This was an experimental study aimed at determining antiplasmodial, hemocompatibility and larvicidal potential of biosynthesized silver NPs using A. cordifolia leaves. The plant was harvested and authenticated taxonomically. Crude extract of A. cordifolia leaves (AC-CE) was screened for phytochemical composition and used for AgNPs synthesis. The optimization of AC-AgNPs was made, and the optimized AC-AgNPs were characterized and tested for antiplasmodial, hemocompatibility and larvicidal potential (Figure 1). The study was approved by ethical committee of the National Institute of Malaria Research (NIMR), India (N°PHB/NIMR/EC/2020/55).

Healthy and fresh leaves of A. cordifolia (AC) were collected at Faculty of Sciences (FS), main campus, University of Douala (UD), Littoral Region, Cameroon (Figure 2). Malaria is highly prevalent in Cameroon, and P. falciparum is the main malaria species. Other species including Plasmodium vivax, Plasmodium ovale spp have also been reported across the country (Kojom Foko et al., 2021; Kojom Foko et al., 2022a). The taxonomic authentication was done by Dr Tchiengue Barthelemy at Cameroon National Herbarium, Yaounde, in comparison with voucher specimen number 9657/SRF/Cam previously deposited.

About 500 g of fresh A. cordifolia leaves were washed with running tap water and distilled water to remove dust and surface contaminant, and thereafter air-dried for 2 weeks at room temperature. The dried material was introduced in an electric grinder to obtain a fine powder. Ten grams of powder was taken in a conical flask containing 100 mL of distilled water, heated at 80°C for 10 min in a water bath under static conditions (Eya’ane Meva et al., 2016). The mixture was allowed to cool at room temperature, and then filtered using a Whatman paper n°1 to remove particulate matter. The filtrate obtained (crude extract, AC-CE) was used to perform phytochemical screening and AC-AgNPs biosynthesis. The AC-CE was not used more than a week following its preparation in order to avoid gradual loss of viability due to long storages (Eya’ane Meva et al., 2016). The AC-CE was lyophilized and stored for biological assays. The yield of extraction of AC-CE was 41% (w/v).

The AC-CE was subjected to gas chromatography-mass spectrometry (GC-MS) analysis to identify the composition and percentage abundance of phytochemical constituents. The GC-MS was carried out on a Perkin Elmer Turbo Mass Spectrophotometer (Norwalk, CTO6859, NY, United States) which includes a Perkin Elmer Auto sampler XLGC. The column used was a Perkin Elmer Elite-5 capillary column measuring 30 m × 0.25 mm with a film thickness of 0.25 mm composed of 95% dimethyl polysiloxane. The carrier gas used was helium at a flow rate of 1.21 mL/min 1 μL sample injection volume was utilized. The inlet temperature was maintained at 260°C. Oven temperature was programmed initially at 100°C for 2 min, and then programmed to increase to 290°C at a flow rate of 10°C/min (Supplementary Figure S1). The total run time was 39.98 min. The Mass Spectrometry transfer line was maintained at a temperature of 200°C. The source temperature was maintained at 220°C. The GC-MS was analyzed using electron impact ionization at 70 eV. Full scan mode was used to detect analytes. Data were evaluated using total ion count for compound identification and quantification. Measurement of peak areas and data processing were carried out by Turbo-Mass-OCPTVS-Demo SPL software, and spectrums of the components were compared with database of spectrum of known components stored in the GC-MS library.

AC-AgNPs were synthesized by blending AgNO₃ aqueous solution with freshly prepared AC-AE and incubated in dark until color change. The determination of optimal conditions for AC-AgNPs biosynthesis was performed by recording UV-Visible spectra of reaction mixtures after varying four parameters, namely, incubation temperature (35°C–85°C), incubation time (10 min–5 h), AgNO₃ concentration (0.5–5 mM), and AgNO₃/AC-CE volume ratio (10–100 µL) as described earlier (Hawadak et al., 2022). Thereafter, the optimized reaction mixture was centrifuged at 15,000 rpm for 10 min, the pellet was washed twice with distilled water and once with 95% ethanol, filtered using sterile syringe filter (MICRO-POR^®, 0.22 µm), and then lyophilized for further AC-AgNPs characterization and biological assays.

