Genetic Diversity and DNA Barcoding of Yam Accessions from Southern Nigeria

Knowledge of genetic diversity and barcoding of yam is lacking in Enugu and Ebonyi States of southern Nigeria. Therefore, DNA barcoding was used to facilitate identification and biodiversity studies of yam species from Southern Nigeria. Seventy five yam accessions were collected from Enugu and Ebonyi States, including International Institute of Tropical Agriculture for DNA extraction and amplification using a chloroplast DNA (cpDNA) ribulose-1,5bisphosphate carboxylase (rbcL) marker. There was high level of similarity among the accessions and presence of 534 conserved and 7 variable sites. A transversional mutation of G/T at a consensus position of 335 was identified followed by transitions at 362 (A/G), 368 (A/G), 371 (C/T) and 391 (C/T) within the accessions. Phylogeny resolved the yam accessions into ten major groups with their bootstrap values ranging from 0 100. Phylogenetic diversity was highest in group X, followed by VII, VI and IX. The inter-group genetic distance based on Kimura 2-parameter model ranged from 0.5000 ± 0.4770 5.0560 ± 2.5760, while the intra-group had 0.5250 ± 0.5000 2.0103 ± 1.2579. The mean genetic diversity within the entire population was 0.7970 ± 0.06910. BLAST analysis of total bit score, query coverage, and percentage identity were in the ranges of 411 1011, 99% 100% and 97% 100%, respectively. However, the rbcL could not resolve the yam accessions well following the comparative assessment of some discrepancies in the detected number of species from phylogenetic groupings, genetic diversity indices and NCBI BLAST hits, thereby, exposing the inefficiency of this marker in disHow to cite this paper: Ude, G.N., Igwe, D.O., McCormick, J., Ozokonkwo-Alor, O., Harper, J., Ballah, D., Aninweze, C., Chosen, O., Okoro, M., Ene, C., Chieze, V., Unachukwu, M., Onyia, C., Acquaah, G., Ogbonna, J. and Das, A. (2019) Genetic Diversity and DNA Barcoding of Yam Accessions from Southern Nigeria. American Journal of Plant Sciences, 10, 179-207. https://doi.org/10.4236/ajps.2019.101015 Received: December 21, 2018 Accepted: January 19, 2019 Published: January 22, 2019 Copyright © 2019 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access DOI: 10.4236/ajps.2019.101015 Jan. 22, 2019 179 American Journal of Plant Sciences

The crop ranks fourth after potato, sweet potato and cassava as the most important food tuber crop in the world [6].Yam is important in the economic and social life of people in West Africa [9] [10].As a starchy food, it provides a major source of cheap caloric energy food for millions of people in the tropical and sub-tropical regions of the world particularly in West Africa, the Caribbean, parts of Asia, South and Central America and the Pacific [8] [11].Yam tubers are rich sources of energy, vitamin C, musin (glycoprotein), minerals (K, P, Ca, Mg, Fe, Cu, Co), phytosterols and steroidal saponins [12].They are converted into different types of food products such as pounded yam, boiled yam, roasted yam, fried yam slices, yam balls, mashed yams, yam chips, and yam flakes [13].
Fresh yam tubers are also peeled, chipped, dried, and milled into flour that is used to prepare dough called "amala" or "telibowo" [14].
Yams are widespread in the tropics and subtropics.Nigeria is the leading producer of yam with 71% of the world production [15] [16] [17].West Africa accounts for over 92% of the world's production (54.2 million tonnes) [6].In Ghana and Nigeria, 26.2% and 31.8% of people, respectively rely on yam species for income generation and food security [4].Despite the increasing demand for local consumption and export of yams, there has been a marginal decline in its production due to lack of proper identification of unique species for biodiversity DOI: 10.4236/ajps.2019.101015G. N. Ude et al.
diversification for resistance to drastic changes in climate, introgression, cross hybridization and conservation processes to reduce genetic erosion [18]- [24].To have adequate knowledge of these yam accessions, characterization to the species level, genetic richness and assessment of phylogenetic diversity (PD) are of utmost importance following the genetic resource preservation roles of PD in crop extinction [25], functionality in ecosystems [26] [27] and abiotic variability [28] and these can be achievable with accurate, sensitive and reliable methods.
Morphotaxonomy, the use of morphological characters to identify and classify plants, is currently the most widely used in yams in Nigeria.It entails using traits such as size, form and number of tubers per plant, bulbil formation, presence of spines on the stem, twining direction, fruit shape, and aerial bulbils, which could lead to misidentification of yam species [1] [3] [29].Further, morphotaxonomy-based method requires cumbersome assessment of whole plants and the importance of this approach declines when specimens/tissue materials are utilized [30].Use of molecular markers has become significant for accurate identification of these yams to the species level and to harness the genetic diversity inherent in them.Different markers including Restriction fragment length polymorphism (RFLP) [31], Random amplified polymorphic DNA (RAPD) [32], Simple sequence repeat (SSR) [32], Inter-simple repeat (ISSR) [12] and Amplified fragment length polymorphism (AFLP) [32] and gene sequencing [33] [34] have been applied in the characterization of yam species.The use of molecular tools to support morphotaxonomy-based identification is important to clear ambiguous species classification.
A DNA barcode facilitates taxonomic identification through the use of a standardized short genomic segment that is generally found in target lineages with adequate variations capable of discriminating living animals to the species level [35].DNA barcoding techniques are useful tools in characterization as they allow more objective and rapid specimen identification, which can be cost-effective in providing a central catalog of species diversity.In general, DNA barcoding can improve biodiversity and genetic resource databases [36] [37].Also, a phylogenetic diversity (PD) method possesses the merits of ease of reconstruction of phylogenetic relationships of species and as such it has resultant potential to enlighten effective taxonomic challenges [38].MatK and rbcL which are the two plant barcode loci have been chosen for phylogenetic studies of Dioscorea [1] [39].In this study, a barcoding marker of rbcL was used for identifica- tion and genetic characterization of Dioscorea accessions cultivated in southern Nigeria.

