Nutritional and phenolic profiles of leaves of fifteen Anchote (Coccinia abyssinica) accessions

Abstract The purpose of this study was to evaluate the nutritional and phenolic compositions of leaves of Anchote (Coccinia abyssinica) accessions grown in East Wollega Zone (Nekemte), Ethiopia. Leaves of fifteen accessions of Anchote were collected from research farm of Wollega University Ethiopia. The dried leaves of each accession were grounded separately into fine powder following standard procedures. Thereafter, the proximate, mineral, and phenolic compositions were determined according to published official standard procedures with grade chemicals and reagents. Mineral ratios were considered to examine mineral-mineral interactions and antinutrient/mineral molar ratios were calculated to predict minerals bioavailability. Principal component analysis (PCA) and cluster analyses were conducted to characterize the accessions based on nutritional and mineral traits. Proximate compositions (% dwb) except moisture content were varied significantly (P < 0.05) and ranged: moisture 9.21–10.15, crude protein 16.88−29.84, crude fiber 8.52–12.07, crude fat 2.03–4.19, total ash 10.06–17.27, utilizable carbohydrate 37.45–46.36 and gross energy 272.67–293.11 kcal/100 g. PCA of proximate traits showed 74.15% variation for two principal components and the accessions categorized into three clusters. Mineral contents (mg/100 g dwb) were significantly varied (P < 0.05) and ranged: calcium (79.66–100.68), magnesium (31.18–68.47), sodium (41.32–71.95), potassium (71.79–111.13), phosphorus (42.07–66.9), Iron (3.68–11.51) and zinc (1.02–3.11). PCA of mineral traits showed 69.30% variation for two principal components and the accessions grouped into three clusters. Results of phenolic analyses were significantly varied (P < 0.05). Range of condensed tannin, phytate, total flavonoids and total phenols for the examined accessions were 129.0–255.32 mg/100 g, 218.46–240.67 mg/100 g, 8.03–12.74 mg/g QE, and 18.43–29.96 mg/g GAE, respectively. The study revealed that Anchote leaves contain appreciable amount of vital nutrients when compared to the commonly consumed green vegetables in Ethiopia. Particularly, those leaves of Anchote accessions that contained significantly higher amounts can be applied for breeding, new product development, and supplementation programs. Thus, the promotion of Anchote leaves for consumption and cultivation should be encouraged.

ABOUT THE AUTHOR Dr. Habtamu Fekadu Gemede is currently working as Assistant Professor of Food and Nutritional Science at Department of Food Technology and Process Engineering, Wollega University. He did his BSc degree at Haramaya University (Ethiopia), his MSc and PhD degrees at Addis Ababa University. Dr. Habtamu Fekadu Gemede has published 32 articles on international reputable journals. He has more than 14 years experiance of research and teaching, and has supervised 29 MSc and one PhD students. Among those, 25 MSc students under his supervision have successfully defended their theses. Furthermore, he is an editorial board member of eleven international repuatable journals. His research interest are areas of nutritional science, food science, food compositional analysis, food chemistry, Nutritional assessment, Antioxidant analysis etc.

PUBLIC INTEREST STATEMENT
Anchote leave is one of the less known and neglected and native to Ethiopia particularly in Western part. In Ethiopia, it is growing in limited parts of the country and it is high productive when compared with the common edible leaves. Therefore, this neglected crop and traditional food for some specific part of the community in Ethiopia should be further studied and investigated for its nutritional and antinutritional compositions which will promote the crop for further cultivation and consumption.

Introduction
The introductions of indigenous vegetables widely across rural and urban areas add variety and improve nutritional value of the diet (Fao, 2014).Most African countries, blessed with a variety of natural surroundings, varying climates and seasons, have a number of edible indigenous vegetables such as Anchote. Anchote (Coccinia abyssinica) is an endemic and potentially valuable crop of Ethiopia, primarily categorized under root and tuber crops (Ayalew et al., 2017;Holstein, 2012). It belongs to the order Cucurbitales, family Cucurbitaceous (Asfaw et al., 1992;Fekadu et al., 2013). There are about 10 species of Coccinia in Ethiopia; however, only Coccinia abyssinica is cultivated for human consumption (Bekele, 2007).
Anchote is cultivated mainly for its tuber though it has an edible leaves, which is used as nutritious vegetable served after being cooked (Abera, 1995;Ayalew et al., 2017). Anchote tuber is rich in protein, calcium, iron and phosphate contents with minimum antinutritional content, with the latter further reduced during cooking processes, the character that made it a nutritionally recommended food (Fufa & Urga, 1997;Tolera, 2017).However, it is such an underutilized vegetable that the compositions of different accessions of Anchote leaves were not thoroughly studied.
Hence, the populace is unaware of the value of Anchote leaves and rather it is regarded as traditional food mainly for its tubers and for low-income earners. Thus, they have not benefitted from the same level of research attention given to other vegetables and studying compositions of different accessions of Anchote leaves could provide awareness for promoting as an alternative means for dietary diversification. Therefore, the nutritional and phenolic compositions of leaves of Anchote accessions, which were identified, were studied.

