Molecular characterization of a species in the genus Rubus in Boyacá, Colombia

Colombia is home to blackberry genetic resources which present a morphological diversity. The relevant characteristics related to its diversity are the presence of prickles, the shape of its leaves, the number and color of its fruits, and its enormous agro-industrial potential due bioactive compounds such as polyphenols. These plants can grow between 1,700 and 3,400 m asl and are cultivated in the central region of the country. The study evaluated 13 wild and cultivated plants from the genus Rubus. A molecular characterization was carried out using 16 SSR microsatellite markers, all of which produced positive amplification generating 23 loci and 26 alleles. The AMOVA indicated a molecular genetic differentiation of 23% between the groups which corresponded to the geographic location of the sample. The greatest contribution to variance is found within the groups (76%), possibly because each of them is composed of different cultivated species and wild relatives of the genus Rubus. This suggests that the grouping of the genotypes studied doesn’t necessarily correspond to geographical origin. However, the findings show high genetic variation, with an Fst value of 0.27. This may be useful in breeding programs where genetic diversity, morphological characteristics of the fruits, and the molecular identification of the fruits are taken into account.


Introduction
There are between 750 and 800 species of the genus Rubus, including raspberry, blackberry, some native species, and other species considered invasive (CLARK et al., 2013;MORENO-MEDINA et al., 2018). However, this genus of plants is difficult to identify due to its frequent hybridization and apomixis. This leads to cultivars of similar morphology which makes testing for SSR markers (Simple sequence repeats) in this genus of plants of enormous interest (LÓPEZ et al., 2019). This procedure has contributed to the selection of species that present agronomic potential, provide reliable identification, and allow the genetic diversity of populations to be measured (ROA et al., 2014;MARULANDA et al., 2012).
In Colombia, the most important commercial cultivar is the blackberry (R. glaucus), however, there are other species such as R. alpinus, R. alutaceus, R. floribundus, R. bogotensis, R. giganteus, R. megalococcus, and R. nubigenus which can grow between 1,700 and 3,400 meters above sea level. Departments such as Antioquia, Boyacá, Caldas, Cauca, Cundinamarca, Santander, and Valle del Cauca all contribute to the national production of blackberries throughout the year; however, most of the crops established in these regions do not have suitable agronomic management. These problems include adjustments to the growth and fruiting habits of the species  and the occasional lack of pre and postharvest management with the purpose of improving yields or increasing the quality of the fruit (MORALES; VILLEGAS, 2012; MORENO -MEDINA; DEAQUIZ, 2016;MORENO -MEDINA et al., 2020). Colombia has valuable genetic resources for blackberries in its cultivars and natural populations. This includes a wide morphological diversity such as prickles, the shape of the leaves, the number and color of the fruits, and an enormous agro-industrial potential due to its sweetness and phenolic compounds.
The genetic diversity and identification of any species can be measured through morphological descriptors and molecular markers. Morphological descriptors have some limitations including their complexity, access to taxonomic keys, subjective results due to plant phenology and environmental factors, and inaccurate information. (CASTILLO et al., 2010). Molecular markers represent a fast, objective method, use simple techniques, are not affected by environmental conditions, are used in plants regardless of their phenology, and are efficient when discriminating genotypes in the same species (FAQIR et al., 2017). Using molecular markers for the characterization of a species are a useful instrument to describe morphologically similar individuals with agronomic potential that present a certain capacity to adapt to climate change (EL KADRI et al., 2019). SSR markers (Simple Sequence Repeats) are small sequences of DNA repeated in tandem which are widely used in plants to perform a genetic diversity analysis. These markers have codominant characteristics and a high level of allelic diversity at different loci (KALIA et al., 2011). In the case of the Rubus genus, microsatellites (SSR) have been used to perform a polymorphism analysis to evaluate the genetic diversity of raspberry cultivars with different fruit colors using structural genes and regulators of flavonoid biosynthesis (LEBEDEV et al., 2019). Studies were performed to identify the genetic diversity of cultivars in different locations in Venezuela (ROA et al., 2014) and from 32 Rubus cultivars in Europe and the United States with differences in ploidy and berry color. Additionally, LÓPEZ et al. (2019) recently evaluated the genetic diversity of parental plants in Colombia, which are used in the process of reproduction or development of parental populations, based on their morphology and tolerance to C. gloeosporioides.
The objective of this study is geared towards the molecular characterization of the species in the genus Rubus located in Colombia. The genetic information provided by wild and cultivated blackberries species found in province of Alto Ricaurte, in Boyacá, Colombia would provide numerous benefits. These benefits include promising production alternatives, recovering cultivation areas, increasing employment in rural areas, and establishing species with adequate agronomic management and minimum environmental impact.