Sample Collection
Different yams were sampled from different locations across Eastern and Western Nigerian, including the ones in the germplasm collection at the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria (   eleven Local Government Areas (LGAs), cutting across three States including Oyo (where IITA, Ibadan is located), Enugu and Ebonyi States were used for the yam collection (Figure 1).The IITA, Ibadan, has Genetic Resources Unit that contains many yam species from other parts of Nigeria.

DNA Extraction
Fresh young leaves of yam species weighing from 0.1 -0.2 g were collected for DNA extraction using Silica resin method standardized by the DNA Learning Center (http://www.dnabarcoding101.org/lab/protocol-2.html) [40] In brief, fresh young yam leaf samples were weighed and homogenized in 300 µL of lysis solution using sterile mortar and pestle followed by incubation in a heat block at 65˚C for 10 minutes.Next, samples were centrifuged in a balanced configuration at maximum speed (13,000 rev/min) for 1 min to pellet debris.A 150 μL sample of the supernatant was transferred to fresh micro centrifuge tubes, being careful not to disturb the debris pellet.A 3 μL silica resin, was subsequently added to the respective supernatants, mixed well by pipetting up and down, and placed for 5 minutes in a heat block at 57˚C.The silica resin is a DNA binding matrix which in the presence of lysis solution binds readily to nucleic acids.After incubation, tubes were subject to centrifugation, with cap hinges pointing outward, for 30 seconds at maximum speed to pellet the silica resin, which was now bound to nucleic acid.Using a micropipette with fresh tip the supernatant was removed and 500 μL of ice cold wash buffer added to the pellet.The silica resin bound to nucleic acid was re-suspended by vortexing and centrifuged to repeat the wash procedure.The wash buffer removes contaminants from the samples while nucleic acids remain bound to the resin.A dry spin step after wash was performed to remove any remnant drops of supernatant with a micropipette.Finally, 100 μL of distilled water was added to the silica resin, mixed well by vortexing and incubated at 57˚C for 5 minutes.Samples were then centrifuged for 30 seconds at maximum speed to pellet the resin.This time 90 μL of the supernatant was transferred to fresh tubes as the nucleic acids eluate from the resin.The eluted DNA was stored to proceed to PCR step.

