Transferability of Sorghum Microsatellite Markers to Bamboo and Detection of Polymorphic Markers

INTRODUCTION

Bamboos are members of the sub-family Bambusoideae within the grass family Poaceae [1] in which most important cereal crops such as rice, wheat, sorghum, maize and barley are also grouped. Bambusoideae is a large sub-family containing more than 70 genera with over 1450 species [2]. It is a fast growing wood grass distributed all over the world, but major contributors of bamboo resource are Asia (particularly China, India Japan, Myanmar and Malaysia) and South America mainly Brazil, Venezuela and Colombia [3]. Ethiopia ranks first in Africa which comprises 67 % of the total bamboo area coverage of the continent [4].

Bamboo is an important multipurpose genetic resource in the world for its wide range of economic values. It is the source of building materials, high quality furniture, food, fiber, fodder, pulp, paper, board and charcoal [5-8]. It also serves as ornamental plant. The plant also plays a significant role in ecological applications because of unlimited environmental values [2]. Due to its fast growth nature, it can be harvested within short period of time without depletion and deterioration of the soil. It can grow on marginal land which is not suitable for agriculture or forestry. In addition, the plant serves as source of feed for many wild animals.

Despite its huge importance of the plant, it is less studied and its genetic diversity is little understood. Besides, very wide area coverage in western and south western Ethiopia currently are severely affected as a result of mass flowering and seed setting of the plant [9]. It demands extensive research to exploit the effective methods for characterization and conservation of this important genetic resource. Morphological traits are the most commonly used approach in bamboo classification [3]. Although morphological traits remain to be major characterization tools, the recent advance in molecular genetics has enabled in refining classification and within species diversity analysis at higher resolution.

Various marker systems such as random amplified polymorphic DNA (RAPD), restriction fragment length polymorphism (RFLP), amplified fragment length polymorphism (AFLP), simple sequence repeat (SSR), diversity arrays technology (DArT) and single nucleotide polymorphism (SNP) have been used in the study of various plant genetic resources. Owing to the technical simplicity in their application, recently, SSRs have become major tools in population genetics, genetic mapping and other plant genomic studies [10]. These markers are highly reproducible, which is highly important in genetic analysis. They are highly polymorphic and produce very high allelic variations even among very closely related varieties. The co-dominance nature of the marker also helps to analyze the segregating populations or paternity testing. The SSR markers are abundant and well distributed throughout the nuclear genome of the species [11, 12]. However, the development of SSR markers requires relatively high cost and time consuming [13, 14]. On account of their transferability across related families, SSR developed for one species can be used for other species within same family.

Microsatellites are usually transferable across grass family due to the evolutionary relationships among species belong to the same family. SSR markers particularly expressed sequence tag (EST)-derived SSRs have shown to possess a higher potential for inter-specific transferability than genomic SSRs [15]. Several studies have been conducted on the transferability of SSR markers across species or genera of several cereal crops which belong to grass family such as rice [16], wheat [17], wheat, rye and triticale [18], barley [19], sugar cane [20, 21], sorghum [22, 23] major cereal to minor grass [24] and pearl millet [25]. There are also reports about transferability of SSRs from cereals such as rice and sugar cane to bamboo species [26, 27]. Sharma et al. [15] also conducted an experiment on identification and amplification of EST-SSR markers in different bamboo species. Sorghum is one of the cereals crops with already developed large number of SSRs markers and it is genetically closer to bamboos than other cereals such as rice. However, there has been no report on transferability of sorghum SSR markers to bamboo species. The objective of this study was to perform the transferability test of SSR markers from sorghum to bamboo species and to employ the transferable markers in the genetic diversity analysis of selected bamboo species.

