Genetic diversity for nutritional traits in the leaves of baobab , Adansonia digitata

1 Department of Plant Science, Institute for Agricultural Research, Ahmadu Bello University, Zaria, Nigeria. 2 Department of Crop Protection, Institute for Agricultural Research, Ahmadu Bello University, Zaria, Nigeria. 3 Department of Agronomy, Institute for Agricultural Research, Ahmadu Bello University, Zaria, Nigeria. 4 Department of Biological Sciences, Kaduna State University, Kaduna, Nigeria.


INTRODUCTION
Adansonia digitata (L.) commonly referred to as baobab belongs to the family of bombacaceae (Keay, 1989), consisting of about eight species (Baum and Oginuma, 1994), and that only one of the species exists in Africa.A. digitata originated in Madagascar and was introduced to the rest of African countries by long distance dispersal before the breaking of West Gondwanaland block through Continental drift at Cretaceous.(Assogbadjo et al., 2006;Baum et al., 1998).In Nigeria, A. digitata is found in all the ecological zones.Keay (1989) reported that the tree is evenly distributed in arid and semi-arid zones of Nigeria.It is reported to be widely distributed in savannah regions of Nigeria, particularly north west and north east regions, where it is called Kuka tree (Esenowo, 1991;Baum 1995;Ibrahim and Otegbeye, 2004).It is also found in the southern region of Nigeria where the Yorubas call it Ose tree, Nupe (Muchi), Edo (Usi) (Keay, 1989).The tree is used in parkland agro-forestry system in the arid and semi regions of Nigeria where it grows undistributed and survives until their natural death.The tree is also grown in courtyards of villages and cities of Northern Nigeria.Baobab seems to be the most protected tree in the arid and semi-arid zones of Nigeria.Esenowo (1991) noted that the tree is protected by those who know its value while ICRAF (2003) stated that in hot and dry part of the world, where the average per capita income may not reach 300 USD a year, the tree is vital for food security and that the tree is cherished and even worshipped.
The tree contributes significantly to the socio-economic activities of the people of Northern Nigeria.Recently, attention has been diverted on the use of indigenous trees which may be due to the problems obtained in the use of exotic trees agro-forestry system.The problems which, resulted into poor acceptability of the exotic trees by the farmers, could be, among many, unfavorable competition with the crops for nutrient, water, and sunlight, more importantly, is the problem of allelopathic influence (Momodu et al., 1997).In contrast, indigenous multipurpose tree species (MPTS) such as A. digitata have evolved and adapted over time to the local farming systems and the additional advantages of multipurpose uses in the production of fuel-wood, edible fruit, leaves, oils, tannin and medicaments (Ibrahim and Otegbeye, 2004).Many publications and calls have been made in recent times on the need for the domestication of indigenous multipurpose tree species (Momodu et al., 1997;ICRAF, 1997;ICRAF, 2003;Ibrahim and Otegbeye, 2004), but there is no any practical action taken in this direction, as there is little or no research study on the genetic and nutritional improvement of this tree in Nigeria (Ibrahim and Otegbeye, 2004).This is probably the first genetic and variability study on an indigenous tree species in Nigeria.Most of the edible indigenous MPTS are becoming endangered due to large-scale deforestation of the natural ecosystem.This problem may affect the reproductive potential, species extinction and genetic resources, which would be used for future, tree improvement (ICRAF, 2003).According to Stephenson et al. (2000), malnutrition is the most important cause of mortality in the global human population.More than a quarter of children less than five years old suffer from protein-energy malnutrition, as determined by rates of stunting and underweight.Of these, 70% are in Asia, 26% in Africa, and 4% in Latin America.Stunting in poor-resource population is usually associated with reduced mental development A. digitata leaves have been reported to be a low to moderate source of pro-vitamin A (Delisle et al., 1997;Nordeide et al., 1996) and very rich in calcium (Glew et al., 1997;Becker, 1983).Leaf crude protein values for A. digitata have been reported in the range of 10.6% (Yazzie et al., 1994) to 12.5% (Kerharo, 1980).Leung et al. (1968) also reported that the nutritional composition of fresh baobab leaves per 100 g edible portion is: water 77 g, energy 289 kJ (69 kcal), protein 3.8 g, fat 0.3 g, carbohydrate 16.1 g, fibre 2.8 g, Ca 402 mg, P 65 mg, ascorbic acid 52 mg.Vos et al., (1995) used amplified fragment length polymorphism (AFLP) analysis to find the intra-specific genetic diversity of those locally recognized morphotypes and on the whole populations in order to assess the genetic diversity and differentiation within and between baobabs in Benin.Assogbadjo et al. (2010) used this AFLP technique to investigate the traditional classifications of A. digitata confirmed by genome-level genetic differentiation and also if there is within the species some genetic variations which should be conserved for the benefit of local people.
A lot of researches have been conducted on the nutritional value of baobab leaf compared to other fruit trees.This study evaluates the nutritional variability and genetic diversity among different populations of A. digitata trees at seedling stage with the aim of evaluating if the seedlings could be used as garden crop.

