Differentiation of Urochloa brizantha cultivars by inter-simple sequence repeat ( ISSR ) markers in seed samples

Approximately 80-90% of cultivated grassland areas in Brazil are covered by Urochloa brizantha [syn. Brachiaria brizantha (Hochst. ex A. Rich.) Stapf.]. Some genotypes of Urochloa have being widely used with a wrong nomenclature, like species and cultivars. In this way, the Urochloa cultivar identification is primordial for breeding programs and seed production. Considering the importance of genetic purity in commercialized seed lots, the present work aimed to evaluate the use of inter-simple sequence repeat (ISSR) markers in six cultivars of U. brizantha (Xaraés; Piatã; Basilisk; MG4; MG5 and Marandu) to discriminate and determine the contamination in seed batches. Results showed that it is possible to discriminate all cultivar with only two primers in pure samples. Basilisk was confirmed as a U. brizantha cultivar. ISSR markers showed a low polymorphism level. It was not possible to separate samples intentionally contaminated even at 5%.


INTRODUCTION
The Urochloa genus (syn.Brachiaria) belongs to the Poaceae family and was introduced in Brazil from Africa, having as its main centre of origin and diversification, the east of the continent (Vilela, 2005).In Brazil, sixteen species were introduced from Africa and the most important were Urochloa brizantha, Urochloa decumbens, Urochloa ruziziensis and Urochloa humidicola (Karia et al., 2006.), which represent the majority of commercial seeds.In addition to its use as pasture, forage species have been used to cover the soil in crop-livestock integration (Crusciol et al., 2009).Currently, it is estimated that more than 50% of the area cultivated with pastures in the Center-west region of the country are cropped with this grass (Macedo, 2006).*Corresponding author.E-mail: ceci@unoeste.bror nbmneto@unoeste.br.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License There are seven major worldwide collections of Urochloa, all ex situ, which have a total of 987 accesses of 33 described species (Keller-Grein et al., 1996).However, problems with incorrect identifications are still frequent among Urochloa species, as well as in the accessions of germplasm collections (Assis et al., 2003).Presently, Embrapa Gado de Corte is responsible for the largest and most important resources of Forage Germplasm Bank of the Urochloa used for the plant breeding efforts (Valle et al., 2008).
Currently, nine cultivars of U. brizantha are registered, but just Marandu; Xaraés; BRS Piatã; BRS Paiaguás; Basilisk; MG4; MG5 Vitória and MG-13 Braúna are commercially available (Brasil, 2016), and there are still doubts about the identification of some of these cultivars.An important cultivar, originally introduced as Urochloa decumbens, is Basilisk, collected originally in Uganda, sent to Australia and subsequently introduced from International Research Institute (IRI) to São Paulo in the early 1960s and was the first Urochloa cultivar (Karia et al., 2006).According to Renvoize et al. (1996), Basilisk was wrongly identified as a U. decumbens but it is actually an U. brizantha, data is supported by Ambiel et al. (2008Ambiel et al. ( , 2010)).
The cultivars, MG-4, MG-5 "Vitória" and MG-13 "Braúna" were released by Matsuda company after an accession selection from CIAT genebank (International Center for Tropical Agriculture) (Matsuda, 2016).Despite some data pointing that Xaraés and MG-5 are not the same material (Ambiel et al., 2008(Ambiel et al., , 2010)), for the National Cultivar Register, they are under the same number (Brasil, 2016).Assis et al. (2003) succeeded in using morphological markers to differentiate six species of Urochloa (U. brizantha, U. humidicola, U. decumbens, Urochloa jubata, U. ruziziensis and Urochloa dictyoneura) using vegetative and reproductive characteristics and pubescence.However, there is still no standard for varietal discrimination of U. brizantha cultivars, which can result in commercial seed lots with high varietal mix due to technical difficulties to identify the degree of contamination (Zanine et al., 2007).
For the seed production and selling of U. brizantha, the minimum seed purity percentage is 60% and contamination with other cultivars, must be below 0.15% (Brazil, 2008).Therefore, techniques for discrimination and for varietal purity are important for quality control in many points of the production chain, because of the easy identification for synonyms cases and homonyms between cultivars and to assist in individual selection that can be used to compose germplasm banks or in breeding plans (De Paula et al., 2012).Besides the aspect related to seed quality control, another reason for the significant increase is the interest in the cultivar characterization and identification for protection of commercial cultivars in an increasingly competitive markets.
