LEVELS OF MAJOR AND TRACE METALS IN THE LEAVES AND INFUSIONS OF CROTON MACROSTACHYUS

The levels of essential metals (Ca, Mg, Fe, Mn, Zn, Cu, Co, Cr, Ni) and toxic metals (Cd, Pb) were determined in the leaves of Croton macrostachyus (traditional medicinal plant) collected from four different regions of Ethiopia (Akaki, Abomsa, Bonga and Dilla) and also in the infusions of leaves collected from Akaki using flame atomic absorption spectrometry. An optimized digestion procedure took 2.5 h for the digestion of 0.5 g of powdered sample with 2 mL of HNO3 and 2 mL of HClO4 at 270 C while 2 h were needed to digest 25 mL infusion with 4 mL of HNO3 and 1 mL HClO4. The recoveries of metals were in the range 92–103% for the leaves powder and 94–105% for the infusion samples. The mean concentration ranges (g/g) were Ca (5,823–12,040), Mg (1,971–4,961), Fe (192–581), Mn (157–1,770), Zn (19.5–60.5), Cu (6.31–18.6), Co (1.97–3.45), Cr (2.13– 8.75), Ni (2.15–3.80), Cd (0.75–1.08) and Pb (1.05–2.19) for the leave powders and Ca (716–1,776), Mg (16.7– 80.9), Fe (1.39–3.34), Mn (2.17–3.40), Zn (0.674–7.88), Cu (1.94–2.31), Co (0.157–0.224), Cr (0.144–1.23), Ni (0.203–0.267), Cd (0.05–0.098) and Pb (0.0148–0.185) for the infusion samples. Among the quantified metals in the infusion samples, Ca showed the highest leaching rate (61.5%) in the 24 h infusion while Fe exhibit the lowest (1%) leaching rate in the 3 h infusion.


INTRODUCTION
The use of plants as treatment of diseases and as food dates beyond recorded history perhaps as old as the history of mankind. Old civilisations in China, India, Egypt and Greece had a rich knowledge and expertise in using many types of plants. It has been estimated that more than 80% of the world population utilizes plants as primary source of medicinal agents, largely due to the high cost of Western pharmaceuticals, but also because the traditional medicines are generally more acceptable from a cultural and spiritual perspective. Even in the Western world, the use of herbal medicines is steadily growing with approximately 40% of the population reporting use of herbs to treat medical illness [1,2].
It is known that many countries in Africa, Asia and Latin America use traditional medicine to meet some of their primary health care needs. In Africa, up to 80% of the population uses traditional medicine for primary health care. Traditional medicine has maintained its popularity in all regions of the developing world and its use is rapidly spreading in the industrialized countries. In China, for example, traditional herbal preparations account for 3050% of the total medicinal consumption. In Ghana, Mali, Nigeria and Zambia, the first line of treatment for 60% of children with high fever resulting from malaria is the use of herbal medicines at home [3].
Many of the plants species used for this purpose have been found to contain therapeutic substances which can be extracted and used in preparation of drugs, but the plant itself can also be used either directly or as an extract for medication, a practice that is particularly popular in developing countries [47].
The literature survey revealed that there is no report on the levels of metals in leaves of C. macrostachyus. Thus the objectives of this study were: (i) to determine major and trace essential (Ca, Mg, Fe, Mn, Zn, Cu, Co, Ni, Cr) and toxic metals (Cd, Pb) in leaves of C. macrostachyus and its infusion grown in Ethiopia by flame atomic absorption spectroscopy, (ii) to compare the levels of essential (Ca, Mg, Fe, Mn, Zn, Cu, Co, Ni, Cr) and toxic (Cd, Pb) metals in the leaves and infusions of C. macrostachyus, (iii) to estimate the amount of elements released from plants into the water extracts, and (iv) to look some correlation trends among the element concentrations in the different area of study by applying simple statistical tools.