MATERIALS AND METHODS

Selection of Sorghum SSR Markers and PCR Analysis

A pair of 90 SSR polymorphic sorghum primers [28-31] that are evenly distributed across the whole sorghum nuclear genome were selected to test the transferability of the marker across the selected bamboo species (Table 2). PCR amplification was performed in a 10 µl reaction volume comprising of 1 x PCR buffer (20 mM Tris-HCl, pH 7.6; 100 mM KCl; 0.1 mM EDTA; 1 mM DTT; 0.5% (w/v) Triton X-100; 50% (v/v) glycerol), 2 mM MgCl₂, 0.16 mM dNTPs, 0.16 µM fluorescent labeled M13-forward primer, 0.04 µM forward primer, 0.2 µM reverse primer, 0.2 units of Taq DNA polymerase (SibEnzyme Ltd, Russia) and 30 ng of template DNA. All forward primers contained an M13-tag (5’- CACGACGTTGTAAAACGAC - 3’) on the 5’ end that was fluorescently labeled to allow detection of amplification products. The Forward primers were labelled with FAM, PET, NED or VIC (Applied Biosystems). PCR reactions were carried out in a GeneAmp^® PCR System 9700 thermal cycler (Applied Biosystems, USA) programmed for initial denaturation at 94°C for 15 min, followed by second denaturation at 94°C for 30 sec, annealing temperature ranging from 50°C to 55°C for 1 min, extension at 72°C for 2 min and final elongation at 72°C for 20 min.

Successful amplification was confirmed by running 2.0 µl of the PCR products on a 2% (w/v) agarose gel stained with GelRed^® (Biotium) and visualized using gel image system (UVtech, UK). Annealing temperature for the selected bamboo species were optimized in order to obtain clear bands.

Clear and strong bands were observed at annealing temperature 55°C (Fig. 1) while the optimum annealing temperature for most sorghum polymerase chain reaction is 50°C. The success of PCR amplification is largely dependent on the annealing temperature of the PCR program. Optimization of the program for the selected experiment is a crucial step during the analysis. The optimized annealing temperature (55°C) can be applied to amplify the DNA of various bamboo species and other related polyploidy species.

Table 2.