Experimental materials
Anchote leaves were collected from an identified fifteen accessions from Wollega University experimental field located in Nekemte, Ethiopia. The leaves samples were coded, packed in polyethylene bags, kept in an icebox and were transported to laboratory of Food Technology and Process Engineering, Wollega University. Analyses were conducted according to published official standard procedures with grade chemicals and reagents.

Sample preparation
Samples of leaves were cleaned with a dry cloth and moisture content was determined on a wet basis. The rest of the samples of leaves were rinsed with distilled water and were then air-dried at room temperature. The dried leaves were grounded separately into fine powder using electric grinder until to pass through 0.5 mm sieve mesh size. The powdered samples were stored in desiccator until analyses done after sealed into airtight polyethylene plastic bags.

Methods of analyses
2.2.1.1. Proximate compositions. The proximate analyses were conducted for contents of moisture, crude protein, crude fiber, crude fat, and total ash according to AOAC (2000). Official method 925.10 (air-oven drying method) used for moisture content determination while ES ISO 1871:2013 method [by Kjeldhal method through which Digestion (H 2 SO 4 ), Distillation (NaOH) and Titration (HCl) processes involved] used for crude protein determination. The crude fiber content was determined according to official method 962.09 using acid/base digestion. Soxhlet extraction method (official method 920.39) used to extract and determine the crude fat content. Total ash content was determined according to official method 923.03 by incineration in a muffle furnace.
Utilizable carbohydrate content was calculated by difference basis (Gemede et al., 2016;Manzi et al., 2004). Atwater's conversion factors [16.7 kJ/g (4 kcal/g) for protein, 37.4 kJ/g (9 kcal/g) for fat and 16.7 kJ/g (4 kcal/g) for carbohydrate] used for calculating the energy value and expressed in calories(Kcal per 100 g) (Gemede et al., 2016). The utilizable energy due to protein percentage [UEDP % = 60% * Percentage of Energy due to Protein (PEP) %)] was also calculated to indicate the contribution of protein to the total energy (Ilesanmi & Jonathan, 2018 Latta and Eskin (1980) method as later modified by Vaintraub and Lapteva (1988) was used for determination of phytate after extraction of sample with 10 ml of 0.2 N HCl for 1 h, centrifuged, and reacting sample extract (3 mL) with 2 mL of Wade reagent (0.03% FeCl 3 .6H 2 O and 0.3% sulfosalicylic acid in distilled water).The absorbance of sample was measured at 500 nm using UV-Visible spectrophotometer (DU-64 spectrophotometer, Beckman). The amount of phytate content (mg/100 g sample) was calculated using phytic acid standard curve prepared in the same condition. Burns (1971) method as modified by Maxson and Rooney (1972) was used for determination of condensed tannin using vanillin-HCl reagent, and D-catechin as standard. One mL extracted sample (1 g of sample and 10 ml of 1%HCl in methanol, for 24 hr) reacted with 5 ml of vanillin-HCl reagent (8%concentrated HCl in methanol and 4% vanillin in methanol, 50:50, v/v), and the absorbance read at 450 nm using UV-visible spectrophotometer (DU-64 spectrophotometer, Beckman).Tannin concentration was calculated from the linear regression equation obtained and the result was expressed as condensed tannin content in mg/100 g dry weight.

Condensed tannin content determination.
2.2.2.3. Total flavonoidsdetermination. Dowd method as adopted by Arvouet-Grand et al. (1994) was used to determine total flavonoids content colorimetrically using AlCl 3 and Quercetin as standard. One ml of 2% Aluminum trichloride (AlCl 3 ) in methanol mixed with the same volume of methanolic extracts (1000 µg) and after 10 min, the absorbance of the sample and blank (1 ml of extract solution with 1 ml methanol, without AlCl 3 ) measured. A quercetin calibration standard was prepared from 12.5 to 100 µg/ml of quercetin standard in methanol from 0.5 mg/ml stock.

