Findings

Background

The genus Prosopis L. belongs to the Leguminosae botanical family, which contains 44 species. Prosopis L. is predominantly restricted to the neotropics [1]. Prosopis rubriflora[2] and Prosopis ruscifolia[3] are tree species known locally as “espinheiro” and “algarroba,” respectively. These species are important both economically and ecologically. For example, the fruits and seeds of P. ruscifolia are reported to be good sources of nutrition for humans and animals [4], and the flowers of P. rubriflora, which are present throughout the year, provide important food resources, such as pollen and nectar, for the local fauna [5]. P. rubriflora has a narrow distribution range and is limited to Paraguay and Brazil, but P. ruscifolia is also found in Argentina and Bolivia [6, 7].

In Brazil, P. rubriflora and P. ruscifolia are associated with Chaquenian areas [8] and are limited to the southern portion of the Pantanal [9, 10]. Both species are excellent indicators of Chaquenian areas in Brazil; P. rubriflora is usually associated with arboreal physiognomy, and P. ruscifolia is frequently associated with forest physiognomy. Both species can be used as models for genetic studies of diversity in these areas.

While estimating genetic diversity, the use of molecular markers has been helpful in defining alleles and studying genetic flow, population structure, paternity, inheritability, genetic maps and conservation genetics [11]. Simple sequence repeat markers (SSRs), commonly referred to as microsatellite markers, are desirable tools because they are co-dominant in nature, multi-allelic and widely distributed in the genome; they are also currently cheap, reproducible and relatively easy to analyze [12]. This work reports the development, characterization and transferability of microsatellite markers for P. rubriflora and P. ruscifolia.

Construction of a microsatellite-enriched library

DNA was extracted from P. rubriflora and P. ruscifolia using the DNeasy® Plant Mini Kit (Qiagen, Hilden, DE) according to the manufacturer’s instructions. Microsatellite-enriched libraries for P. rubriflora and P. ruscifolia were constructed as described by Billote et al.[13]. The genomic DNA was digested with Afa I after enrichment with streptavidin-coated magnetic beads (Streptavidin MagneSphere Paramagnetic Particles, Promega, Madison, WI); biotinylated (CT)8 and (GT)8 microsatellite probes were added for the dinucleotide-enriched library. The fragments were amplified by PCR and cloned into the pGEM-T vector (Promega, Madison, WI). XL1-Blue (Escherichia coli) competent cells were transformed with the recombinant plasmids and then cultivated on agar medium containing ampicillin (100 mg/ml), X-galactosidase 2% (100 μg/ml) and IPTG (100 mM). The selected clones were added to a Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) and sequenced using an ABI 377 sequencer (Applied Biosystems, Foster City, CA). The sequences were aligned and edited using SeqMan Software (DNAStar, Madison, WI), and the adapters and restriction sites were removed using Microsat Software (A. M. Risterucci, CIRAD, personal communication). To identify microsatellite-enriched regions, we used the Simple Sequence Repeat Identification Tool (SSRIT) [14] and defined the following numbers of repeats/motifs: five/dinucleotides, four/trinucleotides and three/tetra- or pentanucleotides. After these steps, primers were designed using the PrimerSelect software (DNAStar, Madison, WI).

Fragment amplification

The fragments were amplified using polymerase chain reactions containing 8 ng of template DNA, 2 mM MgCl2, 50 mM KCl, 20 mM Tris–HCl (pH 8.4), 0.2 mM dNTPs, 0.19 mg/ml BSA (bovine serum albumin), 0.15 mM of each primer and 1 U of Taq DNA polymerase; the reactions were then brought to a final volume of 20 μl with ultrapure water. To define the temperatures for the PCR reactions, we adopted the guidelines described by Mottura et al.[15]; for the annealing temperatures, we used a gradient program with temperatures ranging from 65°C to 55°C. The samples were collected in the Chaco remnants of Corumbá and Porto Murtinho, Mato Grosso do Sul, Brazil. Twenty P. rubriflora samples were collected in each of two Chaco remnant locations: Fazenda São Manoel (FSM) (21°47′44.5″S; 57°39′34.6″W) and Fazenda Santa Vergínia (FSV) (22°06′40.5″S; 57°49′57.6″W). Twenty-three P. ruscifolia samples were collected in Estação do Carandazal (ECD) (19°48′33.2″S; 57°10′11.0″W), and 25 samples were collected in Fazenda Retiro Conceição (FRC) (21°42′23.7″S; 57°45′58.2″W). The cross-amplification of the markers was evaluated in 5 P. rubriflora samples obtained from FRC (21°41′00.7″S; 57°46′43.8″W) and 5 P. ruscifolia samples from Chácara Jacaré (21°41′20.1″S; 57°52′15.5″W) using the same conditions as for the native species. The amplified samples were genotyped by vertical electrophoresis using denaturating polyacrylamide gels (6%), and DNA bands were visualized using silver nitrate [16]; the sizes of the resulting fragments were estimated by comparison to a 10-bp DNA ladder (Invitrogen, Carlsbad, CA). Statistical analyses were performed using Microsatellite Toolkit v.3.1.1 [17] to calculate the expected heterozygosity (He), observed heterozygosity (Ho) and polymorphism information content (PIC). The Genepop software v.1.2 [18] was used to estimate adherence to Hardy-Weinberg (HW) equilibrium and possible linkage disequilibrium (LD), and the frequency of null alleles was estimated using FreeNA [19].