Apparatus and instrument
Teflon (PTFE, China) knives were used to cut the C. macrostachyus leaves in to pieces and an oven (Digitheat, J.P. Selecta, Spain) was used for drying the samples placed on porcelain dish. Blending device (Moulex, France) and ceramic pestle and mortar were used for grinding and homogenizing the C. macrostachyus leaves samples. C. macrostachyus samples were digested on Kjeldahl heating apparatus (Gallenkamp, England) using round bottom flasks with ground glass joint (100 mL) fitted with reflux condenser. Borosilicate volumetric flasks (25,50 and 100 mL), pipettes (Pyrex, USA), measuring cylinders (Duran, Germany), and micropipettes (Dragonmed, 1−10 µL, 100−1000 µL, Shanghai, China) were used for measuring different quantities of sample solution, acid reagents and metal standard solutions. Metal concentrations were determined by flame atomic absorption spectrophotometer (Buck Scientific, Model 210VGP AAS, East Norwalk, USA) equipped with deuterium background corrector and hollow cathode lamps with air-acetylene flame.

Chemicals and reagents
All the reagents were of analytical grade. 6970% HNO 3 (Supreme Enterprises Cantt, India) and 70% HClO 4 (Aldrich, UK) were used for the digestion of the C. macrostachyus leaves and infusion samples. Lanthanum nitrate hydrate, (99.9%, Aldrich, USA) was used to prevent the chemical interference on Ca and Mg in the sample solution during the analysis of C. macrostachyus samples. Stock standard solution of concentration 1000 mg/L in 2% HNO 3 of the metals Ca, Mg, Mn, Fe, Zn, Cu, Zn, Cr, Ni, Co, Cd and Pb (Buck Scientific Puro-Graphic tm ) from which 10 mg/L of intermediate standard obtained were used for the preparation of the calibration standards of each metal. Working standards were prepared from intermediate standards of each metal. Deionized water was used for sample preparation, dilution, and rinsing apparatus prior to analysis.

Description of the study area
The samples were collected from the selected sites of four different regions namely: (i) Abomsa, a town in central Ethiopia  located in the Arsi Zone of the Oromia region, administrative center of Merti Woreda, (ii) Dilla, a market town in southern Ethiopia  the administrative center of the Gedeo Zone in the Southern Nations, Nationalities and People Region (SNNPR), and located on the main road from Addis Ababa to Nairobi, (iii) Akaki, one of the outskirt sub-cities of Addis Ababa and (iv) Bonga, the administrative center of Kefa zone in the Southern Nations, Nationalities and People Regional state. The selection of these sites was based on: firstly, availability of ethnobotanical information, secondly, the need to compare metal accumulation in this plant from industrially and agriculturally polluted areas with that of none polluted environment; thirdly, availability of the plant and its popularity in using as medicinal plant by the local peoples around. The geographical location of the sampling sites and their distance from Addis Ababa (capital city of Ethiopia) are given in Table 1. Table 1. The geographical location of the sampling sites and their distance from Addis Ababa (capital city of Ethiopia).

Sample collection and treatment
Representative amount of leaves were collected from four different regions. The leaves were collected from a minimum of 16 mature plants, four plants per region and eight leaves per plant starting from the bottom to the tips by stalk position. The C. macrostachyus samples were put in clean plastic bags labeled according to their region and brought to the laboratory for further pretreatment. The leaves of the C. macrostachyus were separated with Teflon knives, washed with a running tap water so as to remove adsorbed soil particulates and then rinsed with distilled deionized water and air dried. About 300 g of the C. macrostachyus leaves from each region were put on acid-washed and labeled porcelain dish and dried in an oven at 80 o C for 48 h till it got brittle and crisp. After cooling to ambient temperature, the dried samples were ground in to fine powder with blending device, mortar and pestle, and sieved (1 mm). The powdered sample was then placed in pre-cleaned screw capped polyethylene container and stored in desiccators with calcium chloride to keep to constant dry weight till digestion. The C. macrostachyus leaves infusion was made from C. macrostachyus leaves collected from Akaki site only. By varying the time, four C. macrostachyus leaves infusion samples were prepared. As there is no specific method of preparing the C. macrostachyus leaves infusion in Ethiopia, usually 25 g of fresh leaves (8 to 14 leaves) were ground in the mortar with a pestle and were added in to a 250 mL tap water (at room temperature, 25 o C). This was done on the habits of traditional medician, i.e. they keep the leaves in the water for half a day or for complete day depending on the ages of the patient. So, in this study, to compare and consider the amount of metals extracted in to the supernatant, the infusions were prepared for 3, 6, 12 and 24 hours. Finally, pure infusion sample was obtained by filtering and the decrease in volume was compensated by adding water up to the mark.