Data Analysis
The sequencing results generated from the Applied Biosystems Genetic automated sequencer (ABI Prism 3130X1, Froster City, CA 94404, USA) at Genewiz LLC were uploaded in the blue line of DNA Subway (https://dnasubway.cyverse.org/),which is an intuitive interface for analysing DNA barcodes.Using the Blue Line, the assembled sequences were end-trimmed, paired in their respective forward and reverse sequences to build consensus sequences.The consensus sequences from DNA subway were further edited, filtered and assembled for polymorphism detection using BioEdit software (BioEdit sequence aligner editor, version 7.6.2.1).Sequence alignment and percentage similarity searches were compared with GeneBank databases using NCBI web-based site, BLAST.Multiple alignments were done using the ClustalW [41] [42].Phylogenetic tree reconstruction was performed using MEGA 6 software [43].Phylogenies were constructed using the Maximum Parsimony and Maximum Likelihood options [44] [45] and the effectiveness of the trees was determined by bootstrapping up to 1000 replicates [46].DOI: 10.4236/ajps.2019.101015

Sequence Alignment of Sequences Generated from Dioscorea Spp. Using rbcL Barcoding Marker
A total length of sequence alignment, conserved sites, and variable sites of 525, 534 and 7 were respectively identified among the sequenced yam species.Different regions of polymorphisms and conserved regions at nucleotide level across the sequences exhibited variations among them.At a position of 335, 62_3LeavedYam_Ono and 76_Ona_TDd possessed a transversional mutation by having G nucleotide, while other samples had a T nucleotide (Additional file 1: Figure S1).At a consensus position of 362, yam species such as 3_TDa3050, 4_TDb3050, 5_TDb3044, 6_TDb2857, 7_TDb3058, 8_TDb3690, 61_6-EDO, 83_WaterYam-_Mbuna and 85_AerialYam_Edugbe showed a transitional mutation of A nucleotide, while the rest of the accessions had a G nucleotide.At a position of 368, accessions such as 1_TDa85.00250,3_TDa3050, 4_TDb3050, 5_TDb3044, 6_TDb2857, 7_TDb3058, 8_TDb3690, 61_B-6-EDO, 83_WaterYam-_Mbuna, 85_AerialYam_Edugbe had a transitional mutation of A in place of G nucleotide possessed by other accessions at the same consensus position.. Also at 371 position, accessions including 35_Pepa, 36_Ke-emi, 37_Ame, 38_TDr.89.002665, 39_AlataTda98.01176,40_TDa00.00.94, 42_OgojaVariety.

Discussion
DNA barcoding has become an effective method for species discrimination of flowering plants in the Polygonaceae [35] [47] and Fabaceae families [39], and other land plant species [35] [42] [48] [49].While mitochondrial cytochrome oxidase 1 (CO1) has proven a standardized animal DNA barcoding for necessary discrimination, no single barcode sequence works across all plants [49].In the present work, the candidate barcoding marker, rbcL satisfied the DNA barcoding process, regarding the ease of amplification and sequencing Hollingsworth et al. [49].However, this barcoding marker, rbcL, was not able to achieve the basic quality of discriminating different yam species in this study.Sequence alignment showed low degree of polymorphisms among the sequences.This study of genetic diversity in yam accessions is also dependent on the nucleotide variations occurring within the genome that are informative for the identification of different species.The discriminatory level of the rbcL marker has been linked to other researches, which contradict its potential for use as a universal DNA barcode for plants [50] [51] [52].This low resolution of different accessions of yams into their respective species level could be attributed to the poor efficiency of rbcL marker when not jointly applied with other plastid markers.It has been reported that the joint application of rbcL+matK as a marker of choice in species resolution was based on clear recovery of the region of rbcL and discriminatory efficiency of fast evolving coding region of matK [53].