Total phenol determination.
Total phenols content was determined by Folin-Ciocalteu method as described by Singleton et al. (1999) using Gallic acid as standard. A total of 5 ml of 80% aqueous methanol was added, and the suspension stirred slightly. Tubes were sonicated for 40 min at 40°C in a sonicator bath and centrifuged (14,000 rpm for 10 minutes). Supernatants were collected and the amount of total phenolics in the extract was determined according to the Folin-Ciocalteu procedure. Absorption was read at 765 nm using UV-Visible spectrophotometer. Series of standard solutions were prepared using pure Gallic acid. Finally, the concentration was expressed as Gallic acid equivalents in mg/g of dry material.

Statistical analysis
Analyses were done in duplicates and the results obtained were statistically analyzed for one-way ANOVA using SPSS version 20.0 (SPSS Institute Inc., Cary, NC) for windows. Duncan multiple range tests were performed for analyzing significant differences (P < 0.05) among means as outlined by Obi (2002). Factor analysis based on principal component analysis (PCA) and cluster analyses were performed to characterize the accessions in relation to the most discriminating nutrient traits (Kubie, 2013).

Principal component analysis of proximate nutrient traits
The results of the principal component analysis (PCA) of the six proximate values measured (Table  2) showed that the first two components contributed 74.15% of variability among the 15 accessions evaluated. PC 2 comprises three accessions (P97-Acc147, P85-Acc5, and P55-Acc111) while PC 1 consists of the others with 31.79% & 42.36% of variation, respectively.
The first principal component had high positive loading for crude fat, utilizable carbohydrate & total ash. The second principal component weighed the highest and positive moisture, crude fiber, and influenced by crude protein content. In this study, principal component analysis (PCA) was performed to define the existing pattern of variation among the accessions. Prasad et al. (2010) reported that PCA helps to identify traits that have substantive and meaningful contribution towards the observed variations. The scatter plots of the accessions on the principal component were presented in Figure 1. As shown in Figure 1, cluster1 contained accessions (P3-Acc118, P10-Acc151, P80-Acc201, and P90-Acc162) with higher loading for total ash and lowest in crude fiber (P10-Acc151 and P90-Acc162). Cluster 2 consisted of accessions (P96-Acc69, P83-Acc186, P18-Acc79, P109-Acc149, P93-Acc128, P29-Acc178, P4-Acc40, and P25-Acc140) with higher crude fiber and crude fat (except P25-Acc140) contents whereas total ash (in P83-Acc186, P25-Acc140, P96-Acc69 & P29-Acc178) and crude protein (except in P18-Acc79) contents were the lowest value. Cluster 3 had accessions (P97-Acc147, P85-Acc5, and P55-Acc111) with high loading for crude protein content while lowest values for crude fat were observed. It was also reported that cluster analysis could be done for combining observations into homogenous group with respect to certain characteristics (Bozokalfa et al., 2011).

Mineral compositions
All mean results of mineral compositions of the accessions, which were selected based on the total ash content as an indicator for better mineral composition and studied, are compiled in Table  4.The concentrations of Calcium in the samples varied from 79.66 mg/100 g to 100.68 mg/100 g in "P90-Acc162" and "P4-Acc40" accessions, respectively. Accession "P4-Acc40" had the highest Calcium content (100.68 mg/100 g) which was significantly (P < 0.05) higher than all the accessions except accession "P55-Acc111" (100.63 mg/100 g). Accession "P90-Acc162" had significantly (P < 0.05) lower (79.66 mg/100 g) Calcium content than all the accessions except "P80-Acc201" (79.86%) on dry weight basis. The mean value of calcium content was 89.36 mg/100 g, which was comparable to leaves of Swiss chard (Beta vulgaris) 85.0 mg/100 g as reported by Gobezie et al. (1997). However, the mean value was higher than the leaves of commonly consumed vegetables in Ethiopia: Cabbage (Brassica oleracea) 43.0 mg/100 g, Cauliflower (Brassica oleracea var. botrytis) 30.0 mg/100 gand Lettuce (Letuca sativa) 22.0 mg/100 g as reported by Gobezie et al. (1997). On the other hand, Gobezie et al. (1997)showed some of the commonly consumed vegetables leaves such as Celery (Aplum graveolens) 317.0 mg/100 g, Ethiopian kale (Brassica carinata) 260.0 mg/100 g, Spinach (Spinacea oleracea) 122.0 mg/100 g and Yam (Dioscorea spp.) 119.0 mg/100 g had a higher mean value than the accessions of Anchote leaves under study. Ayalew (2016) reported the mean value of calcium content 147.8 mg/100 g for the leaves of Anchote accessions. Calcium is an important mineral for human beings, which provides good strength of bones and teeth (Amagloh & Nayarko, 2012;Saha et al., 2015). It plays an important role in blood clotting, muscles contraction, and neurological function and helps in

