Genetic diversity and relationships among Nopalea sp. and Opuntia spp. accessions revealed by RAPD, ISSR and ITS molecular markers

Dactylopius opuntiae (Cockerell) (carmine cochineal) is an insect pest highly noxious that has spread through cactus pear crops in the Brazilian semiarid region. Knowledge of diversity and genetic relationships of the cactus pear accessions is fundamental to create new varieties resistance to carmine cochineal. Therefore, this investigation was undertaken to assess the genetic diversity and genetic relationships that existed among cactus pear accessions of Nopalea sp. and Opuntia spp. with contrasting resistance to D. opuntiae. We conducted a molecular analysis in seven cactus pear accessions from the “reference collection” of the Agronomic Institute of Pernambuco, Brazil using RAPD, ISSR and ITS molecular markers. A total of 242 bands were detected from 26 polymorphic primers. The high percentage of polymorphism by RAPD (89.8%), ISSR (81.2%) and ITS (75%) markers suggests that the cactus pear accessions have high genetic diversity. The combined analysis of markers systems enabled the accessions discrimination of about the genus and ploidy, but were incongruous in relation to resistance level to D. opuntiae. Genetic diversity, discrimination of about the genus and ploidy was confirmed by merging information from ISSR, RAPD and ITS markers systems. The IPA-200016, IPA-200149, IPA-100004, IPA-200205 accessions are genetically divergent, therefore could be potentially incorporated into any further breeding programs directed to create new varieties of cactus pear resistant to D. opuntiae.


Introduction
The cactus pear Opuntia and Nopalea are widely cultivated for forage purposes in several arid and semi-arid regions of the world [1]. The succulent stems (cladode) of cactus pear are used to feed domestic ruminants. They are rich in water, sugars, ash, vitamins A and C, but low in crude protein and fiber [2].
In the Brazilian Semiarid Region, cactus pear became the basic food for domestic ruminants. However, since 2001 the main cultivated varieties of the Opuntia (Giant, Round and IPA-20) have been decimated by carmine cochineal (Dactylopius opuntiae Cockerell), a pest insect that has caused great economic losses to local cattle breeders [3,4].
The Agronomic Institute of Pernambuco (IPA), a Brazilian research and rural extension agency located in the country semi-arid region, has several accessions of cactus pear with contrasting resistance to carmine cochineal in its germplasm bank [5], but its diversity and genetic relationships have not yet been sufficiently clarified. Therefore, it is vital that the germplasm of this collection is properly characterized, as they will be useful to plan future improvement strategies aimed at creating new varieties resistant to carmine cochineal.
The cactus pear germplasm characterization has been enhanced with the use of polymerase chain reaction (PCR)based molecular markers such as Inter Simple Sequence Repeat (ISSR) [6] Random Amplified Polymorphic DNA (RAPD) [6,7] and Internal Transcribed Spacer (ITS) [8], with applications extensively in studying genetic diversity, genetic relationship, germplasm management and genetic breeding. However, when we study genotypes close related, the analysis of variability could require more than one DNAbased technique, because the markers systems combinate can provide extensive coverage of the genome [9]. Combined marker approach was more informative than use of individual markers for diversity analysis in Mangifera indica L. [10], Salvia L. [11], Dysosma tonkinense [12], Dactylis glomerata L. [9]. In these studies, the increased information content that resulted from the use of combined markers, did indeed provide a better resolution the genetic diversity.
Therefore, the use RAPD, ISSR and ITS combinate markers systems presents an alternative method in detection of polymorphism in cactus pear. This investigation was undertaken to assess the genetic diversity and genetic relationships that existed among cactus pear accessions of Nopalea sp. and Opuntia spp. with contrasting resistance to D. opuntiae.

Sampling of accessions
Seven cactus pear accession (Table 1) from the germplasm bank of the Agronomic Institute of Pernambuco (IPA), Arcoverde city, Pernambuco state, Brazil was evaluated. The classification of chromosome number was based to Majure et al. [13] and the level of resistance to carmine cochineal was based on Silva et al. [14].

Genomic DNA extraction
Genomic DNA was extracted from tissue samples from the young stems epidermis based on the protocol of Doyle and Doyle [15]. The DNA concentration in the samples was estimated by electrophoresis on a 0.8% agarose gel, comparing the fluorescence intensity of the ethidium bromide-stained DNA bands with a known DNA standard (Invitrogen™ 1 Kb Plus DNA Ladder). Ratio absorbance 260/280 and 260/230 nm of more than 1.8 were considered as a standard for purity analisys.

DNA amplification via PCR using RAPD, ISSR and ITS primers
PCR-RAPD reactions were performed using 12 primers ( Table 2) from Operon Technologies Inc., Alameda, CA, USA (OP06, OPG07, OPG10, OPC11, OPS03, OPG19, OPA02, OPG03, OPG13, OPG15, OPG06 and OPM12). RAPD reactions were prepared with a final volume of   The protocol for DNA amplification consisted of an initial denaturation of 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s; annealing the primers at 52 °C for 45 s; extension at 72 °C for 2 min and final extension at 72 °C for 5 min using a thermal cycler model TC-Plus Techne Bibby Scientific Ltd.
The ITS region of nuclear ribosomal DNA was amplified using the ITS1 and ITS4 primers ( Table 2)

Amplicons separation by gel electrophoresis
Amplicons separation was performed by horizontal electrophoresis in agarose gel (1.5%) with ethidium bromide dye (0.5 ug ml −1 ). Amplicon migration was performed in 0.5× TBE running buffer (Tris, boric acid, EDTA) at a voltage of 100 V for 120 min. A 100 bp (bp) molecular weight standard (Invitrogen™ 1 Kb Plus DNA Ladder) was used. The gel was visualized under ultraviolet light and photodocumented with the aid of the Gel Logic 112Pro imaging system (Carestream, Rochester, NY).