Among the available molecular methods, those based on the polymerase chain reaction (PCR) has advantages over the other methods because they use reduced quantities of DNA and electrophoretic profiles are obtained faster (Ferreira and Grattapaglia, 1998;Ramos et al., 2006).The development of variants of this technique enabled the emergence of various types of molecular markers, including inter-simple sequence repeat (ISSR).This type of marker has been widely used in studies for varietal identification, and also in the seed technology, being a simple and efficient technique to generate high levels of polymorphism (Reddy et al., 2002).ISSR molecular markers were developed by Gupta et al. (1994) and Zietkiewicz et al. (1994).The amplified products of the PCR reaction produced in the ISSR reaction correspond to sequences of different sizes that are located between identical microsatellite repeat regions and oriented in opposite directions.In addition to presenting high levels of polymorphism, they are robust due to the fact that they have greater anchoring surface and also higher temperatures reassociation, thus having high reproducible products (Lin et al., 2010;Haq et al., 2011).
The advantages of using the ISSR technique is that it requires small DNA amounts per reaction, the speed to obtain relevant genetic information for studies of population diversity is low, and also they require little infrastructure equipment for laboratories, when compared with other markers (Satya et al., 2012) and seed can be used as a DNA source, even if the extracted amounts are low, but the quality is satisfactory.ISSR have been widely used to detect the intraspecific polymorphisms in plants (Pharmawati et al., 2004).ISSR can be used to analyze multiple loci in a single reaction (Marotti et al., 2007) and to produce fragments with high reproducibility as compared to other non-specific PCR-based markers such as random amplified polymorphic dna (RAPD) (Wolfe and Liston, 1998).Saini et al. (2004) obtained in rice, the highest percentage of polymorphic ISSR markers as compared to amplified fragments length polymorphism (AFLP) markers, but smaller than simple sequence repeats (SSR).Azevedo et al. (2011) investigated the genetic variability within genotypes of U. ruziziensis by ISSR.These markers are shown to be effective for such evaluation, suggesting that the populations of this Urochloa species retain a wide genetic variability.In the case of U. brizantha, despite the commercial importance of this species as a forage plant, there are still relatively few studies on the genetic diversity within existing accesses in germplasm collections in Brazil.Also, unlike U. ruziziensis, most commercial cultivars, U. brizantha have an apomictic reproduction which justifies the lowest variability among the few available accesses.
evaluated with these markers in accesses and wild species, to obtain data for breeding assisted programs as in Silva et al. (2011) that used ISSR markers to identify inter-and intraspecific variability in accessions of Manihot (Manihot esculenta, Manihot caerulescen, Manihot dichotoma and Manihot flabellifoli).Also, several studies used ISSR markers as a tool for the delimitation of plant species (Dogan et al., 2007;Wood and Nakazato, 2009;Anand et al., 2010).
The DNA molecular markers have been used to discriminate cultivars and evaluate their genetic purity in seed samples.Due to the advances in molecular marker techniques, and the greater demand of the seed market, these tools have become a viable alternative for routine analysis.The ability to access the genetic variability directly from DNA has increased the process of intellectual protection of genetic materials, in association with other discriminatory analysis (Almeida et al., 2009(Almeida et al., , 2011)).While in Brazil, the cultivar registration do not require molecular characterization, in other countries, with consolidated certification systems of plant material, molecular techniques have allowed developing a varietal plant molecular fingerprint, which facilitates the control of the quality at every stage in the process (Bianchi et al., 2004;Wickert et al., 2007;De Paula et al., 2012).
As exemplified above, the use of ISSR has specific characteristics to be explored in genetic analyses and varietal discrimination.Thus, the present study aimed to investigate the potential of using ISSR to discriminate the six cultivars of U. brizantha available and to check if this type of molecular marker could potentially determine the degree of varietal contamination among seed lots of this species.

