Genotype Effect on Proximate and Mineral Analysis of Safflower as a Green Leafy Vegetable

Two field trials were carried out at the Botswana University of Agriculture and Natural Resources Content Farm (5924′S, 9525′E and 993 m above sea level) in Southern Region of Botswana, to evaluate the effects of genotypes on proximate and mineral composition of safflower leaves. Five safflower genotypes (Kiama composite (local), Sina-PI-537598, Gila-PI-537692, PI-537636 and PI-527710) were evaluated in a randomized complete block design (RCBD) with three replications. During the rosette stage safflower leaves (both petiole and blade) were harvested, dried, and ground for analysis. The results showed that safflower genotypes significantly varied in the leaf proximate content of crude protein (24-28%), crude fibre (8-14%), moisture content (86-87%) and dry matter content (13-14%) in both winter and summer growing seasons. The genotype ‘Sina’ had the highest crude fibre content compared to other genotypes. The average leaf mineral content significantly (p < 0.05) varied from 2-3 mg g phosphorus, 3-4 mg g calcium, 5-6 mg g sodium, 15-17 mg g magnesium and 15-18 mg g potassium. The Na:K and Ca:P ratios ranged between 0.18 and 3.41. All the safflower genotypes evaluated had sufficient nutritional content to be used as a green leafy vegetable for human consumption and food security.


Introduction
Safflower is an all year round, drought, heat, cold and saline tolerant oil seed crop. It is a bushy, greatly branched, herbaceous thistle-like annual with long spiny or non-spiny leaves (Singh & Nimbkar, 2006;Weiss, 2000). It belongs to the family Asteraceae/Compositae with other crops of economic importance such as artichoke (Cynara), lettuce (Lactuca), endive (Cichorium) and salsify (Tragopogon) used as vegetables and sunflower (Helianthus) commonly used as a vegetable oil crop (Archibald, 2011). Safflower plant parts are commercially used for different purposes, for example the seed is used for extraction of high quality cooking oil (Liu et al., 2016;Emongor & Oagile, 2017), petals are ground for herbal and medicinal purposes (Li & Mündel, 1996), food colorants, carthamin and carthimidin dyes (J. Cannon & M. Cannon, 2003), stems are used as cut flowers (Kizil et al., 2008), and leaves are used as vegetables (Sigh et al., 2017). Countries such as India utilise seedlings, thinnings, and bottom leaves of safflower as a leafy vegetable (Suneel-Kumar et al., 2016a, 2016b. Safflower leaves have been recommended for consumption from 30 to 70 days after sowing depending on whether the genotype is spiny or non-spiny (Suneel-Kumar et al., 2015;Suneel-Kumar et al., 2016). As a leafy vegetable, safflower is reported to be palatable and highly nutritious for human consumption with high fibre, minerals, vitamins and antioxidants (Gopalaan, 2004;Sigh et al., 2017). Despite its nutritional status, safflower leaf utilisation as a green leafy vegetable is still minimal and/or unknown in most countries. However, there is an increased interest in research and cultivation of safflower due to its adaptive and nutritive characteristics. Studying nutritional composition of safflower leaves can promote its cultivation and consumption as a leafy vegetable. This can improve household diets and human health nutrition, and contribute to farmer income from early stages of production especially in African countries. The objective of the study was to evaluate the effects of genotypes on proximate and nutritional analysis of safflower leaves during the rosette stage.

Experimental Location
Two field trials were carried out in the Botswana University of Agriculture and Natural Resources Content Farm (59 o 24′S, 95 o 25′E and 993 m above sea level). The soils are shallow, ferruginous tropical soil, mainly consisting of sandy loams (De Wilt & Nachtengale, 1996;Emongor & Mabe, 2012). The mean rainfall is 538 mm per annum. The winter and summer temperatures ranges between -1 o C (morning) to 30 °C (afternoon) and 20 o C (morning) to 37 o C (afternoon), respectively (Burgess, 2006).

Treatment and Experimental Design
The trials were laid in a randomized complete block design (RCBD) with three replications, the experimental units were 5 m × 5 m. The treatments comprised of five safflower genotypes (Kiama composite (local/control), Sina-PI-537598, Gila-PI-537692, PI-537636 and PI-527710) grown in winter and summer. These genotypes are the most adaptive and performing genotypes in the semi-arid regions of Botswana.

