Chemical and toxicological study of leaf extracts from Deinbollia boinensis Capuron (Sapindaceae), a malagasy medicinal plant

 Corresponding author: Maholy Pricille Ratsimiebo Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. Chemical and toxicological study of leaf extracts from Deinbollia boinensis Capuron (Sapindaceae), a malagasy medicinal plant Maholy Pricille Ratsimiebo 1, 2, *, Lolona Rakotobe 1, 2, Mihajasoa Stella Razanatseheno 1, 2, Herizo Lalaina Andriamampianina 1, 2, Lovarintsoa Judicaël Randriamampianina 1, 2, Hanitra Ranjàna Randrianarivo 1, 2, Danielle Aurore Doll Rakoto 1, 2 and Victor Louis Jeannoda 1, 2 1 Laboratory of Applied Biochemistry to Medical Sciences, Fundamental and Applied Biochemistry Department, Faculty of Sciences, University of Antananarivo, P.O. Box 906, Antananarivo 101, Madagascar. 2 Life and Environmental Sciences Doctoral School (SVE), University of Antananarivo, P.O. Box 906, Antananarivo 101, Madagascar. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 Publication history: Received on 03 January 2025; revised on 15 February 2025; accepted on 18 February 2025 Article DOI: https://doi.org/10.30574/wjbphs.2025.21.2.0186 Abstract The aim of this study was to investigate the toxicity of Deinbollia boinensis, a Sapindaceae endemic to Madagascar. A purification process involving precipitation with 50% ethanol, dialysis and fractionation with n-butanol was used to obtain a partially purified toxic extract (PPE) from the cold crude aqueous leaf extract (LCE). The active compounds were thermostable and soluble in water, ethanol and n-butanol. They could be precipitated by neutral lead acetate and were adsorbed on activated charcoal. Phytochemical screening of LCE and PPE revealed the presence of tannins, polyphenols, saponins, deoxyoses, triterpenes and sterols. PPE caused symptoms in mice that suggested damage to the nervous system, as well as tissue lesions in some organs characterised mainly by capillary dilatation and hemorrhagic areas in the brain, lungs and kidneys. In the liver, architectural destruction was observed. The PPE LD50 was in the order of 60.25 to 66.55 mg/kg of mice. PPE was also toxic to other warm-blooded animals such as guinea pigs, rats, chicks and cold-blooded animals including mosquito larvae, carp alvins and frog tadpoles. It caused lysis of sheep red blood cells. LCE inhibited the germination of seeds of some plants and the growth of young maize and cowpea bean seedlings. LCE had a stimulating effect on the development of axillary buds, while PPE had an inhibitory effect. LCE and PPE had no effect on the microorganisms tested. Keywords: Deinbollia boinensis; Sapindaceae; Leaf toxicity; Warm-blooded animals; Cold-blooded animals; Effects on plants 1. Introduction Medicinal plants are widely used as a source of medicines, especially in developing countries. The World Health Organization (WHO) estimates that up to 80% of these countries rely on locally available plant resources for their primary health care because they are easily accessible and less expensive [1]. Despite their therapeutic properties, medicinal plants must be used with extreme caution as they can be toxic [2]. The WHO recommends strengthening research and evaluating the safety and efficacy of herbal products [1]. Among the many plant families with medicinal uses is the Sapindaceae family, which contains several genera, including Deinbollia. This genus has 41 recognised species, including one shared between Africa, La Réunion and Madagascar, 5 endemic to Madagascar and 35 restricted to sub-Saharan continental Africa [3]. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 508 Certain species of Deinbollia are used in a number of empirical applications. The leaves of D. pinnata are used to treat sickle cell disorder [4]. The roots and leaves of this species are used in folk medicine as a febrifuge and analgesic, and to treat intercostal bronchitis, intestinal pain, jaundice, cough, asthma and infections [5, 6, 7]. The roots of D. oblongifolia macerated in water are drunk to treat stomachache [8], and boiled roots heal diarrhea [9]. A decoction of D. borbonica roots is used to treat diabetes. Dried leaves and roots are used externally to accelerate wound healing [10]. A few species of Deinbollia were the subject of biological studies. D. pinnata, the most studied species was widely reported for its antibacterial [11], anti-inflammatory [12] and antioxidant [13] activities. D. oblongifolia has antibacterial activity [14]. Deinbollia boinensis, a Malagasy medicinal species, has been used by aged traditionalist for the treatment of various ailments such as fever, headaches and stomachaches and it is also cultivated as an ornamental plant in gardens and as a hedge plant [15]. However, rumours about its side-effects and toxicity persist, limiting its use as a medicine. D. boinensis was chosen as study material in the present work. This choice was made for a number of reasons. Firstly, it is a plant that has never been the subject of any research other than botanical research. Secondly, it is used in traditional medicine, but rumours circulate about its toxicity, which were confirmed during preliminary toxicity tests carried out on mice in our laboratory. The aim of our research was to determine the chemical composition and to study the toxicity of D. boinensis leaf extracts on various organisms. 