2.21
Note: Number of sample size is fifteen and replicates are three. Magnesium contents in the accessions are listed in Table 4. The contents of Magnesium varied from 31.18 mg/100 g in "P109-Acc149" to 68.47 mg/100 g in "P55-Acc111". Accession "P55-Acc111" had the highest Magnesium content (68.47 mg/100 g) which was significantly (P < 0.05) higher than all the accessions and was followed by "P85-Acc5" (60.95 mg/100 g), "P10-Acc151" (58.59 mg/100 g), "P3-Acc118" (54.02 mg/100 g) and "P4-Acc40" (53.84 mg/100 g) in that order. However, the accession "P109-Acc149" had significantly (P < 0.05) lowest (31.18 mg/100 g) Magnesium content on dry weight basis. The mean value of magnesium content in this study was 47.24 mg/100 g, which was higher than the leaves of Amaranthus hybridus (34.12 mg/100 g) reported by Nyonje (2015) but comparable with Mucuna poggei leaves content (56.05 mg/100 g). Ayalew (2016) reported the mean value of magnesium content 45.04 mg/100 g for the leaves of Anchote accessions. Magnesium, which is an essential to the function of several enzyme systems, is important in maintaining electrical potential in nerves and membranes, is involved with liberation of energy for muscle contraction, and is required for normal metabolism of calcium and phosphorus. It is also an important mineral element in connection with its role in circulatory diseases such as ischemic heart disease and calcium metabolism in bone (Hassan & Umar, 2006;Ishida et al., 2000). Table 4. In this study, the Sodium content varied from 41.32 mg/100 g "P93-Acc128" to 71.95 mg/100 g "P109-Acc149". Sodium content of accession "P109-Acc149" was significantly (P < 0.05) highest (71.95 mg/100 g) while accession "P93-Acc128" had the lowest (41.32 mg/100 g) Sodium content which was significantly (P < 0.05) different from all accessions on dry weight basis. The mean value of Sodium in the study was 55.03 mg/100 g. Ayalew (2016) also reported the mean value of sodium content 54.46 mg/100 g for the leaves of Anchote accessions. Sodium is an important mineral, which assists in the regulation of the body fluids and maintenance of electrolyte balance in the body. It is the chief extracellular ions of the body but excessive intake of can contribute to elevating blood pressure (Saha et al., 2015). Table 4. The value of potassium was with the range of 71.79 mg/100 g for P93-Acc128 accession to 111.13 mg/100 g for accession P109-Acc149. The Potassium content of accession "P109-Acc149" had significantly (P < 0.05) higher (111.13 mg/ 100 g) than all accessions except "P97-Acc147" (109.22 mg/100 g), whereas accession "P93-Acc128" had significantly (P < 0.05) lower (71.79 mg/100 g) content than all accessions except "P3-Acc118" (73.44 mg/100 g) on dry weight basis. This study showed a mean value of 95.29 mg/100 g. Saha et al. (2015) reported on his study on leaves of eight green leafy vegetables that Moringa oleifera contained 75.33 mg/100 g and Chenopodium album 70.70 mg/100 g, the rest had the potassium contents in the range of 71.36-74.46 mg/100 g, which was comparable to the present study. Ayalew (2016) also reported the mean value of potassium content 139.82 mg/100 g for the leaves of Anchote accessions. Potassium, which is the principal intracellular cation, is involved with cellular enzyme function and with Sodium helps to regulate osmotic pressure and pH equilibria. It is also essential for life and high amount of potassium increases the Iron utilization and is beneficial to the people taking diuretics to control hypertension as reported by Archana et al. (2012).

Potassium contents of the accessions are listed in
Phosphorus contents of the accessions are presented in Table 4. In this study, the phosphorus content ranged from 42.07 mg/100 g "P93-Acc128" to 66.90 mg/100 g "P85-Acc5". Accession "P85-Acc5" had the highest Phosphorus content (66.90 mg/100 g), which was significantly (P < 0.05) highest and was followed by "P80-Acc201" (62.94 mg/100 g), "P109-Acc149" (60.79 mg/100 g) and "P10-Acc151" (54.18 mg/100 g) in that order. However, the accession "P93-Acc128" had significantly (P < 0.05) lowest (42.07 mg/100 g) Phosphorus content on dry weight basis. The mean value of the study was 53.22 mg/100 g, which was comparable to the value of leaves of Celery (Aplum graveolens) 52.0 mg/100 g as reported by Gobezie et al. (1997). However, Gobezie et al. (1997) reported lower values than the present study for leaves of Cabbage (Brassica oleracea) 29.0 mg/100 g, Swiss chard (Beta vulgaris) 41.0 mg/100 g, Lettuce (Letuca sativa) 31.0 mg/100 g and Yam (Dioscorea spp.) 13.3 mg/100 g. The mean value of the present study on Anchote leaves was lower than the values of Cauliflower (Brassica oleracea var. botrytis) 62.0 mg/ 100 g, Ethiopian kale (Brassica carinata) 64.0 mg/100 g and Spinach (Spinacea oleracea) 110.0 mg/ 100 g leaves (Gobezie et al., 1997). Ayalew (2016) also reported the mean value of phosphorus content 64.63 mg/100 g for the leaves of Anchote accessions. Phosphorus is an essential part of every living cell. It is involved in the enzyme-controlled energy-yielding reactions of metabolism. Phosphorus also helps to control the acid-alkaline reaction of the blood. Highest requirements for calcium and phosphorus are for the young, pregnant, and nursing mothers (Lyaka et al. (2014).