Materials and methods
Processing and analysis of samples: The 13 samples used in this characterization were taken from the completely expanded leaves of cultivated and wild blackberry species collected in the municipalities of Arcabuco and Gachantivá, Colombia ( Table 1). The samples were conserved on silica gel for processing and were sent to the Plant Biotechnology Laboratory at the Universidad Tecnológica de Pereira. Two more samples corresponding to the species Rubus glaucus Benth were added as an external group (outgroup). Amplification: Homologous SSR markers developed by the Universidad Tecnológica de Pereira for the R. glaucus species and heterologous SSR markers developed for other species of the Rubus genus were used. The amplification reactions were carried out using the conditions described by MARULANDA et al. (2012) and a "touchdown" amplification profile of 32 cycles with denaturation at 95°C for 1 minute. The annealing procedure lasted 1 minute, with a decrease in temperature of 1°C every two cycles from 63°C to 58°C and with 10 cycles at 59°C and 10 cycles at 58°C. The elongation procedure was carried out at 72°C for one minute. The amplified products were separated by electrophoresis in 6% denaturing polyacrylamide gels, which were run in a vertical Sequi-Gen Sequencing Cell chamber (BioRad, Hercules, California, USA) at 110 W and 50°C for an hour. The gels were stained with silver nitrate according to the protocol developed by BENBOUZA et al. (2006).

Microsatellite markers:
The markers used for this characterization are shown in table 2. Seven of them were homologous and coded by CL and Rg, used by LÓPEZ et al. (2019); MARULANDA et al. (2012) and developed by the Universidad Tecnológica de Pereira. Four microsatellite markers were used from the "Rubus" series, developed by CASTILLO (2006) for the species R. idaeus and the cultivar called Marion corresponding to a Rubus hybrid. Three microsatellite markers from the RhM and RiM series developed by GRAHAM et al. (2002GRAHAM et al. ( , 2004 were also used for R. idaeus. Finally, 2 microsatellite markers from the mRaCIRRI series developed by AMSELLEM et al. (2001) were used for the species R. alceifolius.

Statistical Analysis
Measures of diversity, variability, and genetic distance were determined using the GenAlEx 6.5 program (PEAKALL and SMOUSE, 2012). A cluster analysis was performed using the paired group algorithm while the Dice similarity coefficient was calculated using the PAST tool (Paleontological Statistics Software Package for Education and Data Analysis) (HAMMER et al., 2001). A population analysis was carried out while taking into account that the samples belong to two different localities (A-Gachantivá and B-Arcabuco).

