Senegalia Senegal response to inoculation with rhizobial strains vary in relation to seed provenance and soil type

Niokhor Bakhoum, David W. Odee, Dioumacor Fall, Fatou Ndoye, Aboubacry Kane, Jacinta M. Kimiti, Alzouma M. Zoubeirou, et al.

Plant and Soil An International Journal on -Soil Relationships

ISSN 0032-079X Volume 398 Combined 1-2

Plant Soil (2016) 398:181-193 DOI 10.1007/s11104-015-2655-6

1 23 Your article is protected by copyright and all rights are held exclusively by Springer International Publishing Switzerland. This e- offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23 Author's personal copy

Plant Soil (2016) 398:181–193 DOI 10.1007/s11104-015-2655-6

REGULAR ARTICLE

Senegalia Senegal response to inoculation with rhizobial strains vary in relation to seed provenance and soil type

Niokhor Bakhoum & David W. Odee & Dioumacor Fall & Fatou Ndoye & Aboubacry Kane & Jacinta M. Kimiti & Alzouma M. Zoubeirou & Samba Nd. Sylla & Kandioura Noba & Diégane Diouf

Received: 22 April 2015 /Accepted: 25 August 2015 /Published online: 2 September 2015 # Springer International Publishing Switzerland 2015

Abstract Results Nodulation and growth of S. Senegal var- Aims The focus of the study was to determine the sym- iedinrelationtorhizobialstrain,provenance,soil biotic and growth response of three Senegalia Senegal type, and their interactions. Generally, nodulation (Syn. Acacia Senegal, gum arabic tree) provenances, was higher in Dahra than Goudiry soils, while namely Dahra (Senegal), Tera (Niger) and Makueni Makueni provenance was the most compatible (Kenya) to inoculation with selected S. Senegal- host. Inoculation had a significant effect on all nodulating rhizobia in soils from Dahra and Goudiry parameters measured in Dahra field soil. By con- regions of Senegal, representing typical soil and trast, inoculation had a significant effect on height enviromental conditions for establishing gum arabic (shoot length), and shoot, and total dry matter production plantations. but not on nodulation. In the two field soils, seed Methods A greenhouse experiment was performed to provenance effect was significant for all parame- evaluate the effect of 11 rhizobial strains on nodulation ters measured. The interaction between inoculation and growth of three S. Senegal provenances in two field and provenance showed a significant effect on all soils, differing in nutrient status and indigenous parameters measured except nodule number in rhizobia. After 4 months, were harvested for Dahra field soil while in Goudiry, the interaction determination of nodulation, shoot and root dryweight. had a significant effect on seedling height and

Responsible Editor: Henk Schat.

N. Bakhoum (*) : F. Ndoye : A. Kane : S. N. Sylla : D. W. Odee K. Noba : D. Diouf Centre for Ecology and Hydrology, Bush Estate, Penicuik, Département de Biologie Végétale, Faculté des Sciences et Midlothian EH26 0QB, UK Techniques, Université Cheikh Anta Diop, BP 5005 Dakar, Senegal D. Fall e-mail: [email protected] Institut Sénégalais de Recherches Agricoles/Centre National de Recherches Forestières, Route des Pères Maristes, BP 2312 Dakar, N. Bakhoum : D. Fall : F. Ndoye : A. Kane : S. N. Sylla : Senegal D. Diouf LCM-Laboratoire Commun de Microbiologie IRD/ISRA/UCAD, J. M. Kimiti Centre de recherche de Bel-Air, BP 1386, CP 18524 Dakar, South Eastern Kenya University, P.O. Box 170-90200, Kitui, Senegal Kenya

D. W. Odee A. M. Zoubeirou Kenya Forestry Research Institute, P.O. Box 20412-00200, Faculté des Sciences, Université Abdou Moumouni, BP Nairobi, Kenya 10662 Niamey, Niger Author's personal copy

182 Plant Soil (2016) 398:181–193 shoot, root, and total dry matter but this effect was species is believed to have reduced over the past years. not significant with nodulation parameters. The species remains under pressure as a result of its Conclusions S. Senegal is variable in its response to overexploitation by human population, shortage of rain- inoculation, it is therefore advantageous to select and fall in the Sahel, overgrazing. In addition, it is due to match effective rhizobia-provenance symbionts for each inappropriate agricultural practices, leading to the deg- site. radation and/or lack of regeneration of S. Senegal park- lands. Therefore, there is a need to conserve and sus- Keywords S. Senegal . Provenance variation . tainably manage the species if they are to meet the Environmental conditions . Inoculation . Acacia . increased demand for fuelwood, fodder, soil improve-

Mesorhizobium . Senegal . Rhizobia ment through N2 fixation, protection of the environment and to cater for gum production, which is an important

source of cash (Fagg and Allison 2004). The N2-fixing Introduction capacity of a legume tree is often used to explain its ability to grow better on and restore the fertility of N- In the arid and semi-arid lands of Africa, low and erratic depleted soil (Dommergues 1995). rainfall, high temperatures and poor soil water and nu- S. Senegal is a promiscuous species that could be trient availability limit agricultural productivity (Mertz nodulated with various rhizobial taxa and strains et al. 2012). Thus, multipurpose trees such as Senegalia (Bakhoum et al. 2014; de Lajudie et al. 1998;Fall (Acacia) species that provide a means to maximise et al. 2008;Nicketal.1999; Njiti and Galiana 1996; agricultural potential and stabilise yields under stressful, Odee et al. 1995;Odeeetal.1997;Sarretal.2005b). unpredictable growing conditions are important for re- Nevertheless, previous studies have shown that forestation and reclamation of marginal lands, for fuel S. Senegal is mainly nodulated in Senegal by rhizobial wood, timber, shelterbelts and soil improvement strains phylogeneticaly close to Mesorhizobium (Midgley and Bond 2001; Raddad and Luukkanen plurifarium (Bakhoum et al. 2014; Fall et al. 2008; 2007). Previous phylogenetic studies indicated that Sarr et al. 2005b). Recent studies showed that inocula- Acacia Miller s.l. is polyphyletic. Recently, tion with Mesorhizobium strains significantly improved Kyalangalilwa et al. (2013) segregated genera for nodulation and S. Senegal plant growth under water- Acacia s.l. and proposed new combinations for the limited conditions (Fall et al. 2011), and enhanced plant African species in Senegalia and Vachellia. The nutrient content and rhizospheric soil fertility of Senegalia clade is represented in Africa, Central and S. Senegal plants (Bakhoum et al. 2012). While in South America, and Asia with more than 60 species. another study, inoculation improved plant nodule num- S. Senegal (L.) Britton & P. Wilson [Syn. Acacia ber but not shoot N content (Ndoye et al. 2012). Like Senegal (L.) Willd.] is a complex group formed by several other African acacias, S. Senegal has the poten- closely related species widely distributed through the tial to fix N2 under a range of soil and environmental arid and semi-arid lands of sub-Saharan Africa (Odee conditions if nodulated by effective rhizobia (Gray et al. et al. 2015). This tree is adapted to survive under harsh 2013; Ndoye et al. 1995; Raddad et al. 2005). S. Senegal environmental conditions such as low and erratic rain- is morphologically variable. It has four distinct varieties, fall, intense solar radiation, and high wind velocity namely vars. Senegal, kerensis, leiorhachis and rostrata (Cossalter 1991). (Fagg and Allison 2004), of which three (Senegal,

