Introduction

Finger millet, Eleusine coracana L. Gaertn., is a tetraploid crop (2n = 4x = 36; genome constitution AABB) belonging to the grass family Poaceae, subfamily Chloridoideae. The crop is adapted to a wide range of environments, can withstand significant levels of salinity, is relatively resistant to water logging, and has few serious diseases. Finger millet is grown mainly by subsistence farmers and serves as a food security crop because of its high-nutritional value and excellent storage qualities. Genetic research in finger millet has been limited to studying the mode of inheritance of a few qualitative traits reviewed by Rachie and Peters [1] and biodiversity analyses. Isozyme and DNA marker analyses have indicated that cultivated finger millet has a narrow genetic base and most likely went through a bottleneck during domestication [2–5]. As expected, variation in the wild subsp. africana was considerably higher [1, 6]. Although the finger millet germplasm pool remains largely uncharacterized, small-scale analyses of the nutritional value of seeds of wild and cultivated E. coracana lines have shown a wide variation in protein, calcium and iron content [7, 8]. Phenotypic variation for blast resistance, early vigor and other yield-related characters has also been observed.

Molecular marker technology provides information that can help to define the distinctiveness of germplasm and their ranking according to the number of close relatives and their phylogenetic position. DNA marker is a new approach based on DNA polymorphism among tested genotypes, and thus applicable to biological research. Several molecular markers viz. RFLP, RAPD [9], SSRs [10], amplified fragment length polymorphism (AFLP) and SNPs are presently available to assess the variability and diversity at molecular level [11]. Simple sequence repeat (SSR) markers or microsatellites are tandem repeats interspersed throughout the genome and can be amplified using primers that flank these regions [12]. More recently molecular markers, such as SNPs and SSRs, which are genetically linked to fragrance and to identify the nature of the locus (homozygous or heterozygous condition), and have the advantage of being inexpensive, simple, rapid and only requiring small amount of tissue, may also be useful for the rapid incorporation of the scent character into breeding lines [13]. On the other hand random amplified polymorphic DNA (RAPD) is the widely used molecular marker where DNA fragments are amplified by the polymerase chain reaction (PCR) using short (usually 10 bases in length) synthetic primers of random sequence. RAPD markers tend to estimate intra- or intergenetic distances among more distantly related individuals. Inspite of many weaknesses, it is relatively easy, speedy, high degree of polymorphisms as well as virtually inexhaustible pool of possible genetic markers make the technique advantageous over other molecular techniques [14]. Randomly amplified polymorphic DNA (RAPD) have been extensively used for the assessment of genetic diversity in a variety of plants like Saxifraga cernua [15], Zea mays [16], Ziziphus spp. [17], Saccharum and Erianthus [18], Panax quinquefolius [19], etc. Inter-simple sequence repeat (ISSR) markers are much more informative than RAPDs and have been used for the analysis of genetic diversity in Cicer sp. [20], Morus alba [21], Pisum sativum [22], Asparagus acutifolius [23], Corchorus species [24] and others. AFLP has helped unravel genetic diversity in Azadirachta indica [25], Brassica nigra [26], Ranunculus acris [27], Nicotiana attenuate [28], Brassica rapa [29], Cicer sp. [30], Z. mays [31], Cynodon [32], Glycine soja [33], Myricaria laxiflora [34], Gardenia jasminoides [35], Chimonanthus spp. [36] and others.

Another molecular technique that has proved to be useful in typing crop genotypes is sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of whole seed proteins, wherein differences seen in protein bands have been successfully used to group the genotypes. SDS-PAGE is used due to its validity and simplicity for describing genetic structure of crop germplasm, but its implication has been limited mainly to cereals due to less polymorphism in most of the legumes [37]. Seed storage proteins have been used as genetic markers obtained by electrophoresis to resolve the taxonomic and evolutionary problems of several crop plants [38, 39]. Researchers can use genetic similarity information to make decisions regarding the choice for selecting superior genotypes for improvement or to be used as parents for the development of future cultivars through hybridization.

The present study was initiated to study genetic diversity on the basis of seed protein profile and its relationship with protein content in E. coracana. RAPD and SSR markers were also used to study the genetic diversity and relatedness of 52 finger millet genotypes in relation to variation in protein content.

Materials and methods

Germplasm collection

A total of 52 genotypes were used in the present study. Seed of 52 genotypes of E. coracana (collected from different districts of Uttarakhand were obtained from Ranichauri Hill Campus G. B. Pant University of Agriculture and Technology). Protein of each sample was estimated by Kjeldhal method. The pass port data of 52 genotypes is presented in (Table 1).

