Transferability and characterization of microsatellite markers from Byrsonima cydoniifolia A. Juss. (MALPIGHIACEAE) in seven related taxa from Cerrado biome reveal genetic relationships

Byrsonima Rich. is one of the largest genera of the Malpighiaceae family with 97 species occurrence in Brazil and multiple potentialities, including pharmaceutical and food industries. In this study, 17 microsatellite markers characterized in Byrsonima cydoniifolia were tested for seven related taxa, all species are native to Brazil and four are endemic. Genomic DNA was extracted from leaves tissues and 17 microsatellite markers were used to cross-amplification of microsatellite regions. Polymorphism and genetic diversity were evaluated for B. intermedia, B. verbascifolia, B. laxiflora, B. subterranea, B. umbellata, B. linearifolia. from 16 individuals and for B. viminifolia from 14 individuals. Transferred microsatellite markers panels ranged from 11 (64.8%) in B. viminifolia to 6 (35.2%) in B. umbellata. The total number of alleles per locus ranged from 5 (B. linearifolia) to 8 (B. subterranea) alleles. B. umbellata showed lower values of observed and expected heterozygosity (HO = 0.312; HE = 0.436) and B. subterranea presented the highest values (HO = 0.687; HE = 0.778). A greater number of microsatellite markers should be developed for B. umbellata. The microsatellite marker panels transferred to the species B. intermedia, B. verbascifolia, B. laxiflora, B. subterranea, B. viminifolia and B. linearifolia are very informative, with a high combined probability of exclusion of paternity (Q ≥ 0.976) and the low combined probability of identity (I ≤ 9.91 × 10–6), potentially suitable for future genetic-population studies, supporting strategies for maintaining the genetic diversity and for exploration of Byrsonima species as genetic resources.


Introduction
Byrsonima Rich., one of the largest genus belonging to the Malpighiaceae family, contains approximately 150 species distributed from Mexico to South America. From these about 97 species occurs in Brazil with a wide distribution in Cerrado Biome, popularly known as muricizeiro or murici [1]. Rarely cultivated, Byrsonima species have an enormous agricultural economic potential can be used in ornamentation, reforestation of degraded areas, and forage in times of lack of pasture [2,3]. Leaves are used in folk medicine and in the tannery industry. Fruits and the oil extracted from the seeds are used in the food and pharmaceutical industry [4][5][6]. Fruit extracts present a potent anti-inflammatory and antihyperalgesic activity [5,7] and antibacterial activity [8].
In order to carry out an effective conservation and domestication strategy for these species as genetic resources and make feasible its economic use in a non-extractive way, 1 3 knowledge about the genetic variability of Byrsonima species it's necessary [3,6]. A highly useful tool that allows the determination of genetic variability and genetic identity is the molecular marker, such as microsatellite markers. SSR markers are widely used for detecting genetic diversity and population structure, gene flow patterns, the incidence of genetic drift, paternity analysis, linkage of phenotypic and genotypic variation, identification of conservation units, creation of germplasm banks and for breeding programs [9,10]. These markers are potentially transferable among related taxa, due to the homologous nature of the DNA sequence in the flanking regions of the microsatellites. As these regions are highly conserved between taxa, this allows cross-amplification between related species to be more costeffective [11,12].
Aiming at the importance of developing future strategies of conservation and economic exploitation of the numerous species of the genus Byrsonima sp. (murici), in the present study we reported the ability of SSR markers developed for B. cydoniifolia [13] to amplify SSR loci in other seven Byrsonima species. Moreover, we characterized the genetic variability of the microsatellite regions transferred for each species, in order to determine how informative are the transferred loci. These polymorphic markers can be used to conduct effective conservation strategy, domestication planning, parentage studies, and creation of germplasm bank using Byrsonima sp. as a genetic resource, and moreover to unveil phylogenetic problems for its related taxa. Genomic DNA was extracted from leaves tissues of each individual of Byrsonima sp. following the cetyltrimethylammonium bromide (CTAB) 2% protocol [14]. DNA yield and quality were visually assessed on 1% agarose gels.