S. Senegal is a N2-fixing shrub or tree with consid- leiorhachis and kerensis) are found in East Africa and erable economic and ecological importance, producing one (Senegal) in West Africa. Rangewide genetic stud- a natural gum (gum arabic) widely used in the and ies of the species also show differentiations among beverage industry, pharmaceuticals, other technical ap- varities and provenances across its native range, with plications and provisioning of several ecosystem ser- clear genotypic distinction between west African and vices in the drylands of tropical Africa (Ballal et al. east and southern Africa (Chevallier and Borgel 1998; 2005;Gaafaretal.2006;Grayetal.2013;Odeeetal. Odee et al. 2012;Odeeetal.2015). Therefore, there is 2011; Omondi et al. 2010) and a well-established tradi- an important need to select appropriate plant tional agroforestry tree component (Raddad et al. 2005). phenotypes/genotypes and rhizobia that are the most However, in Senegal the number of Senegalia (Acacia) compatible to each other. The essential requirement to Author's personal copy

Plant Soil (2016) 398:181–193 183 realize this objective is to increase understanding of the medium (Vincent 1970). S. Senegal seeds of the Dahra effect of abiotic factors such as soil characteristics and provenance were inoculated with soil samples and climatic conditions on nodulation and growth of differ- grown in a controlled environment (Easy-lighting, ent plant provenances. 200 W 8 U 8500LM 6400 K° blue – 2700 K° red, Cis As part of an international consortium aimed at im- products, Paris, France) for three months with a photo- proving growth and sustainable production of gum ara- period of 16 h (under daylight) and eight hours (night), bic, we evaluated the symbiotic and growth response of temperature of 30 ± 1 °C (night), relative humidity of S. Senegal provenances (Dahra, Senegal; Tera, Niger 70 ± 5 % and a photosynthetically active radiation and Makueni, Kenya) inoculated with 11 (PAR) of 120 μmol m2-1 s−1. Mesorhizobium (rhizobial) strains and grown in two Senegalese (Dahra, arid and Goudiry, semiarid) soils Rhizobial strains used under greenhouse conditions. Table 1 shows the 11 rhizobial strains strains used in this study. They were all isolated from S. Senegal in Senegal Material and methods and selected on the basis of their symbiotic infectivity and effectiveness (Bakhoum et al. 2012;Falletal.2008; Samplingandanalysesofsoils Sarr et al. 2005a). S. Senegal nodulating rhizobial strains used in this study have identical nodA, nodC, and nifH Composite soil samples were collected from Dahra gene sequences, and are closely related to (15°21′ N, 15°29′ W) and Goudiry (14°11′ N, 12°43′ Mesorhizobium plurifarium (Bakhoum et al. 2015; Fall W), in the northern and the southern part of Senegal, et al. 2008). respectively. The climate is influenced by a strong north-south dominated precipitation gradient, resulting Plant test in about 400–500 mm and 800–1200 mm per year at Dahra (arid) and Goudiry (semiarid), respectively The three S. Senegal provenances tested originated from (Bakhoum et al. 2012). Soil samples were collected in Makueni County, Kenya (2° 9′ S, 37° 46′ E); Tera, Niger April 2008 from the top 0–25-cm-deep of rhizosphere (14° 0′ N, 0° 45′ E), and Dahra, Senegal (15° 21′ N, 15° soil of S. Senegal trees grown in plantations at Dahra 29′ W). Germination of the seeds was done as described and Goudiry. The soils were passed through a coarse previously (Fall et al. 2008). Pre-germinated seedlings sieve (2 mm mesh) to remove stones and large pieces of were transplanted into 12 cm × 8 cm (height x diameter) organic matter, and stored at 4 °C. The physical and plastic bags filled with 800 mL of field soil from Dahra chemical soil properties were analyzed at LAMA or Goudiry. The eleven strains were grown in glass (Laboratoire des Moyens Analytiques,IRD,Dakar, flasks containing liquid yeast extract mannitol (YEM) Senegal). The total amount of carbon and nitrogen was medium (Vincent 1970)at28°Cfor2daysonanorbital determined by the combustion system ThermoFinnigan shaker. Seedlings were inoculated during transplanting Flash EA 1112 (ThermoFinnigan, France). The colori- with 5 ml of the rhizobial culture in YEM liquid con- metric determination of total and available phosphorus taining approximately 109 cells ml−1. Non-inoculated was performed according to the method of Dabin et al. treatments received 5 ml of autoclaved YEM medium. (1965). Soil pH was determined in 2 M KCl suspensions at a solid liquid ratio of 1:2.5. Soil physical characteris- Experimental design tics were determined according to the method of Gee and Bauder (1986), and exchangeable cations followed The experimental design was a randomized complete themethodofThomas(1982). block at Bel Air Station, Senegal. Each block was The most probable number (MPN, Brockwell 1980) divided into seven plots; two plots represented soil method was used to estimate the number of S. Senegal- origins (Dahra and Goudiry); three plots represented nodulating rhizobia (per g−1 soil) indigenous to Dahra seed provenances (Dahra, Senegal; Tera, Niger and and Goudiry field soils. The seeds were then transferred Makueni, Kenya); two plots represented the inoculation in aseptic conditions into Gibson tubes (four replicates treatment (inoculated separately with eleven rhizobial per soil) containing a sterile Jensen nitrogen-free strains and non-inoculated control). Each plot had Author's personal copy

184 Plant Soil (2016) 398:181–193

Table 1 List of Senegalese Mesorhizobium strains originally isolated from rhizosphere soils of S. Senegal and used in this study

Rhizobial strain GenBank accession Site in Senegal Climatic zone Reference number (16S rRNA)

