Acute toxicity and hypoglycaemic activity of the leaf extracts of Persea americana Mill. (Lauraceae) in Wistar rats

1 Département de Pharmacologie Clinique, UFR Sciences Médicales Bouaké, Université Alassane Dramane Ouattara, Côte d’Ivoire. 2 Département de Pharmacologie Clinique, UFR Sciences Médicales Abidjan, Université Félix Houphouët Boigny, Côte d’Ivoire. 3 Unité de gestion des Ressources Animales, Institut Pasteur (UGRA-IPCI), Côte d’Ivoire. 4 Laboratoire de Chimie Bio organique et de Substances Naturelles (LCBOSN), UFR-SFA, Université Nangui Abrogoua, Côte d’Ivoire.


INTRODUCTION
According to the World Health Organization (WHO), diabetes mellitus is a chronic disease that occurs when the pancreas does not produce the required amount of insulin or when the insulin released in the body is not adequately used.As a result, an increase of glucose serum level (hyperglyceamia) occurs, bringing about complications such as diabetic neuropathy, retinopathy and cardiovascular diseases (Kumar et al., 2011;WHO, 2015).Thus, in 2014, the estimated people living with diabetes reached 387 million and the figure is projected to reach 592 million in the coming twenty years (IDF, 2015).To illustrate, the diabetes prevalence in sub-Saharan Africa, which is currently set at 12.1 million, is expected to rise to 23.9 million by 2030 (Keter and Mutiso, 2012).From this perspective, diabetes mellitus constitutes a global health concern.Therefore, the discovery of a sustainable and adequate treatment for this disease appears to be a challenge to scientists.Studies conducted earlier reported on the use of herbal medicines and traditional remedies to cure diabetes.For example, in China and in Africa, traditional medicine accounts respectively for about 40 and 80% of the proposed treatment of the population meeting their health care needs including treatment of diabetes (WHO, 2002(WHO, , 2013)).
Additionally, records show that roughly 800 plants are used worldwide to treat diabetes mellitus (Alarcón et al., 1993).Among them is Persea americana mill.The avocado is described as a medium to large tree, 9-20 m high, and belongs to the Tracheophyta division of the Magnoliopsida class en the Lauraceae family, and endemic to Mexico (Central America).P. americana can also be found in most sub-tropical and tropical countries like Côte d'Ivoire.The number of species identified in the family is about 500 (Adeboye et al., 1999, Yasir et al., 2010).
Even though previous investigations have reported its biological activities, including anticancer, anti-diarrhea, analgesic and anti-inflammatory properties, little is known about its toxicity and hypoglyceamic activity (Adeyemi et al., 2002;Butt et al., 2006;Odo et al., 2014).The study aforementioned showed lower toxicity and hypoglyceamic effect.However, the mode of action was not explored in depth.
Therefore, the present study aimed at evaluating the acute toxicity from oral route uptake of three extracts of P. americana leaves and the comparative hypoglyceamic and anti-hyperglyceamic activities in normoglyceamic rats.

Extraction method
Fresh leaves of the plant were collected, washed, shade dried, cut into small pieces, and powdered in a Phillips ® blender to obtain a weight of approximately 0.5 kg.

Aqueous extract (AE) of the leaves of P. americana
Fifty grams (50 g) of P. americana leaves powder was boiled in one liter (1 L) of distilled water for 30 min, allowed to cool at room temperature (25°C) and filtered with a Fisherbrand ® paper.The decoction was then lyophilized using the Martin Christ ® ALPHA 2-4 LDplus, Germany and stored at 2-4°C until use for the bioassays and phytochemical analysis.

Ethanol and methanol extracts (EE, ME)
The dried powder of P. americana leaves was macerated three times with absolute ethanol (10%, w/v), Prolabo ® , France for 48 h at room temperature and shaken occasionally.The ethanol fraction was pooled after filtration and dried in an oven (Memmert ® , Germany) at 40°C.Similar extraction procedure was repeated for methanol extract using absolute methanol (Quimicen ® , Spain).

Phytochemical screening
The phytochemical screening was carried out to investigate the presence of active ingredients of extracts AE, EE and ME.The prospective compounds to be identified were generally alkaloids, flavonoids, tannins, polyphenols derivatives, triterpenoids/steroids and coumarins.These identification are performed by the appropriate extraction solvent.So, alkaloids were isolated through the Draggendorff test (Gidwani et al., 2011), flavonoids identified by the Shinida test (Vinod et al., 2010), tannins and polyphenols compounds by the FeCL3 (Acros organic ® , Belgium) test (Békro et al., 2007), saponins determined by foam formation and foaming index (Dohou et al., 2003), steroids and tripenoids were revealed by the Liebermann-Bürchard test (Békro et al., 2007), and coumarins were detected using NaOH (Carlo Erba ® , France) reagent.The compounds aforementioned were visualized under UV-Visible light (λ=366 nm) (Rizk et al., 1986).