2. Materials and methods 2.1. Plant Material 2.1.1. Deinbollia boinensis D. boinensis (Figure 1), a shrub of 3 to 6 m, grows on calcareous soils in the west of Madagascar from Antonibe peninsula to the Onilahy river basin. It is known under the vernacular names such as Ampelamainty, Ampoly fotsy, Dovy, Fandriatotoroka, Kamoty, Tsiramiramy. (Source: the authors) Figure 1 Deinbollia boinensis: the whole plant and leaves Fresh leaves of D. boinensis were collected in Marofandiliha forest, located in the west of Madagascar, in the Morondava region, in September during its vegetative stage. Shade-dried leaves were ground into a fine powder and then stored at -20 °C. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 509 2.1.2. Plant Seeds Seeds came from the collection of National Research Center for Farming (FOFIFA, Antananarivo) (Table 1). Table 1 Plants whose seeds were used for germination assays Plant families Monocotyledons Dicotyledons Common name Poaceae Zea mays Maize (corn) Oryza sativa Rice Apiaceae Daucus carota Carrot Petroselinum crispum Chinese parsley Asteraceae Lactuca sativa Lettuce Brassicaceae Brassica sp Tissam white Cucurbitaceae Cucurbita pepo medullosa Pumpkin, Zucchini Fabaceae Phaseolus vulgaris Bean Pisum sativum Pea Vigna unguiculata Cowpea bean Vigna subterranea Bambara pea Lamiaceae Ocimum basilicum Basil Liliaceae Allium cepa Onion Solanaceae Solanum nigrum Black nightshade Solanum tuberosum Potato 2.2. Animals 2.2.1. Mice OF-1 strain Albino mice (Mus musculus), weighing 25 ± 2 g, came from the Pasteur Institute of Madagascar (IPM) breeding farm. 2.2.2. Tadpoles Apode frog tadpoles (Ptychadena mascareniensis) were harvested from the ponds in the vicinity of the Antananarivo University site. 2.2.3. Alvins The 2-month-old Cyprinus carpio alvins were provided by a fish farmer in Manjakandriana, 45 km east of Antananarivo. They were allowed to acclimatize to the aquarium conditions for a few days before testing. 2.2.4. Mosquito Larvae Stage 3 mosquito larvae, Culex quinquefasciatus, came from stagnant water around the Antananarivo University Campus. 2.2.5. Chiks One day-old chicks (Gallus gallus domesticus), Hubbard classic strain, were provided by poultry farmer. 2.2.6. Other Species Tricoloured guinea pigs (Rattus norvegicus) came from an approved private supplier and white rats (Rattus rattus) of the WISTAR strain from the breeding farm of the Department of Animal Physiology of Antananarivo University. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 510 2.3. Microorganisms Strains Microorganisms used in this study were supplied by the Environmental Microbiology Laboratory (LME) of the National Environmental Research Centre (CNRE). They consisted of 7 strains of bacteria including 1 Gram (+) and 6 Gram (-) and 1 yeast (Table 2). Table 2 List of germs used Microorganisms Strains GRAM Bacteria Staphylococcus aureus + Yersinia pestis - Alcalescens dispar - Escherichia coli - Shigella sonnei - Klebsiella oxytoca - Klebsiella rhinoscleromatis - Yeast Candida albicans 2.4. Preparation of Extracts 2.4.1. Cold Aqueous Extraction Leaf powder was suspended in distilled water in a ratio of 1/10 (w/v). The mixture, subjected to magnetic stirring for 3 h at room temperature, was then left to macerate overnight at 4°C. The macerate was filtered through four layers of gauze to remove insoluble residues. The filtrate obtained was centrifuged at 3,000 rpm for 30 min. The supernatant was recovered and the pellet discarded. 2.4.2. Purification Crude extract was purified using Razanatseheno et al. [16] purification methods, based on solubility, molecular weight or electric charge of active principles. Partially purified extract obtained of this purification was named PPE. 2.5. Phytochemical Screening The detection reactions of chemical groups on leaf powder and extracts were carried out according to the methods of Fong et al. [17] and Marini-Bettolo et al. [18]. 