ORS 3573 JQ039728 Dahra Arid Bakhoum et al. (2014) ORS 3574 JQ039729 Dahra Arid Bakhoum et al. (2014) ORS 3588 JQ039735 Goudiry Semiarid Bakhoum et al. (2014) ORS 3593 JQ039736 Goudiry Semiarid Bakhoum et al. (2014) ORS 3600 JQ039741 Goudiry Semiarid Bakhoum et al. (2014) ORS 3604 JQ039739 Goudiry Semiarid Bakhoum et al. (2014) ORS 3607 JQ039737 Goudiry Semiarid Bakhoum et al. (2014) ORS 3610 JQ039732 Goudiry Semiarid Bakhoum et al. (2014) ORS 3628 JQ039740 Goudiry Semiarid Bakhoum et al. (2014) ORS 3416 EU584256 Kamb Arid Fall et al. (2008) CiradF 300 Unknown Kebemer Semiarid Sarr et al. (2005b) twelve replicates. All plants were grown in a greenhouse Results (daylight approximately 10 h, average daily temperature 25 °C day, 20 °C night) and watered regularly with Soil characteristics tapwater. After 4 months of growth, seedling height measurements were taken, then plants were uprooted, Soils from the arid Dahra and semi-arid Goudiry regions their root systems gently washed with tap water and the of Senegal used in this study were both sandy (Table 2). nodules counted. The oven dry weight (80 °C for 72 h) However, soil from the semi-arid Goudiry had a higher of the shoots, , and nodules were recorded. percentage of clay and silt than the soil from Dahra.

Table 2 Physical and chemical characteristics of Senegalese Statistical analysis Dahra (arid) and Goudiry (semiarid) soils used. For each param- eter analyzed, means followed by the same letter on each row are not significantly different according to Newman-Keuls test at 5 % Data on seedling height, nodule number, and shoot, root level. Means ± SE (n =3) and nodule dry matter were statistically analyzed using one - and two-way ANOVA with XLSAT software Soil characteristics Dahra soil Goudiry soil version 2010. Student-Newman-Keuls range test a b % Clay 3.57 ± 0.50 6.87 ± 0.267 (P < 0.05) was performed to indicate the level of differ- % Silt 10.30 ± 1.14a 19.10 ± 1.51b ences between the means. The means of soil physical % Sand 85.30 ± 1.00b 73.87 ± 1.56a and cemical characteristics of the two soil sources were % Total C 0.52 ± 0.06a 0.77 ± 0.04b compared using unpaired t-test.The hierarchical clas- % Total N 0.05 ± 0.01a 0.07 ± 0.01b sification associated with correlation matrix were Available P (mg kg−1) 8.01 ± 1.05a 8.29 ± 0.01a donewithRsoftware(643.1.0)toshowthe Total P (mg kg−1)49.00±7.55a 79.33 ± 4.70b clustering charateristics based on the correlation % Ca (meq) 0.92 ± 0.11a 1.33 ± 0.07b between nodulation (nodule number, nodule dry a a matter), shoot and root characteristics (root, shoot % Mg (meq) 0.42 ± 0.03 0.42 ± 0.07 a b and total dry matter, and shoot length) parameters % K (meq) 0.20 ± 0.01 0.28 ± 0.02 a a measured in each soil type. A principal component % Na (meq) 0.11 ± 0.05 0.15 ± 0.02 analysis (PCA) was carried out in each soil type to pH H2O5.975.96 4 determine the correlation between inoculation treat- MPN* 34 4.02x10 ment, plant provenance and soil parameters using * Most probable number (MPN) estimates of rhizobia (g soil−1 ) XLSAT software version 2010. able to nodulate S. Senegal Dahra provenance Author's personal copy

Plant Soil (2016) 398:181–193 185

Soils from Dahra can be characterised as poorly devel- from Makueni (Kenya) showed better nodulation than oped soils formed on sandy parent material of dunes or the West African provenances, Dahra (Senegal) and fluvial deposits (with less than 3 % clay). These soils are Tera (Niger), especially in Dahra field soil. Generally, reddish and have previously been classified as for each provenance, the nodule number and nodule dry Arenosols (Batjes 2001). The soils from Goudiry are matter was higher in Dahra field soil than in Goudiry classified as high in ferric lixisols, with clay-enriched field soil. Thus, Dahra field soil was more responsive to lower horizon (FAO 1995, 2003). Soil pH was slightly rhizobial inoculation. Compared to uninoculated plants, acidic in both sites and did not vary significantly. Total significantly high (P < 0.05) nodule dry matter were C, N, P, contents, percentages of Ca, K were significant- obtained by strains CiradF300 and ORS 3610 on Dahra ly higher in Goudiry than in Dahra field soil (P <0.05). (Senegal) provenance in Dahra and Goudiry field soils, By contrast, the difference of available P, percentage of respectively; strains ORS 3604 and ORS 3416 on Tera Mg and Na were not significant between Dahra and (Niger) provenance in Dahra field soil, and strain ORS Goudiry field soils. 3607 on Makueni (Kenya) provenance in Dahra field The number of rhizobia able to nodulate S. Senegal soil. The highest mean nodule number and dry weight (MPN) was also higher in Goudiry (4.02 ×104 cells g−1) were recorded in Dahra field soil on Makueni (Kenya) compared to Dahra (34 cells g−1) field soil (Table 2). provenance plants inoculated with strains ORS 3600 (8.25 nodules plant−1) and ORS 3607 (51.3 mg plant−1), respectively (Table 3). Effect of rhizobial inoculation on nodulation

Uninoculated plants were nodulated except plants of Effect of rhizobial inoculation on plant shoot and root S. Senegal provenance grown in Goudiry field soil, dry weight and plants of Tera (Niger) provenace in Dahra field soil, thus reaffirming the presence of compatible indigenous Makueni (Kenya) and Dahra (Senegal) provenances rhizobia (Table 3). Interestingly, S. Senegal provenance showed contrasting shoot and root dry matter

Table 3 Nodulation (mean nodule number, nodule dry matter each soil type and for each column, means of values (n =10)with plant−1)ofS. Senegal seedlings of three provenances (Dahra, the same letter are not significantly different according to Senegal; Tera, Niger and Makueni, Kenya) grown in two different Student-Newman-Keuls range test (P < 0.05). Nod number: non-sterilised soils (Dahra and Goudiry, Senegal) after four nodule number plant−1; NDM: nodule dry matter plant−1 months in greenhouse conditions at Bel Air Station, Senegal. For

Dahra provenance Tera provenance Makueni provenance

Dahra soil Goudiry soil Dahra soil Goudiry soil Dahra soil Goudiry soil

Treatments Nod NDM Nod NDM Nod NDM Nod NDM Nod NDM Nod NDM number (mg) number (mg) number (mg) number (mg) number (mg) number (mg)