Cross-species amplification of SSR markers and genotyping
All seventeen primer pairs developed to amplify microsatellite regions for the species B. cydoniifolia [13] were used to perform the cross-amplification tests on the species of Byrsonima. DNA amplification using each microsatellite primer pair was performed in a GeneAmp PCR System 9700 thermocycler (Applied Biosystems). The PCR reactions were performed in a total of 10 µl reaction volume: 0.23 mM of primer (foward+reverse), 0.23 μM of dNTP, 3.25 mg of bovine serum albumin (25 mg/ml), and 1 × reaction buffer (10 mm Tris-HCl pH 8.3, 50 mm KCl, 1.5 mm MgCl 2 ), 0.75 U of Taq DNA polymerase (5U-Phoneutria, Belo Horizonte, Brazil), and using 3.75 ng of template DNA. The PCR program performed the amplification following the protocol: initial denaturation at 94 °C for 5 min, followed by 35 cycles at 94 °C for 1 min, specific annealing of the primer at X °C (this annealing temperature varies according to the primer 50-62 °C) for 1 min, extension at 72 °C for 1 min and final extension at 72 °C for 45 min.
The PCR products were submitted to electrophoresis on a 3% agarose gel stained with ethidium bromide (10 mg/ml), in the presence of 1X TBE buffer, for 60 min at a voltage of 70 V, and with a molecular weight marker 100 bp. PCR was repeated for microsatellite loci with amplification products, optimizing the minimum annealing temperature of the primer pairs at 48 °C.
Locus with the cross-amplification pattern was analyzed by capillary electrophoresis. Genotype information was obtained by fluorescence detection based on the separation of fluorophores labeled microsatellite fragments organized in multiplex panels by capillary electrophoresis using an automatic DNA analyzer, ABI-3500 automated sequencer (Applied Biosystems, Foster City, California, USA) and the ROX internal size standard GS500. Forward primers were pre-labeled with fluorophores at their 5′end, having one of the fluorophores described below: green (HEX), blue (6-FAM), and yellow (NED). PCR products were pooled in multiplex panels created according to amplified fragment sizes, could be the same dyes as long as the alleles presented different sizes of at least 10 bp, and different dyes to minimize costs and maximize efficiency. Allele binning and calling was performed using GENEMAPPER 5.0 software (Applied Biosystems), using a detection window (bin) of 1.99 bp.

Validation of transferred SSR markers and genetic diversity analysis
The individual discrimination power was calculated through the genetic identity index (I) [15]; and the paternity exclusion index (Q) (Weir, 1996) using the IDENTITY 1.0 program [16] from the allelic frequencies for each locus and for the total set of loci. All microsatellite loci transferred were evaluated for their use as polymorphic markers for population genetics studies, so analysis of genetic diversity, i.e., the number of alleles per locus (A), observed heterozygosity (H O ), and expected heterozygosity (H E ) were performed by the program FSTAT 2.9.3.2 [17]. Also, tests of Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (P > 0.05) in all pairs of loci, in both, using Bonferroni correction.
For each locus the values of the effective number of alleles per locus (A E ) were measured, calculated from the formula: A E = 1/(1−H E ), and the values of maximum possible heterozygosity, based on the number of alleles per locus (A), from the formula: H E máx = (A−1)/A. The proportion of the maximum genetic diversity obtained by the transferred markers was calculated by H E /H E máx (%) [18].

Cross-species amplification of SSR markers
In general, we obtained success in the cross-amplification of seventeen SSR markers developed for B. cydoniifolia A. Juss in the other seven related taxa. Data with the name of locus, primer sequence and number of Genbank accession are available in Supplementary information. For the 17 pairs of tested primers, the presence of satisfactory amplification, i.e. polymorphic loci without nonspecificity, ranged from 11 loci (64.8%) in B. viminifolia to 6 loci (35.2%) in B. umbellata. Only the species B. laxiflora presented two monomorphic loci (11.8%). The presence of non-specific amplification was found for almost all species, with the exception of B. umbellata and B. viminifolia, ranging from three loci (17.7%) in B. intermedia to one locus (5.8%) in B. subterranea and B. linearifolia (Table 1).