No.	Marker Name	Forward Primer	Reverse Primer
1	gpsb151	ATACCAAGTTTCCTTTACCT	GTTGGGGGAGAGTTTT
2	SbAGE01	GACCGATCTAATGATGCAG	ACGGTAGAGAAAGACCCATC
3	SbAGH04	GGCACTCATGGAGTCACA	TTTATCCAAATCAAACCGG
4	SbKAKG1	AGCATCTTACAACAACCAAT	CTAGTGCACTGAGTGATGAC
5	Xcup02	GACGCAGCTTTGCTCCTATC	GTCCAACCAACCCACGTATC
6	Xcup16	TGCAGTGCTAGCTCATGGTC	CTTTCCAGCCTCCCATATCC
7	Xcup24	AAACTGGATGCCACACCAAG	AGCTATACCAACACGGGCAG
8	Xcup33	GCGCTGCTGTGTGTTGTTC	ACGGGGATTAGCCTTTTAGG
9	Xcup36	TGAGCTGATAATGGCTGCTG	GCGTCACGGAAGTTGGAC
10	Xcup40	ACGGAGAATAGAAAGTGGCG	TTGAGCATGCAACCACCTAC
11	Xcup50	TGATTGATTGAGGCAGGCAC	TTCCGGTCTCTGTCCATTTC
12	Xgap001	TCCTGTTTGACAAGCGCTTATA	AAACATCATACGAGCTCATCAATG
13	Xgap084	CGCTCTCGGGATGAATGA	TAACGGACCACTAACAAATGATT
14	Xgap342	TGCTTGTGAGAGTGCCTCCCT	GTGAACCTGCTGCTTTAGTCGATG
15	Xisep0101	CAGATCTCCGGTTGAAGAGC	TGAGCCGAGCTCAACATACA
16	Xisep0108	GTACGTTCCCCATCCTTCCT	CTCCTGTTCTCTCCGCATTC
17	Xisep0110	GAGGGGAAGCTGGAGACC	TCAAGTGTACAACGCATCCAG
18	Xisep0138	GAGATCGAGAGGCACTTTGG	CAGCGACAAGCCAATACCA
19	Xisep0224	ACTGGGGTTCCTTTTCCTGT	TCCCTGATTTCCCCTCTTTT
20	Xisep0327	CTGTTTGTGCTTGCAACTCC	TCATCGATGCAGAACTCACC
21	Xisep0422	TGCCCGTAATTAAGCCCATA	CCCACTGCTCCAGGTAAGAA
22	Xisep0429	GTCGTCTGGAAGCAACAGC	TGGGGGTAGTTGGTGGTG
23	Xisep0443	TCATGTACAGAGGCGACACG	AGGTCGCAACAGACACCTTC
24	Xisep0449	CCGCTCATCAGTCATCACAT	ACAAAATCCATCCCACAACG
25	Xisep0511	CCTCGCCCAAAACCCTAC	GAGGATCACCTCATCGTGCT
26	Xisep0539	GACCCCATCTCTTCCTTTCC	AGACTGAGGCACCGCTTG
27	Xisep0543	CGAGGGTTTTTCTTCTGTGG	GACATCGGAGACCTTGAGGA
28	Xisep0612	CTCTCTGTCCTCCTCGTCGT	CCTGCTTCTTGGACACCTTC
29	Xisep0617	GGCTGGGAGAGCTAGGAAGA	GACGGCTCGTCCATCATC
30	Xisep0622	GAGGATCGGAGGAAGAGACC	TCTCCCATTCTCCCCTCTTT
31	Xisep0639	TCGGACGGAGTCATCAGATA	GCCTTCGTGTCTTCTGTCCT
32	Xisep0648	GAGAAGTTGGAGCGGAGGA	AACACCCAGATCAGCGAAAC
33	Xisep0805	CTCCCCCGTGATTTGATCT	TAAGCAAAAGCACCATCAGC
34	Xisep0831	TCCATGACCTTGAGGAGGAG	TTGAAGCAGGACAACACACC
35	Xisep0839	TACGCATAGCGCCTTTCAAT	ATTTCATTATGCCGGTCTCG
36	Xisep0938	TGCTGTTCTTGAACGTGTTTG	TTTTGCACAAAGTTGCGTGT
37	Xisep1029	GACCCTCCTCCTCAACCACT	CATGCATGCACAAGCAGATT
38	Xisep1032	GCAAGCTCTACGGGATCTTC	GCAGCTGGAAAATAATCGAAA
39	Xisep1039	GTGGATTCAAATCCGCTGAC	GGCAATTTGGCAAGCAAT
40	Xisep1042	GGAGGCAAGTTCAGGAAGTG	TGTGTGCAGTGCATGCTTAG
41	Xisep1107	GGATAATCTGCAGGCGACTT	CCATCTGCTGCTCTGACTTG
42	Xisep1208	TCCAAACACACAGACCGTTT	TCCGATGGTTGAGAGCTTGT
43	Xisep1231	CTGCTTATGCGCTTCGATTT	CATAATGGGTGCACTCTAGCC
44	Xtxp008	ATATGGAAGGAAGAAGCCGG	AACACAACATGCACGCATG
45	Xtxp012	AGATCTGGCGGCAACG	AGTCACCCATCGATCATC
46	Xtxp014	GTAATAGTCATGACCGAGG	TAATAGACGAGTGAAAGCCC