MATERIALS AND METHODS
Genetic seeds of cultivars, U. brizantha Marandu, BRS Piatã, Xaraés, Basilisk, MG4, MG5 (Table 1) were provided by EMBRAPA -Campo Grande (MS) and Matsuda Seeds.MG-13 Braúna was not analysed because it was released after the start of this work.

DNA extraction and PCR
After the removal of the glumes, DNA was extracted from seeds of each cultivar as in Ambiel et al. (2008Ambiel et al. ( , 2010)).The pellet obtained was then washed twice with 100 µl of 70% ethanol and after drying, it was resuspended in 100 µl -1 1× TE buffer (2.5 mM Tris-HCI and 0.25 mM EDTA).The DNA quantification was performed in a spectrophotometer (Eppendorf BioPhotometer Plus), measuring the absorbance at a wavelength of 260 and 280 nm.The integrity of the purified DNA was assessed by electrophoresis in a 0.8% agarose gel using ethidium bromide as stain.
Initial selection was performed with 28 ISSR primers (Table 2), whereas only ten of these were useful for determining the cultivars electrophoretic pattern.PCR reactions were performed in a final volume of 25 µl containing: 25 ng of DNA, 2.5 µl (10×) Taq buffer (Invitrogen), 2 µl MgCl2 50 mM, 0.25 µL dNTPs (1.25 mM), 0.5 µM of each primer (Fermentas) and 2.5 U of Taq polymerase.The amplifications were performed in a thermocycler (PCT -100, MJ Research) programmed for one cycle at 94°C for 1.5 min for initial denaturation, followed by 40 cycles at 94°C for 40 s, 45°C for 45 s, and 72°C for 1.5 min, the final cycle at 94°C for 45 s, 44°C for 45 s and 72°C for 5 min for final extension or elongation.
Separation of the amplified fragments was performed in a 1.2% agarose gel with 0.5 × SB buffer (10 mmol L -1 NaOH, adjusted to pH 8.5 with boric acid) (Brody and Kern, 2004).Gels were stained with ethidium bromide and visualized using Electrophoresis Analysis System (Biosystems).Each amplification reaction was repeated at least twice and only clearly distinct and reproducible bands were scored.The analysis of the bands was performed with the Quantum program -Capt (Vilber -Lourmat) to determine the electrophoretic pattern of the six U. brizantha cultivars.