The characteristics of green synthesized AC-AgNPs (i.e., surface plasmon resonance-SPR, size, shape, aggregation, functional chemical groups and crystallinity) were determined using several techniques (Figure 1). The formation of AC-AgNPs was monitored by visual inspection of the solution and then followed by UV–Vis spectrum measurement using a double beam spectrophotometer (Model n.o., BRI-2700, BR BIOCHEM Life Sciences Pvt., Ltd., India) operating at 1 nm resolution. Milli Q ultrapure water was used as blank. The selected area electron diffraction (SAED) and powder X-ray diffraction (XRD) were used to determine the physical nature of the AC-AgNPs. The PXRD was made at 45 kv voltage, 40 mA current, 2θ range of 10–80 and speed of 2°/minute (PANanalytical, Xpert Pro model). The PXRD patterns of optimized AC-AgNPs were compared to Joint Committee on Powder Diffraction Standards files (JCPDS 65-2871 and 31-1238). The size, shape and aggregation patterns of AC-AgNPs were determined using scanning electron microscopy—SEM coupled with EDX (Bruker AXS Microanalysis GmbH Berlin, Germany) and transmission electron microscopy—TEM coupled with SAED (TECNAI TF20, Fei, Electron Optics, Oregon, United States) operating at a potential of 20 kv and 200 kv, respectively. The size of NPs was calculated using the Scherrer equation: D = K λ β   Cos θ , where D is diameter (nm) of the crystallite (i.e., NPs in this regard), K is the Scherrer constant (range values = 0.68–2.08) depending on shape of nanoparticles (e.g., K = 0.94 for spherical NPs); λ is the X-ray wavelength (in our study PXRD analysis was performed at wavelength for copper, CuK_α = 1.5406 Å), β is the line broadening at full width at half maximum (FWHM) which is expressed in radians, and θ is the Bragg’s angle of PXRD-related peaks which is expressed in degrees (Muniz et al., 2016). The atomic composition of the NPs was determined using energy dispersive X-ray (EDX). Fourier transformed infrared spectroscopy (FTIR) was used to determine functional chemical groups capped on the AC-AgNPs surface through potassium bromide method. Sample was grinded with KBr in an infrared path and the spectrum was recorded in the range 400–4000 cm^-1 using a FTIR spectrophotometer (Perkin Elmer, Frontier Model). Zeta potential and dynamic light scattering (DLS) were performed to evaluate NPs stability and size distribution using particle size analyzer (Zetasizer nano ZS, Malvern Instruments Ltd., U.K.). In practice, zeta potential of ±30 mV is considered as a good indicator of the stability of colloidal suspensions such as NPs while values outside the range indicate phenomena such as flocculation, aggregation and sedimentation (Kojom Foko et al., 2019).

Chloroquine-sensitive 3D7 and chloroquine-resistant RKL9 of Pf strains were used for antiplasmodial assays for AC-AgNPs, AC-CE, and chloroquine (CQ). The Pf culture was maintained using standard protocols (Trager and Jensen, 1976; Schuster, 2002). Briefly, parasite cultures were maintained in fresh AB positive human erythrocytes suspended at 5% hematocrit in RPMI-1640 culture medium supplemented with L-glutamine and HEPES buffer (0.2% sodium bicarbonate, 0.4% albumax, 50 μg/L hypoxanthine, 200 U/mL penicillin and 200 μg/L streptomycin) and incubated at 37°C under a gas mixture of 1% O₂, 5% CO₂ and 94% N₂. Culture of infected erythrocytes were transferred daily into fresh complete culture medium and checked microscopically for parasite growth.

The in vitro evaluation of antiplasmodial activity was performed using culture-adapted Pf strains: i) 3D7, sensitive to CQ, artemisinin and its derivatives and ii) RKL9, resistant to CQ. Antimalarial drug screening was done based on SYBR green I-based fluorescence assay as described previously (Smilkstein et al., 2004). Parasite culture (0.5%–0.8%) was synchronized at ring stage with 5% sorbitol. A volume of 100 µL of complete medium were introduced into each well of 96-well microplate, then dilutions were performed for AC-AgNPs and AC-CE (4, 8, 16, 31.25, 62.5, 125, 250 and 500 μg/mL) and CQ (4, 6.25, 12.5, 25, 50, 100, and 200 μg/mL) were added. Ten microliters (10 µL) of synchronized blood were thereafter added in each well, mixed and kept in an incubator at 37°C for 48 h in 96-well flat bottom tissue culture-grade plates under reduced O₂ atmosphere. Each experiment was replicated thrice. CQ was used as standard drug, while complete medium was considered as negative control. After 48 h-incubation, 100 µL of SYBR Green I in lysis buffer (0.2 µL of the fluorochrome/mL of buffer) was added into each well, mixed gently twice, and the plate was then covered with foil and incubated in a dark chamber for 1 h at room temperature. The buffer lysis consisted of Triton X-100 (0.08% v/v), Tris (20 mM), EDTA (5 mM), and saponin (0.008% wt/v). The fluorescence counts were read using an ELISA reader (Synergy HTX 1708152, Agilent BioTek, Santa Clara, California, United States) with excitation and emission wavelength bands centered at 485 and 530 nm.