Mineral ratios
The mineral ratios, which was calculated by dividing the first mineral level to the second mineral level (Gemede et al., 2016;Wang et al., 2014), are presented in Table 5.

Calcium to magnesium
Calcium (Ca) to Magnesium (Mg) ratio present in the accessions is shown in Table 5. The Ca to Mg ratio of the leaves of Anchote accessions ranged from 1.4 for accession P85-Acc5 to 3.2 for accession P4-Acc40. The mean Ca to Mg ratio was 2.1, which met the standard with a value greater than two considered as a good mineral balance (Adeyeye et al., 2012). In general, the Ca to Mg ratio in the human body is referred to as the blood sugar ratio. Both minerals are very much interrelated in such a way that calcium is required for the release of insulin from the pancreas whereas magnesium inhibits insulin secretion. Magnesium is also necessary to keep calcium in solution (ARL, 2012).

Calcium to potassium
The Calcium (Ca) to Potassium (K) ratio present in the accessions is shown in Table 5. Ca/K ratios among the accessions ranged from 0.7 to 1.2. Accession P109-Acc162 showed the least ratio while P3-Acc118 had the highest ratio. The mean Ca/K ratio was 1.0, which met the standard, a ratio less than 4 considered as good source of the mineral balance. The Calcium to Potassium (Ca/K) ratio is called the thyroid ratio because calcium and potassium play a vital role in regulating thyroid activity (Gemede et al., 2016;Ojiezeh et al., 2016). Calcium is affected by several hormones and is considered to be under parasympathetic (an involuntary nervous system that serves to slow the heart rate, increase intestinal and glandular activity) control. The elevation of the Ca/K ratio can be indicative of reduced thyroid expression. In contrary, a low Ca/K ratio would indicate an elevation of thyroid expression (Watts, 2010). Watts (2010) also reported that the ideal ratio of Ca/ K is 4:1 with an acceptable ideal range of 2.2 to 6.2. A higher Ca/K levels in foods is required for favorable calcium absorption in the intestine for bone formation (Gemede et al., 2016;Wang et al., 2014).

Calcium to phosphorus
The Calcium to Phosphorus ratio present in the accessions is listed in Table 5. The Ca to P ratio of the leaves of Anchote accessions varied from 1.3 for accessions P80-Acc201 and Acc109 to 2.1 for P55-Acc111 and presented in Table 6. The mean Ca to P ratio of the accessions was 1.7, which was much satisfactory compared to the standard, a ratio greater than 0.5 is a good mineral balance. The recommended Ca to P ratio should be greater than 0.5 (Gemede et al., 2016;Wang et al., 2014). Ca to P ratio greater than 2 also contributes to the absorption of calcium in the small intestine (Adeyeye & Aye, 2005;Oze, 2011). Furthermore, food is considered as good if Ca to P ratio is greater than 1 and poor if this ratio is less than 0.5 (Gemede et al., 2016;Oze, 2011).A higher calcium to phosphorous level in foods is required for favorable calcium absorption in the intestine for bone formation (Adeyeye et al., 2012). According to Ijarotimi et al. (2013), diets rich in protein and phosphorus may promote the loss of calcium in the urine. The Ca to P ratio in this study indicates that the leaves of Anchote accessions would help calcium absorption in the body. The   high Ca to P ratio observed in this study is also of nutritional benefit, particularly for children and the aged who need higher intakes of calcium for bone formation and maintenance (Gemede et al., 2016;Ijarotimi et al., 2013).