Discrimination and varietal purity analysis
To test the potential of ISSR markers to discriminate the U. brizanta cultivars, the pattern of amplified products were tabulated as presence (1) or absence (0) for each polymorphic fragment.One genetic similarity matrix was generated using the Jaccard coefficient and a dendrogram constructed by clustering method unweighted pair-group method using arithmetic averages (UPGMA) using the software NTSYS 2.1 (Rohlf, 2004).
In order to simulate different levels of contamination varietal, a simulation of DNA contamination was performed on Marandu extracted DNA with DNA from the other cultivars in 1, 2.5 and 5% concentration.This experiment was performed in triplicate, with extraction of DNA from three samples of each studied cultivar.
After that, a PCR was performed with polymorphic primers that discriminated the Marandu cultivar from the other cultivars (P4, P7, P9, P10, P15, P23 and P24).The PCR products, of intentionally contaminated DNA samples, were separated on 1.5% agarose gel to verify the possible presence of the band belonging to the contaminant cultivar.The analyses were performed using the program Quantum -Cap, as mentioned above.(04509) in the National Register of Cultivars (RNC)being maintained by Embrapa and Matsuda.However, this data show that these cultivars have some genetic divergence (Figure 2), which was also verified by Ambiel et al. (2008Ambiel et al. ( , 2010) ) using the RAPD markers.Also, it was observed that the cultivar Basilisk, erroneously classified as a U. decumbens genotype (Renvoize et al., 1996) was grouped with U. brizantha cultivars, which proved to be very similar to cultivar MG4, based on the polymorphisms obtained with the selected ISSR primers.These results confirm what was observed by others using RAPD markers (Ambiel et al., 2008;Almeida et al., 2011).Renvoize et al. (1996) suggested that the cultivar 'Basilisk', widely used and commonly identified as U. decumbens is actually a cultivar of U. brizantha.So, this error of taxonomic identification was also confirmed in this study using ISSR markers.It is important to mention that it was possible to identify and distinguish all cultivars of Urochloa studied in this work using the 10 selected ISSR primers.
Ten primers were used to discriminate the cultivars, but a single primer that was sufficient per se to identify all cultivars was not found.P4, P10 and P24 primers shown to be polymorphic among Marandu and MG4, P7 primer differentiates Marandu and Xaraés, the P9 and P23 primers discriminates Marandu from MG5 (Figure 1A and B), the P15 primer differentiated Marandu from BRS Piatã while the primer for P23 was polymorphic for Marandu and Basilisk (Figure 1B).Thus, even with the low variability among commercial cultivars U. brizantha, this work showed that the use of ISSR primers allowed differentiation between these genotypes.Silva et al. (2011) used ISSR molecular markers for the genus Manihot, which allowed the observation of highly reproducibility markers, revealing large intra-and interspecific divergence among accessions of Manihot species.Also, in studies with orchids, phylogenetic analysis showed that for genetic diversity, ISSR can respond optimally to evolutionary issues in complex species consistently and with high reliability values (Rodrigues et al., 2015).Vigna et al. (2011) developed 15 microsatellite markers to study the genetic variability within U. brizantha in a germplasm collection and observed that this collection does not display a considerable variability.Thus, this work showed that the use of ISSR markers can complement the information obtained with other markers.

Genetic purity of seed lots
The second objective was to investigate the ISSR potential as markers to determine the degree of contamination of seed batches of U. brizantha.For this analysis simulation, a contamination of 1, 2.5 and 5% of DNA from each genotype with U. brizantha cultivar Marandu was made.
Several authors have studied levels of varietal purity in commercial species.Jorgensen et al. (2007) analysed the purity in batches of Brassica napus by ISSR, and their results indicated that there was the possibility of separating the varieties with these markers and to estimate the purity in most cases.These authors observed that in harvested seed analysed, three cultivars were contaminated with other varieties over the allowed limit (0.03% for food and <1% for feed).
The ISSR markers used in this study to explore crosscontamination with DNA, between cultivars were not efficient for determining the required levels of contamination (Figure 3) where the samples mixed with Marandu DNA did not group together; just the contaminated samples of Piatã grouped in just one cluster, but apart from the pure sample, the others are mixed with another samples or with the pure material.According to Brasil (2008), for the purity analysis of U. brizantha (Hochst.Ex A. Rich) Stapf., the maximum allowed level of other cultivar at every 180 g of seeds is 30 seeds (Brasil, 2008 and2011), for example between 0.11 and 0.13%, therefore less than 1% (considering the average number of seeds in a U. brizantha seed gram ranging between 123 and 145, approximately) (Brasil, 2009).In this work, even with levels higher than allowed, 5% (Figure 3), ISSR primers were able to differentiate the contamination but not identify it.Therefore, this PCR showed no efficiency in determining the contaminants from the cultivars in any tested level of contamination.

Conclusions
ISSR molecular markers proved to be useful for varietal identification of U. brizantha cultivars and may determine genetic differences between commercially used cultivars.The ISSR markers were not efficient to determine contamination levels at 1, 2.5 and 5% DNA between genotypes of U. brizantha.It was possible to reconfirm the cultivar identification of Basilisk as a cultivar of U. brizantha.

Table 1 .
List of U. brizantha cultivars used in this study.

Table 3 .
Primer fragment length, number of bands (total and polymorphic) used in ISSR analysis of U. brizantha cultivars.
Figure 2. Grouping of six cultivars of U. brizantha by UPGMA method using polymorphic ISSR.