Control 0.17a 1.55a 0a 0a 0a 0a 0.17a 2.06a 1.08a 12.85a 1.08a 3.48a ORS 3573 1.42a 3.26ab 0.33a 0.12a 2.58bc 3.36ab 0a 0a 5.33abc 18.36a 1.36a 6.76a ORS 3574 1.58a 3.75ab 1a 0.53a 0.91abc 4.46ab 0a 0a 4.25abc 18.6a 0a 0a ORS 3588 0.42a 0.65a 0.5a 2.53ab 2.92c 8.49b 0a 0a 5.7abc 25.04a 2.45a 8.54a ORS 3593 1.83a 5.25ab 0.42a 0.08a 0.58ab 3.4ab 0.1a 0.54a 5.08abc 22.05a 1.27a 8.67a ORS 3600 1.08a 3.18ab 0.83a 1.87ab 0.08a 0.08a 0.75ab 3.97a 8.25c 33.5ab 0.78a 0.48a ORS 3604 0.58a 3.39ab 0.42a 1.62ab 1.75abc 8.34b 1.08b 3.56a 2.73ab 13.34a 0.08a 0.31a ORS 3607 1.67a 6.81ab 1a 1.47ab 0.42ab 0.55a 0.09a 1.81a 7.5bc 51.65b 0.45a 5.76a ORS 3610 1.5a 3.9ab 0.64a 4.82b 0.67abc 1.85ab 0.17a 0.04a 5.09abc 28.77a 1.91a 8.78a ORS 3628 0.83a 1.24a 0.5a 0.35a 0a 0a 0a 0a 3.6abc 24.67a 0a 0a CiradF300 1.08a 8.7b 0.67a 1.22ab 2.42bc 7.2ab 0a 0a 5.55abc 16.5a 1.18a 4.07a ORS 3416 0.17a 0.7a 0.25a 0.29a 1.33abc 8.74b 0.1a 0.31a 4.14abc 26.37a 1.45a 7.65a Author's personal copy

186 Plant Soil (2016) 398:181–193 accumulation in Dahra and Goudiry soils (Fig. 1a, b). best nodule dry weight (Table 3). Strain CiradF 300 Makueni provenance had better shoot than root growth, significantly (P < 0.05) improved the root dry weight while Dahra provenance had better root than shoot of Makueni provenance by 47 % compared to uninocu- growth in both soils. However, shoot and root growth lated plants. In Goudiry soil, inoculation with rhizobial of Tera (Niger) provenance did not show any differences strains ORS 3416, ORS 3607, and ORS 3593 between the two field soils. signicantly (P < 0.05) increased shoot dry weight of In Dahra soil, the best inoculation response was provenance Makueni (Kenya). These strains showed recorded with the strains ORS 3607 which showed the hagh nodule dry weight (Table 3). All rhizobial strains

A Senegal Niger Kenya Shoot ) -1 Dry maer (g plant

Root

Control ORS 3573 ORS 3574 ORS 3588 ORS 3593 ORS 3600 ORS 3604 ORS 3607 ORS 3610 ORS 3628 CiradF 300 ORS 3416 Inoculaon treatments B Shoot ) -1 Dry maer(g plant

Root

Control ORS 3573 ORS 3574 ORS 3588 ORS 3593 ORS 3600 ORS 3604 ORS 3607 ORS 3610 ORS 3628 CiradF 300 ORS 3416 Inoculaon treatments Fig. 1 Shoot and root dry matter yield of Dahra, Senegal; Tera, matter taken separately, bars with the same letters are not signifi- Niger and Makueni, Kenya S. Senegal provenances grown in Dahra cantly different according to Student-Newman-Keuls range test arid (a) and Goudiry semi-arid (b) Senegalese non-sterilised soils (P < 0.05) for each S. Senegal provenance. Error bars are standard inoculated with selected rhizobial strains. For shoot and root dry errors of the mean (n =10) Author's personal copy

Plant Soil (2016) 398:181–193 187 significantly increased root dry weight in Makueni nodulation, root, shoot and total dry matter of seedlings (Kenya) provenance. In Dahra soil, rhizobial strains (Table 4). Provenance also had a significant effect on all ORS 3574, ORS 3593, ORS 3604, ORS 3607, parameters measured. The interaction between inocula- CIRAD F300 and ORS 3616 significantly (P <0.05) tion and provenance showed a significant effect on increased shoot dry weight of Dahra (Senegal) prove- all parameters measured except nodule number. nance plants compared to uninoculated plants (Fig. 1a). Regarding Goudiry soil, inoculation had a signifi- All of them showed high nodule dry weight (Table 3). cant effect on the seedling height, and shoot, root Nevertheless, no significant effect of inoculation was and total dry matter. However, inoculation had no observed on root dry weight. In Goudiry soil (Fig. 1b), effect on nodulation. Seed provenance effect was all rhizobial inoculation treatments significantly in- significant for all parameters measured. The inter- creased the shoot and root dry weight of Dahra prove- action between inoculation and plant provenance nance (Senegal) plants, except the strain ORS 3628. In had a significant effect on the height and shoot, Dahra soil, inoculation with rhizobial strains ORS 3573, root, and total dry matter but this effect was not ORS 3574, ORS 3588, ORS 3604, ORS 3610, ORS significant with nodulation parameters. 3628 and ORS 3588, ORS 3604 to Tera (Niger) prove- nance, significantly improved shoot and root dry Hierarchical classification and correlation weight, respectively, in comparison with uninoculated between inoculation and plant growth parameters plants (Fig. 1a, b). In Goudiry soil, inoculation with the strains ORS 3604 and ORS 3610 increased significantly Hierarchical classification associated with correlation (P < 0.05) the shoot dry weigth of plants by 40 % and matrix are represented in Fig. 2a, b. In our study, we 32 %, respectively. In contrast, strains ORS 3604 and used this method to identify the impact of field soil on ORS 3610 showed low nodule dry weight (Table 3). the hierarchichal clustering and the correlation of plant Only the root dry weight of plants inoculated with the parameters measured. Results of Dahra soil showed two strains ORS 3604 was significantly increased by 82 % clusters in relation to the correlation of parameters: the compared to uninoculated plants.. first comprises correlationbetweenrootdrymatter (RDM) and total dry matter (TDM), and the second Interactions and correlations between factors tested formed by nodule dry matter (NDM), nodule number (NN), shoot dry matter (SDM) and seedlings height The two-way interaction of plant provenances and strain (shoot length) which were correlated. Three clusters were significant (P < 0.05) for most parameters, except were revealed in Goudiry (Fig. 2b): correlation between nodule number in Dahra soil, and both nodule and total dry matter (TDM), shoot dry matter (SDM) and nodule number in Goudiry soil (Table 4). For Dahra seedlings height (shoot length); among nodule number soil, ANOVA test with two factors showed that inocu- (NN) and nodule dry matter (NDM); and root dry matter lation had a significant effect on height (shoot length), (RDM). In Dahra field soil, the improvement of TDM

