Genetic variability among cassava accessions based on SSR markers

The aim of this study was to characterize and estimate the genetic similarity among 93 cassava accessions. The DNA amplification was performed with 14 microsatellite primers. The amplification products were separated by a polyacrylamide gel electrophoresis, showing a polymorphism formation, through which the accessions were discriminated against. The genetic similarity among accessions of cassava was estimated by the Dice coefficient. Cluster analysis was carried out using the UPGMA method. The polymorphic primers amplified a total of 26 alleles with 2-4 alleles per loci. The genetic similarity ranged from 0.16 to 0.96. The average values for obser ved and expected heter ozygosity wer e 0.18 and 0.46, r espectively . Twenty genetic similarity clusters wer e determined, demonstrating diversity among accessions, suggesting the possibility of heterotic hybrid generation.


INTRODUCTION
Cassava (Manihot esculenta Crantz) is the staple food of nearly 700 million people worldwide.It is the only species of the genus Manihot grown commercially for the production of edible roots.It adapts to different climate and soil conditions and can be found from the Southern U.S. to northern Argentina.Generally, it is planted in marginal areas with low productivity (Fukuda and Iglesias 2006).The species has a large genetic diversity due to the ease of pollination, high heterozygosity and sudden dehiscence of the fruit, leading to a continuously infinite number of new genotypes.In Brazil, the wide genetic variability of cassava is kept in on-site collections and germplasm banks distributed throughout the country.This variability is represented mostly by native varieties or those naturally selected by farmers.
The conservation of cassava germplasm is essential in reducing the genetic erosion of species and genetic diversity available to help improve the culture.In cassava this erosion is mainly due to the expansion of agricultural frontiers, the advancement of urbanization, and the biotic and abiotic stresses and to a lesser extent, by the replacement of traditional varieties by improved varieties (Fukuda et al. 1996).The characterization of the species diversity is important for germplasm conservation.Estimating the degree of relatedness between accessions makes it possible to eliminate possible duplicates and more efficient use of them in breeding programs.
The DNA markers are useful in studies of evolution, domestication, ecology and phylogeny of plants and also for genetic mapping and cloning of genes.Therefore, allowing a short term evaluation of a large number of genotypes without environmental influence (Costa et al. MNO Ribeiro et al. 2003).The use of molecular markers in assessing the genetic diversity of cassava germplasm has been increasing in recent years, mainly due to the large amount of information concerning the genome of this species available (Colombo et al. 2000, Mühlen et al. 2000, Carvalho and Schaal 2001, Elias et al. 2001, Mba et al. 2001, Elias et al. 2004, Zaccarias et al. 2004).As cited by Faraldo et al. (2002), the molecular markers are very important techniques for the identification of accessions held by germplasm banks, whose purpose is to maintain genetic variability for subsequent use of the accessions.One type of molecular marker that is more informative in evaluating cassava germplasm is a microsatellite (Chavarriaga-Aguirre et al. 1998).
The aim of this study was to estimate the genetic variability among 93 accessions of cassava from the germplasm bank at the Federal University of Lavras for 14 microsatellite primers, enabling the identification of genetically divergent clones for polycrossing, aiming at the development of new clones.

MATERIAL AND METHODS
Part of the accessions held in the germplasm bank at UFLA was provided by Embrapa Cassava and Fruit Crops, Cruz das Almas, BA.These clones were used in field polycrossing, from where the new clones were obtained, which together with some commercial cultivars were evaluated in this study (Table 1).
Molecular analyses were performed at the Molecular Genetics Laboratory, Department of Biology (DBI) Federal University of Lavras.DNA samples of each accession were extracted according to the procedure used by Pereira et al. (2007).Subsequently the DNA was used for PCR amplification using 14 microsatellite primers (GA-5,  preestablished by Chavarriaga-Aguirre et al. (1998).
Each amplification used 20 ng DNA, 100 mM of each dNTPs, 1U of taq DNA polymerase, buffer consisting of 50 mM TRIS pH 8.3, 20 mM KCl, 2 mM MgCl2, 10 mg BSA, 0. 25 % Ficoll 400, 10 mM of tartrazine and pure water.The final volume for each reaction was 12 ìL.The amplification was performed in Eppendorf Mastercycler® Thermal Cyclers, which was used in the following program: one step at 95 o C for two minutes for initial denaturation, followed by 32 cycles at 94 o C for 20 seconds each for DNA denaturation, 40 seconds for annealing primer at 47 ºC, 40 seconds for extension at 72 °C, followed by a final extension at 72 o C for four minutes.
After amplification, the reaction products were separated by electrophoresis in polyacrylamide gel (19:1 -Acrylamide: Bis) 6 %, non denaturant, in TBE buffer at 130 V for two hours.For revelation of the gels a staining method was used with silver nitrate.After electrophoresis, the plates were separated and the gel was immersed in 1 liter of a fixative solution (10 % ethanol, 0.5 % acetic acid) and kept under slow agitation for 15 minutes.Then, the gel was submerged in 1 liter of a silver nitrate solution (AgNO3) (0.2 %) under slow agitation for 10 minutes.The gel was then washed with water and slowly stirred in a solution of revelation (3 % NaOH, 0.5 % formaldehyde) until complete visualization.The bands were then photographed using a Sony Cyber-shot DSC-W30 for later analysis.
The identification of DNA fragments amplified by microsatellite markers was done by performing a visual analysis.Each microsatellite band was evaluated for the 93 accesses, identifying the presence of a band by 1 and for an absence by 0. From the matrix 0 and 1 genetic similarity (sg ij ) was estimated between all pairs of genotypes using the Dice coefficient (Dice 1945): sg ij =2a/(2a+b+c), which a corresponds to the presence of a band on the individual i and j; b represents the presence of the band in i and absence in j; and c represents the absence of the band in i and presence in j.Analyses of similarity were performed using the program NTSYS version 2.1 (Rohlf 2000).Based on the coefficient similarity, the genotypes were clustered by UPGMA and represented in a dendrogram.The genetic difference among accessions was identified in the dendrogram given the estimate of the maximum significant similarity (sgm).The sgm was estimated by the t-test obtained by the expression: sgm=1-( ), where t is the tabulated value of t with n-2 degrees of freedom and is the average error of sg ij (Hagiwara et al. 2001).
The genetic diversity parameter: number of alleles per locus and heterozygosity (observed and expected) were estimated using the software Popgene version 1.31 (Yeh et al. 1999).