Cultural Practices
The land was cleared, ploughed followed by disc harrowing to a fine soil tilth. Soil was sampled to determine mineral composition prior to planting. The general fertilizer application for basal dressing was 60 kg/ha nitrogen (N); 30 kg/ha phosphorus (P) and 20 kg/ha potassium (K) (FAO, 2013). Safflower seed was sown directly at a rate of 2 seeds per hill at a depth of approximately 4.5 cm. This was followed by thinning 15 to 20 days after emergence. All necessary management practices were undertaken to enhance good growth and development. The amount of water applied was according to crop water requirements (ETm) as related to reference evapotranspiration.

Data collection: Proximate and Mineral Analysis
During rosette stage safflower leaves (both petiole and blade) were sampled for proximate and mineral analysis. The rossette stage is an early vegetative growth after emergence with an active leaf growth where leaves accumulate a greater leaf size and number. It is the longest vegetative stage that lasts for 30-40 days after sowing depending on season and genotype (Moatshe & Emongor, 2019)

Determination of Moisture Content
The clean porcelain crucibles were oven dried at 105 o C for 24 hrs, then cooled in a dessicator and weighed (W 0 ). Approximately 5.0 g sample was weighed into a crucible and weighed again as crucible + sample (W 1 ). The crucible with sample was oven dried at 105 o C for 48 hrs and cooled in a desiccator before they are reweighed (W 2 ). The percentage moisture content was calculated as:% moisture content = (W 1 -W 2 )/(W 1 -W 2 ) × 100

Determination of Dry Matter Content
Leaf dry matter (DM) was determined according to AOAC (1996). The dry matter (DM) was determined by weighing approximately 1 g of ground sample into pre-weighed crucibles and placed in an oven set at 66°C for 72 hours. Dry matter was determined as the difference between initial sample weight and moisture weight and expressed as a percentage.

Crude Fibre Determination
This was determined according to (AOAC, 1996). The dried sample was ground and approximately 1.0 g (W 0 ) was weighed into a fritted glass crucibles and hydrolysed with boiling 0.128M sulphuric acid, followed by boiling in 0.223M potassium hydroxide solution in hot extractor. The residue was washed with preheated distilled water before being transferred to a cold extractor and washed with acetone. The residue and crucibles were oven dried at 105 o C overnight and weighed (W 1 ) before being ignited into a muffle furnace at 450 o C for 8 hrs. The residual ash was first cooled in an oven at 105 o C overnight, then cooled to room temperature in a dessicator and weighed (W 2 ). The percentage of crude fibre was calculated as:% Crude Fibre = ((W 1 -W 2 )/W o ) × 100.

Crude Protein Determination
The ground dry sample was digested in a BD block at 330 o C for 7 hours. After digestion, nitrogen (N) was determined through distillation and titration using the micro-kjeldahl method (AOAC, 1996). The crude protein content was calculated by multiplying percentage N content by a factor of 6.25 (AOAC, 1996).

Mineral Analysis Procedure
The leaf samples were oven-dried at 66 o C to constant weight. The dried samples were ground using a sieve of size two and 1.25 g composite sample digested in 20 ml sulphuric acid (98%) and 4 ml hydrogen peroxide (30%) in a BD block at 330 o C for 7 hours. Phosphorus (P) was determined calorimetrically using sodium phenol and ammonium molybdate plus ascorbic acid method (AOAC, 1996). The absorbance of phosphorus, calcium (Ca), magnesium (mg) and potassium (K) was read in an Optimal Emission Spectrophotometer using Inductively Coupled Plasma of model Optima 7300 DV. Data was expressed as total mineral content in mg g -1 on dry weight basis.

Statistical Analysis
Analysis of variance was performed on the data collected using general linear model (PROC GLM) procedure of Statistical Analysis System (SAS 2009, Carey, NC) program package. Multiple comparisons among means was done using Protected Least Significant Difference (LSD) at P = 0.05.
The crude fibre content (CFC) of safflower genotypes on study ranges between 7 to 16% irrespective of season (Table 1). For the genotypes studied, average crude fibre content of safflower leaves was higher in winter than summer by 36% difference, genotype 'Sina' produced more CFC compared to other genotypes (Table 1). The crude protein content (CPC) of safflower leaves significantly (P < 0.05) ranged between 21 to 37% and 19 to 30% in winter and summer grown genotypes, respectively (Table1). Comparing winter genotypes, 'Sina' and 'Kiama' had the lowest and highest protein content of 21 and 37%, respectively. However, genotypes 'Sina' and 'PI-537636' or 'Kiama' and 'PI-527710' did not significantly differ (Table 1). Whereas in summer genotypes 'PI-537636' and 'PI-527710' had the lowest and highest protein of 19 and 30%, respectively (Table 1). Genotype 'PI-537636' had no significant difference with all other genotypes studied with exception of 'PI-527710' and 'Gila' (Table 1).

