Spatial Scale Patterns of Genetic Diversity and Gene Flow in Populations of Sweet Detar (Detarium microcarpum Guill. & Perr.; Fabaceae)

The main objective of this study is to investigate the patterns of genetic diversity and phylogenetic relationships within populations of Detarium microcarpum (Fabaceae) relative to different spatial conditions. Seventy-eight (78) accessions of D. microcarpum belonging to six populations (Phytogeographic districts) were sampled. In order to have very good quality DNA for molecular analysis, an optimization of the DNA isolation protocol was made. The molecular analysis of the accessions was carried out using 7 chloroplast microsatellite markers. The polymorphism rate (P) is 85.71% and the Polymorphism Information Content (PIC) was in the range of 0.43 (Ntcp_9) to 0.73 (Ccmp_2) with an average of 0.59. Allelic richness (A) ranged from 1.41 to 2.85 with an average of 2.04. The observed heterozygosity (Ho) ranged from 0.23 to 0.60 with an average of 0.39. The expected heterozygosity (He) ranged from 0.43 to 0.60 with a mean of 0.50. Wright's xation index (F IS ) ranged from − 0.17 to 0.47. The effective allele (Ae) is between 1.77 and 2.53 with an average of 2.02. Wright differentiation index (F ST ) was 0.024. Phylogenetic analysis revealed that the N ST value was signicantly higher than the G ST value (N ST = 0.452; G ST = 0.190; P < 0.05). A relatively low h d haplotype diversity is obtained (H d = 0.320). AMOVA analysis showed that 17.35% of the variation existed within populations but 45.80% among populations within the species. Neighbor-Joining phylogenetic tree of D. microcarpum revealed three non-distinct clusters haplotypes showing the existence of gene ow between populations of the species. Our ndings of genetic structure and gene ow of D. microcarpum populations based on different spatial conditions is caused by evolutionary forces such as scattering and pollination.


Introduction
Detarium microcarpum Guill. & Perr. (Fabaceae) is a diploid forest plant species (2n = 22 chromosomes) used in human and animal nutrition, traditional medicine, crafts, household energy and also for medico-magic purposes (Kouyaté 2005;Agbo et al. 2019a). It is involved in the treatment of several diseases including sexually transmitted diseases (Nacoulma 1996). The avonoids in methanol extracts of the plant have been shown to have potent inhibitory effects on HIV-1 or HIV-2 infection (Mahmood 1993;Doulougou 2002;Nikiema et al. 2010). Detarium microcarpum (sweet dattock or sweet detar) therefore constitutes an important source of income for the local communities which exploit its resources daily for their needs.
Unfortunately, due to overexploitation of its timber and the effects of climate change, the species has become vulnerable in parts of West Africa (Adjahossou et al. 2016;Agbo et al. 2017). It is therefore urgent to develop strategies for its conservation in order to avoid its disappearance. The conservation of a forest species requires knowing the information about the level of genetic diversity and the extent of genetic differentiation within and between natural populations of that species (Assogbadjo et al. 2010). Several studies have been carried out on systematics and parataxonomy (Calvin 2007;Houénon et al. 2021), ethnobotany (Kouyaté et al. 2009;Agbo et al. 2019a), spatial and temporal distribution (Agbo et al. 2019b), phenology (Kouyaté 2005), morphological variability (Kouyaté 2005;Agbo et al. 2018) and pharmacology (Sani et al. 2014) of D. microcarpum. However, studies relating to the assessment of its genetic diversity using molecular genetic markers remain unavailable. Furthermore, knowledge of the level of genetic diversity and of the population structure of a plant is one of the very important aspects in programs for the management, conservation or genetic improvement of genetic resources Jahan et al. 2020). Loading [MathJax]/jax/output/CommonHTML/jax.js The development of effective strategies for the conservation and sustainable management of forest genetic resources requires an analysis of the diversity and genetic structure of these plant species which requires the use of highly informative genetic molecular markers (Liu et al. 2012). DNA molecular markers, including RAPD (Ercisli et al. 2008;Vanijajiva 2011), AFLP (Schenk et al. 2008;Assogbadjo et al. 2010) and RFLP (Bhattacharjee et al. 2002;Yan et al. 2005), were used to assess genetic diversity at the molecular level of many populations of plant species. In recent years, microsatellite markers (SSR) have become popular and powerful tools used for the assessment of genetic diversity due to their codominance, high rate of polymorphism, and