Table 1 List of genotypes used in the present study and their protein content
Full size table

DNA extraction and PCR amplification

The genomic DNA of different accessions of finger millets were isolated by standard method [40] quantified and analyzed on agarose gel electrophoresis [41].

RAPD and SSR markers assay

A total of 21 random primers and 24 SSR primers were used for the polymorphism survey (Table 2). PCR amplification was performed as per the standard protocol using 50–100 ng of template DNA, 30 ng of primer (Life Tech), 0.1 mM dNTPS, 1.5 U Taq DNA polymerase (Bangalore Genei pvt. Bangalore, India), 1× PCR buffer (10 mM Tris pH 8.0, 50 mM KCl and 1.8 mM MgCl2) in a volume of 25 μl. Amplification was performed with thermal cycler (Eppendorf Germany). The standardized amplification was: Initial denaturation 95°C for 5 min followed by 40 cycles of denaturation 94°C for 1 min; Primer annealing based on T m value for 1 min; primer extension at 72°C for 2 min; and final primer extension at 72°C for 7 min. The annealing temperatures of the cycling parameter were readjusted for each microsatellite primers according to their calculated melting temperature (T m) based on the sequence composition [T m = 4°C (G + C) + 2°C (A + T) − 3°C].

Table 2 Primers sequences used in RAPD-PCR and SSR-PCR
Full size table

PCR amplified products of all the primers were subjected to gel electrophoresis using 1.8% agarose gel in 1× TAE buffer at 100 V. The fragment sizes, ranged from 0.3 to 4.0 kb were detected by comparing the amplicons with a 100 bp DNA ladder and EcoRI/HindIIIdouble digest marker (Genei Pvt., Bangalore, India) and the ethidium bromide stained gels were documented using Alpha Imager 1200™ (Alpha Innotech Corporation, USA). Duplicated independent DNA preparations for each sample were done and only major bands consistently amplified were scored.

Statistical analysis

DNA fingerprints were scored for the presence (1) or absence (0) of bands of various molecular weight sizes in the form of binary matrix. Data were analyzed to obtain Jaccard’s coefficients [42] among the genotypes by using NTSYS-pc (version 2.11 W; Exeter Biological Software, Setauket, NY, [43]. The SIMQUAL program was used to calculate the Jaccard’s coefficient, a common estimator of genetic identity and was calculated as follows: Jaccard’s coefficient = N AB/(N AB + N A + N B).

Similarity matrices were utilized to construct the UPGMA (unweighted pair- group method with arithmetic average) dendrograms. Polymorphic information content (PIC) or average heterozygosity was calculated as per the formula: PIC = 2fi (1 – fi), where ‘fi’ is the frequency of the amplified allele and ‘1 – fi’ is the frequency of null allele. Principal coordinate analysis was performed in order to highlight the resolving power of the ordination. To determine robustness of the dendrogram, the data were bootstrapped with 2000 replications along with Jaccard’s coefficient by the computer programme WINBOOT [44].

Protein profiling

All 52 genotypes of finger millet were tested for their protein profiles. Total proteins were extracted by grinding seed (50 mg) with 2% (w/v) SDS, 5% (v/v) 2- mercaptoethanol, 10% (v/v) glycerol, 0.0625 M Tris–HCl, pH 6.8 (1 ml) followed by boiling for 5 min and centrifuged at 10,000 rpm for 15 min. Total protein in the form of supernatant was collected and resuspended in 50 μl of 2× sample buffer (0.5% sodium dodecyl sulphate, 1.25% 2-β-mercaptoethanol, 0.03% bromophenol blue, 2.5% glycerol in 15 mM Tris–C1 at pH 6.8) and incubated in a dry bath at 98°C for 15 min. Approximately 25 μg of the protein sample was taken and subjected to SDS–polyacrylamide gel electrophoresis (SDS-PAGE) containing 5% stacking and 15% of resolving gels and separated based on Laemeli discontinuous buffer system (Harlow and Lane 1988). After electrophoresis on a vertical slab unit under a constant voltage of 150 V for 7 h, the gels were stained with coomassie brilliant blue R-250 (HiMedia, Mumbai, India). A medium protein marker calibration kit (Bangalore Genei, Bangalore, India) was used to estimate the molecular weight of protein bands.