Table 4 Significance level obtained from two-way ANOVA test- (semi-arid) inoculated with Mesorhizobium strains on three ing the effects of inoculation and provenance level on different S. Senegal provenances cultivated during four months at green- parameters measured on field soils, Dahra (arid) and Goudiry house conditions

Soils Factors tested Nod numb NDM SDM RDM TDM Height (shoot length)

Dahra Inoculation * * *** ** *** *** Provenance *** *** *** *** *** *** Inoc*Prov NS *** *** ** ** ** Goudiry Inoculation NS NS *** *** *** *** Provenance ** ** *** *** *** *** Inoc*Prov NS NS * ** ** ***

Significant values are indicated: *P <0.05;**P < 0.01; ***P < 0.001; NS, not significant to student-Newman-Keuls test. Inoc: inoculation treatments; Prov: provenances; Nod number: nodules number; NDM: nodules dry matter; SDM: shoot dry matter; RDM: root dry matter; TDM: total dry matter Author's personal copy

188 Plant Soil (2016) 398:181–193 A

B

Fig. 2 Hierarchical classification associated with correlation ma- gradations from black to light shades correspond with high to low trix of nodulation (NN, NDW), shoot and root characteristics correlation between the parameters. The letters are defined as (RDM, SDM, TDM and Height (shoot length)) of three follows: NN, nodule number per plant, NDW, nodule dry weight S. Senegal provenances inoculated with selected rhizobial strains per plant, RDM, root dry weight per plant, SDM, shoot dry weight in Dahra (a)andGoudiry(b) Senegalese soils. The colour per plant, TDM, total dry weight plant per plant

was linked to RDM; however, in Goudiry field soil, it identify which rhizobial strain inocultated to a was correlated to SDM and height (shoot length). There S. Senegal provenance is able to improve plant is an influence of the soil type on plant growth parameters measured in relation to the soil type. parameters. PCA showed that variables were condensed into two principal components that together were ex- PCA distribution of inoculation and provenance tracted and accounted for 90 % and 82 % variance treatments, and plant growth parameters for Dahra (Fig. 3) and Goudiry (Fig. 4) soils, respectively, suggesting that rhizobial inoculation To reveal the similarities and differences between and provenance treatments had positive effect on samples and to assess the relationships between nodulation and plants growth parameters measured. the observed variables, principal component analy- Inoculation effects changed significantly depending sis was performed. We used this method to on soil type and S. Senegal provenance. However, Author's personal copy

Plant Soil (2016) 398:181–193 189

Biplot (axes F1 et F2 : 89,74 %) 4

TDM DKCiradF300 DS3593 DS3604 DS3574 Height 2 RDM DSCiradF300 DS3607 SDM DK3607 DS3416 DN3573 DS3573 DN3604 DS3588 DS3628 Dk3574 DK3610 DS3610 DS3600 NN 0 DN3610 DK3416 DN3574 NDM DK3593 DN3588 DK3604 DN3628 DN3416 DK3573 DN3593 DN3607 DKC DSC

F2 (27,25 %) (27,25 F2 DN3600 -2 DNCiradF300 DK3628 DK3588 DK3600

-4 DNC

-6 -6 -4 -2 0 2 4 6 8 10 12 F1 (62,49 %) Fig. 3 Principal component analysis representing the relationships given in parenthesis. The letters are defined as follows DS, Dahra between nodulation (NN, NDW), shoot and root characteristics soil associated to Dahra (Senegal) provenance; DN, Dahra soil (RDM, SDM, TDM and Height (shoot length)) of three associated to Tera (Niger) provenance; DK, Dahra soil associated S. Senegal provenances inoculated with selected rhizobial strains to Makueni (Kenya) provenance, and C, control in Dahra soil. The % variance explained by each component is the provenance impact was most pronounced in In Goudiry soil two clearly distinct clusters can be Dahra than in Goudiry soil. identified: The cluster A is represented by an association In Dahra soil, three major clusters were clearly sep- of TDM, SDM, NDM, NN and height with treatments arated: Cluster A represented by nodule number (NN), of provenance Makueni (Kenya) inoculated with ORS nodule and shoot dry weight (NDM and SDM) and the 3573, ORS 3588, ORS 3593, ORS 3607, ORS 3610, height (shoot length) values correlated with the Makueni CiradF 300, ORS 3416 and treatment of provenance (Kenya) provenance inoculated with ORS 3573, ORS Tera (Niger) inoculated with ORS 3604 in the positive 3574, ORS 3588, ORS 3593, ORS 3600, ORS 3604, values of F1 axis. The cluster B was formed with treat- ORS 3607, ORS 3610, CiradF 300 and ORS 3416 in the ments of provenance Dahra (Senegal) inoculated with positive values of F1. Cluster B consisted inoculated ORS 3573, ORS 3588, ORS 3593, ORS 3604, CiradF plants from Dahra (Senegal) provenance with ORS 300, treatments of provenance Tera (Niger) inoculated 3573, ORS 3588, ORS 3600, ORS 3610, ORS 3628, with ORS 3607, ORS 3610, CiradF 300 and treatment ORS 3416 and inoculated plants from Tera (Niger) of provenance Makueni (Kenya) inoculated with ORS provenance with ORS 3574, ORS 3588, ORS 3593, 3604, associated with RDM in the positive values of F2 ORS 3600, ORS 3607, ORS 3610, ORS 3628, ORS axis. 3416 wich were linked to root dry matter (RDM) in the negative values of F1. The plants of provenance Dahra (Senegal) inoculated with ORS 3574, ORS 3593, ORS Discussion 3604, ORS 3607, CiradF 300 and plants of provenance Tera (Niger) inoculated with ORS 3573, ORS 3604 Our results demonstrate provenance variation in symbi- constitued the Cluster C, which is correlated to total otic association with selected Mesorhizobium strains as dry matter (TDM) in the positive values of F2. influenced by soil characteristics (nutrient status and Author's personal copy

190 Plant Soil (2016) 398:181–193

Biplot (axes F1 et F2 : 82,47 %) 2 GNCiradF300 RDM GK3604 GS3604 GS3416 GN3610 GK3628 GS3593 GSCiradF300 GN3604 GN3416 GS3573 GS3588 TDM GN3593 GN3607 GS3600 Gk3574 Height GS3607 GN3600 0 GKCiradF300 GK3607 GN3574 GS3610 SDM GK3593 GS3574 GN3588 GK3600 GK3416