47	Xtxp023	AATCAACAAGAGCGGGAAAG	TTGAGATTCGCTCCACTCC
48	Xtxp024	TTGTGTAGTCCATCCGATGC	TTCTAAGCCCACCGAAGTTG
49	Xtxp026	AAGTGTAGTAGCAGTTTAGTCTC	TAGGTATCAAAGGACCAAGG
50	Xtxp030	AAAAAGGACGCGCAGCTG	CTGGTCTCCACCATCCGTAG
51	Xtxp033	GAGCTACACAGGGTTCAAC	CCTAGCTATTCCTTGGTTG
52	Xtxp034	TGGTTCGTATCCTTCTCTACAG	CATATACCTCCTCGTCGCTC
53	Xtxp040	CAGCAACTTGCACTTGTC	GGGAGCAATTTGGCACTAG
54	Xtxp057	GGAACTTTTGACGGGTAGTGC	CGATCGTGATGTCCCAATC
55	Xtxp088	CGTGAATCAGCGAGTGTTGG	TGCGTAATGTTCCTGCTC
56	Xtxp094	TTTCACAGTCTGCTCTCTG	AGGAGAGTTGTTCGTTA
57	Xtxp113	CTCAGCTAATTTAGCCATAG	CAAGTAATAGACGAGTGAAAG
58	Xtxp141	TGTATGGCCTAGCTTATCT	CAACAAGCCAACCTAAA
59	Xtxp159	ACCCAAAGCCCAAATCAG	GGGGGAGAAACGGTGAG
60	Xtxp205	CCTGCCGTGTCTTCC	TATATGCATGCCGTAGATTT
61	Xtxp211	TCAACGGCCAATGATTTCTAAC	AGGTTGCGAATAAAAGGTAATGTG
62	Xtxp258	CACCAAGTGTCGCGAACTGAA	GCTTAGTGTGAGCGCTGACCAG
63	Xtxp270	AGCAAGAAGAAGGCAAGAAGAAGG	GCGAAATTATTTTGAAATGGAGTTGA
64	Xtxp274	GAAATTACAATGCTACCCCTAAAAGT	ACTCTACTCCTTCCGTCCACAT
65	Xtxp279	ATTCTGACTTAACCCACCCCTAAA	AGCTCATCAATGTCCCAAACC
66	Xtxp283	CGCCCGAACTCTTCTTAAATCT	ATTATGCCCTAACTGCCTTTGA
67	Xtxp284	CCAGATTGGCTGATGCATACACACT	AAGGGTAATTTATGCACTCCAAGGTAGGAC
68	Xtxp285	ATTTGATTCTTCTTGCTTTGCCTTGT	TTGTCATTTCCCCCTTCTTTCTTTT
69	Xtxp286	AGCAGCAGCAGCAACAG	GCGTGGTCTTTGTGGTTC
70	Xtxp295	AAATCATGCATCCATGTTCGTCTTC	CTCCCGCTACAAGAGTACATTCATAGCTTA
71	Xtxp312	CAGGAAAATACGATCCGTGCCAAGT	GTGAACTATTCGGAAGAAGTTTGGAGGAAA
72	Xtxp320	TAAACTAGACCATATACTGCCATGATAA	GTGCAAATAAGGGCTAGAGTGTT
73	Xtxp348	CGACATCAGCGTTGTCTTTCTA	GCTTACGAATAGGGCAAAAGAACT
74	Xtxp355	TGGTTGGAAAGATAATCAAG	GCCCTAATGAGTCCTCAC
75	Xtxp357	CGCAGAAATACGATTG	GCTATCTGGAGTAACTGTGT
76	Xiabtp 346	CCGTCTCCACAAGCTTCTTC	GACTGTGCCAGCTGTCTTCC
77	Xiabtp 452	CATGGATCGACCCTTTTGTTC	AGCGACCGAGCACAACTTATC
78	Xiabtp 175	ATGCACCAGCTCAAGGCTATC	ACGTGCATACGTGCAGAGTC
79	Xiabtp 361	TTTGCTGGACACTGAAAACG	ATCATGCATCCATGTTCGTC
80	Xiabtp 310	CACAAGACACGCACAAAGGTG	GAGAAGCATTTGCCATGGATG
81	Xiabtp 240	GAGCACAGATTTCTTCCTCCTC	TTATCTGCGACAAACCCACA
82	Xiabtp 26	CGCACTCGATCGACTAGCTTA	CGAGAGAGGCGATAAGGATC
83	Xiabtp 106	CACCCTCCTCTCATCCTCAC	GCGCCGTAGTCGTAAAGG
84	Xiabtp 424	ACCTGCAGCCTACCACAGATC	AATCCTTTTTCGTTGCCAGA
85	Xiabtp 247	GTCCAAGGCTTTCAATCTGC	TGGGACAGCAGATTACGACA
86	Xiabtp 340	CATTGCTCACTGCTCAGTTCA	CCATCGATCGAGCTCTCTG
87	Xiabtp 103	CACTTCGTTGTGCCACATTC	ACTCTGCCACCCAATCAAAC
88	Xiabtp 168	AAGAAGGACAAGAAGCGGAAG	GCTCTTCACGGTCTCCTCTG
89	Xiabtp 378	GAGCAGOGGATAAGTTCGAG	TAACCACATGGCCCGTTC
90	Xiabtp 69	TGCACAAGAGGCAGGATATG	AGCCCATGAGAAAAAGCTCA