Seed storage proteins (Albumin, Globulin, Prolamin and Glutelin) were also extracted and subjected for their protein profiles. Prolamins were extracted from ball-milled seed (10 g), which was defatted with chloroform (2 × 100 ml) and air-dried. Albumins and globulins were extracted by stirring with 1 M NaCl (2 × 100 ml) for 1 h and centrifuged (10,000×g for 15 min), the supernatant solutions were dialysed and freeze-dried. The pellet was washed with water and prolamins extracted with 70% (v/v) aqueous ethanol (2 × 100 ml for 1 h each), followed by 50% (v/v) aqueous propan-1-ol, 2% (v/v) acetic acid and 2% (v/v) 2-mercaptoethanol (100 ml for 1 h). The respective supernatants were dialysed in a low Mr cutoff membrane (Spectra/Por 3, Pierce and Warriner) and freeze-dried. Glutelin-alkali soluble fraction, the insoluble residue obtained after the above extraction was extracted with 20 ml of 0.2% NaOH [45]. Proteins were analysed on 15% (w/v) acrylamide SDS–PAGE gels, based on the system of Laemmli [46].

Results

RAPD analysis

In this study, RAPD-PCR fingerprints were generated for 52 genotypes of finger millet. Eighteen randomly designed 10-mer oligonucleotide primers were initially used for screening DNA samples to obtain reproducible RAPD fingerprints. RAPD-PCR was run thrice to evaluate and check for the repeatability of the fingerprints generated. Out of the 21 primers tested, only 20 primers provided consistent well resolved and reproducible band patterns and were therefore selected for further analysis.

The total number of fragments observed among the finger millet genotypes based on RAPD analysis with 20 polymorphic primers was 146. The number of scorable fragments produced per primer ranged from 3 to 15 and size of the products ranged from 100 to 3034 bp. A representative RAPD profile obtained by primer RAPD-10 is shown in (Fig. 1a). Of a total of 11 bands (0.15–2.5 kb), 7 were polymorphic (64%). Marked ‘A’ a 0.8 kb band and marked ‘B’, a 0.18 kb band, is unique to genotypes containing high protein. The similarity coefficients based on RAPD markers ranged from 0.64 to 0.999 with an average value of 0.819. The PIC values, a reflection of allele diversity and frequency among the varieties, were not uniformly higher for all the RAPD loci tested. The PIC value ranged from 0.141(RAPD-09) to 0.5 (RAPD-030) with a mean of 0.351.

Fig. 1
figure 1

a UPGMA dendrogram b Bootstrap analysis for Eleusine coracana genotypes generated by the RAPD-PCR profiles, Arrows indicate the unique bands observed in the fingerprints

Full size image

Cluster analysis of RAPD primers generated RAPD profiles separated the genotypes at an average similarity values of 73% respectively (data not shown). A dendrogram based on the similarity matrix generated with the RAPD primers is presented in Fig. 1a. The dendrograms at an average similarity value of 73% grouped all genotypes in different clusters showing high diversity in profiles. Besides, the RAPD profiles also enabled a few of the genotypes to be discriminated based on their protein content. The remaining clusters consisted of mixed genotypes. The dendrogram generated were also support bootstrap value (Fig. 1b) which indicates the accuracy of the tree.

SSR analysis

A total of 168 scorable markers were yielded by the 21 polymorphic primers with an average of 08 bands per primer. 112 (66.6%) with an average of 5.3 per primer were polymorphic. A representative fingerprint pattern generated by primer SSR, UTR-36 is shown in (Fig. 2a). Out of 11 alleles generated by this primer (size range 0.1–2.0 kb), four were monomorphic. A 0.2 kb allele ‘A’ and 0.1 kb allele ‘B’ are present in genotypes containing high protein but absent in genotypes containing low protein content. The PIC value ranged from 0.274 (SSR-10) to 0.758 (SSR-02) with a mean of 0.557. The similarity coefficients based on SSR markers ranged from 0.55 to 0.999 with an average value of 0.774. Jaccard’s pair-wise similarity coefficient values ranged from 0.255 to 0.950 with an average value of 0.602. Cluster analysis of SSR primers generated SSR profiles separated the genotypes at an average similarity values of 75% respectively (data not shown). A dendrogram based on the similarity matrix generated with the SSR primers is presented in Fig. 2a. The dendrograms at an average similarity value of 75% grouped all genotypes in different clusters showing high diversity in profiles. The dendrogram generated were also support bootstrap value (Fig. 2b) which indicates the accuracy of the tree.

Fig. 2
figure 2

a UPGMA dendrogram b Bootstrap analysis for Eleusine coracana genotypes generated by the SSR-PCR profiles, Arrows indicate the unique bands observed in the fingerprints

Full size image

Protein profiling

On 15% SDS-PAGE electrophoresis, the analysis of total seed crude protein in 52 genotypes of finger millet yielded approximately 15–25 clear and distinct polypeptide bands with molecular weights ranging from 10 to 100 kDa. The total seed crude protein banding patterns were observed to be identical for all the genotypes tested. However, in few genotypes, an additional band of 32 kDa was detected (Fig. 3a). Although, significant difference was found on the basis of comparative quantitative analysis of total seed protein content of 52 finger millet genotypes (Fig. 3b).