GSC GNC NDM GK3573 GN3628 GS3628 NN GN3573 F2 (18,97 %) (18,97 F2

-2 GK3610

GK3588 GKC

-4 -8 -6 -4 -2 0 2 4 6 8 10 F1 (63,50 %) Fig. 4 Principal component analysis representing the relationships given in parenthesis. The letters are defined as follows: GS, between nodulation (NN, NDW), shoot and root characteristics Goudiry soil associated to Dahra (Senegal) provenance; GN, (RDM, SDM, TDM and Height (shoot length)) of three Goudiry soil associated to Tera (Niger) provenance; GK, Goudiry S. Senegal provenances inoculated with selected rhizobial strains soil associated to Makueni (Kenya) provenance and C, control in Goudiry soil. The % variance explained by each component is indigenous rhizobia). Several authors have shown sim- inversely related to soil N. This could be attributed to ilar results, for example, in the common bean, that the act that nodule number and N2 fixation are Phaseolus vulgaris (Cardoso et al. 2009) and several regulated in response to the N status of the plant as woody legumes (Elbanna et al. 2009;Mnasrietal.2007; described by Ruffel et al. (2008). Dart (1974)showed Odee et al. 1995; Sanginga et al. 1991). In our study, that N compounds like nitrates may affect nodulation Makueni provenance had the best nodulation response regardless of plant age, size or prior to inoculation when inoculated seedlings were grown in arid Dahra status. Plants use the available nitrogen in soil and form soils that had low nutrients status and number of indig- nodules to complement the quantity of nitrogen re- enous S. Senegal-nodulating rhizobia (Table2). These quired, thus the observed nodulation may also reflect results also suggest that Makueni provenance has a differences in the relative N-limiting status between the higher N demand compared to the West African prove- soils. However, indigenous rhizobia could also be re- nances. This is also corroborated by a previous study sponsible for the observed diffierence in nodulation of that reported higher shoot N contents of the Makueni S. Senegal plants grown in Dahra and Goudiry soils. provenance (Kenyan) than the West African Singleton and Tavares (1986) and Turk et al. (1993) demostrating differences in their N requirements indicated that the response of rhizobial inoculation (Bakhoum et al. 2012). Thus, high nodulation capacity mostly occurs when the indigenous population densities may indicate higher N-demand in the Makueni than are <50 rhizobia g−1 of soil; Dahra soil rhizobial MPN Dahra and Tera provenances. (Table 2) was within this threshold. On the other hand, Nodulation tended to be higher in Dahra soils poor in MPN of rhizobia was high in Goudiry soil (4.02 × 104 nutrients compared to fertile Goudiry soils. This was g−1 of soil), a large number which could also outcom- probably due to the differences in level of available N in pete the inoculant strain. In studies solely dependent on the soils, which was higher in Goudiry than Dahra indigenous rhizobia, it has been shown that low rhizobia (Table 2). These results showed that nodulation was counts in the soil reduce nodule numbers and biomass Author's personal copy

Plant Soil (2016) 398:181–193 191 while high rhizobial counts in soil enhance nodulation, nodulation parameters between Dahra and Goudiry soils for example cowpea (Kimiti and Odee 2010), Acacia (Table 4). Corollary to this result, PCA showed that in saligna (Benbrahim et al. 1998) and Cajanas cajan Goudiry soil, the rhizobial strain ORS 3604 inoculated (Mapfumo et al. 2000). Therefore, the inoculation re- to Tera (Niger) provenance seedlings had improved sponse in Goudiry soil cannot only be explained by growth parameters except root dry matter. In contrast, available N in soil, but also the number of competitive the strain was only correlated with the total dry matter in indigenous rhizobia in soil. Notwithstanding the differ- Dahra soil. This reaffirms the importance of soil type, ence in MPN estimates of indigenous rhizobia capable hence plant available N on the nodule development and of nodulating S. Senegal in Dahra and Goudiry soils, functioning. In our study, Dahra provenance generally inoculation in most cases improved nodulation. In addi- performed better in produced more roots biomass than tion, indigenous rhizobial strains originally isolated other provenances, especially in Dahra soil, indicating from Goudiry generally performed better than exoge- its adaptation grow in poor soils, by growing an exten- nous strains in Goudiry soil despite high indigenous sive root system in order to get nutrient from wider soil populations, indicating the importance of selection and area. Therefore, these results implied that nodulation re-inoculation with an effective indigenous strain as and thus effectiveness of symbiosis is regulated by plant previously demonstrated in Sesbania sesban by provenance, rhizobial strain and soil origin. Makatiani and Odee (2007). Besides natural adaptation, the re-inoculated strain is also expected to reduce competion for nodulation from other compatible indig- Conclusions enous soils. Another important finding of this work is that nodu- The nodulation and growth of S. Senegal seedlings was lation (nodules number and nodule dry matter) were variable and dependent on complex interactions of rhi- correlated with shoot dry matter and seedlings height zobial strain inoculation, plant provenance and soil type. in Dahra and Goudiry soils indicating effectiveness of This study has shown that it would be advantageous to the symbioses and contribution of N2 fixation to the select effective combinations of rhizobia × provenances growth of S. Senegal seedlings. Nodule dry weight and in relation to soil and environmental conditions where numbers were negatively correlated with root dry matter they are to be planted. in Dahra soil and not in Goudiry soil. These results suggested that the control and biomass partitioning for Acknowledgments This work was supported by nodule development in S. Senegal is driven by the soil ACACIAGUM INCO STREP project Contract Number 032233. available N, but other factors such as host provenance NB is grateful to the French Embassy in Senegal for SCAC and rhizobial strain may also be equally important. Fellowship. Other workers (e.g. Laguerre et al. 2007;Rodiñoetal. 2011) have reported variability of nodulation, root and shoot characteristics in relation to rhizobial strain and References plant genotype. Our results showed that inoculation with rhizobial strains significantly improved growth of S. Senegal Badji S, Ducousso M, Gueye M, Colonna JP (1988) and cross inoculation with Rhizobium from two seedlings. These results are in agreement with several acacia species producing arabic gum: Acacia Senegal L. previous studies in nursery conditions which reported Willd and Acacia laeta R. Br. ex Benth. C R Acad Sci Paris enhanced growth of S. Senegal species due to inocula- 307 (Serie III). tion with effective microsymbionts (Badji et al. 1988; Bakhoum N, Galiana A, Le Roux C, Kane A, Duponnois R, Ndoye F, Fall D, Noba K, Sylla SN, Diouf D (2015) Räsänen et al. 2001). Nevertheless, there was high var- Phylogeny of nodulation genes and symbiotic diversity of iability in plant development among the provenances Acacia Senegal (L.) Willd. and A. seyal (Del.) studied. It is important to note that effectiveness of mesorhizobium strains from different regions of Senegal. – rhizobial strain on improving plant growth parameters Microb Ecol 69:641 651 varied according to provenance and soil source. This is Bakhoum N, Le Roux C, Diouf D, Kane A, Ndoye F, Fall D, Duponnois R, Noba K, Sylla SN, Galiana A (2014) also reflected in the variable interaction effects of inoc- Distribution and diversity of rhizobial populations associated ulation × plant provenance on the various growth and with Acacia Senegal (L.) Willd. Provenances in Senegalese Author's personal copy

