Anthonotha P. Beauv. (Fabaceae) is an African native genus belonging to the monophyletic tribe Detarieae. Anthonotha species are found in evergreen to deciduous tropical African forests. Breteler (2010) recognizes 17 species almost completely confined to the Guineo-Congolian region, but species distinction is not always easy without flowers. Among these 17 species, A. macrophylla P. Beauv. is the most common and frequently collected species of the genus. It is a shrub or tree that usually grows 4–20 m tall and is one of the forest tree species found in the Holocene Climate Optimum forest relics in the Dahomey Gap. Its wide and nearly continuous distribution from Guinea to the Democratic Republic of the Congo (west and central African rainforest) should be useful to study the impact of past climate change on tropical African forest from genetic diversity pattern and phylogeographic and demographic inferences. To date, no microsatellite resources have been developed for Anthonotha species.
In this paper, we isolated and characterized a set of 18 polymorphic microsatellite markers for Anthonotha. These markers will complement the ones developed for Terminalia superba Engl. & Diels (Demenou et al., 2015) to study the history of fragmentation of the tropical African rainforest in the Dahomey Gap. We also attempted cross-amplification in 15 congeneric Anthonotha species.
METHODS AND RESULTS
Microsatellite development —Total genomic DNA of A. macrophylla was extracted (ca. 5 µg) from 30 mg of silica gel–dried leaf collected from a sample coded OH3840 (2.30018°N, 25.02499°E; Appendix 1) from the Democratic Republic of the Congo using a cetyltrimethylammonium bromide (CTAB) method (Fu et al., 2005). The extracted DNA was used to prepare a DNA genomic library without enrichment, following the protocol of Mariac et al. (2014), and sequenced using the Illumina (San Diego, California, USA) MiSeq platform (sequencing performed at CIRAD, Montpellier, France) as described in Demenou et al. (2015), which generated 28,902 150-bp-long paired-end reads. After assembling the paired reads with PANDAseq (Masella et al., 2012), the identification of simple sequence repeats (SSRs) and design of primers were performed with the bioinformatics pipeline QDD (Meglécz et al., 2014) following three steps: (1) selection of sequences containing SSRs, (2) elimination of redundant sequences, and (3) primer design. We detected 1109 loci (>7 repeats) between 3246 reads containing microsatellite motifs. From these, we selected 48 primer pairs representing the longest dinucleotide repeats with PCR product length ≥100 bp and flanking region length of at least 15 bp from the microsatellite. Finally, using an M13-like protocol of Micheneau et al. (2011), we attached one of the four possible linkers (Q1–Q4) to the 5′ end of the forward primer of each locus to label PCR products with the distinct fluorochromes FAM, NED, VIC, and PET.
Amplification for each pair of designed primers was evaluated in three individuals of A. macrophylla from Benin (EE271; 6.96013°N, 2.67641 °E), Cameroon (BS102; 5.10500°N, 11.40056°E), and Côte d'Ivoire (GK1034; 6.42321 °N, 7.48098°W) (Appendix 1). PCR reactions (13 µL) were performed using 1 µL of DNA (ca. 50 ng/µL), 1.5 µL PCR buffer (10×), 0.6 µL MgCl2 (25 mM), 0.45 µL dNTP (10 mM each), 0.3 µL of each primer (0.25 µM), 0.08 µL TopTaq DNA polymerase (5 U/µL; QIAGEN, Venlo, The Netherlands), and 8.77 µL H2O using the following conditions: an initial step at 94°C for 4 min; followed by 30 cycles of 30 s at 94°C, 45 s at a primer annealing temperature of 55°C, and 1 min at 72°C; and a final extension of 10 min at 72°C. PCR products were visualized on a 1% agarose gel and stained with SYBR Safe (Invitrogen, Merelbeke, Belgium).
All but two of the 48 primer pairs amplified consistently. Polymorphism was assessed on the same three previously amplified individuals (Appendix 1). For this step, PCR amplification was performed for each of 46 loci with fluorescent labeling in a total volume of 15 µL, combining: 0.3 L of the reverse (0.2 µM) and 0.1 µL of the forward (0.07 µM) microsatellite primers with a Q1–Q4 universal sequence at the 5′ end, 0.3 µL of Q1–Q4 labeled primer (0.2 µM each), 3 µL of Type-it Microsatellite PCR Kit (QIAGEN), H2O, and 1.5 µL of DNA. Cycling conditions were as described above with 30 cycles and primer annealing temperature of 55°C. A mix of 1 µL of each PCR product with 12 µL of Hi-Di Formamide (Life Technologies, Carlsbad, California, USA) and 0.3 µL of Map-Marker 500 labeled with DY-632 (Eurogentec, Seraing, Belgium) was run on an ABI3730 Capillary Sequencer (Applied Biosystems, Lennik, The Netherlands). Electropherograms were analyzed with GeneMapper version 3.7 (Applied Biosystems). Twenty-eight loci were discarded because of lack of amplification, genotyping difficulties, or unreadable electropherograms. The remaining 18 selected polymorphic loci were combined into three multiplexed reactions (Table 1) using Multiplex Manager 1.0 software (Holleley and Geerts, 2009).
