0291-B2

Phylogenetic Relationships among African Populations of Faidherbia albida (Del.) A. Chev: a Review.

Otto George Dangasuk ¹

ABSTRACT

Genetic variability studies of 16 provenances of Faidherbia albida from different regions of Africa were conducted based on the characteristics of seeds, seedlings growing at two sites in Kenya, six months and four years old trees growing in semi arid Baringo District of Kenya, and enzyme electrophoresis. The primary objectives of the studies were (1) To determine the phylogenetic relationships among the Western, Eastern and Southern African provenances of F. albida, and (2) To determine the extent of genetic diversity in F. albida populations that occupies different regions of Africa. Such information could be useful for management, conservation and tree improvement of this endangered tree species in different regions of Africa. However, during the course of these studies unexpected relationship was discovered in that Bignona (Senegal) a coastal, lowland and high rainfall provenance of West African Origin was genetically closer to East and Southern African than fellow West African provenances. In the majority of cases it was genetically similar to Rama (Ethiopia) a riparian, highland and moderate rainfall provenance of East African origin.

INTRODUCTION

Faidherbia albida a leguminous woody species is distributed throughout the arid and semi arid lands (ASALs) of Africa. F. albida is an important agro-silvo-pastoral tree species in Sahelian Africa, due to its unique reverse phenology of shedding leaves during the rainy season at the time of higher microbial activities in the soil and when such litter decomposes, they improve the soil structure, stability and permeability, thus providing micro-environment favorable for crops. The retained leaves in the dry season provide shade and green fodder rich in protein and carbohydrates for livestock and also mulch which reduces evaporation thus conserving the available soil moisture. F. albida has a remarkable capacity for recycling nutrients from underground to the surface due to its very deep root system (Le Houerou, 1980). F. albida does not compete with inter-planted crops for soil nutrients as it enters a period of physiological rest during the normal crop growing season (ICRAF, 1989). The tree also stabilizes sand dunes and prevents soil erosion (Dancette and Poulain, 1969). In addition increase in yield from crops grown below the trees has been attributed to increased fertility due to nitrogen fixation, dung from stock browsing and fallen leaves and pods (Radwanski and Wickens, 1967). It is also used for timber, fuelwood, medicinal purposes and seeds are food for human (Wickens, 1969).

However, drought and human activities have endangered the genetic resources of F. albida warranting its inclusion among endangered species and giving it priority status for conservation and study (FAO, 1977). Owing to its importance in the ASALs of Africa, an attempt to domesticate it was once made in West Africa (Kirmse and Norton, 1984), but this failed, probably due to lack of understanding of its genetic diversity and breeding biology. Conservation of germplasm of F. albida requires a closer assessment of its genetic variation within and between populations covering its natural distribution in Africa. In the previous studies we tried to understand the phylogenetic relationships among the Western, Eastern and Southern African provenances of F. albida, and to determine the extent of genetic diversity among populations that occupy different regions of Africa. This was intended to form the basis of the long term goal of developing F. albida ecotypes adapted to different agro-ecological zones in order to provide adequate fodder, soil improvement, fuelwood, medicinal extracts and food in the ASALs of East African region. This paper reports on an important discovery during the course of these studies of the unique, persistent and consistent genetic closeness of Bignona (Senegal) provenance from West Africa to East African provenances in general and Rama (Ethiopia) from East Africa in particular in all the experiments conducted.

MATERIAL AND METHODS

Plant materials: sampling strategy and origin

Bulked seed of 16 provenances of F. albida from different geographic regions of Africa were used in this study (Fig.1 and Table 1). Seeds were collected from 25 or more trees spaced at least 100m apart from which comparable amount was bulked to provide 5 - 10 kg of seed.

Morphological variation in seed characteristics among 12 F. albida provenances

The seeds were oven dried at 35°C for 24 hours to create uniformity in moisture content. Two hundred and fifty randomly selected seeds formed an experimental unit in a completely randomised design (CRD) replicated four times. Length, width and thickness of 1000 seeds per provenance were measured using vernier callipers to the nearest 0.02mm. Weight of 1000 randomly selected clean seeds per provenance was determined using a weighing balance to the nearest 0.0001gm (Dangasuk et al. 1997).