192 Plant Soil (2016) 398:181–193

Arid and Semiarid Regions. Open J Forest 4:136–143. doi: genotypic characteristics of Acacia Senegal (L.) Willd. root- 10.4236/ojf.2014.42019 nodulating bacteria isolated from soils in the dryland part of Bakhoum N, Ndoye F, Kane A, Assigbetse K, Fall D, Sylla SN, Senegal. Lett Appl Microbiol 47:85–97. doi:10.1111/j.1472- Noba K, Diouf D (2012) Impact of rhizobial inoculation on 765X.2008.02389.x Acacia Senegal (L.) Willd. growth in greenhouse and soil Fall D, Ourarhi M, Missbah EIM, Bakhoum N, Zoubeirou AM, functioning in relation to seed provenance and soil origin. Abdelmoumen H, Diouf D (2011) The efficiency and com- World J Microbiol Biotechnol 28:2567–2579. doi:10.1007/ petitiveness of three Mesorhizobium sp. strains nodulating s11274-012-1066-6 Acacia Senegal (L.) Willd. under water deficiency conditions Ballal ME, El Siddig EA, Elfadl MA, Luukkanen O (2005) in the greenhouse. Symbiosis 54:87–94 Relationship between environmental factors, tapping dates, FAO (1995) 2003) the digitized soil map of the world and derived tapping intensity and gum arabic yield of an Acacia Senegal soil properties. In: In: FAO (ed) land and water digital media. plantation in western Sudan. J Arid Environ 63:379–389 Rome, FAO Batjes NH (2001) Options for increasing carbon sequestration in Gaafar AM, Salih AA, Luukkanen O, El Fadl MA, Kaarakka V West African soils: an exploratory study with special focus (2006) Improving the traditional Acacia Senegal-crop system on Senegal. Land Degrad Dev 12:131–142. doi:10.1002/ldr. in Sudan: the effect of tree density on water use, gum pro- 444 duction and crop yields. Agroforest Syst 66:1–11 Benbrahim FK, Ismaili M, Salema MP (1998) Improving biological Gee GW, Bauder JW (1986) Particle-size analysis. In: Page AL nitrogen fixation of Acacia saligna (Labill) H. Wendl by con- (ed) Methods of soil analysis part 1: physical and mineralog- centrated inoculant, repeated inoculation and phosphorus addi- ical methods, 2 edn. American Society of Agronomy, tion. In: Mpepereki, S., Makones, F. (Eds.), Harnessing Madison Biological Nitrogen Fixation in Africa, Challenges and opportu- Gray A, David O, Cavers S, Wilson J, Telford A, Grant F, Diouf nities. Harare Zimbabwe, University of Zimbabwe Publications. M, Ochieng J, Grant H, Stott A (2013) Does geographic Brockwell J (1980) Experiment with cropand pasture legumes: origin dictate ecological strategies in Acacia Senegal (L.) principles and practice. In:Bergersen, F. J. (Ed), Methods for Willd.? Evidence from carbon and nitrogen stable isotopes. Evaluating BNF. Wiley. New York. Plant Soil 369:479–496 Cardoso JD, Gomes DF, Goes KP, Junior FNS, Dorigo OF, Kimiti JM, Odee DW (2010) Integrated soil fertility management Hungria M, Andrade DS (2009) Relationship between total enhances population and effectiveness of indigenous cowpea nodulation and nodulation at the root crown of peanut, soy- rhizobia in semi-arid eastern Kenya. Appl Soil Ecol 45:304– bean and common bean plants. Soil Biol Biochem 41:1760– 309 1763 Kyalangalilwa B, Boatwright JS, Daru BH, Maurin O, Van Der Chevallier M, Borgel A (1998) Diversité génétique des Acacias. Bank M (2013) Phylogenetic position and revised classifica- In: C Campa, C Grignon, M Gueye, S Hamon (eds) l’Acacia tion of Acacia s.l. (: ) in Africa, includ- au Sénégal. Collection Colloques et Séminaires ORSTOM ing new combinations in Vachellia and Senegalia.BotJLinn Paris, France. Soc 172:500–523 Cossalter C (1991) Acacia Senegal : gum tree with promise for Laguerre G, Depret G, Bourion V, Duc G (2007) Rhizobium agro-forestry. In: N 91-02 (ed) Nitrogen Fixing Tree leguminosarum bv. Viciae genotypes interact with pea plants Association (NFTA), Waimanalo, Hawaii. in developmental responses of nodules, roots and shoots. Dabin B, Brion J-C, Verdoni J, Moreau D, Ratovo J (1965) New Phytol 176:680–690 Application des dosages automatiques à l’analyse des sols : Makatiani ET, Odee DW (2007) Response of Sesbania sesban (L.) essais effectués au laboratoire des services scientifiques Merr. To rhizobial inoculation in an N-deficient soil contain- centraux de l’ORSTOM à bondy avec l’autoanalyseur ing low numbers of effective indigenous rhizobia. Agroforest technicon : 1ère partie. Cahiers ORSTOM Série Pédologie Syst 70:211–216 3:335–366 Mapfumo P, Mpepereki S, Mafongoya P (2000) Pigeon pea Dart PJ (1974) The biological nitrogen fixation.Oxford:North- rhizobia prevalence and crop response to inoculation in Holland Publishing Company. Amsterdam. Zimbabwean smallholder-managed soils. Exp Agric 36: de Lajudie P, Willems A, Nick G, Moreira F, Molouba F, Hoste B, 423–434 Torck U, Neyra M, Collins MD, Lindstrom K, Dreyfus B, Mertz O, D’Haen S, Maiga A, Moussa IB, Barbier B, Diouf A, Gillis M (1998) Characterization of tropical tree rhizobia and Diallo D, Da ED, Dabi D (2012) Climate variability and description of Mesorhizobium plurifarium sp. nov. Int J Syst environmental stress in the Sudan-Sahel zone of West Bacteriol 2:369–382 Africa. Ambio 41:380–392 Dommergues YR (1995) Nitrogen fixation by trees in relation to Midgley JJ, Bond WJ (2001) A synthesis of the demography of soil nitrogen economy. Nutr Cycl Agroecosys 42:215–230 African acacias. J Trop Ecol 17:871–886. doi:10.1017/ Elbanna K, Elbadry M, Gamal-Eldin H (2009) Genotypic and S026646740100164X phenotypic characterization of rhizobia that nodulate snap Mnasri B, Mrabet M, Laguerre G, Aouani ME, Mhamdi R (2007) bean (Phaseolus vulgaris L.) in Egyptian soils. Syst Appl Salt-tolerant rhizobia isolated from a Tunisian oasis that are Microbiol 32:522–530 highly effective for symbiotic N2-fixation with Phaseolus Fagg CW, Allison GE (2004) Acacia Senegal and the gum arabic vulgaris constitute a novel biovar (bv. Mediterranense) of trade: monograph and annotated bibliography. Oxford Forestry Sinorhizobium meliloti. Arch Microbiol 187:79–85 Institute, University of Oxford, Oxford, United Kingdom Ndoye F, Kane A, Bakhoum N, Sanon A, Fall D, Diouf D, Sy MO, Fall D, Diouf D, Ourarhi M, Faye A, Abdelmounen H, Neyra M, Noba K (2012) Response of Acacia Senegal (L.) Willd. Sylla SN, Missbah El Idrissi M (2008) Phenotypic and seedlings and soil bio-functioning to inoculation with Author's personal copy