The SYBR Green based antiplasmodial assay was validated by inspecting microscopic slides of parasite cultures treated with negative control, standard drug, AC-CE and AC-AgNPs (Kaushik et al., 2015; Hawadak et al., 2022). After 48 h-incubation, thick and thin blood films were made, air-dried and stained with 10% Giemsa stain for 20 min. The number of schizonts with ≥2 nuclei out of 200 asexual parasites was noted. Also, fluorescence counts of untreated and treated Pf cultures were compared to detect any quenching effect-related measurement artefacts which may due to chemical compounds of AC-AgNPs and AC-CE (Kaushik et al., 2015).

The method described by Wang and others was used to evaluate hemocompatibility of biosynthesized AC-AgNPs (Wang et al., 2010). Human red blood cells (RBCs) were obtained from the ICMR-NIMR malaria parasite bank, washed with incomplete media, and diluted with phosphate-buffered saline (PBS) to obtain a suspension (Hematocrit = 1%). Different concentrations (2, 4, 8, 16, 30, 62.5, 125, 250 and 500 μg/mL) of AC-AgNPs and AC-CE were incubated with RBCs in Eppendorf tubes (20 µL of each concentration in 180 µL blood) at 37°C for 30 min and 24 h at pH of 7.40. The reaction was stopped by placing tubes at 4°C for 15 min. The mixtures were then centrifuged at 3,000 g for 4 min, and 100 µL of supernatant was loaded into a 96-well plate to measure the released hemoglobin at 540 nm (SPECTROstar ^Nano , BMG LABTECH GmbH, Ortenberg Germany). Saponin was used as positive control, inducing 100% hemolysis, while PBS was considered as negative control. The experiment was performed in triplicate. RBCs hemolysis at each concentration after 30 min and 24 h was calculated as follows: %   h e m o l y s i s = A s − A N C A P C − A N C × 100 where A_S, A_NC and A_PC are the absorbance of the sample, negative control (PBS) and positive control (saponin).

The eggs of An. stephensi, Cx. quinquefasciatus and Ae. aegypti were obtained from NIMR Insectarium, New Delhi, India. The characteristics of mosquitoes used are as follows: An. stephensi—laboratory strain collected from Sonepat, Haryana, India (established in 1996; black and brown, malathion-deltamethrin-susceptible and DDT–resistant strain), Cx. quinquefasciatus—laboratory strain collected from Sonepat, Haryana, India (established in 1999; selected for permethrin resistance and is resistant to DDT, malathion and deltamethrin), and Ae. aegypti—laboratory strain collected from Delhi, India (established in 2006; DDT–malathion-deltamethrin strain). Adult Ae. aegypti were derived from batches of 100 eggs in 18 cm × 13 cm × 4 cm trays containing 500 mL of boiled and cooled water in a laboratory maintained at 25°C–29°C temperature and 65%–70% Relative humidity; 12:12 h Light/Dark photoperiod. Eggs were fed daily with TetraBits fish food (Tetra GmbH, Herrenteich, Germany), and late 3rd and 4th instar larvae were used for larval bioassays.

The protocol described by the World health Organization (WHO) was used for this experiment (WHO, 2016). Late 3^rd and 4^th instar larvae were exposed to the AC-AgNPs with different concentrations (0–50 μg/mL). Each concentration was tested in triplicate comprising of 25 larvae placed into plastic bowls (8 cm diameter, 300 mL capacity) containing distilled water. The larval mortality was monitored after 24 h, 48 h and 72 h post-treatment periods, and the lethal concentrations to cause 50%/90% mortality in treated larvae (LC₅₀/LC₉₀) and percentage mortality after post-treatment periods were calculated as described previously in the WHO procedures (WHO, 2016). Distilled water was used as control. All experiments were performed under laboratory conditions as described above.