Genetic variation in the nursery seedling traits among 16 provenances at Marigat

Marigat is located at 00°28'N, 36°00'E and 1000m.a.s.l. With a mean annual rainfall of 650 mm, mean annual maximum and minimum temperatures are a bout 30°C and 18°C, respectively. Sixty clean seeds per provenance were treated by nipping manually and each seed sown in a polythene bag measuring 5.5 x 18 cm, filled with forest soil and sand mixed in a ratio of 2:1, respectively. The trial was laid out as CRD consisting of 30 seedlings per provenance and with the16 provenances giving a total of 480 experimental units. Seedling height was measured using a meter rule 30, 60, 90 and 120 days after germination. Stem collar diameter was measured using vernier callipers only in the third and fourth month (Dangasuk, 1999).

Genetic variation six months and four years after planting in the field at Noiweit

Noiweit located at 0°10'N, 36°00'E and 1500 m.a.s.l. The mean annual rainfall ranges between 500 and 800 mm and the area has a mean temperature of 21°C. The seedlings were planted in a randomised complete block design (RCBD) with repeated observations. There were five blocks (replicates) and each block had 16 experimental units representing each of the 16 provenances, each unit had two seedlings of the same provenance. In total there were ten seedlings per provenance in the whole design at a spacing of 3m x 3m.

Six months after planting height was measured using a meter ruler, stem collar diameter was measured using vernier callipers and branches were directly counted (Dangasuk, 1999). Four years after planting, height was determined using a clinometer and a tape measure. Diameter at breast height (DBH) was measured using a DBH meter. While survival percentage was assessed by direct count of live tree per provenance, which was then computed as a percentage of the total number of trees per provenance planted (Dangasuk and Wachira, 2001).

Genetic variation based on allozyme analysis of 16 F. albida provenances

The enzymes were extracted according to Murphy et al (1996) and Wendel and Weeden (1989). Briefly, sixty clean seeds per provenance were sterilized by immersing in hydrogen peroxide (30%) for 30 min., rinsed with distilled water, nipped to permit water absorption and germinated at a controlled temperature of 30°C for 48hours in a Sanyu Growth Cabinet (Model: MLR 3500). The germinants with tap roots greater than 1.5 cm were washed in double distilled water and homogenized in extraction buffer Tris-HCl pH 7.8 (0.1M Tris base, 0.01M potassium Chloride, 0.01M Magnesium Chloride, 0.001M EDTA Sodium salt, 5% PVP-40, 20% PVpP, 1% BSA and 1% Mercaptoethanol). Discontinuous starch gel electrophoresis of Murphy et al. (1996) and Wendel and Weeden (1989) were employed. Electrode buffer consisted of Tris-Citrate pH 7.0 (0.125M EDTA, 0.05M L-Histidine HCl and 0.1M Tris base) and gel buffer was made up of EDTA-HCL-Tris pH 7.0 (0.0014M EDTA, 0.05M L-Histidine HCL and 0.1M Tris base).

Six enzyme systems including alcohol dehydrogenase (ADH EC 1.1.1.1), malate dehydrogenase (MDH EC 1.1.1.37), isocitrate dehydrogenase (IDH EC 1.1.1.42), 6-phosphogluconate dehydrogenase (6-PGDH EC 1.1.1.44), glucose-6-phosphate dehydrogenase (G-6-PDH EC 1.1.1.49) and alpha esterase (a-EST EC 3.1.1.1) were assayed in a 12% starch gel electrophoresis system. The isozymes were visualised according to the staining procedures described by Wendel and Weeden (1989). The locus specifying the most anodally migrating isozyme was designated as 1 the next as 2 and so on according to Murphy et al (1996). Within each locus the most common allozyme was given the arbitrary score value of 100 and the other allozymes were given score values according to their migration relative to the common allozyme (Murphy et al. 1996). Only strongly staining and consistent bands were scored (Dangasuk and Gudu, 2000).

Statistical analysis

One way ANONA was used for the analysis of variance and whenever the F-value was significant at (P < 0.05), Duncan's Multiple Range Test (DMRT) were used to indicate levels of differences between means for laboratory, nursery and field data using SAS, 1991 package. BIOSYS-1 statistical package was used to analyse the isozyme data.

RESULTS

Morphological variation in seed characteristics among 12 provenances

Although seed length showed significant differences (P < 0.05) between Bignona (Senegal) and Mwembe (Tanzania), Bignona was closer to Mwembe than to the next fellow West African provenance of Tera (Niger) (Table 2). Seed width showed significant difference (P < 0.05) between Bignona (Senegal) and Rama (Ethiopia), but the data showed that Bignona was closer to Rama than Tera (Niger) from West Africa (Table 2). As regards seed thickness, Bignona was completely integrated among the East and Southern African provenances and significantly different from the fellow West African provenances. Again it was evident in seed weight that Bignona was closer to Rama (Ethiopia) than the nearest West African provenance of Moulvouday (Cameroon).

