Obtaining the fruits

To obtain the sample, a batch of 100 fruits (5.73 kg) was selected at random, with a maturity index of 26.60. The fruits were collected at the 2018 January-March period in the town of Actopan, Veracruz, located at the latitude: 19°27´36´´, longitude: 96°49´48´´ west 1060 m.a.s.l. After harvest, fruits were washed and disinfected with 0.01% sodium hypochlorite. Manually obtained pulp (1.04 kg) was lyophilized until it became a powder. The taxonomic identification of the fruit was confirmed by biologist Manuel Carlos Durán Espinosa. A copy of the reference fruit, with Num. of herbarium copy: NJuárezT2 (XAL) was deposited in the Xalapa herbarium of the Institute of Ecology A. C (^{Juárez-Trujillo et al., 2018}).

Preparation of experimental animals

The experimental animals used were 21 male Wistar rats, weighing 300 - 359 g and between 3 to 5 months of age from of Neuroethology laboratory of the Universidad Veracruzana. Six groups of n=3 rats corresponding to the different concentrations (10, 100, 1000, 1600, 2900, and 5000 mg/kg weight body) and a control group (without administration) were randomly formed. The rats were housed in transparent plastic boxes, three rats per box inside the bioterium, with light / dark cycles of 12:12 h at a room temperature of 25 ± 2 ºC, and a humidity between 50 - 60 %. The experimental animals carried a standard diet with water and pelletized food ad libitum. They were allowed seven days for acclimatization before the commencement of the research. In this study, all animals were handled in accordance with the guidelines for the Care and Use of Laboratory Animals of Universidad Veracruzana based in the National Institutes of Health.

Acute toxicity study

The dried pulp was mixed at different concentrations with a solution of tween 20 (5%), propylene glycol (5%), and purified water. Doses were prepared according to guide 423, the Organization of Economic Co-operation and Development (^{OECD, 2001}). The groups were administered with extract by intragastric cannulation in a single dose of 5000, 2900, 1600, 1000, 100, and 10 mg/kg body weight for each of the treatment groups, respectively, while the control group received only tween 20, propylene glycol and water used as a vehicle. LD₅₀ was estimated according to the guideline 423 from Organization of Economic Co-operation and Development (^{OECD, 2001}).

The animals were monitored during the first 24 hours after cannulation with special emphasis on the first 30 minutes, and with 15-minute videos taken at 2, 4, 6, 8, 12, and 24 hours in search of apparent modification of toxicological behavior (^{Cavanaugh, 2003}). The symptoms analyzed were mucus, ataxia, salivation, diarrhea, seizures, respiratory disorders, hyperactivity, and paralysis (^{Pérez, 2017}). The weight of the rats was recorded daily for 15 days. On day 15, all rats used in the experimental procedures were sacrificed with an overdose of sodium pentobarbital (60 mg/kg).

A blood sample was obtained by cardiac puncture and collected in plastic tubes containing EDTA. Erythrocyte count, hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, platelet count, and leukocyte count were evaluated using a veterinary hematology analyzer (Abaxis Vatscan HM2). All rats were sacrificed on day 15 and subsequently perfused and dissected; the extraction and weighing of organs, liver, kidneys, pancreas, and spleen were performed.

Histopathology study

Selected organs, including pancreas, liver, kidney, and spleen of the treated and control animal groups, were fixed in 10 % buffered formalin in labeled bottles and subsequently processed for histological examination. The tissue was processed using an automated tissue processor (Leica 820, Germany) and embedded in blocks of paraffin. Stained tissue sections were examined under a light microscope (Leica DM 750, Germany). Number and area of Langerhans islets in the pancreas were determined in at least five islets per section at a 40x magnification. Full images of the selected Langerhans islets were obtained by photomicrographs with Adobe® Photoshop CS6 software, and their analysis was performed using the ImageJ Java® software in which the area, thickness, density, and volume data of the selected islets were obtained according to the methodology reported by ^{Cavanaugh (2003)}. The numerical concentration of the cells in the islet was determined by manual counting by differential hematoxylin and eosin phloxine (H&E) staining (^{Coggehall, 1992}; ^{Mohammadi and Naik, 2012}).

Antimicrobial activity of the aqueous extract of pulp of Randia monantha fruit

The Randia monantha pulp extract antimicrobial activity was evaluated using the broth dilution method. Active cultures of Listeria monocytogenes (ATCC 19115), Staphylococcus aureus (ATCC 25923), Escherichia coli (ATCC 43895), and Salmonella typhimurium (ATCC 14028) were prepared according to ^{Balouri et al. (2016)}. For the above, dilutions were prepared at 0.5, 1.0, 1.5 and 2.0 mg/mL in a liquid growth medium, distributed in tubes containing a volume of 2 mL. Subsequently, each tube was inoculated with a microbial inoculum prepared in the same medium, which was previously standardized to the Mcfarland scale of 0.5. Each tube was then shaken and incubated for 24 hours at 37 °C. After 24 h, the absorbance λ= 600 nm was read.

