpenoid scaffold and the RRM1 pocket (Figure 6).

In the RRM2 binding region, a more restricted interaction pattern is observed: the compound establishes a π-alkyl interaction with TRP233, contributing to hydrophobic stabilization, and forms a conventional hydrogen bond with THR232 (Figure 7).

4. Discussion

The rich phytochemical diversity of both T. riparia and T. poggei revealed by qualitative screening is consistent with their traditional medicinal uses. The co-occurrence of flavonoids, polyphenols, tannins, and alkaloids suggests marked antioxidant, antimicrobial, and antiparasitic potential [27,28]. The exclusive detection of triterpenoids in T. riparia and free quinones in T. poggei indicates distinct metabolic specialization between the two species. Our results for T. riparia are consistent with Hamilton-Amachree et al. (2024) [20], with a minor divergence regarding anthraquinones.

T. poggei showed significantly higher polyphenol and flavonoid contents than T. riparia. These values exceed those reported by Chepng’etich et al. (2018) [29] and Ngolo et al. (2025) [19] for T. riparia. The variability may be attributed to harvest season, soil pH, drying conditions, geographic location, and extraction method [30]. The particularly high flavonoid content of T. poggei is consistent with the dense array of galloylated catechin derivatives subsequently identified by UPLC-MS. Phenolic compounds are known to modulate gene expression and exhibit antioxidant, antimutagenic, and antiparasitic activities [13].

UPLC-QTOF-MS analysis revealed complementary chemical profiles between the two species. In T. riparia, rosmarinic acid is the dominant polyphenol, already identified as a major constituent by Ngolo et al. [19]. Luteolin, acacetin, and astragalin are well-documented antioxidants. The terpenoid fraction containing diterpenoid SP-II, asiatic acid, and the abietane diterpene, represents structurally diverse scaffolds not previously described by UPLC-MS in this species. In T. poggei, the dominance of EGCG, catechin 3-O-gallate, and (−)-epicatechin-3-(3-O-methyl)-gallate is noteworthy; the galloyl ester moiety of these compounds is a well-established structural determinant of both antioxidant capacity and protein-binding affinity [31,32]. This chemical profile has not previously been reported for T. poggei [17].

Percolated extracts consistently outperformed decocted extracts in both DPPH and ABTS assays, likely reflecting the superior extraction efficiency of the dichloromethane/methanol solvent system for polyphenolic and terpenoid compounds. The higher DPPH activity of T. poggei (IC50 = 4.141 µg/mL) correlates with its greater polyphenol and flavonoid content, and with the strong hydrogen-donating capacity of its galloyl ester compounds [33]. The divergence between DPPH and ABTS rankings reflects the mechanistic differences between the two assays; DPPH primarily measures hydrogen atom transfer from hydrophilic compounds, while ABTS captures both hydrophilic and lipophilic antioxidants, explaining the relatively higher ABTS performance of T. riparia, which contains lipophilic diterpenoids [34]. The IC50 of T. riparia percolated extract by DPPH (7.211 ± 0.222 µg/mL) is superior to the value reported by Ngolo et al. (2025) [19], a difference attributable to seasonal, geographic, or protocol variations. Phenolic compounds particularly rosmarinic acid, gallic acid, caffeic acid, and the galloylated catechins are known to be the primary determinants of antioxidant capacity [35], consistent with our positive correlation between polyphenol/flavonoid content and IC50 values.

Importantly, many of the identified antioxidant compounds in both species are the same molecules that show the strongest binding affinities in the molecular docking analysis. This dual antioxidant–antiparasitic profile is mechanistically coherent; the structural features responsible for antioxidant activity including multiple hydroxyl groups, aromatic rings, and electron-donating substituents are the same features that drive hydrogen bonding and aromatic stacking interactions with RRM domain residues. EGCG and catechin 3-O-gallate are among the best-characterized RRM-binding polyphenols in the literature [31,32]. In T. poggei, (−)-epicatechin-3-(3-O-methyl)-gallate achieved the highest RRM1 binding energy (ΔG = −7.3 kcal/mol; Ki = 4.424 µM). Interaction analysis of EGCG at RRM1 reveals classical hydrogen bonds with GLY186, TYR177, ARG180, THR182, and GLU185, along with π-sigma (GLU181), π-alkyl (LEU179, ARG178), and T-shaped π–π (PHE188) interactions. The involvement of PHE188, a conserved aromatic residue in the RRM β-sheet that normally participates in RNA base stacking, strongly suggests direct competition with RNA substrates [31,32]. At RRM2, catechin 3-O-gallate forms an extensive network including a π-cation interaction with ARG247 and π–π stacking with TRP233.

In T. riparia, diterpenoid SP-II (ΔG = −7.3 kcal/mol) and asiatic acid (−7.2 kcal/mol) achieved RRM1 affinities comparable to the best polyphenols, through a different mechanistic basis: primarily hydrophobic and van der Waals contacts rather than hydrogen bonding. Diterpenoid SP-II at RRM1 forms alkyl interactions with LEU179 and hydrogen bonds with GLY186 and THR182; at RRM2, it engages π-alkyl with TRP233 and a hydrogen bond with THR232. This terpenoid binding mode is consistent with recent evidence that natural diterpenoids interact with RNA-binding proteins through hydrophobic surface complementarity [38]. The convergence of comparable RRM1 binding energies from structurally distinct chemical classes suggests that DRBD18 accommodates a broader range of ligand scaffolds than expected from a purely polar binding interface.

These in silico findings require experimental validation. Key limitations include the use of an AlphaFold structural model rather than an experimentally resolved crystal structure, the static nature of docking, which does not account for protein flexibility or solvation effects, and the absence of mammalian homologue selectivity data. Future studies should include in vitro RNA-binding inhibition assays, cellular anti-trypanosomal activity testing, and in vivo evaluation.

5. Conclusions

This study provides the first integrated antioxidant, phytochemical, and in silico characterization of Tetradenia riparia and Tetracera poggei as sources of dual-action compounds with antioxidant and antiparasitic potential. Both species displayed significant antioxidant activity (all IC50 < 100 µg/mL), with T. poggei percolated extracts achieving the highest DPPH activity (IC50 = 4.141 ± 0.175 µg/mL). UPLC-QTOF-MS identified complementary metabolite profiles including rosmarinic acid, diterpenoids, and flavonoids dominating in T. riparia, and galloylated catechin derivatives in T. poggei. Molecular docking against DRBD18 identified diterpenoid SP-II and (−)-epicatechin-3-(3-O-methyl)-gallate as the top candidates (RRM1: ΔG = −7.3 kcal/mol; Ki = 4.424 µM), with a consistent structure-activity relationship linking polyphenolic and terpenoid complexity with RRM binding affinity. The dual antioxidant–antiparasitic profile of the identified compounds provides a mechanistic basis for the traditional use of these plants and a rational foundation for the development of plant-derived trypanocidal agents targeting DRBD18. Experimental in vitro and in vivo validation remains an essential next step.