In order to perform the genetic diversity and relatedness among the target species, the successfully transferable markers were genotyped using capillary electrophoresis. Depending on the nature of fluorescent label and strength of the amplification bands used, a volume ranging from 2.5 µl to 3.5 µl of four different amplification products were co-loaded along with the internal size standard, GeneScan™ -500 LIZ^® (Applied Biosystems) and Hi-Di™ Formamide (Applied Biosystems, USA). The fragments were denatured at 95°C for five minutes using GeneAmp^® PCR System 9700 thermal cycler. The fragments were separated by capillary electrophoresis using an ABI Prism^® 3730 Genetic analyzer (Applied Biosystems) in order to determine the exact allele size of each fragment.

Data Analysis

PCR fragment size was manually scored using GeneMapper v4.0 software. The allele size in terms of base pairs was compared to the internal size standard, GeneScan™ -500 LIZ^® (Applied Biosystems, USA). Percentage of transferability was computed by dividing the number of the SSR markers amplified in non-donor (bamboo) species to the total number of tested SSRs. Fragments in base pair were converted to binary data because of the complex amplification pattern in bamboo genome [27]. The level of polymorphism of the entire markers across the selected bamboo species was computed. Only polymorphic markers were selected to analyze the genetic diversity and relatedness among the target species. The proportions of polymorphic loci were calculated as:

Where,

P = proportion of polymorphic loci

n_pj = number of polymorphic loci

n_total = total number of loci

Power Marker v.3.25 [32] was used to compute PIC and allele frequencies. Rare, common and abundant were manually computed in MS Excel (Microsoft Inc., Seattle, USA). Polymorphism information content (PIC) was calculated using the method of Botstein et al. [33].

Where, p_i and p_j are the frequencies of alleles i and j, respectively

Markers with a PIC value of more than 0.5 were considered highly informative, between 0.25 and 0.5 as informative and less than 0.25 as less informative.

Pair-wise genetic differentiation between the selected bamboo species, a genetic dissimilarity matrix was calculated using simple matching with DARwin v5 software (available at http://darwin.cirad.fr/darwin/Home.php). Individual relations were analyzed with a tree construction based on Neighbor Joining (NJ) method, as implemented in DARwin v5.

RESULTS AND DISCUSSION

The availability of large number of sorghum SSR markers as a community resource and the consideration of sorghum as a model crop to study other grass family created a scope of opportunity to exercise similar methodology in non-model plants like bamboo. The current advancement in genomic technology will also help to set the best conservation strategies particularly for the endangered plant genetic resource as well as economically important plants like bamboo. To achieve this magnificent plan of genetic resource conservation, it is imperative to clearly characterize the core collection prior to conservation practices.

Among 90 tested sorghum SSR markers, 62.2% of them scored successful amplification in a bamboo representative species, Arundinaria alpine (Table 3). However, transferability rate differed among bamboo species. The lowest transferability rate that accounted only 40% was observed in Gigantochloa atter. Markers with high rate of transferability include Xtxp023, Xtxp205, Xtxp283 and Xcup16 where as Xiabtp346 and Xisep0327 showed reduced transferability rate. Markers such as Xiabtp424, Xise0443 and Xisep0648 did not score any amplification in either of the bamboo species. The transferability rate obtained in this study is higher than the one reported for various grass family species including bamboo using [18, 26, 27]. However, the rate is slightly lower than the rice SSR transferability to bamboo that was reported to be 68% [26]. It is possible to improve the transferability rate of markers by using markers that were developed from expressed sequences [18].

The scored fragment size ranged between 104 bp and 486 bp. Most of these fragments were amplified higher than the size of the donor fragment, which is in agreement with the previous study implying that the amplified fragments from the selected species have different allele sizes than those of the donor species [26, 24, 27].