Fig. 3
figure 3

a Representative protein profiles of Eleusine coracana genotypes. M, protein marker; lanes 1–20, finger millet genotypes, Arrows indicate the 32 kDa unique band observed in the fingerprints, b Graphical representation of total seed protein variation in the 52 genotypes of Eleusine coracana collected from Uttarakhand

Full size image

In the previous studies calcium content of 52 finger millet genotypes was estimated using atomic absorption Spectrophotometry (AAS) [47].

The genotypes of finger millet collected from different districts of Uttarakhand grouped according to high, medium and low calcium contents, estimated by atomic absorption Spectrophotometry (AAS). These results were also supported by RAPD, SSR and cytochrome P450 gene based DNA marker profiles. In E. coracana analysis of all the three markers (RAPD, SSR and cytochrome P450 gene based markers) grouped the finger millet genotypes into three distinct clusters. The first cluster had genotypes containing low calcium (100–200 mg/100 g). Second cluster included genotypes containing high calcium (300–450 mg/100 g). Third cluster included genotypes containing medium calcium (200–300 mg/100 g).

In the present study protein content of all 52 genotypes was estimated by Kjeldhal method. When the total calcium content of 52 genotypes was compared with total crude protein content, it was found that negative correlation was present between the total calcium and protein content of 52 finger millet genotypes (Table 3).

Table 3 Relationship of seed protein vs seed calcium content in 52 genotypes of finger millets
Full size table

In terms of the total crude protein, 52 genotypes were divided into three groups (High, Medium and Low). In these three groups, the first group had 36 genotypes having protein content ranging from 10 to 14%. The protein content of first group of genotype GPHCPB-1 was found highest and three genotypes (GPHCPB-16, GPHCPB-32 and GPHCPB-43) were found with lowest protein content. In second group, there were 12 genotypes having protein content ranging between 8 and 10% and it was observed that in this group genotype GPHCPB-52 had higher protein content while two genotypes (GPHCPB-18 and GPHCPB-49) with minimum protein content. In third group, there were four genotypes with protein content below 8% and it was observed that in this group the protein content was maximum in genotype GPHCPB-12 and minimum in genotype GPHCPB-45 (Fig. 3b).

Comparative seed storage protein profiling: Seed storage protein fractions i.e. Albumin, globulin, prolamin and glutelin were separated on the basis of their solubility in their respective solvent. SDS-PAGE based seed storage proteins generated profiles showed no major difference in banding pattern of 52 finger millet genotypes (Fig. 4a) while quantitative estimation of seed storage protein fractions using Lowry method revealed that glutelin was highest followed by prolamin, globulin and albumin (Fig. 4b).

Fig. 4
figure 4

Representation of (a) qualitative and (b) quantitative protein profiles of different genotypes of Eleusine coracana; M, protein marker; lanes 1–20, finger millet genotypes of seed storage Proteins (Albumin, Globulin, Prolamin and Glutelin), Comparative graphical representation of 52 genotypes

Full size image

All the fractions (albumin, globulin, prolamin and glutelin) were divided into three groups on the basis of percentage of crude protein. In the first group, it was observed that the albumin content was maximum in genotype GPHCPB-4 and minimum in genotype GPHCPB-19, globulin content was maximum in genotype GPHCPB-23 and minimum in genotype GPHCPB-20, prolamin content was maximum in genotype GPHCPB-21 and minimum in genotype GPHCPB-1 and glutelin content was maximum in genotype GPHCPB-4 and minimum in genotype GPHCPB-20.

Similarly in the second group, the albumin content was highest in genotype GPHCPB-6 and lowest in GPHCPB-17, globulin content was highest in genotype GPHCPB-6 and lowest in GPHCPB-22, prolamin content was highest in genotype GPHCPB-22 and lowest in GPHCPB-47 and glutelin content was highest in genotype GPHCPB-28 and lowest in GPHCPB-30.

On the other hand, in the third group the albumin content was highest in genotype GPHCPB-46 and lowest in GPHCPB-45, globulin content was highest in genotype GPHCPB-46 and lowest in GPHCPB-45, prolamin content was highest in genotype GPHCPB-45 and lowest in GPHCPB-12 and glutelin content was highest in genotype GPHCPB-45 and lowest in GPHCPB-12.