Data was keyed into an Excel spreadsheet (Microsoft Office, United States) and then exported to statistical package for social sciences v16 (SPSS, IBM, Inc., Chicago, United States), and GraphPad v5.03 (GraphPad PRISM, Inc., San Diego, California, United States) for statistical analysis. Using GraphPad software v8.03 (GraphPad PRISM, Inc., San Diego, CA, United States), fluorescence counts of antiplasmodial assay were used to plot graph of percent inhibition of Pf parasite growth against concentrations of AC-AgNPs, AC-CE, and CQ to determine 50% inhibition concentration (IC₅₀). The dose/time mortality response data of larvicidal assays was analyzed using log-probit regression model to determine LC₅₀ and LC₉₀ with their confidence interval at 95% (95% CI). The Abbott’s formula was used to correct mortality rate if comprised between 5% and 20% in the negative control group (Sun and Shepard, 1947). Experiments were considered invalid when mortality rate in negative control group was >20%. Regarding hemocompatibility assay, the amount of NPs required to lyse 50% of RBCs (hemolysis concentration, HC₅₀) was determined. Quantitative and qualitative variables were presented as mean ± standard deviation (SD) and percentages, respectively. One-way analysis (ANOVA), McNemar’s and Pearson’s independence chi square tests were used to make comparisons. The level of statistical significance was set at p-value <0.05.

GC-MS chromatogram of AC-CE revealed several peaks which represent different compounds as shown in Supplementary Figure S1. A total of 42 compounds were identified in AC-CE after comparing the peaks with database of spectrum of known components stored in the GC-MS library (Table 1). Two compounds were predominantly represented, namely, 2-hexadecen-1-ol, 3,7,11,15-tetramethyl-, [R-[R*,R*-(E)]]- and phytyl tetradecanoate, with proportions of 25.14% and 14.53%, respectively (Supplementary Figure S2; Table 1).

The synthesis of AC-AgNPs was noted after 2 minutes following the incubation of plant extract and AgNO₃ solution as a dark brown color was observed (Figure 3A). The UV-Vis spectrum analysis revealed a SPR at 445 nm wavelength (Figures 3B–E). The SPR did not change with the variation of four parameters used to optimize AC-AgNPs synthesis (AgNO₃ concentration, incubation time, incubation temperature and volume of plant extract). In contrast, the amplitude of UV-Vis curves gradually increased with increasing values of each parameter (Figures 3B–E). Thus, the optimization of AC-AgNPs was achieved for the following parameters: 100 µL of fresh plant extract was mixed with 900 µL of AgNO₃ (5 mM), and then incubated at 85°C for 5 h under static conditions.

Analysis of SEM and TEM micrographs of AC-AgNPs is depicted in Figure 4. Based on SEM, agglomeration of AC-AgNPs was observed (Figures 4A, B). TEM images of silver colloidal solution exhibited that AC-AgNPs were polydispersed, predominantly spheroidal with various sizes (Figures 4C, D). The size distribution when a section of these NPs is considered is presented in Figure 4E. Following the digitization phase of various images, size distribution using ImageJ software was found to be within 5–25 nm range. The distribution of AC-AgNPs size was large, with mean size ±SD of 10.89 ± 5.67 nm.

The PXRD patterns outline that AC-AgNPs are face-centered cubic. The intense and narrow diffraction peaks revealed the formation of pure crystals of silver and silver chloride. The nanosilver crystal peaks obtained at 2θ values of 38.07°, 46.20°, 64.33° and 77.40° which correspond to the (111), (200), (220) and (311) planes of the face-centered cubic (fcc) structures, respectively (JCPDS file 65-2871). Additional peaks corresponding to silver chloride nanocrystallites were observed at 2θ values of 27.8°, 32.2°, 54.8°, 57.4° and 67.4° indexed to (111), (200), (311), (222) and (400) planes, respectively (JCPDS file 31-1238). SAED suggests that the NPs are polycrystalline with diffraction rings associated due to their stacking each other due to their magnetite phase (Figures 5A, B). The crystallinity percentage of AC-AgNPs was 84.13%. Using data from PXRD, the size of silver nanocrystals and silver chloride nanocrystals based on the Scherrer formula was 13.47 ± 5.81 nm and 10.42 ± 3.34 nm, respectively (Table 2). The overall mean size of AC-AgNPs was 11.77 ± 5.57 nm.

The EDX profile of AC-AgNPs showed a strong signal due to silver atom (Ag) which was involved in AC-AgNPs at a percentage of 58.31%. Other signals due to chlorine (Cl), cadmium (Cd), carbon (C) and oxygen (O) were also observed at 5.56%, 0.36%, 9.31% and 26.46%, respectively (Figure 5C). The identity of functional chemical groups at the interface of AC-AgNPs were determined using FTIR which revealed strong signals at 3,416 cm⁻¹, 1630 cm⁻¹, and 1023 cm⁻¹ which are characteristics of alcohols (O-H stretch), alkenes (C=O stretch) and alkyl and Aryl Halides (C-F stretch), respectively. Smaller signals corresponding to alkanes (C-H stretch) at 2927 cm⁻¹, alkanes/aldehydes/alkenes (C-H stretch, C-O stretch) at 2857/1739 cm⁻¹, nitriles (C≡N stretch) at 2373/2323 cm⁻¹, aromatic compounds (C=C stretch) at 1458 cm⁻¹, and nitro compounds (NO₂ stretch) at 1377 cm⁻¹ were also seen (Figure 5D; Table 3).