putative in uence on transcribed genes (Benemann et al. 2012;Huda et al. 2019;Karam et al. 2019). Several nuclear microsatellite markers (SSRs) have been used to assess the genetic diversity of many plant species in Benin, including Sorghum bicolor , Pentadesma butyracea Sabine (Houédjissin 2016), Pennissetum glaucum , etc. In addition, SSRs markers developed on chloroplast genomes (SSRcp) constitute a powerful tool in the analysis of phylogenetic relationships because they provide information on the structure of variability and allow the ow of genes to be traced back to the scale of the natural area (Andrianoelina Andrianaivo 2009;de Cesare et al. 2010;Phumichai et al. 2015). They make it possible to assess the impact of Human and animal activities on the diversity of species at the level of populations and sub-populations studied (Nanema et al. 2010;Vihotogbé et al. 2015). As of today, no molecular genetic marker have been developed on D. microcarpum. Genetic characterization and assessment of the genetic diversity of a species involves the extraction of the genomic DNA of that species. However, DNA extraction kits are very expensive and their use requires expertise. Hence the need for the establishment of a DNA extraction protocol. The effectiveness of DNA isolation protocols depends not only on the type of plant (woody or not) of the organ serving as biological material (leaves, root, ower, etc.) and on its state of conservation, but also of the chemical composition of the biological material; which means that several protocols have to be tested in order to retain the ideal (Chabi Sika et al. 2015;Adjé et al. 2016).
The objective of this study is to investigate the patterns of genetic diversity and phylogenetic relationships within populations of Detarium microcarpum relative to different spatial conditions.

Materials And Methods
Plant material sampling -The plant material consists of seventy-eight (78) samples of young leaves of D. microcarpum (taken from 78 trees) collected in six (06) phytogeographic districts in Benin (Bassila, Zou, South Borgou, North Borgou, Atacora Chain and Mekrou-Pendjari), at the rate of thirteen (13) trees per phytogeographic districts, spaced at least 50 meters apart. These phytogeographic districts belong to Sudanian and Guineo-Sudanian and were chosen on the basis of the occurrence of natural stands of the species ( Fig. 1; Table 1) (Agbo et al. 2019b). Samples are georeferenced using a GPS and those belonging to the same phytogeographic district are considered to be from the same population (Table 2)  Optimization of genomic DNA isolation protocol of D. microcarpum -Molecular Genetics and Genome Analysis Laboratory of University of Abomey-Calavi (Benin) served as a study setting for molecular analysis of D. microcarpum samples. Several genomic DNA isolation protocols (Table 3) were tested on different quantities of young fresh leaves (0.1g, 0.15g and 0.2g) on the one hand and on dry young leaves on the other hand of D. microcarpum. The composition of the isolation buffer was adjusted in order to nd the appropriate protocol, relatively fast, inexpensive and allowing an average yield of good quality genomic DNA. DNA Ampli cation by Polymerase Chain Reaction (PCR) -Dilutions of the DNA extracts were made to obtain the concentrations necessary for the ampli cation of DNA by PCR. A concentration of around 150 ng / µL was obtained from the suspensions previously obtained after DNA extraction. To amplify sequences of the D. microcarpum DNA, seven (07) chloroplast microsatellite markers (SSRcp) out of the ten tested were used. These are six (06) Ntcp primers (Nicotiana tabacum chloroplast) developed on tobacco (Bryan et al. 1999) and a Ccmp (Consensus chloroplast microsatellite primers) primer, a universal primer developed by Weising and Gardner (1999). The primers used are: Ntcp_8, Ntcp_9, Ntcp_33, Ntcp_36, Ntcp_37, Ntcp_39 and Ccmp_2 (Table 4). These primers have been used successfully on other speculations such as Helianthus annuus (Wills et al. 2005), Solenostemon rotundifolius (Nanema et al. 2010) and Solanum tuberosum (Zeka et al. 2014  (TBE) buffer, pH 8) at constant power of 60 W for 1 h 30 min to 2 h, depending on the expected product size. The detection of electrophoretic plates was carried out with silver nitrate (AgNO 3 ) according to Creste et al. (2001). The different bands or allelic level are determined by scoring as a function of the level of migration of the bands obtained.
Data analysis -The bands obtained at each locus were recorded as allelic compositions: absence and presence of band were respectively coded as 0 and 1. A binary matrix was constructed for statistical analysis.