Optimization of digestion procedure
Before applying the digestion procedure for the analysis of samples, digestion time, reagent volume ratio and temperature were optimized by varying one parameter at a time for the digestion of C. macrostachyus leaves powder and its infusion. Thus, for a complete digestion of 0.5 g of the dry C. macrostachyus leaves sample, 2 mL HNO 3 (69-72%) and 2 mL HClO 4 (70%) for a total of 2.5 h at 270 o C was selected as optimized procedure while 1 mL HClO 4 (70%) and 4 mL of HNO 3 (69-72%) for a total of 2 h at 240 o C for the digestion of C. macrostachyus leaves infusion samples.

Digestion of the C. macrostachyus leaves and its infusion
Applying the optimized procedure, exactly 0.5 g of the dried and ground C. macrostachyus leaves sample was accurately weighed and transferred quantitatively in to a 100 mL round bottom digestion flask. To the C. macrostachyus leaves sample, 4 mL of freshly prepared 1:1 mixture of 70% of conc. HNO 3 and 70% of conc. HClO 4 was added. The mixture was then heated on Kjeldahl heating apparatus fitting the flask to a reflux condenser for the first 30 min by setting the temperature at 120 o C and followed by heating at 270 o C as maximum temperature for the other 2.0 h. The digested solutions were allowed to cool for 20 min without dismantling the condenser from the flask and for an additional 10 min after removing the condenser. To the cooled solution, two 5 mL of de-ionized water was added and the solution was slowly shacked to dissolve precipitates remained on the wall of the flask and to minimize dissolution of filter paper by residue. Subsequent rinsing the flask with 5 mL portions of distilled-deionized water followed until the volume reached the mark. For each bulk samples, triplicate digestions were carried out. The digested and diluted sample solutions were stored in tightly capped polyethylene bottles and kept in refrigerator until analysis time.
To determine the amount of metals extracted from the C. macrostachyus leaves to the leaves infusion (to the water at room temperature, 25 o C) 25 mL of the clear infusion (after filtration) was transferred quantitatively to the 100 mL round bottom flask and heated on the hot plate to evaporate until the sample was decreased to around 3 mL. This was done to avoid dilution of the acid mixture added for digestion. The solution was allowed to cool before the addition of the acid mixture to avoid potentially explosive reactions that may occur. A 5 mL portion of 4:1 HNO 3 and HClO 4 was added and the contents swirled gently to homogenize then fitted to a reflux condenser and digested continuously for two hours on a Kjeldahl digestion block. The temperature was adjusted at 120 o C for the first 30 min and finally to the maximum temperature of 240 o C for 1.5 h. The contents were diluted to 50 mL as described above. The digestion gave a clear colorless solution. Each C. macrostachyus leaves infusion sample was digested in triplicate and hence a total of twelve digests were made for C. macrostachyus leaves infusion samples prepared by infusing for 3, 6, 12, 24 hours. All the final digest solutions were kept in a refrigerator until the analysis for the metals.
Digestion of a reagent blank was also performed in parallel with the C. macrostachyus samples keeping all digestion parameters the same. For the analysis of the C. macrostachyus samples 9 reagent blank samples were prepared. All the digested samples were stored in refrigerator until analysis. The solutions were used to determine concentration of the 11 elements by flame atomic absorption spectrometry.