Sodium to potassium
The Sodium to potassium ratio present in the accessions is shown in Table 5. The Na to K ratio of the leaves of Anchote accessions ranged from 0.5 to 0.7 with the mean ratio of 0.6, which was below the standard, a ratio less than one considered as a good mineral balance. The accessions P55-Acc111 and P4-Acc40 had the lowest ratio while P3-Acc118 gained the highest value. The recommended Na to K ratio should be less than one (Gemede et al., 2016;Wang et al., 2014). Ijarotimi et al. (2013) also reported that the Na to K ratio less than one is recommended for diets, particularly for hypertensive patients. Therefore, the observed Na to K mineral ratio showed that leaves of Anchote would help to prevent hypertension and might lower blood pressure in hypertensive patients. It may be also suitable for people who have the risk of high blood pressure. This ratio is also referred as the life-death ratio in humans physiology because it is so critical in their interdependence since the Sodium pump mechanism and the electrical potential of cells is regulated by Sodium and potassium levels (Gemede et al., 2016;Wang et al., 2014). Sodium is normally extracellular while potassium is normally intracellular. If the ratio of these minerals is unbalanced, it indicates important physiological malfunctions within the cells (ARL, 2012).

Iron to zinc
The Iron to zinc ratios of the leaves of Anchote accessions are presented in Table 6. The Fe to Zn ratio of the accessions varied from 1.7 for P10-Acc151 to 11.3 for P97-Acc147. The mean Fe to Zn ratio of the accessions was 3.9. Pérès et al. (2001) reported that Iron did not impair zinc absorption up to an Iron: zinc ratio of 2:1. Beside on this report, one can conclude that the Iron present in the accessions did not impair zinc absorption. Iron functions as hemoglobin in the transport of oxygen and in cellular respiration, as it functions as essential component of enzymes involved in carbohydrates, protein, and fat oxidation to control body weight, which is very important factor in diabetes (Moses et al., 2012). Vitamins A and E metabolism and bioavailability are dependent on zinc status (Ibrahim et al., 1999;Soetan et al., 2010).

Mineral safety index
All the calculated MSI values were less than standard MSI (Table 6) meant that none of the minerals would constitute mineral overload or became toxic to the leaves consumers. All the differences between the standard and calculated MSI values were positive which implied that the body might not be overloaded with the listed minerals (Nieman, 1992).

Principal component analysis of mineral traits
The principal component analysis (PCA) result of the mineral traits is listed in Table 7. The PCA showed that two Eigen vectors were required for reaching a total variance of 69.30% among the 10 Anchote accessions evaluated. The first component accounted for 40.14% of the total variation and the second contains 29.16%. PC 1 comprises five accessions, which are P90-Acc162, P80-Acc201, P93-Acc128, P3-Acc118, and P10-Acc151. PC 2 also consists of the other five accessions, which are P109-Acc149, P97-Acc147, P85-Acc5, P55-Acc111, and P4-Acc40.
Scatter plot of the accessions on PCs based on mineral traits showed that Cluster 1 contained the accessions P10-Acc151, P90-Acc162, P80-Acc201, P3-Acc118, and P93-Acc128, with high loading for zinc and low value for Sodium, potassium and Iron. Whereas cluster 2 consisted of accessions, which were P4-Acc40, P55-Acc111 and P85-Acc5, with high loading for calcium. P85-Acc5 had highest value for magnesium and phosphorus whereas iron (in P4-Acc40) was recorded as lowest value. The accessions P97-Acc147 and P109-Acc149 belong to the third cluster with highest loading for iron, sodium, and potassium and the lowest value in calcium, magnesium, phosphorus (P97-Acc147) and zinc (P97-Acc147) (Figure 2).

Phytochemical contents 8.2.1. Condensed tannin content
The results of the condensed tannin content of the accessions evaluated are listed in Table 9. The condensed tannin content of the accessions ranged from 129.0 ± 6.89 mg/100 g to 255.32 ± 4.62 mg/100 g for the accessions P4-Acc40 and P10-Acc151, respectively. A significant difference (P < 0.05) was observed in between the ranges and P85-Acc5 had a significant difference (P < 0.05) with the rest of the accessions. The mean value of condensed tannin content was 189.72 mg/100 g. Ayalew (2016) also reported higher condensed tannin content of 216.53 mg/100 g for the leaves of Anchote accessions. According to Essack et al. (2017), tannins are plant polyphenols that are capable of forming complexes with metals ions and macromolecules like proteins and polysaccharides. Tannins affect the nutritional value of food products by chelating metals like Iron and zinc and reducing the absorption of these nutrients as well as forming complexes with protein thereby inhibiting their digestion and absorption (Olawoye & Gbadamosi, 2017). The antinutritional/toxicity effects of tannin depend upon their chemical structure and dosage (Fekadu et al., 2013). Therefore, the condensed tannin content of leaves of Anchote accessions obtained from this study may not lead to toxicity since the total acceptable tannic acid daily intake for a man is 560 mg/100 g and further reduced during traditional processing (Fekadu et al., 2013).Condensed Tannins have a wide range of biological and pharmacological activities including antioxidative, cardio-protective, antitumor, antibacterial, antiviral, anti-inflammatory and immune-modulatory. These protective effects are related to their capacity to: (a) act as free radical scavengers; (b) activate antioxidant enzymes (Kumari & Jain, 2012).
Means within the same column with different superscripts were significantly (P ≤ 0.05) different. Values are presented as Mean± SE of replicate determinations n = 3 and the number of sample size is fifteen.