Determination of total polyphenol compounds
Total polyphenol compounds' content from leaf extracts of P. americana were determined using the Folin-Ciocalteu colorimetric method (Singleton and Rossi, 1965).The named method is *Corresponding author.E-mail: kouame_mat@yahoo.fr.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License developed as follows: Primarily, 1 ml of plant extract (0.02 mg/ml) was mixed with 0.5 ml of the reagent.Then, 1.5 ml of Na2CO3 (17% w/v; Carlo Erba ® , France) was added after shaking thoroughly with a vortex (Heidoph ® , Germany).This mixture was incubated at 40°C for 10 min in a water bath.The absorbance of the total phenol compounds were read out using a Cecil ® spectrophotometer (ce1021, UK).Before then, the standard calibration curve was obtained using gallic acid with a concentration range of 1 to 10 μg/ml and the figure recorded were expressed as gallic acid equivalent (GAE) (μg/g) of plant extract.

Determination of total flavonoid compounds
Twenty milligram (20 mg) of each P. americana leaves extract were mixed with 10 ml of ethanol (80%, v/v).Then, after 10 min, 50 μl of the Neu reagent mixed with 12 ml of pure methanol were added to 1 ml of the crude extract.In addition, the mixture was shaken using a vortex, and the absorbance was read at a wavelength of 404 nm using the Cecil ® spectrophotometer.For this experiment, quercetin solution (0.05 mg/ml) was used as positive control and the percentage of total flavonoids was expressed as quercetin equivalent according to the following relationship (Dohou et al., 2003): F (%): Percentage of total flavonoids in crude extract, Aext: absorbance of crude extract, Aq: absorption of quercetin (standard pure compound), Cext: plant leaves extract concentration (2 mg/ml).

Biological material
Wistar albinos (Rattus norvegicus var.albinus) bred up to 2-3 months at the Animal Resources Unit of the Institute Pasteur Côte d'Ivoire, and weighing 150-250 g were used for the bioassay.Their living condition could be described as follows: the housing was cross-ventilated at mean temperature of 24-28°C with relative humidity of 60-80% and 12 h/day light.Prior to the experiments, rats were allowed to acclimatize for 2 weeks during which they were fed with standard diet and water ad libitum according to the international standards of animal use and care.

Sample preparation
Extracts AE, EE, ME and reference antidiabetic drug (glibenclamide; Daonil ® , Sanofi-Aventis) were prepared in a mixture of water and Tween 80% (2%, v/v) called the vehicle.Then, test samples were taken up orally by force-feeding the animals, which were under mild anaesthesia (lidocaïne oral gel; AstraZeneca ® , UK).The negative control group of animals received the vehicle only orally (10 ml/kg, b.w.).

Acute toxicity study of extracts
Acute oral toxicity was evaluated by following modified OECD test guidelines (OECD, 2001).Previously, research showed that P. americana exhibits low toxicity.Based on that, the "limit test" was used to assess its toxicity level (Figure 1).Thus, the experiment was designed such that, healthy female rats underwent fasting overnight, and they were divided into four groups (n=3) randomly.
Group I received vehicle (10 ml/kg; b.w.) by oral route, was considered as negative control whereas, Groups II, III and IV received 2000 mg/kg, b.w. of AE, EE and ME, respectively.The experiment was repeated 2 weeks later with another group of animals.
After extract administration, the rats were observed continuously every 1h for 4 days.Further observation was carried out every 24 h for 15 days to identify any change such as tremors, convulsions, salivation, diarrhea, lethargy, sleep, coma and decrease of the respiratory rate or any lethality.Thus, after 4 h observation, all rats were allowed to feed themselves and drink water ad libitum.Then, the body weight was monitored every four days.From this experiment, the median lethal oral dose (LD50) was determined with respect to the OECD 423 standards fixed by the Globally harmonized Classification System (GHS) (Figure 1).
At day-15, animals were subjected to fasting overnight, euthanized by cervical dislocation under anaesthesia (Forene ® , Abbott, USA).Blood samples were collected in red topped plastic vials and centrifuged at 1500 g for 10 min.The serum collected out of this step was used for biochemical analysis.A gross necropsy of all animals was also carried out.The heart, liver and kidney were carefully isolated and weighed individually.From the bio-analysis performed, the following parameters were determined: blood urea nitrogen (BUN), creatinine (CREA), aspartate aminotransferase (AST), alanine aminotransferase (ALT), sodium (Na + ), potassium (K + ), chloride ions (Cl -), using biochemical kits (Hitachi 704R ® ) and electrolyte analyzer (ISE 3000 ® ), respectively.