To analyze the effectiveness of the microsatellite markers used in characterizing D. microcarpum populations, polymorphism rate (P) and Polymorphism Information Content (PIC) were calculated. In order to assess the genetic differentiation and the genetic structure within and between populations of D. microcarpum, following genetic diversity and structure parameters are calculated: Allelic richness ( Polymorphism Information Content of (PIC) was computed by the formula: PIC = 1-∑ 2 where , the allele frequency. It provides information on the discriminating power of the locus (Botstein et al. 1980;. Allelic richness (A) is the sum of alleles for a locus over the total number of loci. It is determined by the formula: A = (1/k) ∑ k n = 1 niwhere k is the total number of loci and ni the number of allele per locus (Bhandari et al. 2017). Allelic richness is particularly important for conservation strategies because it is often used in the management of collections and seed banks (Bataillon et al. 1996). Effective allele number (Ae) was computed by the formula: Ae = 1 / ( ) 2 where is the allele frequency. It provides information about the dispersal ability of the organism and the degree of isolation among populations (Kremer 1994;Govindaraj et al. 2015). Rate of expected heterozygosity (He) or genetic diversity of Nei was calculated by locus according to the formula∶ He = 1-∑ ( ) 2 , where is the frequency of the allele i at the considered locus (Nei 1987). The average expected heterozygosity rate of the loci was calculated by population and also by chromosomal region, in order to determine the level of genetic variability of the microsatellite sequences of the coding and non-coding regions of the cpDNA of D. microcarpum (Zeka et al. 2015). Rate of observed heterozygosity (Ho) was calculated from the measured frequency of heterozygotes by: Ho = Nh/(Nt) where Nh is the number of heterozygous individuals and Nt the total number of individuals (Nei 1987). By comparing these two indices (He and Ho), it is possible to deduce the impact of evolutionary forces in populations, due to self-fertilization or autogamy. Wright xation Index (F IS ) was calculated from the allele frequencies under the Hardy-Weinberg hypothesis according to the formula: F IS = 1 -(Ho / He). It re ects the differentiation of individuals within populations. F IS = 1 mean complete xation (self-fertilization), F IS <1: excess heterozygosity, F IS = 0: Hardy-Weinberg equilibrium population. F IS <0: excess heterozygosity (Wright 1978;Govindaraj et al. 2015). Wright differentiation index (F ST ) is related to heterozygosity and genetic drift. Since inbreeding increases the frequency of homozygotes, as a consequence, it decreases the frequency of Loading [MathJax]/jax/output/CommonHTML/jax.js heterozygotes and genetic diversity. It is commonly used to determine whether there is gene ow between populations. It is obtained with the following formula: F ST = 1 -(H S / H T ). Where S is the average e of subpopulations assuming random mating within each population, and T is the e of the total population assuming random mating within subpopulations and no divergence of allele frequencies among subpopulations. According to Wright (1978), 0 < F ST < 0.05: weak differentiation; 0.05 < F ST < 0.15: moderate differentiation; 0.15 < F ST < 0.25: signi cant differentiation; F ST > 0.25: very important differentiation. In order to establish the phylogenetic relationship within populations of D. microcarpum with a quanti cation of gene ow in its populations, the program PermutCpSSR_1.2.1 (Pons & Petit, 1996)

Results
Genomic DNA isolation buffer of the new protocol - Table 3 showed the composition of the buffer solutions of the different genomic DNA isolation protocols tested. Genomic DNA isolation buffer of the new protocol is composed of: Chloroform Isoamylic Alcohol (24: 1), isopropanol, 70% ethanol, Tris-EDTA (10 mM Tris-HCl, 1 mM EDTA) and 2% CTAB containing 2g of CTAB, 1.4M of NaCl, 100mM of Tris-HCl (1M, pH = 8), 20mM of EDTA (0.5M, pH = 8) supplemented with 2% β-Mercaptoethanol and 0.5% Proteinase K (added after sterilization and just before use in a hood).