Plant Soil (2016) 398:181–193 193

arbuscular mycorrhizal fungi, rhizobia and Pseudomonas provenances in dryland clays of the Blue Nile Sudan estimat- fluorescens. Afr J Microb Res 6:7176–7184 ed by the 15 N natural abundance method. Plant Soil 275: Ndoye I, Gueye M, Danso SKA, Dreyfus B (1995) Nitrogen 261–269 fixation in Faidherbia albida, Acacia raddiana, Acacia Raddad EY, Luukkanen O (2007) The influence of different Senegal and Acacia seyal estimated using the 15 N isotope Acacia Senegal agroforestry systems on soil water and crop dilution technique. Plant Soil 172:175–180 yields in clay soils of the Blue Nile region, Sudan. Agric Nick G, de Lajudie P, Eardly BD, Suomalainen S, Paulin L, Zhang Water Manag 87:61–72 X, Gillis M, Lindstrom K (1999) Sinorhizobium arboris sp. Räsänen LA, Sprent JI, Lindström K (2001) Symbiotic properties nov. and Sinorhizobium kostiense sp. nov., isolated from of sinorhizobia isolated from acacia and prosopis nodules in leguminous trees in Sudan and Kenya. Int J Syst Bacteriol Sudan and Senegal. Plant Soil 235:193–210 4:1359–1368 Ruffel S, Freixes S, Balzergue S, Tillard P, Jeudy C, Martin- Njiti CF, Galiana A (1996) Symbiotic properties and rhizobium Magniette ML, van der Merwe MJ, Kakar K, Gouzy J, requirements for effective nodulation of five tropical dry zone Fernie AR, Udvardi M, Salon C, Gojon A, Lepetit M Acacias. Agroforest Syst 43:265–275 (2008) Systemic signaling of the plant nitrogen status triggers Odee DW, Sutherland JM, Kimiti JM, Sprent JI (1995) Natural specific transcriptome responses depending on the nitrogen rhizobial populations and nodulation status of woody le- source in Medicago truncatula. Plant Physiol 146:2020– gumes growing in diverse Kenyan conditions. Plant Soil 2035 173:211–224 Sanginga N, Manrique K, Hardarson G (1991) Variation in Odee DW, Sutherland JM, Makatiani ET, McInroy SG, Sprent JI nodulation and N2 fixation by Gliricidia sepium/ (1997) Phenotypic characteristics and composition of Rhizobium spp. symbiosis in a calcareous soil. Biol rhizobia associated with woody legumes growing in diverse Fert Soil 11:273–278 Kenyan conditions. Plant Soil 188:65–75 Sarr A, Diop B, Peltier R, Neyra M, Lesueur D (2005a) Effets of Odee DW, Telford A, Wilson J, Gaye A, Cavers S (2012) Plio- rhizobial inoculation methods and host plant provenances on Pleistocene history and phylogeography of Acacia Senegal in nodulation and growth of Acacia Senegal and Acacia dry woodlands and savannahs of sub-Saharan tropical Africa: nilotica. New Forest 29:75–87 evidence of early colonisation and recent range expansion. Sarr A, Neyra M, Houeibib MA, Ndoye I, Oihabi A, Lesueur D Heredity 109:372–382 (2005b) Rhizobial populations in soils from natural Acacia Odee DW, Wilson J, Cavers S (2011) Prospects for genetic im- Senegal and Acacia nilotica forests in Mauritania and the provement of Acacia Senegal: can molecular approaches Senegal river valley. Microb Ecol 50:152–162 deliver better gum yield and quality? In: Kennedy JF, Singleton PW, Tavares JW (1986) Inoculation response of le- Philips GO, Williams PA (eds) Gum arabic. Roy Soc gumes in relation to the number and effectiveness of indige- Chem, Cambridge, pp. 99–109 nous rhizobium populations. Appl Environ Microbiol 51: Odee DW, Wilson J, Omondi S, Perry A, Cavers S (2015) 1013–1018 Rangewide ploidy variation and evolution in Acacia Thomas GW (1982) Exchangeable cations. In: Page AL (ed) Senegal: a north-south divide? Ann Bot Plants. doi:10. Methods of soil analysis part 2. American Society of 1093/aobpla/plv011 Agronomy, Madison, WI Omondi SF, Kireger E, Dangasuk OG, Chikamai B, Odee DW, Turk D, Keyser HH, Singleton PW (1993) Response of tree Cavers S, Damase PK (2010) Genetic diversity and popula- legumes to rhizobial inoculation in relation to the population tion structure of Acacia Senegal (L) Willd. in Kenya, tropical. density of indigenous rhizobia. Soil Biol Biochem 25:75–81 PlantBiol3:59–70 Vincent JM (1970) A manual for the practical study of root-nodule Raddad E, Salih A, Fadl M, Kaarakka V, Luukkanen O (2005) bacteria. International Biological Programme. Blackwell Symbiotic nitrogen fixation in eight Acacia Senegal Scientific Publications, Oxford