G. N. Ude et al.
In this study, 525 bp distinct total lengths of sequence alignment, 534 conserved sites, and variable sites of 7 were identified in the sequenced yam species.
The alignment of 525 bp out of the total lengths of 568 bp, followed by the existence of similar regions (conserved sites) and low points of variations (variable areas) among the sequences demonstrate the low level of informativeness of rbcL in DNA barcoding of yam species.These findings are not in complete agreement with a previous report on yam species [30], where 568 bp, 538, and 30 as total lengths of sequence alignment, conserved sites and variable sites were identified among accessions.Also, the sequence alignment length, conserved sites excluding the variable sites detected in this work correlate with the findings of Sun et al. [54] in which 553 bp, 522 bp and 31 of alignment length, conserved sites and variable sites were found among the accessions of Dioscorea species.The difference in the variable sites could have emanated from the number of samples studied.
Phylogenetic reconstruction of the generated Dioscorea species using rbcL marker resolved them into ten groups and this indicates different existing isolated groups inherent in the accessions.The existence of these different accessions among the collections could be attributable to lack of exchange of yam tubers by farmers among villages thereby resulting in a stronger heterozygosity among species compared to wild ones as reported by Ngo Ngwe et al. [24].A contribution of evolutionary biology regarding conservation is the knowledge of diverse phylogenetic diversities, defined by the sums of branch lengths of the evolutionary trees connecting a set of taxa or individuals [55].In this present work, group X had the highest PD value of 88, followed by groups VII, VI, IX, and VIII with their respective PDs of 86, 79, 60 and 51.The highest PD was identified in a group containing wild species of D. aspersa and this is in agreement with a previous report though in a different wild species wild called D. praehensilis [24].When compared with other unrelated crops, the highest was observed in Cocoyam and other crops which were deliberately included to access the accuracy of this marker.The group with the lowest PD value D. rotundata clustered with other species and they were collected from a given single region.
In this way, a given set of taxa will have a greater PD if they are widely spread out on a phylogenetic tree.Lack of or total loss of PD is generally assumed as a declining signal in the degree of biodiversity [56].Furthermore, PD is associated with functional diversity since it is a measure of features also due to the fact that evolutionarily distant species are more likely to possess variable molecular functions in an ecosystem [28] [57] [58].Also in group I, some accessions including 45_Amolavariety, 40_TDa00.00.94, 38_TDr89.002665,16_TDc0497.4,78_Obella, 37_Ame, and 35_Pepa had a PD value of 0 and this could be attributed to lack of sequence divergence.It could also be attributable to occurrence of common ancestral sequence homology [59] or poor resolving power of the rbcL DNA barcoding marker in yams [54].Most of the accessions were accurately grouped according their species.For instance in group VIII, all the D. esculenta species was DOI: 10.4236/ajps.2019.101015grouped with a known reference sequence from NCBI database.Also, group VII had all the accessions classified as D. bulbifera thereby identifying correctly an accession, 3_TDa3050, which was regarded as D. alata.A particular yam species was given different names as Ighu or Una (Ona) at a village in Enugu State but it was found to be just one species called D. dumetorum through the use of rbcL thereby resolving the issue of multiple names for the plant.Group IX had three reference sequences as D. dregeana, D. hispida and D. dumetorum but most of the accessions in the group are D. dumetorum and this could be as a result of their genetic relatedness [59].However, accessions in group I of the trees were not correctly resolved following the existence of different yam species in various distinct subclades and non-grouping of any of the retrieved yam sequences from NCBI database.This may possibly be linked to sample contamination or a deficiency on the part of the rbcL resolution.
The identified genetic distances (0.5000 ± 0.4770 -5.0560 ± 2.5760) based on K2P model regarding the inter-groups were in agreement with the previous works of other researchers in yams [24] [54] and in authentication of native plants [60].High genetic diversity indices were obtained from between group calculations, producing 5.0560 ± 2.5760 with the highest in two combined groups (groups VI and VII) and this demonstrates higher interspecific diversity than intraspecific one within the yam accessions as obtained in an earlier report involving ornamental plants with interspecific value of 3.080 [61].Assessment of genetic diversity within the groups (intra-group genetic diversity) could not be computed in most of the groups except three groups (groups I, VI and X), where group I had the lowest value of 0.5250 ± 0.5000, while X had the highest value of 2.0103 ± 1.2579.These values are higher than the ones obtained by Sun et al. [54].The mean genetic diversity within entire population was 0.7970 ± 0.06910 and this is higher than the one (0.00266 ± 0.0044) obtained by Sun et al. [54].
BLAST hits obtained in this study showed some degrees of similarity matches to the ones already annotated and deposited in NCBI database and some were not purely specific.The percentage sequence identity ranged from 97% -100%, demonstrating low efficiency of this tool in identification of unknowns in yam species.However, some of the yams sampled from different regions were differently identified from what they were previously known to be using this method, indicating the potential of rbcL barcoding marker to resolve misclassification encountered via morphotaxonomy based approach despite the low discriminatory power.For instance, yam accessions with acronyms including TDa, TDc and TDm denoting D. alata, D. cayenensis and D. manganotiana were found to be D. bulbifera, D. rotundata or cayenensis, respectively.Furthermore, 62_3LeavedYam-Ono and 76_Ona_TDd sequences were correctly identified as D. aspersa.In the community where the two species (D. aspersa and D. dumetorum) were collected, they were misclassified by the villagers who generally called them D. dumetorum due to their similar morphological features.According to the villagers, the ones in group X which were later identified as D. aspersa are normally boiled and eaten directly, while the other ones (D. dumetorum, which had similar values of NCBI hits with D. hispida) are usually boiled, processed to remove bitterness in them before they are consumed.The discriminatory level of the rbcL marker in plants as a potential universal DNA barcode is demonstrated in this study as reported in other researches [50] [51].However, some of the yam sequences had two, three or five NCBI hits of different species of yams with synonymous values of total bit score, query coverage, -value and percentage identity with different accession numbers except in one that had five BLAST outputs with three having similar accession numbers and two with different accession numbers.For instance, the yam accession, 59_D-10-White-Nwopoko-Adaka, had five different NCBI hits of D. spicata, D. intermedia, D. wallichii, D. rotundata and D. oppositifolia with three (D.wallichii, D. rotundata, D. oppositifolia) having similar accession number (KY679569), while the re- maining two (D.spicata and D. intermedia) had separate accession numbers of KY457460 and KY457459, respectively, after the BLAST analysis.Also, sequences generated from accessions 45_Amolavariety, 40_TDa00.00.94, 38_TDr89.002665,16_TDc0497.4,78_Obella, 37_Ame, and 35_Pepa hit two (D.cayenensis and D. rotundata) sequences with similar values of query coverage, e-value and percentage identity, while total bit score ranged from 1000-1005.This is possible due to existence of common ancestral homology as opined by Pearson [59] or due to redundancy, which in bioinformatics is observed when one or more homologous or synonymous sequences are found in the same set of data [62].It could also be attributable to the low discriminatory potency of rbcL marker to correctly resolve species as previously reported in yams [54] and ornamental plants [61].