Instrument calibration and method detection limit
For the present study, the pooled standard deviation of the 9 reagent blanks was calculated. The detection limits were obtained by multiplying the pooled standard deviation of the reagent blank by three. The wavelengths at which analysis was done, the correlation coefficients of the calibration curve for each of the metals and method detection limits of each metal are given in Table 2. The results clearly show that the calibration curves with good correlation coefficients and lower method detection limits were obtained during the analysis.

Precision
The precision of the results of the present analysis is reported with corresponding pooled standard deviation of nine measurements for a bulk sample of triplicate reading per sample and relative standard deviation.

Recovery test
The efficiency of the optimized procedure was checked by adding known concentration of each metal to 0.5 g sample. The procedure was as follows: 783 µg of Ca and 0.6 µg of Co were spiked at once in to 0.5 g of C. macrostachyus leave powder sample and the remaining metals Since the concentration of metal in the C. macrostachyus leaves infusion is lower than that of the C. macrostachyus leaves powder, the recovery test for the C. macrostachyus leaves infusion was done by adding smaller amount of the spiked metals than that of the C. macrostachyus leaves powder. Hence, for spiking C. macrostachyus leave infusion sample, 183 µg of Ca, 0.31 µg of Cu, 0.24 µg of Co, and 0.73 µg of Cr were spiked from stock solution of each metal per gram of the sample. The remaining metals (8.50 µg of Mg, 0.98 µg of Fe, 0.34 µg of Mn, 0.17 µg of Ni, 0.11 µg of Cd and 0.16 µg of Pb) were spiked at once in to per gram of the infusion sample and 2.85 µg of Zn was spiked in another flask containing the same amount of infusion sample and the same digestion process was followed. Each sample was analyzed for their respective spiked metals by atomic absorption spectrophotometer. Each recovery test for both samples was performed in triplicates.

Conversion factor determination
To report results obtained on a dry weight basis in terms of fresh weight basis, a conversion factor was calculated by taking the weight of a known quantity of fresh C. macrostachyus leaves before and after drying to a constant weight. This procedure was repeated for all the analysed plant leaves from different areas. Thus 4.17, the mean value was taken as a conversion factor because 4.17 g of the wet sample (on an average) gave 1 g of dry sample.

RESULTS AND DISCUSSION
In this study, the method was validated by the spiking experiment in which known quantities of the metals standard solution were added to three samples for leaves and infusion samples. The spiked samples were then digested using the optimized procedure and calculated the percent recoveries. The percentage recovery of metals in C. macrostachyus leaves powder and its infusion samples were between 92% and 105% (100 ± 10), which are within the acceptable range. The results are presented in Tables 3 and 4.

Levels of major, trace and toxic metals
The accuracy and precision of the method were tested by spiking the samples with a standard of known concentration of the analyte metals. The results indicated that the concentrations of elements determined by the present FAAS method are in agreement (100 ± 10%) within the acceptable range for all metals. Thus, in the present work, the concentration of the eleven metals, essential (Ca, Mg, Mn, Fe, Cu, Zn, Co, Cr, Ni) and toxic (Pb, Cd) in the C. macrostachyus leaves and their infusions were determined by FAAS. The results showed that the samples had variable composition of each analyte metals with wide concentration range. Among the identified eleven elements, all were found to be above the method detection limit except Cd not found in the samples from Dilla and Bonga sites in the C. macrostachyus leaves powder. On the other hand most metals were not detected in the infusion samples prepared for three and six hours. The results are given in Table 5 and 6.