MATERIALS AND METHODS
Locations for the collection of population samples were selected based on the survey conducted by Momodu et al. (1997).These sites were surveyed to have prevalence of high population of A. digitata where high genetic variability is obtained in the arid and semi-arid zones of Nigeria.
Fruits samples were collected from 36 populations in North West and North East Nigeria (Table 1).Eight mother trees were randomly selected from each of the populations and four fruits were collected from round the crown of each of the sampled mother trees (four fruits per tree) in each of the populations.This is to ensure that fruits from different pollen sources were collected.The collected fruits from each population were then pooled together and stored in a well-ventilated seed room at optimum temperature of 25°C after which the seeds were then removed, washed properly and subjected to pre-germination treatment; soaking the seeds in a boiled water containing one teaspoon of potash for 1 h.Seeds from each population were sown in polyethylene pots containing mixtures of cow dung, river sand and top soil in ratio of 2:2:1, respectively, in the nursery of Savanna Forestry Research Station, Zaria.(11° 11'N, 07° 38'E, attitude 688 m above sea level).The experiment was replicated three times in complete randomized design (CRD).
The pots were watered daily morning and evening and number of germinated seeds were counted daily till 30 th days after sowing.Agrocivicultural treatment was applied to promote seedling growth such as weeding, watering and application of Basagram to check termite activities.Proximate analysis of the leaves was conducted at the end of 16 WAS.The lower leaves of the seedlings per population were harvested and bulk and then taken to the Food Science Technology of the Institute for Agricultural Research Ahmadu Bello University, Zaria for the proximate analysis following the method described by AOAC (1980).Each treatment was repeated three times.The data were statistically subjected to analysis of variance and mean ranking using Domcan multiple range test (DMRT) to determine the variability that exists among the populations.The genetic diversity and genetic distance of 36 genotypes (populations) were estimated using Mahalanobis D 2

RESULTS AND DISCUSSION
Highly significant differences (P0.01) was found in all the traits across the populations (Table 2) indicating that variability exists in the nutritional traits of the leaves and as such selection of genotypes with superior nutritional traits in the leaves could be made.This result is in conformity with the works of Sidibe et al. (1996) who noted that variability of vitamin C content exist in the pulp of A. digitata fruits among the trees assessed in Mali.Similarly, Sidibe et al. (1998) recorded difference in vitamin A content and other minerals in the leaves of different trees of baobab.Proximate analysis conducted by Odetokun (1996) shows similar result in the pulp.
The mean separation and ranking among the populations is consistent with the variability observed (Table 3).Jarkasa population had the highest ash content, while the population from Azare had high protein content.Kukawa/Ganjarma and Dutsin Ma populations had high   fiber and carbohydrate, respectively.Multiple crosses can be made from these populations to obtain desirable hybrid.In order to develop genotype with superior nutritional quality for crude protein, carbohydrate and ash content, multiple crosses can be made from the populations such as Jarkasa, Azare, Kukawa and Dutsin Ma.
Mean inter and intra cluster values (D 2 ) and distance (D 2 ) using Mahalanobis D 2 statistics are presented in Tables 4 and 5.The study grouped the 36 populations into four clusters I-IV with 25, 3, 7 and 1 population(s) respectively, as displayed in dendrogram (Figure 1).The highest cluster value, 34.806, was found to be between Clusters IV and I and with intra-cluster distance of 5.900.Clusters II and I had the least values of D 2 and D 2 as 11.602 and 3.406, respectively.Furthermore, cluster I had the highest intra-cluster D 2 value and intra-cluster distance D 2 , 66.534 and 8.157, respectively, and consequently had the highest population (25) followed by cluster III with intra-cluster D 2 and D 2 as 16.88 and 4.109, respectively, with seven populations.However, cluster IV was observed to be an isolated cluster with only 1 population and 0 intra-cluster values.
The inter-cluster proximity of nutritional traits dendrogam (Figure 1) that was found to be minimum in clusters I, II, III and isolated in cluster IV is due to genetic influence as the genotypes in population 4 (Song) might not be affected by migration of the agents of dispersal.This result is in agreement with the result of Zehdi et al. (2005) who reported narrow genetic diversity in date palm and that the clusters were grouped independent of their geographical origin.The isolation of the cluster 4 (Song) could be as a result non palatability of the fruits to agents of dispersal and pollination of the trees.Similarly, Sie et al. (2005) obtained three clusters of 79 genotypes of Cola nitida, which did not merge with the geographical origin,  (1997) on Acacia aulacocarpa, Wickneswari and Norwati (1993) on Acacia auriculiformis and House and Bell (1996) on Eucalyptus pellita had their centre of dispersion merged with their centre of origin, showing that the clusters were grouped alongside their centre of origin.Hybrids of A. digitata could be between clusters that are highly divergent or with higher genetic distance as this will reduce the danger posed by the harsh environment, pest, and diseases and drought.The nutritional problems reported by Obilana (2011) energy and protein deficiencies for Nigeria are 42% stunting, 11% wasting and underweight accounted for 24%.This could be alleviated in these zones by selection and hybridization among the populations with superior traits to improve the nutritional status of the trees.

Figure 1 .
Figure 1.Dendogram showing populations of Adansonia digitata grouped into four clusters based on nutritional traits.

Table 1 .
List of the locations of population samples collected, agroecology coordinates and their serial number.

Table 2 .
Analysis of Variance for Nutritional Traits of Leaves of A. digitata Seedlings at 16 WAS.

Table 3 .
Result of proximate analysis for leaves of A. digitata for all the 36 locations at 16 WAS.

Table 4 .
Mean intra and inter-cluster D 2 values for nutritional traits of the leaves of Adansonia digitata for the 36 populations.

Table 5 .
Mean intra and inter-cluster (D 2 ) values for nutritional traits of the leaves of Adansonia digitat for the 36 populations.showingthatthestudy sites were not dependent on centers of dispersion.Although, contrary to this result,Montagnon et al. (1991)on coffee, McGranahan et al.