New protocol for genomic DNA isolation from D. microcarpum -The new protocol resulted in cleaner, good quality DNA that was easily ampli ed by PCR (Fig. 3). This new protocol is described as follows: A quantity of 150 mg of dry leaves per sample was weighed and ground in a porcelain mortar with 1 mL of CTAB buffer supplemented with 2% β-Mercaptoethanol and 0.5% proteinase K. The ground material obtained was poured into a 2 mL eppendorf tube, homogenized for a few minutes and then incubated in a water bath at 65°C with gentle stirring for 60 min with homogenizations every 10 min. 1 mL of Chloroform Isoamylic Alcohol (CIA) buffer was added to the mixture after cooling followed by gentle stirring for 5 min. The whole is centrifuged at 10,000 rpm for 15 min at 4°C without brake (free deceleration). The upper phase aqueous supernatant is transferred to another 1.5 mL eppendorf tube to which an equal volume (approximately 1 mL) of very cold isopropanol has been added to precipitate the DNA. In order to optimize the quantity of the DNA ball, an incubation is done at -20°C for 30 min. After this precipitation phase, the mixture was centrifuged at 10,000 rpm for 10 min. The supernatant was removed and on the pellet, 500 µL of 70% ethanol was added and the mixture homogenized then left at room temperature for 20 min. The mixture was then centrifuged at 10,000 rpm for 10 Loading [MathJax]/jax/output/CommonHTML/jax.js min then the supernatant carefully emptied and the DNA pellet dried at room temperature on lter paper in order to remove the ethanol (Fig. 2). After drying, the DNA ball was taken up in 50 µL of Tris-EDTA buffer. To be sure of the success of the DNA extraction, the DNA yield was estimated using a spectrophotometer at UV-VIS 230, 260 and 280 nm, and 2 µL of genomic DNA extract were visualized on a 1% agarose gel stained with ethidium bromide (BET). After this con rmation, DNA samples were kept in a freezer at -20°C for molecular analysis. The shape and the relative intensity of the agarose gel migration bands are well parallel to the DNA concentration values extracted for the different samples.
Genetic structure of D. microcarpum using SSR markers -Seven chloroplast microsatellite markers (Ntcp_8, Ntcp_9, Ntcp_33, Ntcp_36, Ntcp_37, Ntcp_39 and Ccmp_2) of ten tested for DNA ampli cation by PCR ampli ed D. microcarpum genome and were polymorphic except Ntcp_33 (which was monomorphic). Polymorphism rate (P) was so 85.71%. Four out of ten (Ntcp_5, Ntcp_15, Ntcp_25 and Ntcp_26) of the primers developed on N. tabacum failed to amplify a speci c region on cpDNA in D. microcarpum accessions. On the chloroplast genome of N. tabacum these primers amplify microsatellite loci at T14 (Ntcp_5), at A10 (Ntcp_15), at A13 (Ntcp_25) and at T10 (Ntcp_26) with bands whose size varies between 120 and 190 bp. In total, twenty-two (22) alleles were detected on the seven ampli ed chloroplast microsatellite loci on D. microcarpum genome. The number of alleles per locus (Na) varied from 1 (Ntcp_33) to 7 (Ccmp_2) with an average of 3.14. Major allele frequency varied from 0.12 (Ntcp_9) to 0.22 (Ntcp_37) with an average of 0.16 (Fig. 4). Ccmp_2 locus was the most polymorphic (Na = 7 alleles) and the most discriminating (PIC = 0.73) while Ntcp_9 was the least polymorphic (Na = 2 alleles) and the least discriminating (0.43) ( Table 4) (Table 6). The classi cation of D. microcarpum accessions showed on the phylogenetic tree constructed by the Neighbor-Joining method three (03) non-distinct Haplotypes characterized by the presence of accessions from different populations (Fig. 5). The distribution of individuals was therefore not made according to the populations considered. There is a mixture of individuals from populations in each haplotype.

Discussion
Genomic DNA isolation of D. microcarpum: an essential step in genetic molecular analysis DNA isolation is a crucial and the rst step of a DNA-based genetic characterization study. The quality and purity of nucleic acids are among the most critical factors for subsequent analysis. Like some plants including Ceratonia siliqua (Konaté 2007), D. microcarpum is a leguminoseae very rich in secondary metabolites such as polysaccharides, polyphenols, tannins, proteins, etc. These secondary metabolisms can bind to the DNA molecule, making it di cult to extract and access enzymes (Lakshmi et al. 2002). This is why our rst intentions and efforts in this study were concentrated on the development of an isolation technique capable of having a good quality of DNA which can be used as a template for the ampli cation of DNA by PCR. The de ciency of DNA isolation protocol adapted to Detarium microcarpum and the expensive cost of DNA extraction kits obliged us to test and modify some DNA isolation protocols including that of Dellaporta et al. (1983), Doyle and Doyle (1987), Benbouza et al. (2006), Adjé et al. (2016) using Cetyl Triammonium Bromide (CTAB) and that of Agbangla et al. (2002) using Mixed Alkyl Timethyl Ammonium Bromide (MATAB). The DNA isolation protocol set up by Dellaporta et al. (1983) and de Doyle and Doyle (1987) using CTAB are very e cient and commonly used to extract genomic DNA from woody plants (Dwivedi et al. 2001;Li & Nelson 2002;Mongkolporn et al. 2003;Konaté 2007).