Discussion

Finger millet is an excellent source of calcium (seven times more than rice) and also has good amounts of phosphorous. Among cereals, it possesses a reasonably high level of methionine, the major limiting amino acid of tropical regions, and the component least correctable by the addition of pulses to the diet [1]. Hence research efforts needs to be directed to utilize the full potential of this crop in terms of seed storage proteins containing high amount of essential amino acids. In the present study molecular characterization of 52 genotypes was carried out by using various markers to understand the genetic basis of this important character.

Molecular characterization techniques are now widely used for the categorization of genotypes on the basis of specific traits and location. In this study, we have evaluated three molecular methods, RAPD-PCR, SSR-PCR and whole seed protein profiling to differentiate genotypes on the basis of protein content. Finger millet genotypes were arbitrarily grouped into three classes viz. very low, moderate, and high in terms of protein content using standard statistical programs. Significant differences between the genotypes were observed for protein content and calcium content. Protein content as high as 14.2% [48] and as low as 5.85% [49] have been reported in finger millet. Wide variations in protein content have also been reported [6, 48]. In the present study total crude Protein content of the 52 finger millet genotypes were found to be ranged from 6.4 to 14%. Thirty-two genotypes possessed significantly higher protein content than the general mean of 10.3 g per 100 g of grain. Calcium content of the 52 finger millet genotypes ranged from 117 to 452 mg/100 g. Twenty genotypes possessed significantly higher calcium content than the overall mean of 260 mg/100 g of grain. The protein content had a negative and highly significant (P < 0.01) genotypic correlation with calcium content [3].

Molecular characterization led to the amplification of various specific bands, like 0.8 and 0.18 kb band amplified by primer RAPD-10, a 1.0 kb SSR band amplified by primer SSR-01, and 0.2 kb band amplified by primer SSR UTR-36 which are present only in genotypes containing high protein but absent in genotypes containing low protein content similarly a 0.1 kb band was amplified by SSR UTR-36 present only in genotypes containing low protein. Dendrograms generated from RAPD and SSR primers data showed similarity in relative placement of genotypes. Cluster analysis was carried out on three sets of marker profiling data based on RAPD and SSR. The results based on these DNA marker profiles analysis broadly grouped the 52 genotypes into distinct clusters showing relation on the basis of protein content. The genotypes containing high protein, medium protein and low protein grouped in different clusters.

These markers demonstrated striking genetic differentiation between pairs of finger millet varieties examined. This study reveals the average genetic variation among the finger millet varieties and emphasizes the need for stock/variety wise cultivation, conservation and propagation assisted rehabilitation and selection of the natural populations of finger millet. These varieties have expressed nearly similar characteristic features to some extent, while molecular markers revealed maximum similarities between high protein content biochemical characteristics. There have been initiatives for finger millet improvement using classical plant breeding approach for high yielding, early maturing, resistance to biotic stress, tolerance to abiotic stresses particularly cold and drought to enhance nutritional quality [55]. The prerequisite for attaining this goal involves screening of different germplasms to obtain desired traits to be utilized in making crosses [56]. The acquisition of primary information about plant genetic diversity is an important fundamental work to sustain genetic conservation i.e., in situ and ex situ for gene bank management. Consequently, exploiting the genetic diversity existing in the available germplasms could be quite beneficial to breeders in crop improvement through genome-based utilization of unexploited gene pools [8, 57] because, so far, a very small fraction of the total available collections of finger millets have been used in the national breeding programs of India [58].

The total seed protein banding patterns were observed to be identical for all the genotypes tested. However, in few genotypes, an additional band of 32 kDa was detected. It is important to note that a low level of intra-specific variation has been reported in various legumes, i.e., chickpea [50], lentil [51, 52], groundnut [53], pigeon pea [54] and black gram [50] but in the case of E. coracana, a considerable low amount of variation was also observed based on SDS-PAGE. Although, significant difference was found on the basis of comparative quantitative analysis of total seed protein content of 52 finger millet genotypes.

SDS-PAGE based seed storage proteins generated profiles showed no major difference in banding pattern of 52 finger millet genotypes while quantitative estimation of seed storage protein fractions using Lowry method revealed that glutelin was highest followed by prolamin, globulin and albumin.

In this study, although SSR-PCR, RAPD-PCR and SDS-PAGE profiles were reproducible and generated several bands, the banding patterns observed with protein profiling were almost similar and not discriminatory as observed with DNA fingerprinting. Thus, it can be concluded that RAPD-PCR and SSR-PCR which is a rapid and simple tool could be used in typing and differentiating a large number of E. coracana genotypes which could be useful in their characterization.