RESULTS AND DISCUSSION
A total of 26 alleles were amplified with 14 SSR loci analyzed in 93 accessions.Of the 14 primers, 12 generated polymorphic (GA-5, GA-12, GA-16, GA-21, GA-57, GA-123, GA-127, GA-131, GA-134, GA-136, GA-140 and GA-161).The number of alleles observed per locus ranged tS sg S sg Genetic variability among cassava accessions based on SSR markers from 2-4 with an average of 2.2 alleles per locus and the most informative primer GA-131 was amplified by having a larger number of alleles (Table 2).
In this study, the findings resemble those of Roa et al. (2000).These authors, working with the amplification of DNA from cassava by means of microsatellite markers, found a higher polymorphism with the same primer GA-131.Mühlen et al. (2000) studied the genetic variability of cassava landraces assessed by microsatellite markers; obtained 97.96 % of polymorphic primers, with the total number 2-8 alleles per locus and an average of 4.5 alleles per loci.
Assessment of diversity using molecular markers has confirmed the wide genetic variability of cassava, but some comparisons with wild species indicate that they may contain as much or more variability than those cultivated.MNO Ribeiro et al.Wanyera et al. (1992) using polymorphism of 13 isozyme loci, analyzed the genetic variability of 20 plants of M. glaziovii, 20 cassava clones and 49 clones of cassava tree, collected in Nigeria, and found higher diversity among cassava trees, followed by M. glaziovii and less variability among the cassava clones.
Heterozygosity can be considered a measure of genetic variability.This measure refers to how much of that variation exists in the population and how that variation is distributed across the alleles of an analyzed locus.Low heterozygosity means little genetic variability.It is considered a polymorphic locus when the most common allele frequency is less than 0.95 (Menezes et al. 2006).The heterozygosity of a marker is the probability of an individual being heterozygous at a marker locus, and this depends on the number of alleles and their frequency in the population.
The observed heterozygosity (H o ) is the proportion of heterozygous individuals in population samples, expected heterozygosity (H e ) is the probability of an individual being heterozygous in any locus.In this study, the highest heterozygosity values observed and expected (Table 3) were achieved with primer GA-131 (H o = 0.56 and H e = 0.69).The observed heterozygosity showed low values in relation at expected heterozygosity values in most of the loci, indicating homozygous in excess.58 % of the polymorphic primers were homozygous and this suggests that sexual reproduction from the 93 clones must have had a high rate of self-pollination.Another possibility is that the clones originated from small populations or high levels of inbreeding.
The accessions showed values among 0.00 and 0.56 to observed heterozygosity and values among 0.34 and 0.69 to expected heterozygosity.The average achieved for these indexes were H o = 0.18 and H e = 0.46.Roa et al. (2000) found for 10 microsatellite loci, levels of expected heterozygosity H e = 0.43 (38 accessions of Manihot esculenta), H e = 0.31 (26 accessions of Manihot carthaginensis), H e = 0.54 (16 accessions of Manihot esculenta ssp.flabellifolia) and H e = 0.32 (Manihot esculenta ssp.peruviana).The index of heterozygosity observed were H o = 0.63; H o = 0.12; H o = 0.44 and H o = 0.31 respectively.Lefèvre and Charrier (1993) found rates of heterozygosity for 17 polymorphic isozyme loci of 0.225 in 365 accessions of cassava and 0.252 in 109 accessions of M. glaziovii.
These genotypes are often given confusing names, sometimes assigning the same name to different access or different names to the same access, which complicates the correct identification of the material and may cause the presence of duplicates in germplasm banks.Thus, based on this same rationale, it is noted in Table 1, different accessions of the same name (accessions 37 and 91 -MOCOTÓ and accessions 50 and 77 -UFLA 60).Those through the dendrogram are in different groups.
Coefficients of similarity between the 93 cassava accessions ranged from 0.16 to 0.96.The lowest genetic similarity (0.16) was observed among accessions UFLA 60 and UFLA 76, while the highest genetic similarity was observed between UFLA 9 and DESCONHECIDO 3; UFLA 25 and LT 3; IAC 12 and UFLA 73 all with a 0. 96 similarity.After estimating the genetic similarities and considering the similarity of 69 % among clones, we observed the formation of 20 groups.
Coefficients of similarity between the 93 cassava accessions ranged from 0.16 to 0.96.The lowest genetic similarity (0.16) was observed among accessions UFLA 60 and UFLA 76, while the highest genetic similarity was observed between UFLA 9 and DESCONHECIDO 3; UFLA 25 and LT 3; IAC 12 and UFLA 73 all with 0. 96 similarity.After estimating the genetic similarities and considering the similarity of 69 % among clones, we observed the formation of 20 groups. The

Table 1 .
Accessions of cassava (Manihot esculenta Crantz) used in this study and their origins