** = Geographic origin of individuals selected for the first screening
Freshly collected leaves were dried in a coffee filter containing 10 g Silica gel. Each filter containing sample was put in a plastic zip bag for transport to the laboratory. Our genomic library was constructed using DNA of P. erinaceus samples from twelve randomly selected individuals among populations (Table 1).
DNA Extraction
Total genomic DNA extraction was performed with a solution of alkyltrimethylammonium bromide (MATAB) using ttwenty milligrams of dried leaves from each tree sample. Extraction protocol used derived from Bousquet et al. (1990) methodology.
The quality of the genomic extracted DNA was controlled on a 1% agarose gel, and quantification done by Hoescht assay using fluoroskan (Fluoroskan™ Microplate Fluorometer).
Construction of the DNA library and validation
The Westburg NGS DNA Library PrepKit was used to prepare the DNA bank with extract mixed DNA of twelve individuals. We started with 1µg ofPterocarpus mixed DNA. This DNA was fragmented, and ligated with Illumina adapters. Purification on magnetic beads (Agencourt AMPure XP beads -A63881-, Beckman Coulter) was performed before and after PCR. PCR amplification was run with 7 cycles.
The quality of DNA library was controlled using an Agilent 4200 TapeStation with a screen tape D5000. The size of the fragments were expected between 100 and 600 pb. The library was quantified using the Takara kit (638324) on a qPCR machine (LightCycler® 480 Real-Time PCR System, Roche Life Science).
Sequencing the DNA library
MiSeq system Illumina sequencer DNA was used to perform DNA sequencing on the genotyping platform at CIRAD-Montpellier. A 500 cycles NANO V2 cartridge Illumina (2 x 250 pb) was used to sequence the library.
Design and choice of primers
A total of 800,000 reads was generated for P. erinaceus samples. Development of optimised and streamlined microsatellites was based on the bioinformatics Galaxy pipeline and its following tools: FASTQ Groomer tool, Filter FASTQ tool and ABySS parallel assembler (Simpson et al., 2009). The MISA MIcroSAtellite identification tool (Thiel, 2003) and primer modelling software Primer3 (Whitehead Institute) were used for the identification and design microsatellites primers in nucleotide sequences generated. A data matrix containing all the microsatellite primers was obtained as output.
Among the 38,715 single sequence repeats identified, primers were designed for 11,718 sequence repeats of which 3,530 were dinucleotide repeats, 2,970 trinucleotide repeats, 2,847 tetranucleotide repeats, 1,001 pentanucleotide repeats, 525 hexanucleotide repeats, and 844 contained complex SSR motifs.
Dinucleotide and trinucleotide microsatellites motifs were selected for SSR-PCR amplification screening. These primers exhibited a minimum of five repetitions of the repeated motif and amplified fragments between 100 and 400 bp in length. Thirty suitable microsatellite were identified and selected for initial screening. This first test was performed on an ABI 3500XL sequencer (Life technologies, Carlsbad, California, USA) using genomic DNA extract from eight individuals selected from different countries (Table 1).
An M13 tailed primer (5′-CACGACGTTGTAAAACGAC-3′), allowing detection of fluorescence was added to the forward primers. Each PCR amplification was performed in a 96-well plate using 10-μL volume reaction containing 20 ng of DNA, 1X PCR buffer (without MgCl2), 0.08 μM of the M13-labeled primer, 0.1 μM of each forward fluorescent primer (FAM, NED, PET, and VIC) and the reverse primer, 0.1 μM of M13 fluorescent primer, 2 mM of MgCl2, 200 μM dNTPs, 0.4X Solution “S” (additive solution that facilitates amplification of difficult templates) and 0.05 U/µL of Taq DNA polymerase. PCR running conditions were: initial denaturation at 94°C for 4 min followed by 36 cycles each at 92°C in 30 s, 1 min at 52°C, 45 s at 72°C and with a final extension step at 72°C for 5 min.
Electropherograms were analysed and allele sizes were determined with GeneMapper® software version 4.1 using GeneScan 600 LIZ as a size standard (Applied Biosystems). Among the 30 primer pairs tested, 17 were selected. Indeed, we eliminated primers with profiles that were difficult to read on GeneMapper®, or with no or little polymorphism. The 17 selected primers are shown in Table 2 and they were used for screening of remaining individuals in order to calculate genetic parameters.
Genetic parameters including alleles’ number per locus, observed (Ho) and expected (He) heterozygosity were computed using GenAIEx software version 6.0 (Peakall and Smouse, 2012). Deviation from the Hardy–Weinberg equilibrium (HWE) was measured for each locus by χ2 tests, and with a Bonferroni correction procedure (Rice, 1989), p-value significance was assessed in the context of multiple testing. Significant linkage disequilibrium was rated among these loci by using GENETIX software version 4.05 (Belkhir et al., 1996). MICRO-CHECKER software version 2.2.3 (Van Oosterhout et al., 2004) was used to check for the null alleles in microsatellite data.
Results and Discussion
A total of 237 alleles were identified for the 17 locus on the 365 characterized trees, with each locus having from 4 to 30 alleles (mean of 13.9 alleles per locus). Mean values for the expected heterozygosity (He) varied from 0.307 to 0.781 (0.571 ± 0.176) and from 0.234 to 0.821 (0.551 ± 0.190) for observed heterozygosity (Ho) for the individuals screened. Evidence of significant linkage disequilibrium, was found for 12 for 136 possible SSR pairwise combinations after Bonferroni corrections. Significant departures from Hardy–Weinberg equilibrium for 14 of the 17 loci was recorded after Bonferroni corrections and presence of null alleles was suggested for all loci excepted for mPeCIR_D2 and mPeCIR_T3. This set of 17 pairs of specific primers ofPterocarpus erinaceus would serve to study the genetic diversity of this species in West Africa.
Table 2 : Characteristics of 17 microsatellite primers designed for Pterocarpus erinaceus Poir.