Statistical analysis

Analysis of variance was carried out to evaluate the results found in the different treatments with pulp extracts, and subsequently, the significant differences were assessed using the Fisher test with a level of significance p ≤ 0.05, using Minitab program version 17, States United.

Body and organs weight gain

The results of the acute toxicity study indicated that the aqueous extract of the R. monantha pulp administered orally to rats at doses between 10-5000 mg/kg body weight in acute toxicity testing methodology did not cause death or significant toxicity in test animals, indicating that the LD₅₀ is greater than the doses administered. Table 1 shows that the groups corresponding to an administration of 2900 and 5000 mg/kg body weight presented a slight increase in weight that ranged from 0.25 to 7.94%. However, in relation to the weight of the organs, it was observed that the pancreas, liver, kidneys, and spleen did not show significant differences (p > 0.05) compared to the respective organs of the control treatment.

Behavior observations

Consistent with these results, the different treatment groups showed no significant difference in the evaluated clinical signs (mucus, ataxia, salivation, diarrhea, convulsions, hyperactivity, paralysis, and changes in breathing), compared to the control treatment.

Macroscopic evaluation of organs

External macroscopic analysis of liver, spleen, and pancreas did not reveal any alteration nor signs of toxicity, without visual differences with the control treatment, indicating the absence of toxicity. However, the kidneys of the group corresponding to an administration of 5000 mg/kg body weight showed yellow spots. The kidney histopathology examination revealed that the yellow areas presented a minor or no blood supply, with a notable decrease in erythrocytes and a decrease in blood flow than the kidney cuts of the control treatment (Figure 1).

Figure 1 Kidney histological examination. A) Control treatment and B) Treatment with a dose of 5000 mg/kg of weight. The arrows point out the blood supply.
Figura 1. Examen histologico del rinon de A) Tratamiento de control y B) Tratamiento con dosis de 5000 mg/kg de peso. Las flechas senalan el aporte de sangre. 

Hematology parameters analysis

The hematology parameters are summarized in Table 2. The hematological analysis of leucocytes, lymphocytes, monocytes, granulocytes, red blood cell, mean cell volume, hemoglobin, hematocrit, mean cell volume, mean corpuscular hemoglobin, erythrosary distribution index, platelets, mean platelet volume, and platelet distribution revealed no significant differences (p > 0.05) in relation to the control treatment and the days analyzed, the values were within the range considered normal (^{Tasić et al., 2021}; ^{Nagamma et al., 2019}; ^{Dantas et al., 2006}). The samples did not show significant differences (p > 0.05) in the triglycerides and cholesterol concentration with the control group, which should be considered in treating dyslipidemias. Treatments with a dose between 1600 and 5000 mg/kg of body weight exhibited at time zero (first day of administration) had an increase in glucose concentration (116-131.69 mg/dL) compared to the control group (62.79-77.63 mg/dL). However, these values are in the range of values considered normal for laboratory animals (^{Lapchik et al., 2017}).

Histopathology examination of pancreatic cells

The size, area, diameter, and volume of the Langerhans islet of the different treatments did not show significant differences (p > 0.05) (Table 3), which is important since its parameters directly influence the maintenance of glucose in the blood and their alteration is related to diabetes (^{Jo et al., 2007}). However, a significantly lower concentration of langerhans islets (643 cell/mm²) was observed in the 5000 mg/kg body weight treatment with compared to the other treatments. The groups that were administered 10, 100, 1000, 1600 and 2900 mg/kg body weight of extract exhibited a concentration of cells in the Langerhans islets that varied from 2008 to 3146 cells/mm2, without significant differences (p > 0.05) with the control treatment (3100 cells/mm2). In contrast, the treatment of 5000 (643 cells / mm2) mg/kg body weight, showed a significant decrease in the number of cells in relation to the control treatment. Consistent with these results, the pancreas cells of the group administered with 5000 mg/kg body weight of extract showed a lower number of cells (alpha, beta, and delta) in the Langerhans islets compared to the control group (Figure 2).

Figure 2 Pancreas histological examination. A) Control treatment and B) Treatment with a dose of 5000 mg/kg of weight. The arrows point out the Langerhans beta cells.
Figura 2. Examen histologico del pancreas. A) Tratamiento de control y B) Tratamiento con dosis de 5000 mg/kg de peso. Las flechas senalan las celulas beta de Langerhans. 

Antimicrobial properties of aqueous extract

The antimicrobial effects of the Randia monantha fruit aqueous extract against Gram (-) and Gram (+) bacteria are shown in Table 4. It can be seen that the aqueous extracts did not present dose-dependent antimicrobial activity in the test concentration range (0.5-2.0 mg/mL) against Gram (+) bacteria, nor Gram (-) test bacteria. These results were similar to those reported for the methanol and ethanol extract of Randia aculeata against Mycobacterium bovis (^{García-Cruz, 2018}), but different from those reported for the crude extract of Randia dumetorum, which exhibited significant antimicrobial activity (^{Kumar et al., 2006}).