Results and discussion

We designed 32 primer pairs: 13 for P. rubriflora and 19 for P. ruscifolia. However, only 10 of the P. rubriflora primer pairs and 13 of the P. ruscifolia primer pairs amplified properly. The nine remaining pairs of primers were discarded because amplification errors were observed in the preliminary tests. Polymorphisms were detected in 9 of the native P. rubriflora markers and 12 of the native P. ruscifolia markers; only one marker from each species had a monomorphic pattern based on the populations analyzed. Eight markers from P. rubriflora successfully cross-amplified and were polymorphic for the tested samples, and 2 markers failed during cross-amplification. Eleven P. ruscifolia markers were successfully cross-amplified; 7 were polymorphic, and 2 failed this analysis (Table 1).

Table 1 Primers developed for Prosopis rubriflora and Prosopis ruscifolia
Full size table

The number of P. rubriflora alleles in the sampled remnants ranged from 3 to 12; the polymorphism information content (PIC) values of these markers ranged from 0.073 to 0.791, the observed heterozygosity (Ho) ranged from 0.000 to 0.850, and the expected heterozygosity (He) ranged from 0.000 to 0.835. No evidence of null alleles was observed, and no departure from Hardy-Weinberg equilibrium was observed (Table 2). No significant linkage disequilibrium (LD) was observed for any of the markers of this species after Bonferroni correction (P-value for 5% = 0.001389). The number of P. ruscifolia alleles in both of the remnants ranged from 3 to 17, the PIC values ranged from 0.289 to 0.883, the Ho values ranged from 0.040 to 0.783, and the He values ranged from 0.275 to 0.884. Possible null alleles were observed for the markers Prsc5, Prsc8 and Prsc9 from one remnant (ECD), and the markers Prsc4 and Prsc10 had possible null alleles in both remnants. A departure from HW was observed for Prsc2, Prsc5, Prsc6, Prsc8, Prsc9 and Prsc11 in one of the remnants (the majority were observed in ECD) and for Prsc4, Prsc7 and Prsc10 in both remnants (Table 3). Significant LD was observed for the loci Prsc5 and Prsc6 after Bonferroni correction (P-value for 1% = 0.00016).

Table 2 Markers developed for Prosopis rubriflora
Full size table
Table 3 Markers developed for Prosopis ruscifolia
Full size table

Higher values of Ho were observed for the Prb1, Prb2, Prb4, Prb6, Prb7, Prsc1, Prsc3 and Prsc5 markers in this study; these higher values may indicate that an insufficient number of samples was collected or may be related to the reproductive patterns of these populations. The ECD populations are highly disturbed, and the FRC population is currently recovering from a relatively recent suppression (within the last 15 years); these factors may underlie the observed departure from HW and the presence of null alleles. A study with new and conserved populations may produce better results for these markers.

These markers are the first microsatellite markers developed for Prosopis rubriflora and Prosopis ruscifolia, and together with the set of P. ruscifolia markers amplified by Bessega et al.[20], they are expected to be useful tools for studies of the conservation genetics, reproductive biology, phylogeography and taxonomy of these species.

Availability of supporting data

The original sequences of the developed markers were submitted to the GenBank database (http://ncbi.nlm.nih.gov), and the registered codes are available in Table 1.

The testimony samples were deposited at Herbarium Universidade Estadual de Campinas (UEC – Campinas, SP, BR) and registered according to the following: P. rubriflora – 74477 (Fazenda São Manoel – Porto Murtinho, MS), 154715 (Fazenda Santa Vergínia – Porto Murtinho, MS); P. ruscifolia – 74469 (Fazenda Retiro Conceição – Porto Murtinho, MS), 37266 (Estação do Carandazal – Corumbá, MS).