The stability of AC-AgNPs was determined using zeta potential, and the analysis revealed a zeta potential value of −18.1 mV which outlines a good stability (Supplementary Figure S3). On analysis of DLS results, the AC-AgNPs had a mean size ±SD of 89.77 ± 16.50 nm, with polydispersity index of 0.242 (Supplementary Figure S4).

High antiplasmodial activity was found for AC-AgNPs against 3D7 (CQ-sensitive) and RKL9 (CQ-resistant) Pf strains. Based on IC₅₀ values, AC-AgNPs exhibited higher antiplasmodial activity as compared to that of AC-CE irrespective of plasmodial strain, and differences were statistically significant (p < 0.0001): 8.05 μg/mL vs. 20.27 μg/mL for 3D7, and 10.31 μg/mL vs. 32.55 μg/mL for RKL9. The standard drug CQ exhibited IC₅₀ values of 0.04 μg/mL and 0.35 μg/mL against Pf strains 3D7 and RKL9 (Figures 6A, B). The SYBR green assay findings were supported by microscopic data. AC-AgNPs and AC-CE elicited no quenching effects as no statistically significant difference was found between fluorescence counts of NPs, standard drug and plant extract (Supplementary Figure S5).

We have noted that hemolysis rates were dependent on substance, dose and time (Figures 7A, B). After 30 min, hemolysis rates elicited by AC-AgNPs and AC-CE were significantly higher than that of CQ at doses ≥62.5 μg/mL. At these concentrations (125–500 μg/mL), hemolysis rates ranged from 6.25%–13.15% for CQ, 14.55%–48.14% for AC-AgNPs, and 5.50%–40.95% for AC-CE (Figure 7A). To be noted, HC₅₀ was not achieved after 30-min incubation as hemolysis rate was below 50% at 500 μg/mL. After 24 hour-incubation, hemolysis rates increased for all substances tested, with the highest values in AC-AgNPs-treated samples (maximum hemolysis of 98.14% at 500 μg/mL). Statistically significant difference between AC-AgNPs, AC-CE, and CQ were seen at doses ≥8 μg/mL (Figure 7B).

Mortality of Cx. quinquefasciatus, Ae. aegypti and An. stephensi larval stages was followed 24h, 48h and 72 h after treatment with AC-AgNPs. Mortality rates of the three mosquito species increased as a function of time and concentration (Supplementary Figure S6). After 48 h incubation, larval mortality rates were 100% at doses 23.5 μg/mL for Cx. quinquefasciatus, 20 μg/mL for Ae. aegypti, and 15 μg/mL for An. stephensi (Supplementary Figure S6). AC-AgNPs were more lethal against An. stephensi regardless of exposure time, with LC₅₀ values of 10.67 μg/mL and 5.85 μg/mL at 24 h- and 48 h-exposure, respectively. These values were 16.71 μg/mL and 7.52 μg/mL for Ae. aegypti; 18.41 μg/mL and 8.97 μg/mL for Cx. quinquefasciatus, respectively. Regardless of exposure time, the larvicidal activity of AC-AgNPs was much higher than that of AC-CE for which LC₅₀ of 231.41 μg/mL, 110.33 μg/mL and 53.15 μg/mL against Cx. quinquefasciatus, Ae. aegypti and An. stephensi were found after 24 h exposure, respectively (Tables 4, 5). Interestingly, AC-CE did not cause any larval mortality at LC₅₀ and LC₉₀ concentrations found for AC-AgNPs.

The stereomicroscopic observations of Ae. aegypti, An. stephensi and Cx. quinquefasciatus larval stages treated with AC-AgNPs are depicted in Figure 8, and revealed the induction of behavioral and morphological changes in mosquito larvae. It was observed that swimming behavior of larvae was reduced, with morbid larvae at the bottom of bowls and unable to swim to the surface. Several morphological changes were noted in AC-AgNPs-treated larvae and these included loss of external hairs/bristles, swelling of the apical cells, pigmentation of the body, shrinkage of the larvae, and necrosis and thickening of the epidermis (Figure 8).