Safflower Leaf Mineral Analysis
Safflower genotypes had a significant (P < 0.05) effect on leaf mineral content (P, K, Mg, Ca, Na) of safflower grown in both seasons (Table 2). Phosphorus content for winter harvested leaves significantly (P < 0.05) ranged between 1-3 mg g -1 , during this season all genotypes were not significantly (P > 0.05) different except for 'Kiama' which had the lowest P value of 1 mg g -1 (Table 2). Summer grown genotypes significantly (P < 0.05) differed in phosphorus content at a range between 1-2 mg g -1 , all genotypes had no significant difference with exception of 'PI-527710' (Table 2). Genotype 'PI-527710' had the highest P value in both seasons (Table 2).
In winter, leaf calcium content significantly (P < 0.05) ranged between 1-3 mg g -1 (Table 2). Genotype 'PI-527710' produced the highest calcium content of 3 mg g -1 , however it did not significantly (P > 0.05) differ from all other genotypes except 'Kiama' which had the lowest calcium value. Safflower calcium level of summer harvested leaves significantly (P < 0.05) ranged between 3-5 mg g -1 among genotypes (Table 2). Genotype 'Sina' had the lowest calcium content compared to other genotypes, however it did not significantly (P > 0.05) differ from 'Kiama' (Table 2). Similarly, calcium content of genotypes 'PI-527710' and 'PI-537636' had no significant difference ( Table 2). The highest calcium content of 5 mg g -1 was produced by genotype 'Gila' in summer ( Table  2).
The sodium content significantly (P < 0.05) ranged between 4-8 and 4-6 mg g -1 in winter and summer conditions, respectively (Table 2). During winter, the lowest and highest leaf sodium content was from genotypes 'Kiama' and 'PI-537636', however 'PI-537636' was not significantly (P > 0.05) different to all other genotypes except 'Kiama' (Table 2). In summer, genotypes 'PI-527710' and 'Kiama' had significantly (P < 0.05) lowest and highest sodium content (Table 2). Similar to winter, all genotypes in summer were not significantly different in leaf sodium content except for 'PI-527710' (Table 2). In general, winter harvested leaves were 19% higher in average leaf sodium content compared to summer across genotypes ( Table 2).
The leaf Na:K ratio significantly (P < 0.05) ranged below 1 in both seasons, however winter genotypes had no significant difference (Figure 1). Minimum and maximum Na:K ratio of summer was from genotypes 'PI-527710' and 'Kiama', all other genotypes (PI-537636, Sina, Gila) were not significantly (P < 0.05) different (Figure 1). Ca:P levels significantly (P < 0.05) ranged from 1-3 irrespective of genotypes in winter and summer (Figure 1). Genotypes did not significantly (P > 0.05) differ in leaf Ca:P ratio except for 'Kiama' which had the highest Ca:P ratio (Figure 1). In summer, the lowest ratio was from genotype 'PI-537636', however it had no significant difference from that of genotypes 'Sina' or 'PI-527710'. Genotype 'Gila' recorded the highest leaf Ca:P ratio of 3 ( Figure 1). Figure 1. Effect of genotype on leaf mineral ratio of safflower planted in winter and summer; bars with the same letter are not significantly different; mean separation by LSD at P = 0.05