Out of the 56 SSR markers that were found to be transferable, 31 (55%) of the showed polymorphism. A total of 147 SSR marker alleles were recovered, at an average rate of 4.5 alleles per marker. The average number of fragments per marker is less than the previous transferability test conducted using rice and sugar cane SSRs markers to study bamboo genetic diversity [27]. The number of alleles per marker ranged from 2 to 12 with the highest number of alleles obtained from Xisep0138 followed by Xcup24 and Xiabtp240 (Table 4). The presence of these multiple amplification products could be due to duplication, high level of polyploidy and genetic variability in bamboo species [12, 27, 34]. Among the polymorphic markers (n=31), 15 of them showed a clear band and highly transferable to all the bamboo species tested in this experiment (Fig. 1). These markers are found to be very effective and useful resources to utilize them for phylogenetic and genetic diversity studies of the different bamboo species. In contrary, very few markers were amplified only in some specific bamboo species ( Supplementary Table S1).

A total of 59 alleles (40 %) were detected as rare alleles with an average of two alleles per marker. The highest number of rare alleles was obtained for marker Xcup24. Similarly, 58 alleles (39.5%) scored common allele frequency considered as common alleles with an average of two alleles per marker. Xisep0327 showed the highest number of common allele. The remaining 30 alleles (20.5%) were found to be abundant allele with an average number of one allele per marker. The higher frequencies of rare and common alleles are very important resources for future bamboo characterization and maintenance program. Even though, the significance of the rare alleles cannot be speculated from the current results, their abundance in the selected bamboo species suggests that this diverse set of genetic resource has not been extensively exploited.

The level of polymorphism detected across species or genera mainly depends on the genetic divergence of species tested and primers used [24]. The PIC values for the SSR loci ranged from 0.19 to 0.37 with an average of 0.29 (Table 4). Five markers such as Xtxp205, Xtxp057, Xiabtp310, Xtxp040 and Xiabtp340 showed the highest PIC value of 0.37 whereas the lowest PIC value of 0.19 was recorded for marker Xcup24. The gene diversity index (expected heterozygosity, He) ranged from 0.21 to 0.49 with a mean of 0.37. Marker Xcup24 presented the lowest gene diversity as well as PIC value. Though the computed average PIC value in this experiment was relatively smaller, it would be sufficient to utilize in characterization of bamboo collection. The wide variation in ploidy level among the different bamboo species has created the generation of multiple alleles derived from a single locus. This has implication that a small number of markers can discriminate the different bamboo species and a useful aspect towards the conservation of the genetic resources with rare allele.

The highest genetic distance (0.90) was observed between Guadua amplexita, Gigantochloa apus and Guadua amplexita, Gigantochloa atter while lowest genetic distance (0.32) was recorded between Gigantochloa apus and Gigantochloa sumatra (Table 5). Relatively higher genetic distance (0.77) was obtained between Ethiopian lowland (Oxytenanthera abyssinica) and highland bamboo (Arundinaria alpine). The distinctiveness of these species requires proper conservation techniques in order to maintain the genetic resources.

Genetic analysis based on the polymorphic markers successfully grouped the different species in accordance to the geographic origin of the species and their genera (Fig. 2). This study revealed that Ethiopian bamboos which exist in different genera are highly differentiated. The dandrogram generated from Neighbor-Joining (NJ) tree analysis was in congruent with the result obtained from the principle coordinate analysis with few exceptions (Figs. 2, 3). The similarity in the clustering of these bamboo species using these two approaches further validates the representative nature of the transferred markers for the diversity analysis of bamboo species. The current study will serve as a foundation to study bamboo genetic resources using the transferable markers and hence exhaustive collection and characterization of both highland and lowland bamboo from the entire growing regions will be necessary in the future.

CONCLUSION

The transferability of SSR markers from related families is an alternative approach to bypass the cost developing species specific markers as well as the complexity of the work. The current screened sorghum SSR markers for the purpose of bamboo research will provide a scope of opportunity to researchers for the proper characterization and conservation of bamboo genetic resources. The adequate rate of polymorphism in the markers helps to characterize large number of materials with limited SSR markers. This work will be a foundation particularly in Ethiopia which contributes 67% of the African bamboo area coverage to develop strategies for characterization, and conservation of bamboo genetic resource using molecular approaches. The higher level of rare alleles across both exotic and Ethiopian bamboo species plays an important role in further study of these alleles in detail. Such markers will also guide the implementation of various conservation decisions that will be necessary to ensure that these rare alleles are not lost.

REFERENCES