Today, medicinal plants are renowned for their therapeutic potential. They are increasingly used in our communities as a source of primary health care. The use of natural substances derived from plants is therefore becoming an obvious solution. The toxicity of medicinal plants needs to be studied.

The toxicity of a substance is its capacity to induce adverse morphological or functional effects in a living organism following administration of a single high dose, or a small dose repeated over a long period. The study of the toxicity of a medicinal plant is the set of pharmacological tests that determine the degree or harmfulness of the plant in order to regulate its use. Toxicity is assessed according to mode of administration, dose, weight variation, mortality rate and variation in biochemical parameters [1] - [3] .

A plant-based recipe is used in Togolese traditional medicine to treat typhoid fever [4] . The use of this recipe for therapeutic purposes in no way guarantees its safety. Thus, in addition to the pharmacological effects obtained, a toxicological study must be carried out to determine its safety.

The aim of this study is to evaluate the acute and sub-acute toxicity and cytotoxicity of a recipe based on three plants used in the treatment of typhoid fever in Togo (Carica papaya L., Cocos nucifera L. and Persea americana M.)

After 14 days of observation ( Figure 1 ), we recorded no deaths or dangerous signs following oral administration of the recipe at a dose of 5000 mg/kg body weight. According to the Globally Harmonized System of Classification and Labelling of Chemicals [6] , the hydroethanol extract of the recipe is non-toxic by the oral route in Wistar rats. This suggests that the LD50 is greater than 5000 mg/kg. From this result, it appears that the recipe can be considered not to cause acute toxicity.

Treated male and female rats received doses of 500 and 1000 mg/kg body weight for 28 days ( Figure 2 and Figure 3 ), while control rats received physiological water only. During this period, no deaths, dangerous signs, changes in locomotor activity, or food and water consumption were recorded. All rats gained weight. Weight gain was slightly higher in treated rats than in control rats after 28 days.

The various organs—liver, kidney, testes, lungs, heart, ovaries and spleen—were sampled, weighed and compared with those of control rats. The results showed that there was no significant difference between the various organs of treated rats compared with controls, except in the liver and lungs of female rats at a dose of 1000 mg/kg body weight. ( Table 1 )

***P < 0.001.

As with the organs, biochemical parameters ( Table 2 ) were measured in the different batches of rats studied. A dose-dependent decrease in creatinemia compared with the control was observed in male rats.

**P < 0.01. Gly: Blood glucose, Creat: creatinemia, ASAT: Aspartate amino-transferase, ALAT: Alanine aminotransferase, GGT: Gammaglutamyl transferase, ALP: Alkaline phosphatase, Bil T: Total bilirubin, Bil D: bilirubin direct, CHOT: total cholesterol, Trig: triglycerides.

Blood cells are called upon when a foreign agent enters the body. Toxicity may induce a change in the number, shape or size of these cells. To this end, we performed blood counts in rats from the three (3) batches studied. No significant differences were observed between treated and control rats, regardless of sex. ( Table 3 )

GB: white blood cells, Lymp: Lymphocyte, Gran: Granulocytes, HGB: Hemoglobin, HCT: hematocrit, GVM: mean corpuscular volume, TMH: mean corpuscular hemoglobin content, CCMH: mean corpuscular hemoglobin concentration, PLT: Blood Platelets.

Cytotoxicity was assessed in vitro on normal NCM 365 colon cells ( Figure 4 ) using the 3-(4,5-dimethylthiazolyl-2-yl-2,5-diphenyltetrazolium bromide (MTT) assay. The results show that the recipe is non-toxic to normal colon cells at a concentration of 100 µg/ml, with a crosis percentage of 9.41% according to NF EN ISO 10993-5 classification. The cytotoxic concentration (CC50) is 59,224 µg/ml or 59.22 mg/ml.

The aim of our work is to evaluate the acute and subacute toxicity and cytotoxicity of a recipe of 3 plants (Carica papaya L., Cocos nucifera L., Persea americana M.) used in traditional medicine to treat typhoid fever.

The pharmacological properties of medicinal plants have led us to evaluate the toxicity of the recipe.

As the extract’s action on pathogens can cause damage to consumers’ organs, it is appropriate to carry out a toxicological study of our recipe. With this in mind, we assessed the acute and sub-acute toxicity and cytotoxicity of our recipe. The acute toxicity results indicate that the LD50 is greater than 5000 mg/kg body weight. According to [7] , any substance with an LD50 greater than 5 g/kg can be considered non-toxic, so the recipe made from Carica papaya roots, Cocos nucifera and Persea americana leaves is not acutely toxic. According to [8] , the limit test is carried out only when there is sufficient evidence that the test substance is not likely to be toxic or is of low toxicity. Consequently, [9] - [11] , obtained LD50 values at 2000 mg/kg and 5000 mg/kg using Carica papaya roots, seeds and leaves respectively. On the other hand, [12] [13] showed that Cocos nucifera roots are not toxic at a concentration of 2000 mg/kg and [14] showed that Persea americana leaves are not toxic at a dose of 2000 mg/kg.

The hydroalcoholic extract showed no sign of significant toxicity for 28 days in the rats studied. The change in body weight is a sensitive indication of the state of health of the animals [15] [16] . Thus, the increase in liver and lung weights in female rats could be explained by weight gain in female rats treated at 1000 mg/kg.

Analysis of blood parameters is important for toxicity risk assessment, as any changes in biochemical and haematological systems have a higher predictive value for toxicity [17] [18] . The absence of significant changes in transaminases and renal function tests suggests that the recipe did not cause liver or kidney damage. Transaminases are enzymes that catalyze the transfer of alpha-amino radicals from alanine and aspartic acid to alpha-ketoglutaric acid [16] . They are synthesized in the cytoplasm and released into the general circulation when cells are damaged [1] [19] [20] . In addition, the dose-dependent reduction in creatinine levels in treated male rats may allow the recipe to be used to moderate creatinine levels in males. The absence of significant variations in haematological parameters suggests that the recipe did not induce a toxic effect on bone marrow, considered to be one of the most sensitive targets of toxic compounds. Moreover, any changes in the haematological systems of treated rats are more predictive of toxicity in humans than in rats when extrapolated from animal study data [17] . In vitro toxicity on normal NCM 365 colon cells showed no cell necrosis.

The results obtained would suggest that plants taken in isolation did not show significant toxicity at repeated dose for 28 days, so [21] [22] , did not observe mortality or abnormalities in liver and kidney markers using Carica papaya leaves similarly [9] [12] used the immature inflorescence of cocos nucifera for 28 days of exposure at a dose ranging from 1.75 to 14 mg/kg body weight, and lastly [23] , obtained no significant signs of toxicity using Persea americana seeds. On the other hand, [24] obtained liver degeneration following administration of 750 mg/kg of Persea americana aqueous extract.

The recipe made from Carica papaya L., Cocos nucifera L. and Persea americana M. is used in traditional medicine to treat typhoid fever. Our study assessed the recipe’s acute and subacute toxicity and cytotoxicity. At the end of the study, it emerged that the recipe, prepared using the traditional method, did not induce any significant toxicity. The results obtained justify the use of this recipe in traditional medicine for the treatment of typhoid fever. These results could constitute a database that could guide the development of new drugs for the treatment of typhoid fever.