Results and discussion
The molecular markers showed positive amplification with the 16 SSRs used, generating 23 loci and 26 alleles, seven of which amplified two loci. These samples were analyzed using SSR markers from other species of the Rubus genus (heterologous markers) and showed positive polymorphic amplification. The average number of alleles was 4.2 and approximately 3 effective alleles were present ( by LÓPEZ et al. (2019) who reported similar genetic diversity parameters in blackberry (R. glaucus) cultivars. These authors found 29 loci and 58 alleles, with 3.4 effective alleles which were very close to those found in this study. LEBEDEV et al. (2019) analyzed the genetic diversity of 21 raspberry cultivars (R. idaeus and R. occidentalis) with SSR. A total of 26 alleles with an amplification of up to nine alleles per locus with an average of 3.7 alleles were found from their study. The similarity of results from previous studies, the current study, and those reported by CASTILLO et al. (2010) and MARULANDA et al. (2012) suggest that there is an adequate transferability or positive response of the markers (SSR) used to characterize the polymorphisms of wild and cultivated species of the Rubus genus in different geographic locations. These findings may be used in genetic improvement programs to encourage agronomic success.  (2017) and LEBEDEV et al. (2020), which found that the H e was higher than H o in two types of wild Chinese brambles of the genus Fragaria and Rubus, in 32 cultivars from Europe and the United States respectively. Consequently, it can be inferred that there is excess genetic variability in the evaluated individuals since there is a higher value in the observed heterozygosity, compared to the expected heterozygosity. This information can be used to understand the morphological diversity of the species in the area where the studied samples come from, particularly in relation to the number of leaves and the color of the fruits.
With the use of the marker RhM011, the greatest contribution to genetic differentiation was observed with an F st value of 0.436. This indicates its importance in the search for differences between the genotypes studied. This result agrees with the study from DOTOR-ROBAYO et al. (2016) who found a marker (F st 0.53) that was useful to differentiate the ecotypes collected in seven municipalities in Boyacá, Colombia, using RAMs. In the current study, the fixation index (Fst) obtained when evaluating cultivated and wild materials from the localities of Arcabuco and Gachantivá had an average value of 0.273. This is an indicator of high genetic diversity as mentioned by WRIGHT. (1978). Some of the plant material collected for the current study is asexual which almost guarantees low genetic diversity (Hu et al., 2012). The wild and cultivated ecotypes collected from the study area must be conserved and taken into account for future plant breeding programs.
Rev. Bras. Frutic., Jaboticabal, 2021, v. 43, n. 2: (e-713) The estimation of genetic diversity using the Dice index allowed the construction of a dendrogram, represented in Figure 1. The 15 genotypes studied including the reference groups (UTP-UTP7), were grouped without strictly depending on the collection site. The groups of plant material included cultivated and wild species whose similarity ranged between 35 and 80%. This indicates that there is genetic diversity among the tested materials which is necessary in various genetic improvement programs. Research carried out by LÓPEZ et al. (2019) in six departments of Colombia reported similarities ranging between 32 and 88%. Research carried out by MARULANDA et al. (2012) found molecular similarities of between 49 and 100% with the species from the genus Rubus in eight Colombian departments.
The cluster analysis separated the genotypes into three groups, which were analyzed from the bottom up in figure 2. The first group was made up of four cultivated species (A-5AL; A-3GL and A-1BL) and one wild species (A -10BS). The color of fruit was different with a node of similarity close to 83% with species that produce a yellow-whitish fruit (A-1BL and A-10BS). The second group consisted of five genotypes (A-4AL; B-7AL; B-6AL; A-12R and A-11S), of which four correspond to locally grown plant materials (R. alpinus) and one wild species. The node with the greatest genetic similarity in this group is represented by samples A-12R and B-6AL even though they differ in the color of the branches (red and purple). The third group is made up of three cultivars (R. glaucus) and a wild species (B-12S; B-9GL; B-8GL and A-2GL). The node with the greatest similarity in this group is made up of cultivated species with similar morphological characteristics (B-9GL and B-8GL) which had a similarity of 73%. Figure 1. Cluster analysis with a paired algorithm for the species of the genus Rubus studied using the Dice distance index. Two outgroups are included as reference samples.
Rev. Bras. Frutic., Jaboticabal, 2021, v. 43, n. 2: (e-713) As described above, the mixture of the genotypes studied, the distribution of the groups, and the behavior of the nodes with greater genetic similarity could be explained through the exchange of plant material propagated by asexual methods which occurs in the area of this study. This occurs as a result of geographical proximity which is similar to what was reported in other departments of Colombia as mentioned by LÓPEZ et al. (2019). It is interesting to note the contribution of wild species to the genetic diversity of each of the groups formed. In figure  1, the samples of the wild species: A-10BS, A-11S, and B13S are located in the three groups contributing to the differentiation of the studied populations and their genetic relationships. This suggests that the genetic diversity of cultivars can be influenced by wild populations which is important due to its wide distribution in the studied localities and can stimulate the transfer of genes through pollination (DOTOR-ROBAYO et al., 2016).
