Vegetables have been one of the regularly recommended cheap solutions to the malnutrition or malnourishment challenge experienced worldwide (Bello et al. 2019). They play a very important role in our diet and are most readily available sources of carbohydrates, fats, proteins, vitamins, minerals and essential amino acids (Fasuyi 2006; Sharma & Kumar 2013). Many vegetables are mainly consumed for their nutritional value without much consideration for their safety and possible contamination with related lethal effects. Aside from the nutritional intake, they have the ability to synthesize several secondary metabolites of relatively complex structures possessing antioxidants (Yadav et al. 2013). These metabolites produce specific effects on the physiology of human beings and other organisms. Recent reports indicate that there is an inverse relationship between the dietary intake of antioxidant-rich foods and the incidence of human diseases (Latifou et al. 2016). Despite their advantages, several studies have established that some vegetable species are potentially toxic to humans and animals because plant chemical compounds are produced as part of the plant’s defence against being eaten by pests and herbivores or to gain an advantage over competing with other plants (Orech et al. 2005; Latifou et al. 2016) with the potential to be toxic to humans and livestock.

However, safe natural products through low or little toxic contents and wide safety margins would benefit society towards further utilisation (Ali-Abdel-Rahman et al. 2011; FAO/WHO 2002; Sanders et al. 2010). Gandhare, Kavimani and Rajkapoor (2013) reported that the purpose of toxicity testing is to provide an adequate database to make decisions concerning the toxicology properties of chemical and commercial products. Its assessment could guide the utilisation or condition of substances including natural products under short- and long-term effects and provides adequate information about the product (Gandhare, Kavimani & Rajkapoor 2013; Obidike & Salawu 2013).

Daucus carota L. commonly called carrot, belongs to the family Apiaceae (Pimenov & Leonov 2004). It is native to Western or near East Asia, Tropical Africa, North and South America (Heywood et al. 2008). The root is a rich source of vitamin A and B that can serve as antioxidants and can inhibit or reduce cardiovascular diseases (Dalvir et al. 2016). Carrot is an important vegetable crop that has contributed to human health (Shelly 2012). Ayeni et al. (2018) reported on the phytochemicals, nutraceuticals and antioxidants potentials of carrots aerial parts and its possible utilisation as vegetables. However, the aerial part of carrot is used as component in livestock feed in some parts of Northern Nigeria without information on its toxicity. The present study was designed to investigate the possible toxic effects of the crude extracts of D. carota aerial parts and to determine safe levels in laboratory rats.

The acute oral administration of up to 5000 mg/kg n-hexane, ethyl acetate and methanol extracts neither showed mortality nor any toxic signs in the experimental rats. The extracts could be assumed as practically safe and non-toxic. This result is in agreement with Olorunnisola et al. (2012), Parasuraman et al. (2014), Ugbogu et al. (2016) and Adesegun et al. (2016) that reported that the acute toxicity of plants could be considered practically non-toxic and safe above the oral administration of 5000 mg/kg.

Oral administration of the extracts (n-hexane and methanol) led to a statistically significant decrease in body weight at 500 mg/kg, 1000 mg/kg and 1500 mg/kg at week 2 when compared with the initial body weight of Wistar rats at week 0 (Table 1), while the ethyl acetate extracts had no significant difference on the body weight. This reduction was a dose-dependent decrease in their body weight and was probably caused by the presence of some phytoconstituents in the extracts that may have interfered with the rat’s appetite and food consumption. The body weight changes serve as a sensitive indication of the general health status of animals. Changes in body weight have been used as an indicator of adverse effects of drugs and chemicals (Grance et al. 2008; Nandy & Datta 2012).