2.6. Acute Toxicity Test in Animals 2.6.1. Effect on Mice Toxic effect on mice was evaluated by intraperitoneal route (i.p). The extract was injected at a volume of 0.3 mL per 25 g of body weight. All changes in intoxicated mice behavior were observed during 24 h after extract administration and other physiological activities or death was noted. The LD50 (24 h) of extract was determined on mice by calculation and graphical methods [19]. Seven different doses of PPE were injected by i.p route on seven groups of five male mice. Another group receiving physiological serum served as control. 2.6.2. Anatomopathological Study Histopathological examination was carried out as described by Rasoatahina et al. [20]. Brain, lungs, heart, stomach, liver, kidneys and intestine of mice were harvested. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 511 The preparation of the organ sections for histopathological examinations was carried out using a classical method including the following steps: body fixation, inclusion, organ sections, preparation by microtomy, glass slide mounting, staining and microscope examination. 2.6.3. Hemolytic Activity Test The hemolytic activity of PPE was evaluated as described previously by Razanatseheno et al. [16] with a slight modification. The red blood cells (RBC) were washed three times with physiological saline. The solution was centrifuged at 3,000 rpm for 5 min. The pellet was recovered and the supernatant removed by aspiration. A 100% red cell suspension was obtained at the end of the third wash. The suspension was diluted with 2% phosphate buffered saline (PBS). The 2% red cell suspension was distributed in the wells of a V-bottom microplate. The extract to be tested, diluted in cascade with PBS was poured into the wells. Two controls were performed: a positive control (C+) and a negative control (C-). After a gentle stirring, plate was incubated in an oven at first at 37°C for 3 h and then in a fridge at 4°C for 24 h. The composition of the medium in each well is shown in Table 3. Table 3 Composition of the medium for the hemolytic test Wells n° C+ C3 4 5 6 7 8 9 10 11 12 Final Concentration of PPE (µg/ml) 0 0 1000 500 250 125 62.5 31.25 15.62 7.81 3.90 1.95 PPE 1 mg/ml (µl) 0 0 50 25 12.5 6.25 3.12 1.56 0.78 0.39 0.19 0.10 PBS (µl) 0 50 0 25 37.5 43.75 46.87 48.44 49.22 49.61 49.81 49.90 2% red cell suspension (µl) 50 50 50 50 50 50 50 50 50 50 50 50 Distilled water (µl) 50 0 0 0 0 0 0 0 0 0 0 0 Final volume of mixture (µl) 100 100 100 100 100 100 100 100 100 100 100 100 2.6.4. Acute Toxicity Test in Cold Blooded Animals Different concentrations of extract were tested in alvins and frog tadpoles according to Razanatseheno et al. [16] techniques. The graphical method of linear regression by Boyd [21] was used to determine the LC50 (24 h) or lethal concentration that killed 50% of the animals tested in 24 h. For the mosquitoes, larvae were placed in crystallizers each containing a final volume of 200 ml of fresh water. Different concentrations of extract were added in the medium. After 24 h, dead and moribund larvae were counted. 2.7. Assays in Plants 2.7.1. Assays on Seed Germination Batches of 10 seeds of each species were used. These seeds were first soaked in water at 30°C in darkness for 48 h. The soaked seeds were transferred on cotton wool, soaked in extract or in water in a Petri dish. Seed germination was observed 72 h later. The lifting or not of seed dormancy was assessed. 2.7.2. Effects on Seedling Growth The effects of extract were studied on epicotyl and hypocotyl growth of rice and bean according to the method developped by Razanatseheno et al. [16]. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 512 2.7.3. Assays on Axillary Bud Growth Assays were realized on 15-day-old pea seedlings previously sectioned above the second axillary bud as described by Rakoto et al. [22]. Effects of extracts were compared with those of the plant growth regulators gibberellins and auxin. 2.8. Antimicrobial Assays Antimicrobial activity was assessed using the methods detailed in our previous articles [23, 24]. They consisted in measuring the inhibition zone diameter (IZD) using the disc diffusion method [23]. The results were interpreted according to the IZD scale [25]: bacteria were considered non-susceptible with an IZD ≤ 8 mm; susceptible with an 9 ≤ IZD ≤ 14 mm; highly susceptible with an 15 ≤ IZD ≤ 19 mm and extremely susceptible with an IZD ≥ 20 mm. Neomycin was used as the reference antibiotic and miconazole as the reference antifungal agent. 3. Results 3.1. Extraction Yields Extraction of the dried leaves (50 g) of D. boinensis gave a dark brown colored leaf crude extract (LCE) with a pH of 4. The extraction and purification processes are summarized in the following diagram (Figure 2). Figure 2 Diagram summarizing the extraction and the purification stages of active principles From 50 g of D. boinensis leaf powder, 8.4 g of LCE and 0.65 g of partially purified extract (PPE) were obtained. The extraction yield was estimated at 16% and the purification yield 1.3%. 