Genetic variation in the nursery seedling traits among 16 provenances at Marigat

Stem collar diameter showed statistical similarity between Bignona (Senegal), Lake Awassa (Ethiopia) and Rama (Ethiopia) ninety days after germination. Like wise Bignona and Rama were statistically similar 120 days after germination (Table 3). Generally, as regards Stem collar diameter, Bignona identified more with East African than fellow West African provenances.

Genetic variation six months and four years after planting in the field at Noiweit

The height growth at six months after planting showed that Bignona (Senegal) statistically identified more with the West African than East African provenances. Similarly, collar diameter development at six months after planting indicated that Bignona was part of the West African group. However, as regards branch number Bignona (Senegal) from West Africa was integrated among East African provenances (Table 4).

Four years after planting all the West African provenances died except Bignona (Senegal) which survived just like the East and Southern African provenances. It was statistically similar to the East and Southern African provenances in height and diameter growth (Table 4).

Genetic variation based on allozyme analysis of 16 F. albida provenances

The UPGMA dendrogram of phylogenetic relationship based on Modified Roger's distance (Wright, 1978) calculated from allelic frequency data showed two distinct clusters (Fig. 2). Group one is composed of East and Southern African provenances except Bignona (Senegal) provenance of West Africa which is peculiarly categorized in this group. It is also evident from the dendrogram that Bignona (Senegal) is genetically closest to Rama (Ethiopia) of Eastern Africa. Group two is composed exclusively of West African provenances except for Debre Zeit (Ethiopia) of East African origin.

DISCUSSIONS

In general, the geographically more distance populations are also more distance genetically and vice versa. The peculiar genetic closeness of Bignona (Senegal) to the East and Southern African instead of fellow West African provenances in F. albida was evident in all the quantitative and qualitative data of this study. Which seems to suggest that Bignona (Senegal) population could indeed be genetically closer to Eastern and Southern African than West African populations. This tends to support the conclusion of Rocha and Lobo (1996) that factors other than geographical isolation are involved in genetic differentiation of populations.

According to Ross (1966) and Wickens (1969) in the East and Southern African distributions range, F. albida is associated with riparian habitats (i.e. high ground water level) and moderate to high amount of annual rainfall. While in the harsher Sahelian conditions of West Africa (i.e. a short rainy season of low and erratic rainfall, followed by a long dry season) it occurs away from rivers and watercourses. Bignona (Senegal) population although found in the Sahelian West Africa, is located at the Sea shore (10 m above sea level) and hence has high ground water level. Secondly, it receives a relatively very high amount of annual rainfall. This seems to suggest that adaptation to the two common factors, namely; association with watercourses (i.e. high ground water level) and high annual rainfall could explain the genetic closeness of Bignona (Senegal) to the East and Southern African provenances of F. albida.

According to Dangasuk (1999) genetic diversity in F. albida is greatest in populations where the mean annual rainfall is between 500 mm and 1000mm and lowest in populations found in extreme environments (i.e. areas of lowest and highest mean annual rainfall). Chase et al (1995) also reported decreasing mean heterozygosity in areas with increasing annual rainfall in Cordia alliodora (Boraginaceae). Since, Joly et al (1992), Harris et al (1997) and Dangasuk and Gudu (2000) all reported higher genetic diversity among West African than East and Southern African populations of F. albida. It could be that adaptation to one single factor, the comparatively high amount of rainfall (1408 mm) in Bignona which might have contributed considerably in reducing the genetic diversity of this West African provenance to the level of Eastern and Southern African provenances.

Based on differences in ecological conditions, environmental factors, geographic locations and distance between Bignona (Senegal) and Rama (Ethiopia) it would be expected that natural selection could generate considerable genetic differentiation and diversity between the two populations. Secondly, Bignona (Senegal) being at the limit of the species' distribution (i.e. bordering the Sahara desert in the north and the Sea in the west) was expected to experience high selection pressure. And therefore should represent a highly specialized population, genetically distinguishable from the rest of the provenances in general and Rama (Ethiopia) in particular. Contrarily, according to this study the two populations are very close genetically for reasons not immediately clear. However, Vandenbelt (1991) and Joly (1992) reported the uniqueness of the Ethiopian provenances of F. albida which made them suspect that Ethiopia is the possible species center of origin containing the greatest genetic diversity of the species. This was further strengthened by the findings of Ibrahim (1996) and Dangasuk et al (1997) that Rama (Ethiopia) seems to share characteristics of both West African and Eastern and Southern African populations. This might be a possible explanation of the genetic similarity between Rama (Ethiopia) and Bignona (Senegal) provenances.