Dose selection
First, a dose of 2000 mg/kg from each extract of the plant was selected for acute toxicity evaluation.Then, both 1/10 th and 1/20 th of the dose, which do not display any behavioural alterations, were considered for anti-diabetec tests (Oliveira et al., 2008).Additionally, the 1/5 th of the dose was also selected for further identical assays.

Acute hypoglycaemic effect on normal rats
The set of healthy males rats was organized by dividing them into 11 groups (n=5 per group), followed by, fasting blood glucose level recording, at an initial time (t=0 h), after the overnight fasting (16 h) and water uptake ad libitum.The negative control (Group 1) received by oral route, 10 ml/kg b.w. of vehicle aqueous Tween 80 (2%; v/v), and the positive control group (2) was administered a standard hypoglycaemic drug (glibenclamide) 10 mg/kg b.w.Groups 3 to 11 received AE, EE and ME at 100, 200, 400 mg/kg for each.Then blood was collected from tail-tip after oral administration of test samples followed by a period of observation which lasted ½, 1, 2, 3 and 5 h.Blood glucose levels were recorded using reactive strips (GOD-POD) and a glucometer (Accu-Chek ® Active, Germany).

Oral glucose tolerance test (OGTT)
Male Wistar normoglyceamic rats were subjected to overnight fasting for 16 h, followed by fasting glycaemia recording before extracts administration, which is called the ½ h glucose administration.Animals were then divided into 11 groups; each of which was made up of five rats (n=5).Group 1 taken as the negative control received the vehicle; Group 2 the positive control was treated with glibenclamide (10 mg/kg, b.w.), and Group 3 to 11 received extracts AE, EE, ME at 100, 200 and 400 mg/kg for each.Each rat was fed 2 g/kg of D(+)-Glucose monohydrate 30 min after administration of extract and was defined as initial time (t=0 h).Then, blood was collected from tail-tips at different time periods -½, 0, ½, 1, 2, 3 and 5 h consecutive to glucose administration.This was followed by glyceamia recording by use of a clinical glucometer (Accu-Chek Active ® , Roche).

Statistical analysis
Results of bioassays were expressed as mean ± S.E.M using one way ANOVA, followed by the Newman-Keul test (GraphPad Prism, version 5.01) and differences were considered significant at p ≤ 0.05.The decrease percentage of glyceamia for each group was calculated as follows: G1/2h: Glyceamia value at ½ h, Gx: glyceamia at x hour; (x = 1, 2, 3 and 5).

Yield of the crude extracts
From the dried plant material, the yield of the crude extract AE was 20.9% (w/w), that of EE was 16.8% and ME 18.6% (w/w).

Phytochemical screening
The phytochemical screening exhibited saponins, polyphenols, flavonoids, alkaloids, sterols/polyterpenoids and coumarins as the ingredients of the plant extract.Moreover, AE was identified to contain gallic tannins, whereas EE and ME contained catechic tanins.The foaming index for AE, EE and ME was respectively 33, 500 and 250.

Determination of total flavonoid compounds
The spectrophotometric determination of total flavonoid compounds revealed a content of 0.543±0.007,

Acute toxicity determination
In this study, all P. americana leaf extracts did not show any toxic effects.Furthermore, neither lethality nor toxic reaction was found at a selected dose of 2000 mg/kg throughout the experiment.Mortality was also not observed when extract were used to treat the rats.With regards to the OEDC guideline, the LD 50 values offered by the extract of P. americana are higher than 5000 mg/kg.In addition, the gross necropsy test performed on all animals revealed neither hypertrophy, lesion, color change and appearance nor gross pathological impairment of organs at a glance.For further studies, concentrations were fixed at 100 (1/20 th ), 200 (1/10 th ) (Oliveira et al., 2008), and 400 mg/kg (1/5 th ) according to the study protocol.

Changes in body weight
Throughout the 15 days observation of animals, body weight of each within a distinct group was considered for comparison.As a result, no significant difference was noticed between both control and treated groups (Figure 2).Nevertheless, at the end point, body weight gain significantly increased by 27.3% for groups fed AE and ME, and by 22.3% for groups fed with EE.

Organ to body weight ratio
Table 1 shows the organ to body ratio of the different groups of animals.From the data, there was no significant difference, except the liver to body weight ratio, which was higher in treated rats in comparison with the control group.Another important fact is that, the relative weights of kidney and liver were constant during the experiment.