Conclusion
The candidate barcoding marker, rbcL, was found to be ambiguously discriminatory in DNA barcoding process of yam accessions.Some of the accessions were not correctly identified to the species level and low polymorphisms were detected and this further demonstrates the low distinguishing potency of rbcL barcoding marker.The use of phylogenetic diversity (PD), which is associated with functionality in biodiversity and which was applied in the computational processes for the estimation of phylogenetic groups with lowest and largest collections in terms of diversity was of great potential.The highest phylogenetic diversity was in D. aspersa, while some were not computable due to the low efficacy of the marker.The group with the lowest PD value, D. rotundata clustered with other indistinguishable species and they were collected from a given single region.The accessions with high PD within the yam accessions should be considered for use in breeding programme to enhance biodiversity of Dioscorea species within the studied region.However, the rbcL could not resolve the yam accessions well following some noted discrepancies in the detected number of species from phylogenetic groupings and NCBI BLAST hits possibly due to inefficiency of the marker.Therefore, the rbcL may not be a marker of choice for species identification, discrimination and estimation of genetic diversity of yam accessions.The marker should be used in combination with other chloroplast markers for accurate DNA barcoding of yams for their improvement and germplasm conservation.

Figure 1 .
Figure 1.Map of Nigeria showing geographical areas for collection of yam accessions.

Table 1 .
List of yam samples collected from different locations and used for DNA barcoding.

Table 2 .
Genetic distances based on Kimura 2-parameter (K2P) between different groups of yam species.

Table 3 .
Genetic distances based on Kimura 2-parameter (K2P) within different groups of yam species.

Table 4 .
BLAST outputs of total score, query coverage, e-value, percentage identity and accession number obtained from different yam accessions. .trifida, D. dregeana, and D. mangenotiana.The total bit score obtained in all ranged from 411 -1011.The query coverage spanned between 99 and 100%, while the expected values (e-values) were 9e -111 or less.The percentage sequence identity ranged from 97% -100%.Some accessions with acronyms including TDa, TDc and TDm denoting D. alata, D. cayenensis and D. manganotiana were found to be D. bulbifera, D. rotundata or cayenensis, respectively.Some of the D

Table S1 .
List of sequenced yam species collected from different locations and their GenBank accession numbers.