However, their application to the leaf of D. microcarpum was inconclusive and the resulting bundle was either very viscous and insoluble or free from any trace of DNA. This led us to the various modi cations to the protocol including the use of β-Mercaptoethanol and proteinase K considered as chemical agents having a positive and effective impact on the dissolution of the tannin-protein complex, responsible for the anchoring of several constituents cells of certain plants and the elimination of plant proteins (Konaté 2007). The results of the DNA sample veri cation testify to the e ciency and cost-effectiveness of the modi ed Doyle and Doyle (1987) protocol on D. microcarpum.
Ampli cation of cpDNA sequences by PCR -Chloroplast microsatellite primers tested on the accessions of D. microcarpum were developed on plant species belonging to the Solanaceae family with the aim of allowing their application in the study of the variability of chloroplast DNA in other species or families of Angiosperms (Bryan et al. 1999;Weising & Gardner 1999). Region-speci c ampli cation of D. microcarpum accessions cpDNA by these primers showed that genetic molecular markers developed in one species can be used in molecular studies of other species. However, the absence of ampli cation by some primers testi es to the limit of the universality of these primers as referred to, but above all the need to develop speci c primers for each species of plant. This lack of ampli cation observed in the present study with certain primers was also observed in other species such as Vitis vinifera (Vitaceae), Helianthus annuus (Asteraceae), Dioscorea sp.
Spatial genetic structuring and phylogeny between D. microcarpum populations -Molecular studies based on chloroplast DNA (cpDNA) focus on the one hand, the study of the mechanisms that control its evolution and on the other hand, the use of its variation as a tool to study the genetic diversity of plant species (Grivet 2002). In Angiosperms, cpDNA is the most used in the study of phylogenetic relationships. Indeed, the small size of cpDNA, its abundance in the plant cell, its maternal single-parent inheritance and especially its conservative rate of evolution (no genetic mixing), make the chloroplast genome a privileged tool for phylogenetic studies ( The highest values of the effective allele, the expected heterozygosity and allelic richness observed in the South Borgou phytogeographic district means that D. microcarpum has a dispersal capacity and a high degree of isolation in this population (Govindaraj et al. 2015). The relatively high value of the expected heterozygosity implies a high degree of dissimilarity within the species studied. Wright xation index recorded in all the different populations are low (F IS 1). In addition, the F IS calculated in the Atacora Chain phytogeographic district is negative (F IS 0) and re ects excess heterozygosity. These low rates of Wright xation index can be explained by the high rate of allofertilization and / or dissemination that exists between the different individuals of D. microcarpum in the same population even though the species is autogamous (Hutchinson et al. 1958 The presence of individuals belonging to different populations of D. microcarpum within the three haplotypes clusters could be justi ed by a probable common origin of these populations or the existence of gene ows between the different natural populations of the species. This nding could be explained by the ow of genes within individuals from one population to another. These results corroborate those of Fontaine et al. (2004) who have shown that gene ow leads to the exchange of genes (or their alleles) between different populations. This gene ow observed in D. microcarpum populations is due to the mode of dispersal of seeds (dissemination) and pollens (pollination) of the plant by various agents. According to Kouyaté (2005), the seeds of D. microcarpum are dispersed by wind (anemochory), animals such as mammals and birds (zoochory) and humans (anthropochory). Although the species is self-pollinating, it possesses a fraction of allogamy. Its owers are pollinated by wind (anemogamy) and insects such as ladybugs, ies and bees (entomogamy).
These various disseminators and pollinators contribute to a wide dispersion of seeds and pollens. This promotes a signi cant ow of genes within populations of the species. Indeed, the mode of seed dissemination and that of ower pollination are the main factors responsible for gene ow within a plant species (Gerber-Danjon 2006).

Conclusion
We found distinct patterns of genetic diversity at different spatial scales.   Analytical electrophoresis pro les on agarose gel of genomic DNA from some Detarium microcarpum samples.  Neighbor-joining phylogenetic tree showing haplotypes clusters from Detarium microcarpum populations. Codes written with the same colour represent individuals of D. microcarpum belonging to the same population.