Biochemical analysis
Table 2 shows the serum content of electrolytic parameters (Na + , K + , Cl -), renal and hepatic function markers (BUN, CREA, AST and ALT) in experimental rat groups as compared to the control.The biochemical parameters of animals treated with AE, EE and ME did not show significant change of level of BUN, CREA, Na + , K + and Cl -in the serum even though increase of hepatic function marker levels was noticed in the serum.They appeared to be more pronounced (AST and ALT) for animals groups treated with AE and ME extract (p < 0.01).

Acute hypoglycemic effect
Considering the normoglyceamic rats experiment, treatment with P. americana extracts at respective concentrations of 100, 200 and 400 mg/kg did not significantly reduce the blood glucose level as compared to that of the control group.Yet, glibenclamide brought about significant reduction of the blood glucose level by 40% (data not shown).

Oral glucose tolerance test (OGTT)
Table 3 shows the blood glucose levels of experimental rats before and after D-glucose (2.0 g/kg, b.w.) administration by oral route.It appeared that glibenclamide induced significant (p < 0.001) reduction of hyperglyceamia within a time period of ½ to 5h.At 5 h, the reduction rate was about 61.5%, whereas no significant decrease was observed in the control group.
On the other side, AE induced significant (p < 0.01) glyceamia depletion within ½ h at a single dose of 100, 200 and 400 mg/kg, after D-glucose ingestion as compared to the control group and 100 mg/kg of AE brought the best reduction (5.6%).Furthermore, the EE (100 mg/kg) was responsible for a significant (p < 0.001) depletion in serum glucose level from ½ to 5 h.The reduction rates were 2.5, 36.8 and 59.6% after 1, 2 and 5 h, respectively.With regard to ME, it induced a significant decrease of hyperglyceamia between 1 and 3 h after glucose administration, except for the 200 mg/kg dose.