Distribution of metals in different C. macrostachyus samples
Plants accumulate minerals essential for their growth from the environment and can also accumulate metals such as Cd, Co, Ag, and Na, which have no known direct benefit to the plant [31]. Therefore, the essential and non-essential metals in C. macrostachyus leaves and their infusion may be affected by different factors. Physical and chemical properties of the soil, taking medicinal samples from the farm areas (where application of natural and artificial fertilizers taking place), increasing industrialization and associated pollution of the biosphere, storage and processing of C. macrostachyus samples, climatic condition of the region and other factors are the main contributors for the mineral contents of C. macrostachyus leaves and their infusion. Table 5 shows the mean concentrations of various metals, in the same C. macrostachyus species plant in different part of the country. From the study, it was revealed that all the metals were accumulated to greater or lesser extents by plant leave studied. Table 5. Mean concentration (X ± SD, n = 9, μg/g dry weight) of essential and toxic metals in C. macrostachyus leaves samples.

Concentration of metals in C. macrostachyus leaves powder
From Table 5, it is clear that Ca was present in all the samples collected from four different regions with concentration ranging from 5,823 to 12,040 g/g. Among all these regions the highest concentration of Ca was found in Akaki C. macrostachyus leave powder and the lowest in Bonga C. macrostachyus leave powder. The concentration of Ca found in Abomsa and Dilla C. macrostachyus leaves powder were 7,976 and 8,202 g/g. The higher concentrations of Ca are very significant because Ca is known to enhance the qualities of bones and teeth and also of neuromuscular systemic and cardiac functions.
The Mg concentration varied from 1,971 to 4,961 g/g. Dilla C. macrostachyus leaves had the lowest Mg concentration (1,971 g/g) while Akaki C. macrostachyus leaves had the highest (4,961 g/g). The Mg concentration of Abomsa and Bonga C. macrostachyus leaves were 2,361 and 2,384 g/g, respectively. Like Ca, Mg is critical to many cell functions.
The abundance of Ca and Mg in the C. macrostachyus leaves is also in agreement with the previous studies on other plants, which indicated that these two elements were the most abundant elements in many medicinal plants and tea leaves [32,33].
The Fe concentration varied in the range from 192 g/g in Dilla C. macrostachyus leaves to 581 g/g in Akaki C. macrostachyus leaves. The concentration of Fe found in Abomsa and Bonga C. macrostachyus leaves powder were 329 and 269 g/g, respectively. The plant in all the four regions accumulates Fe above the limits proposed by FAO/WHO in edible plants (20 g/g). However, for medicinal plants the WHO limits not yet been established for Fe. Sheded et al. [34] reported that the range of Fe in their study was between 261 to 1,239 g/g in selective medicinal plants of Egypt, thus the present result is in good agreement with this range.
Coming to the case of Mn, an appreciable amount of Mn was present in all the areas of the study, its concentration is maximum in Bonga C. macrostachyus leaves 1,770 g/g and  [35]. From Table 5, it is clear that Zn was present in all the medicinal plants with concentration ranging from 19.7 to 60.5 g/g. Among all these study areas the highest concentration of Zn is found in Akaki C. macrostachyus leaves powder (60.5 g/g) and the lowest concentration is found in Dilla C. macrostachyus leaves powder (19.7 g/g). The concentration of Zn found in Abomsa and Bonga C. macrostachyus leaves powder are comparable, 34.5 and 33.6 g/g, respectively. The permissible zinc limit set by FAO/WHO in edible plants was 27.4 g/g. After comparison, metal limit in the studied medicinal plant with those proposed by FAO/WHO it is found that only Dilla C. macrostachyus leave is within this limit while all others study area accumulate Zn above this limit. However, for medicinal plants the WHO limits not yet been established for Zn. According to Bowen and Allaway, the range of Zn in agricultural products was between 15 to 200 g/g [36].