Phytate contents
The phytate contents of the accessions evaluated are shown in Table 9. The range of phytate content in this study was from 188.66 ± 0.81 for accession P85-Acc5 to 206.62 ± 1.01 for accession P10-Acc151 in which a significant difference (P < 0.05) was observed. On this study, the mean value of the accessions was 197.87 mg/100 g. Ayalew (2016) also reported a much higher phytate content of 250.30 mg/100g for the leaves of Anchote accessions. Phytates are a naturally occurring phosphorus storage compound present in green leafy vegetables. It has been shown to inhibit and reduce the absorption as well as bioavailability of minerals and metal ions like zinc, Iron etc. (Essack et al., 2017).Foods high in phytic acid are getting a bad reputation due to its ability to bind to essential minerals such as Iron, zinc, calcium, and magnesium in the digestive tract and inhibit their absorption by the body (Nissar et al., 2017). Recent studies indicate despite being somewhat demonized for its ability to reduce mineral absorption, phytic acid actually has some potentially beneficial properties. On the plus side, phytic acid can act as antioxidant, exhibits anti-cancer properties, and may have a positive impact on cholesterol and blood sugar (Nissar et al., 2017;Onomi et al., 2004). Preparation methods can reduce the phytic acid content in food, as well as adjusting meal times and food choices that can help to have better mineral absorption (Sade, 2009).

Total flavonoids contents
The total flavonoids contents of the accessions evaluated are listed in Table 9. The total flavonoids content range was from 8.03 ± 0.11 mg/g QE for P97-Acc147 to 12.74 ± 0.57 mg/g QE for P3-Acc118 and there was a significant difference (P < 0.05) between the ranges. A significant difference (P < 0.05) between the highest (P3-Acc118) and the two lower (P97-Acc147 & P85-Acc5) were observed. The mean value was 10.43 mg/g QE that was lower than those reported by Nyonje (2015)for Amaranthus caudatus (69.67 mg/g QE). Miean and Mohamed (2001)examined the flavonoids content of 62 vegetables and found that broccoli, cauliflower, cabbage, Chinese cabbage, and kailan contained between 148 and 219 mg/kg offlavonoids.Flavonoids are a ubiquitous group of polyphenolic substances, which have antibacterial, anti-inflammatory, anti-allergic, antineoplastic, antiviral, anti-thrombotic, and vasodilatory activities (Kopsell et al., 2005). Good correlation between the total flavonoids content and antioxidant activity has been show indicating that the flavonoids contribute in free radical scavenging (Cushnie & Lamb, 2005;Nyonje, 2015).