The comparatively, high survival percentage and impressive growth performance of the Bignona (Senegal) provenance of West African origin in the East African environment four years after planting. Coupled with its genetic closeness to the East African provenances increases its importance and potential for domestication and genetic improvement in the East African region for use in agroforestry programs.

CONCLUSIONS

The comparatively high amount of rainfall and nearness to the sea created a genetically unique provenance of F. albida in Bignona (Senegal) relative to other West African provenances. Consequently, Bignona provenance is genetically closer to East and Southern African provenances which are of riparian origin with moderate to high amount of annual rainfall, than to fellow West African provenances which occur away from water courses in areas of limited and erratic rainfall. It could therefore be concluded that distribution of rainfall and ground water exerts considerable influence on the genetic differentiation of populations and distribution of gene pool throughout the natural distribution range of F. albida in Africa.

The genetic closeness of Bignona (Senegal) to Rama (Ethiopia) showed in almost all the morphological and isozyme data. In addition, to the categorization of Debre zeit (Ethiopia) among the West African provenances in the isozyme data, tends to confirm the conclusion that Ethiopia is the species' center of origin and its provenances harbors the greatest genetic diversity in F. albida. Consequently, Ethiopian provenances particularly Rama seems to share characteristics of both West African and Eastern and Southern African provenances.

Such information may be important indicators of the requirements for better establishment and performance of F. albida in different regions of Africa and also useful for germplasm conservation in selecting useful genotypes for domestication and agroforestry programs.

REFERENCES

Dancette, C. and J.F. Poulain, 1969. Influence of Acacia albida on pedoclimatic factors and crop yields. African soils. 14(1-2):143-184

Dangasuk, O.G., P. Seurei and S. Gudu, 1997. Genetic variation in seed and seedling traits in12 African provenances of Faidherbia albida (Del.) A.Chev. at Lodwar, Kenya. Agroforestry Systems 37:133-141

Dangasuk, O.G., 1999. Genetic diversity in Faidherbia albida (Del.) A. Chev. Ph.D. Thesis. Moi University, Eldoret, Kenya.

Dangasuk, O.G. and S. Gudu, 2000. Allozyme variation in 16 natural populations of Faidherbia albida (Del.) A. Chev. Hereditas 133:133-145

Dangasuk, O.G. and M.R. Wachira, 2001. Interaction between soil properties and 16 Faidherbia albida provenances four years after planting in semi arid Baringo district of Kenya. BOT: 421. Special Project Report. Department of Botany, Moi University, Eldoret, Kenya. 29pp.

Chase, M.R., D.H. Boshier and K.S. Bawa, 1995. Population genetics of Cordia alliodora (oraginaceae) a neotropical tree. 1. Genetic variation in natural populations. Amer. Jour. Bot. 82(4): 468-475

FAO (Food Agricultural Organization) 1977. Report of the fourth session of the FAO Panel of Expert on Forest Genetic Resources. FO:FGR/4/Rep. Rome Italy.

Harris, A.S., C.W. Fagg and R.D. Barnes, 1997. Isozyme variation in Faidherbia albida (Leguminosae, Mimosoideae). Pl. Sys. Evol. 207: 119-132

ICRAF (International Center for Research in Agroforestry), 1989. The apple ring. Agroforestry Today, 1:11-12

Joly, H.I., M. Zeh-Nlo, P. Danthu and C. Aygalent, 1992. Population genetics of an African acacia: Acacia albida . Genetic diversity of populations from West Africa. Aust. Jour. Bot. 40:59-73

Le Houerou, H.N., 1980. Chemical composition and nutritional value of browse in tropical West Africa. In: Le Houerou (ed.) Browse in Africa the current state of knowledge. ILCA, Ethiopia, p. 261-289

Wendel, J.F. and N.F. Weeden, 1989. Visualisation and interpretation of plant isozyme. Pages 4-45. In: Soltis, D.E. and Soltis, P.S. (eds.). Isozyme in Plant Biology. Chapman and Hall, Ltd. London.

Wickens, G.E., 1969. A study of Acacia albida Del. (Mimosoideae). Kew Bulletin 23(2):181-202

TABLES

Table 1. Collection site data for 16 provenances of F. albida used in this study.

¹ Department of Forestry, Moi University,
P. O. Box 1125, ELDORET, Kenya.
Tel.: +254-321-63111 ext 364.
Email: [email protected]