Copper was detected in all samples with varying concentrations. Leaves powder of C. macrostachyus from Akaki contain the highest amount (18.6 g/g) of Cu while least concentration was exhibited in the leaves powder of C. macrostachyus from Dilla (6.31 g/g). Abomsa and Bonga C. macrostachyus leaves powder contains comparable amount of Cu, 8.20 and 7.75 g/g, respectively. Although the results are not in the permissible limit of Cu set by FAO/WHO in edible plants 3.0 g/g, the results obtained from our medicinal plant are in the permissible limit of Cu set by China and Singapore in medicinal plants, which were 20 and 150 g/g, respectively. Reddy and Reddy [37] also reported that the range of Cu contents in the 50 medicinally important leafy material growing in India were 17.6 to 57.3 g/g.
The Co content was observed to be the highest, 3.45 μg/g in Abomsa C. macrostachyus leaves powder and relatively comparable in Bonga C. macrostachyus leaves powder (3.36 μg/g). The concentration of Co found in Akaki and Dilla C. macrostachyus leaves powder are 2.18 μg/g and 1.97 μg/g, respectively. There are no established criteria for Co in medicinal plants. Basgel and Erdemoglu [38] determined Co concentration ranged between 0.14 to 0.48 g/g in seven herbs in Turkey.
Another essential element, Cr, acts as a co-factor in the insulin synthesis and in the cholesterol and blood triglycerides control. The Cr content was observed to be the highest, 8.75 μg/g in Akaki C. macrostachyus leaves powder and relatively comparable in Abomsa C. macrostachyus leaves powder (8.15 μg/g). The concentration of Cr found in Bonga and Dilla C. macrostachyus leaves powder are 2.13 and 6.2 μg/g, respectively.
Abomsa C. macrostachyus leave accumulated lowest Ni that is, 2.15 μg/g and Dilla C. macrostachyus leave accumulate maximum that is 3.80 μg/g. The concentration of Ni found in Bonga and Akaki C. macrostachyus leaves powder were 3.16 and 2.62 μg/g, respectively. The permissible limit set by FAO/WHO in edible plants was 1.63 g/g. However, for medicinal plants the WHO limits not yet been established for Ni. Ni toxicity in human is not a very common occurrence because its absorption by the body is very low [39].
It is well known that Pb and Cd are toxic metals because of their effect in disordering the biological activity once taken into the body. In this study, Cd was not detected in Dilla and Bonga C. macrostachyus leaves powder. Elevated Cd concentrations were determined in Abomsa C. macrostachyus leave 0.75 μg/g next to Akaki C. macrostachyus leave 1.08 μg/g. The permissible limit for Cd set by FAO/WHO in edible plants was 0.2 g/g. However, for medicinal plants the permissible limit for Cd set by WHO, China and Thailand was 0.3 g/g. The relatively high concentrations of Cd found in the C. macrostachyus samples are certainly due to industrial pollution as well as the addition of some fertilizers and herbicides on the farm and farm sides.
The content of Pb in the C. macrostachyus leaves analyzed varied between 1.05 and 2.19 μg/g which is lower than the permissible limit for medicinal plant set by China, Malaysia, Thailand and WHO (10 mg kg -1 ). The maximum concentrations of Pb were determined in Bonga C. macrostachyus leaves (2.07 μg/g) next to Akaki C. macrostachyus leaves (2.19 μg/g). The concentration of Pb found in Abomsa and Dilla C. macrostachyus leaves were 1.15 and 1.05 μg/g, respectively.
Most of the studied sites showed more or less comparable pattern in mineral accumulation in the plant. Thus the following pattern in decreasing order was notice: Ca > Mg > Fe > Mn > Zn > Cu > Cr > Ni > Pb > Co > Cd.