Discussion
The genotypic difference in proximate and mineral nutrition of safflower across seasons indicate that nutritional benefit on consumption of safflower leaves varies depending on the genotype. The variation is attributed to genetic differences in the genotypes under investigation in response to environmental conditions during winter and summer. Safflower genotypes differ in growth habits, morphology and physiological performance, this differences affect the source-sink strength hence causing differences in mineral nutrition and its partitioning. However, despite the differences most genotypes had shown to be good source of mineral nutrition indicating a sufficient nutritional distribution among safflower leaves. This was consistent with the findings of Suneel-Kumar et al. (2016) who reported variation on leaf nutrition (calcium, iron, crude fibre, ash, protein and moisture content) as affected by genotype and growth stage. Different authors studying mineral composition of leafy vegetables reported significant genotypic difference, which is influenced by genetic factors, crop growth habits, cultural and environmental conditions (Chope & Terry, 2009;Rouphael et al., 2012;Kunyanga et al., 2013).
Safflower leaves has an average crude protein content of 24-27% with genotype 'Kiama' recording the highest protein level compared to other genotypes in winter or summer. This means 100 g of safflower leaves at dry weight basis is twice the recommended intake for children and 43-79% that of adults. Pearson (1976) confirmed that plant foods should contain more than 12% of its caloric value as a good source of protein. Food and nutrition board for Institute of medicine (2002) reported that the daily required intake of crude protein ranges between 9-13 g and 34-56 g for children and adults, respectively. This indicates that any safflower leaf genotype studied harvested in either winter or summer are nutritionally sufficient as a protein source for consumption to improve protein deficiencies. In both seasons, any safflower genotype studied had an optimal leaf moisture content of a range between 85-89% indicating its freshness and adequacy of water soluble vitamins (Adinortey et al., 2012).
Despite the significant (P < 0.05) variation among genotypes towards mineral composition, potassium and magnesium were higher compared to other elements. The lowest mineral content among the genotypes was phosphorus at an average of 2-3 mg g -1 across seasons. Potassium as the abundant mineral in safflower leaves significantly varied among genotypes with the lowest and highest value ranging between 11-21 mg g -1 in winter and summer. The safflower leaves are a good source of potassium as 100 g leaves of genotypes studied harvested in any season can contribute a range between 60-105% K daily intake as per daily recommended allowance for adults (The National Research Council, 1989). Potassium is beneficial for hypertension patients (Arinathan et al., 2003) and has a synergistic relationship with other essential minerals as it was reported to increase the utilisation of iron in the body (Adeyeye, 2002).
Safflower genotypes can provide a significant amount of leaf dietary calcium. Suneel-Kumar et al. (2016) found the same results and reported 2-3 mg g -1 Ca in safflower leaves. The values are consistent to those of other leafy vegetables studied including okra (1.1-3.1 mg g -1 ) and Amaranthus (2.64 mg g -1 ) (Gemede et al., 2015;Kunyanga et al., 2013). The National Research Council (1989) reported the recommended daily allowance of 8mg per 100g per day of calcium for both adults and children. In the current study, safflower leaves can provide 41-64% of required daily intake in every 100 g depending on the genotype. Calcium concentrations are adequate for blood coagulation and integrity of intracellular cement substances (J. C. Okaka & A. N. O. Okaka, 2011).
Safflower leaves (100 g) provides 18-42% or 26-59% of sodium for adults and children, respectively depending on the genotype and season of growth. This is a sufficient amount to be incorporated in daily diets for human consumption to meet the required intake as reported by Food and Nutrition Board of National Academy of Sciences (2002).
There is a recommended mineral ratios essential for intake restrictions on individuals with nutrition associated diseases. People suffering from acute or chronic renal failure under haemodialysis require controlled potassium intake levels (Copetti et al., 2010). The study at present indicates that safflower leaves contains Na:K ratio of 0.2 to 0.5 among genotypes, this value is within the recommended Na:K ratio in human diets which is reported to be less than 1 (Copetti et al., 2010). Based on this ratios, the consumption of safflower leaves may be adequate for hypertension patients to lower the blood cholesterol (Adinortey et al., 2012;Sodamade et al., 2013) and normal protein retention during growth stages (Nieman et al., 1992). Vegetables are considered adequate for consumption when Ca:P ratio is above 1, while the Ca:P ratio below 0.5 means an inadequate mineral source. Safflower genotypes studied contain Ca:P ratio of 1-3 depending on season. This indicates that safflower leaves are essential for consumption by children and lactating women essential for maintenance of bone, teeth and muscles (Turan et al., 2003;Sodamade et al., 2013).

Conclusion
Safflower as a leafy vegetable is an important food source, adequate for human consumption due to its nutritional value which meets the nutritional recommended daily allowances for different age and/or health groups. All genotypes of safflower studied significantly varied and influenced the leaf proximate and mineral nutrition resulting with a high crude fibre and protein content of 18% and 37% from genotypes 'Sina' and 'Kiama' respectively. In terms of mineral nutrition, genotype 'PI-527710' was the most nutritious with the highest P, K, Ca, Na depending on season as compared to other genotypes. This indicates that safflower genotypes under study can be consumed as a green leafy vegetable in any season or used for value addition to improve quality and nutritional value of other food formulation for infants, adults and nutrition deficient groups. The study suggests that consumption of safflower leaves can enhance food security, increase farmer revenue and be recommended for use in supplementary diets such as hypo-cholesterol diets, consumption by groups with coronary disease problems and protein malnutrition in infants.