R e f e r e n c e s a m p l e s U T P 1 ( r e s i s t a n t t o Colletotrichum gloeosporioides) and UTP7 (moderately resistant to C. gloeosporioides) highlight the differences that exists between the plants cultivated in the department of Boyacá with those found in the regions such as Risaralda and Quindío. These results may be useful for recommendations to progenitors and to initiate plant breeding processes at the national level. This is especially important since genetic distance and morphological characteristics such as the presence or absence of prickles, fruit qualities, and tolerance to phytopathogens must be considered before the development of cultivars in various areas of the country can begin (LÓPEZ et al., 2019).
An important finding which contributes to the wealth of knowledge about the genetic resources available in the Alto Ricaurte Province in Boyacá is based on the group with a high genetic similarity ( Figure 2) created by cultivated R. alutaceus (A-1BL) and wild (A-10BS). This species was reported for the first time by MORENO-MEDINA et al. (2020) and has specific morphological characteristics that are unique to other materials grown conventionally or present in the wild. The fruits of this species are yellow-whitish in color, with a sweetness of 8.8 ° Brix. According to MORENO-MEDINA et al. (2020), this species is intended to be projected as a production line of agricultural interest for Boyacá. The principal coordinate analysis was explained using two components which represented 39.21% of the total genetic variability. Some similarities were observed in relation to the cluster analysis in figure 1 including the weak grouping of species depending on the collection site. In figure 2, three groups can be seen with the first one located in the lower right part of the graph representing the greatest genetic distance. Additionally, within the morphological characteristics of the species that make up this group are purple fruits and wild and cultivated species that have yellow-whitish fruits (R. alutaceus). The second group includes a wild species and the cultivated genotypes characteristic of the area known until now as R. alpinus. The third group consists of cultivated plants from R. glaucus and a wild species. It is important to highlight that the reference groups (Outgroup) used in the current trial correspond to the R. glaucus species that are resistant to Colletotrichum gloeosporioides and moderately resistant to C. gloeosporioides. The current and former studies on this topic allow the consolidation of a group of cultivars that can be utilized to create progenitors in the future to obtain plant populations with high genetic gain (LÓPEZ et al., 2019). In order to calculate the level of genetic differentiation between the species collected in Arcabuco and Gachantivá, a hierarchical analysis of molecular variance (AMOVA) was performed. Table 4 demonstrates that there is molecular genetic differentiation of 23% between the groups corresponding to the geographical location of the sample. In addition, it was observed that the greatest contribution to variance is within the groups (76%), possibly because each of them is composed of different cultivated species and wild relatives of the Rubus genus.
Based on the results of Table 4, the greatest genetic variation is possibly due to the fact that the individuals studied were collected at altitudes between 2,399 and 2,800 meters above sea level. These regions are characterized by having mountains, streams, rivers, and some roads that can act as corridors that determine the behavior and genetic differentiation within the groups or species evaluated (GARRIDO-GARDUÑO; VAZQUEZ-DOMINGUEZ, 2013).
VELJKOVIĆ et al. (2019), mentions that in the genus Rubus, altitude plays an important role in the genetic separation of the populations of raspberries (R. idaeus). It should also be taken into consideration that the species of the genus Rubus are characterized by having a taxa with a wide and successful distribution and can be found in different parts of the world and present morphologically different individuals. It can be inferred that apomixis, allogamy, and polyploidy have maintained a permanent flow of genes between wild, invasive, and cultivated species (SOCHOR et al., 2015). This also extends to causing important adaptive strategies based on environmental conditions. Despite the number of edaphoclimatic conditions previously described in Arcabuco and Gachantivá, both share borders and some of the locations from where samples were taken are close. The genetic variation can therefore be attributed to the fact that the distance between collection sites is short, corroborating the fact that geographical distance is a factor that influences the genetic diversity of plant populations, as stated by WRIGHT, (1943).

Conclusions
The grouping of the species studied does not necessarily correspond to the geographical origin of the genotypes.
The principal and cluster coordinate analysis show the greatest genetic distance for the cultivated and wild species that produce yellow-whitish fruits. It also demonstrates the potential related to genetic diversity and the logical morphological characteristics that can be consolidated in this species as an interesting production line in the department of Boyacá.
The molecular analysis of variance (AMOVA) revealed greater variability between species than between the studied populations.