Concentration of metals in a 3, 6, 12 and 24 hours infusion
As shown in Table 6, the trend of average concentration of metals in the 24 h infusion was Ca > Mg > Zn > Mn > Fe > Cu > Cr > Ni > Co > Pb > Cd. Similarly, the pattern of the average concentration of metals in the 12 h infusion was Ca > Mg > Fe > Zn > Mn > Cu > Cr > Ni > Co > Pb > Cd. Ca was appreciably extracted with the amount, 1,776 µg/g followed by Mg, 80.9 µg/g in the 24 hour infusion sample. Among the trace metals, Zn (7.88 g/g) was leached in larger amount followed by Mn, Fe and Cu with concentrations (3.40, 3.20 and 2.31 g/g), respectively. Other trace metals, Cr, Ni and Co were also extracted in lower concentration in the 24 h and 12 h of extraction time. The lower concentrations of these metals are most likely due to the small intake of these metals by the plant and also due to their low extraction efficiency in the water as these transition metals are known to form stable complexes with organic molecules present in the plant leaves. The toxic heavy metals Cd and Pb were also detected in the 24 hour infusion sample. The amount of Cd and Pb detected in this infusion sample were 0.098 g/g and 0.185 g/g, respectively. Table 6. Mean concentration (X ± SD, n = 9, μg/g wet weight) of essential and toxic metals in infusion samples. The percentage of leaching of the metals was also calculated and results are shown in Figure  1. As can be seen from the figure, each of the metal was extracted from the C. macrostachyus leaves to the water at different rate. For all the metals studied, maximum percentage extraction was observed at 24 h of infusion time followed by the subsequent lower extraction times. This highest leaching rate is appreciable for the essential metals. Whereas, for some of the trace metals and toxic elements, rate of extraction was more pronounced with infusion time. For instance, Cd was not detected in the infusion time of 3 and 6 h, while 2.80% and 15.5% was leached for Pb for the extraction time of 3 and 6 h, respectively. The extraction efficiencies for these two metals proportionally increased with time of infusion. Thus we recommend to the traditional practitioners to infuse their samples at lower period of time as possible so as to minimize the risk of these two heavy metal toxicity. Generally, for most of the metals investigated their concentration in the extract increase with prolonged period of infusion time, i.e. highest concentration were noticed in the 24 h followed by 12, 6 and 3 h. This might be due to the fact that the solvent might get the opportunity to penetrate the matrix and solvate the ions and hence easily leached in to the filtrate. Furthermore, wide variations in extent of extraction were also noticed among metals (Table 6, Figure 1). Range of metals extracted with time for Akaki leaves sample is given in Table 7.

Comparison of metals in the C. macrostachyus leaves powder samples
The concentrations of the elements in the analyzed samples were quite varied. The order of the metal concentration from Akaki C. macrostachyus leaves powder and Abomsa C. macrostachyus leaves powder is almost similar except that the Akaki C. macrostachyus leaves powder with lower mean concentration of Co than Ni and Pb. The trend in metal concentration for that of the Bonga and Dilla C. macrostachyus leaves powder were the same for Ca, Mg and Fe concentration. However, the concentration of Fe was greater than that of Zn and the concentration Cr were lower than the concentration of Ni which were not true for Dilla C. macrostachyus leave powder. Ca was found at highest concentration than other metals in all the four sample sites, with the mean concentration ranging from 5,823 g/g (Bonga C. macrostachyus leaves powder) to 12,040 g/g (Akaki C. macrostachyus leaves powder). Similarly, Mg was the next highest than other metals all the sites. The concentration of Fe was the third next to Mg but Mn was the next highest than the other metals for Bonga (1,770 g/g) and Dilla (421 g/g) sites.
Zn was found to be the third highest amount out of the microelements analyzed with mean concentration ranging from 19.7 g/g (Dilla C. macrostachyus leaves powder) to 60.5 g/g (Akaki C. macrostachyus leaves powder) followed by Cu with mean concentration range 6.31 g/g (Dilla C. macrostachyus leaves powder) to 18.6 g/g (Akaki C. macrostachyus leaves powder). The lowest concentration observed for all the four sites was that of Cd, ranging from 0.75 g/g (Abomsa C. macrostachyus leaves powder) to 1.08 g/g (Akaki C. macrostachyus leaves powder). Overall, the concentrations of each metal in the four sites are comparable.
Generally, Akaki C. macrostachyus leaves powder contained higher concentration of all the metals than other C. macrostachyus leaves samples but its Mn, Co and Ni contents were least. Abomsa C. macrostachyus leaves powder contained higher concentration of Fe, Zn, Cu, Cr and Cd next to Akaki C. macrostachyus leaves powder while Dilla C. macrostachyus leaves powder contained least amount of Mg, Fe, Zn, Cu, Co, Cd and Pb as can been seen from Table 5. This trend of concentration is also similar for the other C. macrostachyus leaves samples.