For cross-amplification parameters, the annealing temperature ranged from a minimum of 50ºC to a maximum of 62 °C, while the allele size range (bp) varied from a maximum amplitude of 136-216 bp to a minimum amplitude of 137-139 bp for the BCY16 locus in B. umbellata. No linkage disequilibrium were detected for any pair of loci for the seven species. After Bonferroni correction, significant deviations from the Hardy-Weinberg equilibrium (P < 0.05) for the fixation index (F IS ) were observed for four locus ( Table 2).
Deviations from HWE occurred due to the sampling effect, in this case, only 16 individuals or 14 individuals per species, since the number of genotyped individuals may not be sufficient to find individuals in all possible genotypic classes, causing deviations in genotypic frequencies. For a locus to be HWE compliant, it must meet many conditions, such as conformance with Mendelian segregation, random mating, absence of recent mutations and genetic drift, and also infinite population. Thus, deviation from the HWE test can indicate that one of the conditions is not being met, in this case, infinite population, since it is usually required a large sample to conform to the 'infinity population' requirement [19].
The detected of maximum possible genetic diversity using transferred SSR markers ranged from 92 (B.   The highest values of genetic diversity were found for B. subterranea (H E = 0.778) using 9 transferred SSR markers and B. viminifolia (H E = 0.774) using 11 transferred SSR markers (Table 3). Therefore, it is noted that despite having a greater number of markers for B. viminifolia, this fact did not demonstrate an increase in the genetic diversity values compared to other species with smaller amounts of 8 or 9 SSR markers.
Greater Although the number of SSR markers transferred for each species varies, an average of 8.43 markers per species, it is noted that when transferred a minimum of 7 markers (B. linearifolia), they proved to be a set of highly informative markers for such species, able to exclude false paternity and demonstrated efficiency in discriminating individuals.
The lowest values of H E and H O were identified in B. umbellata (H E = 0.436; H O = 0.312), probably due to the small number of SSR markers transferred (6). The set of developed SSR markers presented reasonable information (I = 0.778; Q = 2.5 × 10 -3 ). Thus, it is necessary to develop more SSR markers to detect high levels of genetic variability for future advanced studies in population-genetics for this species.

Discussion
One of the limitations of microsatellite markers is their obtaining, since it is necessary to have prior knowledge of the DNA sequence, to develop specific primer pairs flanking each identified microsatellite region [20]. However, an alternative methodology is transferability, since microsatellite markers are potentially transferable to phylogenetically close species, due to the homologous nature of the DNA sequence in the flanking regions of the microsatellites, reducing the costs and long time to develop of this kind of markers [11,12].
In this work, microsatellite markers were obtained for all the seven Byrsonima species tested through transferability. In Brazil, studies related to the development and use of microsatellite markers in plants are approximately two decades old and were first used in studies of forest species such as Ceiba pentandra [21], Araucaria angustifolia [22], and cultivated Oryza glumaepatula [23]. The ability of SSRs to be transferred across species has facilitated several studies of microsatellite markers development for native Cerrado species Hymenaea courbaril [24], Tibouchina papyrus [25], Eugenia dysenterica [26], B. cydoniifolia [13], Dipteryx alata [27], Hymenaea stigonocarpa [28] and Manihot sp. [29]. Thus, this work improve the number of microsatellite markers developed for Cerrado biome species.
Similar results was found only once for another Byrsonima species [30]. Eight SSR markers were developed for the species B. crassifolia showing an average of 5.4 alleles/locus in four populations, ranging from two to 11 alleles. The heterozygosity observed at the loci among the four tested populations of B. crassifolia ranged from 0.000 to 0.933, while the genetic diversity ranged from 0.000 to 0.839. Here, considering the seven species, it is verified that the observed heterozygosity of the loci varied from the minimum value of 0.000 (B. umbellate-locus 09) to the maximum value of 1.000 (B. laxiflora -locus 09), while the genetic diversity varied from 0.063 (B. intermedia) to 0.924 (B. viminifolia).