Characteristics of 18 polymorphic microsatellite loci for Anthonotha macrophylla.
Microsatellite marker data analysis —We evaluated the quality of these 18 microsatellite markers in three populations of A. macrophylla from southern Benin (N = 19), southern Liberia (N = 35), and eastern Cameroon (N = 28) (Appendix 1). Multiplex PCR reactions were carried out as described above to check polymorphism except that we added 3 µL of 5× Q-solution and readjusted the quantity of H2O for a total volume of 15 µL. Multiplex PCR programs consisted of 94°C (5 min); followed by 22 cycles of 95°C (30 s), 56°C (90 s), and 72°C (1 min); followed by 10 cycles of 94°C (30 s), 53°C (90 s), and 72°C (1 min); and a final extension of 10 min at 72°C.
We computed the parameters of allele size range, observed number of alleles (A) per locus, observed (Ho) and expected (He) heterozygosities, inbreeding coefficient (F), and null allele frequencies (r) with INEst 1.0 (Chybicki and Burczyk, 2009) for each locus and population. We also tested deviation from Hardy–Weinberg equilibrium (HWE) for each locus with SPAGeDi (Hardy and Vekemans, 2002).
The number of alleles per locus ranged from two to 24 (average of 11.9 alleles per locus; Table 2). Ho and He ranged from 0 to 0.74 (average: 0.38) and from 0.05 to 0.89 (average: 0.48) for the Benin population, from 0 to 0.86 (average: 0.41) and from 0 to 0.93 (average: 0.58) for the Liberia population, and from 0 to 0.75 (average: 0.43) and from 0.04 to 0.89 (average: 0.63) for the Cameroon population (Table 2), repectively. Significant deviation from HWE (Table 2) was observed for four loci (AntM-ssr08, AntM-ssr09, AntM-ssr27, and AntM-ssr06) in the Benin population, for seven loci (AntM-ssr26, AntM-ssr42, AntM-ssr09, AntM-ssr04, AntM-ssr33, AntM-ssr21, and AntM-ssr06) in the Liberia population, and for nine loci (AntM-ssr26, AntM-ssr08, AntM-ssr42, AntM-ssr27, AntM-ssr39, AntM-ssr02, AntM-ssr43, AntM-ssr16, and AntM-ssr06) in the Cameroon population due to the presence of null alleles. After accounting for the effect of null alleles, INEst inferred no inbreeding across populations (F = 0.00 ± 0.00), indicating an outcrossing mating system. The sequences of the developed microsatellite loci have been deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (Bioproject ID PRJNA352928).
Genetic properties of the 18 polymorphic microsatellite loci for three populations of Anthonotha macrophylla. a
Results of cross-amplification (allele size ranges) of microsatellite loci isolated from Anthonotha macrophylla tested in 15 other Anthonotha taxa.a
Cross-amplification in 15 congeneric Anthonotha species —The selected loci were then tested in one to nine individuals of 15 other Anthonotha species (Table 3) using the PCR conditions described above to check their transferability. Among the 18 loci, six to 17 (mean: 13) successfully amplified depending on the species and displayed one to nine alleles per locus (results not shown). The allelic size varies among species for a given locus, but a few alleles are shared by up to 10 species (e.g., alleles 158 and 160 for locus AntM-ssr21, alleles 205 and 207 for locus AntM-ssr39, alleles 213 and 215 for AntM-ssr15). Private allelic richness (average over loci) computed with HP-rare 1.1 (Kalinowski, 2005) for each species indicated that A. pellegrinii Aubrév. shows the highest value (0.28), followed by A. gilletii (De Wild.) J. Léonard (0.18), A. cladantha (Harms) J. Léonard (0.17), and A. noldeae (Rossberg) Exell & Hillc. (0.17); therefore, these species are likely the most divergent with A. macrophylla. According to the data of this study, no allele of a given locus is shared by all species.
In this study, 18 polymorphic microsatellite markers were developed for A. macrophylla. This set of microsatellite markers showed its tranferability in most of 15 congeneric species. These microsatellite markers and those published on T. superba will be useful for investigating phylogeographic patterns, dispersal patterns, and demographic history of Anthonotha species to provide a better understanding of the fragmentation history of tropical African rainforests in the Dahomey Gap. With them, one can start to disseminate, for example, paleovegetative information for this region.
The authors thank the Fonds pour la Formation à la Recherche dans l'Industrie et l'Agriculture (FRIA, Belgium), the Fonds de la Recherche Scientifique (F.R.S.-FNRS, grant T.0163.13), and Belgian Science Policy (project AFRIFORD) for financial support. The authors also thank Esra Kaymak for help in the laboratory, Ebenezer Ewedje and Armel Donkpegan for technical assistance, and Jan Wieringa for allowing herbarium sampling in Leiden (Naturalis Biodiversity Center).