Data analysis
The bands generated by each primer were classified as monomorphic and polymorphic and their presence/absence computed for all sampled. Subsequently, the total number of amplified bands, number of monomorphic bands, number of polymorphic bands and percentage of polymorphism for each primer were calculated.
The presence/absence data were organized into a binary matrix (1 = presence and 0 = absence) which was used to determine the binary Sokal distance using the Genes software [16]. Using the Sokal binary distance values, a dissimilarity dendrogram was constructed based on Ward clustering method. The definition of the number of groups was performed based on the criterion proposed by Mojena [17] using bootstrap analysis to verify and provide statistical support to the internal nodes of the dendrograms. The validation of the clusters was determined by calculating the cophenetic correlation coefficient [18]. Additionally, a principal component analysis was performed using the Past program [19].

Results
The 26 primers (12 RAPD, 12 ISSR and 2 ITS primers) used for the evaluation of DNA polymorphism in cactus pear accessions produced 242 polymorphic reproducible bands. The amplification patterns of some of the RAPD (OPG15), ISSR (UBC 810) and ITS (ITS 1 and ITS4) primers can be seen in Fig. 1.
RAPD primers amplified from 5 to 12 bands with a mean polymorphism of 90.6% (66.7-100%) and generated 98 bands of which 88 (89.8%) were polymorphic. The ISSR primers amplified from 8 to 19 bands with a mean polymorphism of 78.7% (25-100%) and generated 140 bands, of which 115 (82.1%) were polymorphic. ITS primers amplified 4 bands of which 3 (75%) were polymorphic. These results present that the set of cactus pear accessions has high molecular polymorphism.
The Sokal binary distance between accession pairs ranged from 0.46 to 0.65 with a mean value of 0.57, indicates high genetic dissimilarity between accessions. However, the In the cluster analysis, it was found that the cophenetic correlation coefficient obtained from the genetic distance matrix and the cophenetic distance matrix was high and significant (r = 0.80; P ≤ 0.01), indicating that there was high consistency in the grouping patterns [18]. Furthermore, the distortion (10.04%; P ≤ 0.01) and stress (7.10%; P ≤ 0.01) values according to the Kruskal scale [20] also indicate that the graphical representation of the dissimilarity in the dendrogram was highly faithful to the matrix of dissimilarity.
Taking the dissimilarity of 1.359 as the cutoff point in the dendrogram, the accessions formed two main clusters (Fig. 2). This grouping pattern was consistent with the genus of accessions, as those to Opuntia (IPA-100003 (Fig. 2).
The distribution of accessions in the components (CP1 × CP2) of the Principal Component Analysis (Fig. 3) corroborated with the Cluster Analysis, as the accessions were also allocated in groups consistent with gender. There was less dispersion among the Nopalea cluster accessions in relation

Discussion
The plant breeding requires a profound knowledge about the diversity and genetic relationships from basic population. Efforts in reaching this goal can be supported by the application of molecular markers. In this study, we used three molecular marker systems (RAPD, ISSR and ITS) to detect genetic polymorphism among forage cactus accessions with contrasting resistance to D. opuntiae.
All RAPD and ISSR primers produced consistent and reproducible amplification profiles and with higher informativity than in other similar studies. For example, the average percentage of polymorphic bands revealed by the RAPD (90.6%) and ISSR (78.7%) primers in our study was higher than that verified in the research by Valadez-Moctezuma et al. [6] (RAPD = 37.0%; ISSR = 28.6%) and Tütüncü et al.  [7] (RAPD = 72%). We also demonstrated that being able to detect greater genetic polymorphism of RAPD (90.6%) was higher in relation to ISSR (78.7%) in cactus pear, corroborating the data reported by Valadez-Moctezuma et al. [6]. The high polymorphism of ITS markers (75%) was also similar to that verified by Lyra et al. [8], who were able to discriminate cactus pear accessions in relation to genus, similarly to what was observed in this study.
The high genetic polymorphism detected by RAPD (89.8%), ISSR (82.1%) and ITS (75%) markers and the high dissimilarity between accession pairs (binary Sokal distance) suggest that there is high genetic diversity between accessions being, therefore, a suitable reference set for breeding programs. The high genetic diversity of this set can be explained by its composition that includes different species. The greatest genetic similarity between accessions IPA-100004, IPA-200021 and IPA-200205 is due to the fact that both belong to the species N. cochenillifera.
The structuring of accessions in clusters according to genus indicates that the analysis of RAPD, ISSR and ITS markers can also be useful for phylogeny studies in the Opuntiae family. The structuring of the accessions was also consistent in relation to the ploidy, because when observing the dendrogram (upper part towards the base) there was an increase in the number of chromosomes in the accessions  (Table 1).
Based on the results, the combined analysis of RAPD, ISSR and ITS molecular markers proved to be a valuable tool to accessing genetic diversity and discriminate cactus pear accessions. We conclude that is an important genetic diversity at the DNA level among the cactus pear accessions from Agronomic Institute of Pernambuco, Brazil, which could be potentially incorporated into any further breeding programs directed to create new varieties of cactus pear resistant to carmine cochineal.