3.2. Phytochemical Screening The phytochemical screening of LCE and PPE is presented in Table 4. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 513 Table 4 Phytochemical screening results of LCE and PPE Chemical families Tests Results LCE PPE Saponins Foam test + + Tannins and polyphenols Gelatin test + + Salted gelatin test + + Ferric chloride test + + Deoxyoses Keller – Kiliani + + Iridoids Hot HCL - - Alkaloids Wagner - - Mayer - - Dragendorff - - Flavonoids and leucoanthocyanins Wilstater - - Bate-Smith - - Steroids and triterpenes Lieberman-Burchard + + Salkowski + + Anthraquinones Bornträger - - The results of phytochemical screening of LCE and PPE revealed the presence of tannins and polyphenols, deoxyoses, saponins, unsaturated sterols and triterpenes. 3.3. Effects on Animals 3.3.1. On Mice Behavior Studies After injection of a lethal dose of PPE (115 mg/kg), the signs of the nervous system disorders were developed by contortion of the abdomen, followed by hyperexcitation. After 20 min, its motor activity suddenly decreased. Incoordination of the limbs was observed. The respiratory systems attacks were manifested after 2 h by reduction of respiration frequency, and cyanosis. Exophthalmos and piloerection were noted. Death occurred after 3 h by ataxia and clonic convulsions. Symptoms were developed after injection of a sublethal dose of PPE (38.5 mg/kg), such as a passive state, incoordination of the limbs during rare movements accompanied by hyperemia and piloerection. A progressive remission was observed after the 10th hour. LD50 Values The LD50 of PPE by i.p route was assessed at 60.25 mg/kg to 66.55 mg/kg body weight. Histopathological Lesions The study of tissue damage due to the effects of PPE was carried out on mice injected by i.p route with a lethal dose of 115 mg/kg body weight. The mice were sacrificed 3 h after the symptoms of intoxication appeared. The organs (brain, heart, stomach, liver, intestine, lungs and kidneys) were removed. The main lesions caused by PPE on each organ are summarized in Table 5 and shown in Figure 3. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 514 Table 5 Main lesions caused by PPE at 115 mg/kg on mice organs Organs Observed lesions Brain Dilated cerebral parenchymal capillaries Liver Dilated sinusoidal capillaries, altered hepatocytes Kidneys Retracted glomerular flocculi, hemorrhagic areas Lungs Dilated interalveolar capillaries Heart Normal myocardial tissue Stomach No histological lesion Intestine No histological lesion Figure 3 Main histological lesions in brain, kidney, lung and liver due to i.p. administration of the PPE at a dose of 115 mg/kg weight (magnification X 400) • DC : dilated capillary; HA : hemolytic area; RGF : retracted glomerular flocculi; AH : altered hepatocyte; DSC : dilated sinusoidal capillary Histopathological lesions were observed in the brain, liver, lungs and kidneys. The main detoxifying organs (liver and kidneys) were the most affected. No histological lesion was observed in intestine, heart and stomach. 3.3.2. Effects of PPE on Sheep Red Blood Cells The hemolytic activity of PPE on sheep red blood cells is shown in Table 6 and Figure 4. World Journal of Biology Pharmacy and Health Sciences, 2025, 21(02), 507-520 515 Table 6 Effects of different concentrations of PPE on sheep red blood cells Concentration (µg/ml) 1.95 3.90 7.81 15.62 31.25 62.5 125 250 500 1000 Effect - - - - - + ++ ++ ++ ++ Figure 4 Hemolytic activity of PPE on sheep red blood cells C+: Positive control; C-: Negative control; TH: Total hemolysis; PH: Partial hemolysis; NH: No hemolysis. The hemolytic activity of PPE varied with concentration. No hemolysis was observed at a concentration ≤ 31.25 µg/ml. Partial hemolysis with sedimentation of intact red blood cells was observed at concentrations between 31.25 µg/ml to 62.5 µg/ml. Hemolysis was total at concentrations ≥ 125 µg/ml. 3.3.3. Effects on Cold-Blooded Animals Effects on Frog Tadpoles Seven concentrations of PPE ranging from 45.63 µg/ml to 61.15 µg/ml (reason 1.8) were tested on seven batches of frog tadpoles. The results of these tests are presented in Table 7. Table 7 Effects of different concentrations of PPE on frog tadpoles Concentration (C) in µg/ml log C Number of dead % of death 61.15 1.786 10 100 58.23 1.765 10 100 55.46 1.743 10 100 52.82 1.722 10 100 50.30 1.701 10 100 47.91 1.680 10 100 45.63 1.659 0 0 According to these results, the PPE effect obeyed the « all-or-nothing law »: at 45.63 µg/ml the mortality rate was 0%, whereas at 47.91% it was already 100%. Effects on Carp Alvins Eight PPE concentrations ranging from 40.65 µg/ml to 45.12 µg/ml, with a geometric reason of 1.015, were tested. The results are shown in Table 8.