Total phenol contents
The total phenol contents of the accessions evaluated are presented in Table 9. The total phenol content of the accessions ranged from 18.43 ± 0.38 mg/g GAE (P85-Acc5) to 29.96 ± 0.48 mg/g GAE for P4-Acc40 in which a significant difference observed. As presented in Table 11, there was a significant difference (P < 0.05) between P4-Acc40 with the rest. The mean value of the study was 21.98 mg/g GAE that was lower than total polyphenols in young pods of raw, cooked and sundried A. esculentus with the values as 55.7 mg/g GAE, 51 mg/g GAE and 51.7 mg/g GAE, respectively (Abera, 2014). Ayalew (2016) also reported a higher total phenol content ranged from 26.42-59.9 mg/g GAE for the leaves of Anchote accessions. According to the study by Ritter et al. (1992), when the total polyphenol content in all the vegetables tested measured, there was a significant inverse correlation with Iron absorption. A number of vegetables associated with low Iron absorption turned bluish-black when Iron added to them, suggesting that the total polyphenol content in them was high. Phenolic compounds are substances that possess an aromatic ring bearing a hydroxyl substituent, including their functional derivatives such as esters, methoxy compounds and glycosides (Park et al., 2018 Table  10. The Phy: Ca ratio ranges from 0.12 for accessionsP4-Acc40 and P85-Acc5 to 0.14 for accession P97-Acc147 through which a significance difference (P < 0.05) was maintained in between the ranges. The mean Phy: Ca ratio value was 0.13 mol/g, which was lower than the reported critical molar ratio, indicating that absorption of calcium would not be affected by phytate in these leaves. Phytic acids markedly decrease Ca bioavailability and the Phy: Ca molar ratio has been proposed as an indicator of Ca bioavailability (Gemede & Fekadu, 2014;Latta & Eskin, 1980). The Phy: Ca molar ratios < 0.24 considered as indicative of good calcium bioavailability.
8.2.5.2. Phytate: zinc molar ratio. The calculated Phy: Zn molar ratios are shown in Table 10. In this study, the Phy: Zn ratio ranged from 6.87 ± 1.53 for P10-Acc151 accession to 19.97 ± 2.58 for P97-Acc147 accession. A significance difference (P < 0.05) between the ranges has been shown. There was significance difference (P < 0.05) observed in between P97-Acc147 and the rest of all the accessions except with P4-Acc40. The accessions P85-Acc5, P10-Acc151, and P3-Acc118 with their respective ratio had a good bioavailability ratio <15 whereas P97-Acc147 and P4-Acc40 which had a ratio > 15 as an indicator of poor zinc bioavailability. In general, the mean value of the study was 12.22 mol/g. It indicates good zinc bioavailability (Gemede & Fekadu, 2014;Morris & Ellis, 1989). The importance of foodstuff as a source of dietary zinc depends on both the total zinc content and the level of other constituents in the diet that affect zinc bioavailability. Phytate may reduce the bioavailability of dietary zinc by forming insoluble mineral chelates at a physiological pH (Gemede & Fekadu, 2014;Oberleas, 1983). Zinc deficiency has been shown to be the cause of dwarfism and hypogonadism among adolescents (Gemede & Fekadu, 2014;Prasad, 1984). Zinc has been described as the essential mineral most adversely affected by phytate and the phytate-to-zinc molar ratio has been proposed as an indicator of zinc bioavailability (Gemede & Fekadu, 2014;Plaami, 1997).

Phytate: iron molar ratio
The calculated Phytate: Iron molar ratios are presented in Table 10. The range of Phytate: Iron ratio was 1.4 for accession P97-Acc147 to 4.54 ± 0.26 for accession P4-Acc40, which implied that the absorption of Iron from the accessions would be inhibited by phytate and as a result, the bioavailability of Iron would be poor. Phytate begins to lose its inhibitory effect on Iron absorption when phytate: Iron molar ratios are less than 1.0, although even ratios as low as 0.2 exert some negative effect (Gemede & Fekadu, 2014;Hurrell et al., 1992).The phytate: Iron molar ratios >1 was regarded as indicative of poor Iron bioavailability (Siegenberg et al., 1991). A significance difference (P < 0.05) between the ranges was observed and there was a significance difference (P < 0.05) in between the accessions P97-Acc147 with the rest except P85-Acc5 and P4-Acc40 with all.  Table 10. In this study, the value ranges from 0.02 ± 0.00 for accessions P10-Acc151 and P3-Acc118 to 0.05 ± 0.00 for both accessions P4-Acc40 and P97-Acc147 in which a significance difference (P < 0.05) between ranges was observed. The [Ca] [phytate]/[Zn] mill molar ratio which was <0.5 mol/kg for all the accessions showed good zinc bioavailability. In general, the mean value of the study was 0.03 mol/kg, which was lower than the critical level. This indicates there was less phytate-induced decrease in zinc bioavailability due to the higher calcium content in the food. The potential effect of calcium on zinc absorption in the presence of high phytate intakes has led to the suggestion that the [Phy] [Ca]/[Zn] mill molar ratio may be a better index of zinc bioavailability than the [Phy]/[Zn] molar ratio alone. High calcium levels in foods can promote the phytate-induced decrease in zinc bioavailability when the [Ca] [phytate]/[Zn] mill molar ratio exceeds 0.5 mol/kg (Gemede & Fekadu, 2014;Gibson, 1994).

Conclusions
In conclusion, the results of this study revealed that there is a significant difference (P < 0.05) in proximate, mineral, and phenolic compositions of leaves of Anchote accessions except in moisture content in dwb. The most notable finding of this study is that leaves of Anchote accessions were found to be good source of vital nutrients like crude protein, crude fiber, calcium, iron and zinc, which are good for human and animal health. The study also revealed that leaves of Anchote could be good source of Phenolics. Moreover, those leaves of accessions that contained significantly higher amounts can be implemented for breeding, new product development, and supplementation programs. Therefore, leaves of Anchote contained vital nutrients and seemed to have the potential to contribute for food security and its promotion for consumption and cultivation should be encouraged. Further study needs to be carried out on the possibility of developing valueadded products by itself or blending with other food ingredients.