Comparison of the metal content in C. macrostachyus leaves and infusion samples with other medicinal plants
Results of the present study were compared with reported data on other medicinal plants from other parts of the world. Results of the comparison are given in Table 8. Looking at the table diversified concentration of metals in different species was noticed. However, the range of metals in the present study falls within the range of the reported data. For some of the species, higher concentrations for some of the metals were noticed compared to the present study while lower concentrations were found for some of the others. These variations might be attributed due to variation in species and family of the plants analyzed. Furthermore, environmental and geographical variations might have significant effect on the nutrients accumulation in the plants.
The metals concentrations in the infusion samples were also compared with other medicinal plants leaves from other parts of the world (Table 9). Looking at the table, extent of extractions from the leaves in to the infusions was comparable with the present study for most of the metals investigated. Higher concentrations in the infusion were noticed for those metals in higher concentration in the leaves for almost all of the plants compared. While small concentrations were observed for those whose concentrations were smaller in the leaves. Generally speaking, the range of mineral concentrations in the infusion of the present study was within the range of the plants compared.

Statistical analysis
Analysis of variance (ANOVA) [41] was used to evaluate the difference in concentration of metals among the sampling sites were significant or not. Except between Abomsa/Dilla C. macrostachyus leaves powder samples, there was a significant difference (p < 0.05) in mean concentrations of Ca at 95% confidence interval between all the four samples of C. macrostachyus leave powder when pair wise comparison was made. For Mg, Fe, Mn and Cd significant difference (p < 0.05) at 95 % confidence interval was observed in their mean concentrations between all the four C. macrostachyus leave powder samples. The same was true for Cu and Ni except between Abomsa and Dilla whilst, no significant difference (p > 0.05) in Zn and Cu concentration were observed for all the pair wise comparison except between Akaki and Abomsa C. macrostachyus leaf samples. Similarly, significant variation was in Cr and Pb concentration for all the pair wise comparison tests except between Akaki and Abomsa for Cr and Akaki and Bonga for Pb. Absence of significant difference in some mineral nutrient among samples from Akaki, Abomsa, Bonga and Dilla may indicate that these areas are under the same geographical location and share common climatic conditions. Similarly presence of significant difference in concentration for some minerals of C. macrostachyus leaves powder might aroused from the chemical composition of the leaves itself, the specific environmental conditions in which the plant was grown or the studied area might contains higher concentration of mineral nutrient in the soil [41].

CONCLUSION
Generally, C. macrostachyus (an Ethiopian traditional medicinal plant) accumulated appreciable amounts of major and trace metals in its leaves. Ca and Mg were the most abundant nutrients among the analyzed metals followed by, Fe, Mn and Zn. All the metals enriched in the extract proportionally with extraction time but extent of leaching in the extract was significantly pronounced for the trace metals and the toxic metals (Cd and Pb) for prolonged period of infusion. Thus, beside its medicinal value, C. macrostachyus leaves can be good source of mineral nutrients if further research is conducted to predict appropriate dose to be taken.
Statistical analysis revealed that significant variation in metal concentrations among the sampling sites was observed for some of the metals when pair wise test was made while insignificant variation were seen for the remaining metals. Existence of significant variation in the level of metals for the samples studied might be difference in environmental and/or climatic conditions. Difference in soil parameters might also have significant effect for the variation. Absence of significant variation might be due to the fact that similarity in the aforementioned parameters among the stated areas.