The haematopoietic system is one of the sensitive targets for toxic compounds and an important index of the physiological and pathological status in man and animals. The analysis of blood parameters is relevant to risk evaluation as the changes in haematological parameters have a high predictive value for human toxicity when the data are translated from animal studies (Olson et al. 2000). The haematological parameters of the extracts of D. carota aerial part showed a statistically significant difference (p ≤ 0.05) on the RBC at 500 mg/kg, 1000 mg/kg and 1500 mg/kg when compared with the control group (Table 2). However, there was a mean decreased of the RBC on the extracts. This reduction of RBC could be the result of fatty acid or the invasion of foreign components like some classes of secondary metabolites present in the extracts that could interfere with the haematopoietic system, erythropoiesis, morphology, or osmotic fragility of the red blood cells because red blood cells are involved in the transportation of oxygen and carbon dioxide into the body (Guyton & Hall 2000; Isaac et al. 2013). Thus, a reduced red blood cell count implies a reduction in the level of oxygen that would be carried to the tissues as well as the level of carbon dioxide returned to the lungs (Isaac et al. 2013; Soetan, Akinrinde & Ajibade 2013). The extracts probably contained some classes of secondary metabolites that could interfere with the haematopoietic system. This result is comparable with the work of Guyton and Hall (2000) that reported a similar reduction in the red blood cells in some plant species. This probably showed that the extracts contained some haematopoietic-like principle that could reduce the red blood cell (Adeneye 2008; Palani et al. 2009; Wararut et al. 2012). However, there was no statistically significant difference (p ≥ 0.05) in the WBC, Hgb and HCT in the extracts; this probably did not affect the immune systems but might be susceptible to infection. However, the results of the haematological parameter are within the internationally accepted reference range for each haematological parameter (Daramola et al. 2005; Kifayatullah et al. 2015; NseAbasi et al. 2014; Research Animal Resources 2009; Uboh et al. 2012). Liver function tests involve evaluating serum ALT, AST, ALP, bilirubin and ALB levels. The most commonly used indicators of liver damage are ALT and AST (Ramaiah 2011). These are enzymes normally found in liver cells that leak out of these cells into the bloodstream following an injury to the liver cells. These liver enzymes become elevated in liver cirrhosis, hepatitis, and hepatobiliary obstruction. ALT is localised primarily in the cytosol of hepatocytes and is considered to be a more sensitive marker of hepatocellular damage than AST (Saad et al. 2006). It can provide a quantitative assessment of the degree of damage sustained by the liver (Al-Mamary et al. 2002).

In this study, the liver function test (Table 3) showed a statistically significant decrease at (p ≤ 0.05) on the ALP of the extracts, but there was no statistically significant difference in the ALT, AST, TP and ALB. The significant decrease in the ALP may be indicative of liver injury in the animals.

Urea and creatinine are considered as good prognostic indicators of renal dysfunction and kidney failure for any toxic compound (Ashour, Yassin & Abu 2006; Gnamani et al. 2008). As the kidneys become impaired, the creatinine level in the blood rises because of damage to the functional nephrons and consequently result in poor clearance by the kidneys (Aliyu et al. 2006). In the present study, the kidney function test (Table 4) of the extracts showed that there was no statistically significant difference at p ≤ 0.05 when compared with the control group. Serum creatinine, urea and electrolytes such as sodium and calcium were not affected by the treatment of the extract. However, there was a statistically significant increase of bicarbonate (mmol/L), which was dose-dependent in the extracts when compared with the control group. This increase may have influenced the regulation in the levels of other kidney parameters (urea, potassium, creatinine and chloride). This is an indication that the D. carota aerial part may not be relatively safe for consumption. Histopathological examination is the gold standard for evaluating treatment-related pathological changes in tissues and organs (OECD 2008). The histopathological effects (Figures 1–6) of the extracts at 500 mg/kg showed moderate and slight tubular necrosis in the kidney, while the liver had slight hepatic necrosis with moderate lymphocyte hyperplasia in all of the extracts. Also, 1000 mg/kg and 1500 mg/kg of the extracts showed moderate lymphocyte hyperplasia in the kidney, while vascular congestion with Kupfer cell hyperplasia was observed in the liver. The 500 mg/kg dose of the extracts showed that the lowest dose produces a slightly detectable toxic reaction which was also reported by Leclair and Willard (1970). Because the toxicity target organs are mainly the liver and kidney, the alteration may cause liver and kidney diseases when consumed even at low doses (Okunlola et al. 2012; Soetan et al. 2013).

The sub-acute toxicity study showed that the aerial parts of the D. carota are not safe for consumption when consumed as a livestock feed for animals or vegetables for man. The doses are capable of causing significant liver and kidney injury and may also have an adverse effect on the haematopoietic system. The study suggested further investigation on the chronic toxicity and isolation of the toxic components in the aerial part.