DISCUSSION
Persea americana Mill.or avocado plant tree is wellknown across both the south American and the sub-Saharan African continent (Adeboye et al., 1999;Ojewole et al., 2007).An ethnobotanical survey carried out in Côte d'Ivoire showed the use of the named plant to treat diabetes (N'guessan et al., 2009).Primarily, the present study led to a phytochemical investigation of three leaf extracts of the plant (aqueous, ethanolic and methanolic).
Secondly, the study also compared the acute toxicity of the plant extracts aforementioned as well as their hypoglyceamic and anti-hyperglyceamic activities.
The chemical investigation of the crude extracts of P. americana revealed the presence of bioactive compounds, namely alkaloids, flavonoids, tanins, coumarine, sterols/polyterpenoids, saponins and polyphenols.These results were in accordance with those reported previously (Adeboye et al., 1999;Adeyemi et al., 2002).The amount of total flavonoids determined in a hydro-ethanolic leaf extract of P. americana by Lima et al. (2012) (0.730 ± 0.005%) was higher than the one obtained in the present study (0.474 -0.582%).This may due to the solvent extraction method using the mixture water/ethanol.
From the acute toxicity evaluation performed according to the self-modified OCDE 423 protocol, on healthy Then, the obtained figures in this study best fit within the latter limit.Furthermore, gaining around 27g in weight by rat was observed after extracts uptake (Figure 2), even though there was no significant difference with the control group.The macroscopic autopsy of the vital organs (liver, heart and kidney) did not show any injury or color change, even though the ratio of organs to body weight index exhibited a slight increase of the treated rat's liver weight.This, according to literature, suggests that the leaf extracts used for the experiments did not show significant toxicity for the liver; consequently would do not exhibit any functional side effect on kidneys (Odo et al., 2014).The biochemical analyses carried out however revealed a significant increase (p < 0.01) in the amount of either AST or ALT, respectively for group AE and ME as compared to the control group.In this study, the ratio AST/ALT, which fell respectively at 2.35 and 1.29 for groups EE and ME, indicates an abnormality as to the hepatic function of the liver; thus, confirming the outcomes of the study performed (Ezejiofor et al., 2013).On the other side, these results displayed the leaves of P. americana to be of lesser toxicity than that of the ripe fruit's seeds extracts.The latter parameter falls within the range of 1200.75 and 1767 mg/kg for the mice investigated (Padilla-Camberos et al., 2013;Pahua-Ramos et al., 2012).All in all, it could be stated that the unique dose of 2000 mg/kg affects in a short term, the liver's metabolism rather than that of the kidney, as confirmed by the biochemical factors (urea, creatinine and blood ionogram).However, chronical toxicity survey needs to be carried out by applying the unique dose of 2000 mg/kg to ascertain its tolerance in the long term.This could be done by assessment of the liver's synthesis function (albumin and triglycerids), cell integrity (transaminase), hepato-biliary channels permeability (alkalin phosphate and bilirubine) and its malignancy (prothrombine time period) because the treatment of diabetes last for life.All in all, the fact that anti-hyperglyceamic effective doses are far from the LD 50 is encouraging.
The present study confirms the hypoglyceamic and antihyperglyceamic effect of the glibenclamide, which are mostly encountered with normal glyceamic rats.These effects are more pronounced than the extracts' doses administered in the course of our experiment.Previous investigation as related to hypoglyceamic and anti hyperglyceamic activity on normoglyceamic rats showed that if P. amaricana did not have any effect on basic glyceamia (fasting glyceamia), it would have significantly stopped the provoked hyperglyceamia.In fact, aqueous, ethanol and methanol extracts of P. americana exhibited an anti-hyperglyceamic activity of variable intensity and life span.
On the contrary, the three doses (100, 200 and 400 mg/kg) of the AE prevented prematurely postprandial or provoked hyperglyceamia after one hour of glucose administration.This suggests that the extracts were responsible for inhibited intestinal absorption of glucose.Consequently, the peak of hyperglyceamia was decreased in comparison with the control and glibenclamide.This phenomenon suggests that the AE could contain bioactive molecules capable of inhibiting αamylase and α-glycosidase.Nevertheless, these results need further investigations such as absorption and inhibition mechanisms of glucose.
The EE and ME also extended the anti-hyperglyceamic activity at a low dose of 100 mg/kg after 3 to 5 h of glucose administration.Therefore, this activity seems not to be dose-dependent and could be correlated with the polyphenol's content and especially with that of flavonoids.EE having significant content of polyphenols and total flavonoids (2952.7 ± 166 μgGAE/g and 0.582%, respectively), as compared to the ME (1873.1 ± 63.5 μgGAE/g and 0.474%), brought about long lasting antihyperglyceamic activity.In fact, the inhibition of intestinal absorption of glucose has already been demonstrated earlier on (Tadera et al., 2006), and seems to be similar to that of hypoglyceamic sulfamids like glibenclamide.
However, extension of the anti-hyperglyceamic activity requires other mechanisms that involve the liver, the pancreas and measles.Mastery of the mechanisms mentioned, led to further studies that identified polyphenols as glucose metabolism regulators; using several mechanisms such as, protection and recovery of β cells' integrity, stimulation of insulin secretion and increase of glucose uptake by cells (Solayman et al., 2016;Vinayagam et al., 2016).Studies carried out previously, displayed an important anti-oxidizing role of polyphenols.Especially, flavonoids exhibited a protecting effect against toxicity and oxydating stress.On the other hand, saponines also demonstrated an antihyperglyceamic effect due to inhibition of glycogenesis (Elekofehinti, 2015;Ezejiofor et al., 2013).
Finally, flavonoids, tannins and saponins also displayed anti-hyperglyceamic properties through inhibition of the sodium-glucose transporter 1 (S-Glut 1) according to Tiwari and Rao (2002).

Conclusion
The findings of the present study demonstrated the antihyperglyceamic potential of the leaf extracts (AE, EE an ME) of P. americana along with their lower toxicity.Consequently, it constitutes justification of the named plant as a medicinal herb.Additionally, diabetes mellitus is a chronic disease, further pharmacological investigations over a long period of time are needed to better assess plant's toxicity and elucidate mode of action.

Figure 2 .
Figure 2. Effect on body mass gain (g) in normal healthy female rats treated with single acute oral dose of 2000 mg/kg AE, EE and ME of P. americana leaves.Values indicate mean±SEM (n = 3; OECD 423); ANOVA followed by Newman-Keuls, p < 0.05.
, a village Kamagate et al. 691 located about 10 km from Abidjan (Côte d'Ivoire).The plant specimen was authenticated at the herbarium of the National Floristic Center of the Félix Houphouët Boigny University under the voucher number 8845 deposed at the national herbarium.It is also registered in the Integrated Taxonomic Information System under the number 18154 (ITIS, 2016).

Table 1 .
Organs to body weight ratio of the different groups of animals after 15 days of observation.

Table 2 .
Biochemical parameters of rats treated with acute oral dose of 2000 mg/kg of P. americana extracts after 15 days of observation.

Table 3 .
Effect of P. americana extracts on fasting blood glucose level (mg/dl) after oral load of D-glucose (2.0 g/kg) in normoglyceamic rats.