In another study [31], they tested the transferability of ten markers developed for B. crassifolia in three other species  B. intermedia, B. verbascifolia, B.  laxiflora, B. subterranea, B. viminifolia and B. linearifolia are very informative, with a high combined probability of exclusion of paternity (Q ≥ 0.976) and the low combined probability of identity (I ≤ 9.91 × 10 -6 ), so these markers are potentially suitable for future genetic-population studies. But, the set of markers transferred to B. umbellata is reasonably informative (Q = 0.778; I = 2.5 × 10 -3 ), so it is necessary to develop more microsatellite markers for geneticpopulation studies for this species.
Recent study suggest that the genus Byrsonima sp. belongs to a recent lineage within the Malpighiaceae [32]. Considering phylogenetic studies, it is known that the genus Byrsonima is a monophyletic group within Malpighiaceae, confirmed by molecular and morphological synapomorphies. The genus Byrsonima is traditionally divided into two subgenera (B. subg. Byrsonima and B. subg. Macrozeugma Nied), initially proposed based on the stamen morphology [33]. Recently, using anatomical characters to Byrsonima taxonomy confirmed that the division by subgenera is consistent [34].
The subgenera can be separated based on the color of the petals, in addition to the morphology of the anther. In the subgenus B. subg. Byrsonima the posterior petal is yellow, and the connective does not surpass the anther locules or surpass only one-quarter of its size. While in B. subg. Macrozeugma the posterior petal is white or pink and the connective surpasses the locules in more than one-quarter of its size [33].
Among the species of Byrsonima sp. in this study, regarding the morphological characteristics, the species B. viminifolia, B. subterranea, B. intermedia, B. laxiflora, B. linearifolia presents terminal inflorescences with yellow colored petals and are included in the subgenera Byrsonima subg. Byrsonima, as also the specie B. cydoniifolia. While, B. umbellata is the only species, in this study, with white petals in the terminal inflorescences and belongs to the subgenera B. subg. Macrozeugma.
Therefore, the reason for the low number of microsatellites markers transferred to B. umbellata (six markers) compared to the other species (above 7-11 markers), is due to the lower efficiency in cross-amplification, probably because of the greater evolutionary distance of B. umbellata with B. cydoniifolia, being included in different subgenera. Once that the rate of successful amplification decreases as the genetic divergence between species increases [11].
Finally, the transferability technique for the development of microsatellite markers between different species of Byrsonima is effective and capable of generating numerous sets of microsatellite markers. This fact is fundamental and relevant for species of genus Byrsonima, Given the scope of this complex of species in Brazil for which microsatellite markers have not yet been provided and and is such an extremely important tool for plant molecular studies [36].
So, the markers developed in this study for Byrsonima species by transferability can provide relevant information in advanced studies in population-genetics for these species, once they are capable of detecting different magnitudes of genetic variability within and between populations, so can be used for identification of conservation units and for the investigation of genetic processes that occur in populations, such as the genetic diversity of populations, patterns of gene flow, the incidence of genetic drift and generation of the genetic neighborhood. So, different strategies for in situ or ex-situ conservation can be admitted and also for these species domestication, as a genetic resource.
Funding Work by V. Bernardes was supported by a fellowship from CAPES, and work by M. P. C. Telles, was also supported by productivity grants from CNPq. This study was partially supported by a Programa de Apoio a Núcleos de Excelência (PRONEX) grant from CNPq/FAPEG (AUX PESQ CH 007/2009) and by grants to the research network GENPAC (Geographical Genetics and Regional Planning for natural resources in Brazilian Cerrado) from CNPq/FAPEG (Project No. 201110267000125 and 563839/ 2010-4-563973/2010-2). Our current research in Genetics and Genomics is developed in the context of National Institutes for Science and Technology (INCT) in Ecology, Evolution and Biodiversity Conservation, supported by MCTIC/CNpq (Proc. 465610/2014-5) and FAPEG.