• Open Access

Genetic analysis of Indian mulberry varieties through molecular markers


K. Vijayan, Seri-Biotech Research Laboratory, CSB Campus, Carmelram (P.O), Sarjapur Road, Kodathi, Bangalore 560035, Karnataka, India-560 035. E-mail: kvijayan01@yahoo.com


India is one of the countries where sericulture is being practiced traditionally. Due to the higher economic return and the greater employment potential, attempts are being made to increase the productivity by developing high yielding mulberry varieties. At the present, Mysore local, Bomaypiasbari, Kanva-2, Bilidevalaya, Kajli, S1, BC259, C776, RFS-175, S-36 and Victory-1 are being cultivated extensively in different parts of India for rearing the silkworm Bombyx mori L. Using 17 random amplified polymorphic DNA (RAPD) and 11 inter-simple sequence repeat (ISSR) primers the genetic relationships among these varieties were analyzed. The RAPD and ISSR primers revealed more than 75% polymorphism among the varieties. The genetic similarity estimated from RAPD markers varied from 0.645, between Kajli and Victory-1 to 0.887, between Kanva-2 and Bilidevalaya. Similarly, the genetic similarity estimated from the ISSR markers ranged from 0.600, between Kajli and Victory-1, to 0.873 between Kanva-2 and BC259. The dendrogram constructed from these markers grouped the varieties into three major groups comprising the low yielding, medium yielding and high yielding. The low genetic similarity between the group of varieties originating from the eastern regions with that of the southern region encourages formation of extensive breeding programs between these groups as to transfer the high yield potential of the southern varieties to the low yielding but highly adaptive eastern varieties.

Mulberry is a perennial, deep-rooted, widely adaptable, fast growing tree plant belonging to the genus Morus L. under the order Urticales and the tribe Moraceae (Hooker 1885). Takhtajan (1980), while classifying the plants on the basis of their relative degree of evolutionary advancement, placed this genus under the family Moraceae belonging to the order Urticales and super order Hamamelidanae. Koidzumi (1923) identified 25 species under the genus Morus and classified them into 2 sections based on the length of the style as Dolichostyle and Macromorus. Mulberry is being extensively cultivated in sericulturally important countries of both tropical and temperate regions to feed the monophagous silkworm insect Bombyx mori L. Though mulberry is basically a tree, in sericulture it is being maintained as small bushes through repeated pruning and training. The chief mode of propagation of this highly heterozygous crop in tropical countries like India, Pakistan, Bangladesh is chiefly through stem cuttings while in temperate countries the seed is the major source of propagation. In India, the cost of mulberry leaf production is reported to be covering nearly 60% of the total expenditure of silkworm cocoon production (Das and Krishnaswami 1965). Hence, attempts have been made regularly to improve the leaf productivity per unit area. As a result, a number of mulberry varieties have recently been developed in India for its cultivation at farmers level. However, most of these varieties are suitable to specific regions rather than for a wider locality with varying climatic conditions as India experiences wide range of agro-climatic conditions (Chakraborti et al. 1999). Similarly, sericulture areas in northern, eastern and northeastern parts of India experience extreme seasonal variations in the climatic conditions. Thus, the leaf productivity of these varieties, in this part of the region, varies drastically during different seasons and the low leaf productivity during winter affects the bivoltine (rearing of superior silkworm races) sericulture severely (Chakraborti et al. 1999). Hence, it is urgently required to develop varieties with wider adaptability and higher yield potential to sustain the profitability of sericulture industry in these areas. Since prior knowledge on the genetic relationships of the breeding stock at hand is an essential factor deciding the success of any breeding programs, the present study was undertaken with an objective to unravel the phylogenetic relationship among these mulberry varieties, so that appropriate breeding strategies could be formulated to improve the genetic base of these varieties.

PCR based DNA markers such as RAPD (Williams et al. 1990) and inter-simple sequence repeat (ISSR) (Zietkiewicz et al. 1994; Prevost and Wilkinson 1999) are known to be very useful for investigating genetic relatedness and diversity in plant populations and cultivars (Yu and Nguyen 1994; He et al. 1995; Ramser et al. 1996; Tsumura et al. 1996; Karp et al. 1997; Jones et al. 1997; Ghislain et al. 1999; Deshpande et al. 2001; Li and Nelson 2001; Bornet et al. 2002). Mulberry molecular markers have been used to study the genetic relationships of cultivars differing in the ploidy (Lichun et al. 1996; Lou et al. 1998; Zhang et al. 1998; Weiguo et al. 2000; Bhattacharya and Ranade 2001). Similarly, our earlier study with 11 genetic stocks of the land races indicated that ISSR primers could be used successfully to unravel the genetic relationships in mulberry with both ease and speed (Vijayan and Chatterjee 2003). Thus, in the present investigation, RAPD and ISSR marker systems were used to study the genetic relationships among 11 selected varieties, which are grown extensively at different parts of India under divergent climatic conditions so as to formulate a breeding strategy to evolve mulberry varieties with wider adaptability and higher leaf yield potential.


Plant material

Eleven mulberry varieties namely Victory-1, C776, BC259, S1, Bilidevalaya, Kanva-2, Bombay piasbari, RFS-175, Mysore local, Kajli and S36, were selected for the study. These varieties are cultivated extensively at different agro climatic zones by farmers of India to rear the silkworm insect, Bombyx mori L. (Table 1). Some of the leaf yield attributing characters of these varieties are given in Table 1 and were taken from the catalogues published by the Central Sericultural Germplasm Resource Center, Hosur, Tamil Nadu, India (Thangavelu et al. 1997), as these data were recorded from plants of same age under similar climatic conditions. For DNA extraction young leaves from three clonally propagated plants were collected for each variety.

Table 1.  Average leaf yield characters of the 11 mulberry varieties.
VarietySpeciesRegions of cultivationLeaf lobationNo. of branchesPlant height (cm)Leaf yield (kg/plant/yr)
Mysore localMorus indicaSouthern plateauLobed13.00163.671.23
Bombay PiasbariMorus indicaGangetic plainsLobed28.002000.001.04
Kanva-2Morus indicaSouthern and central plateaus, west coast plains and Ghat regions, Gujarat plains and hills regionsUn lobed24.00151.001.77
BilidevalayaMorus indicaSouthern plateauLobed16.00141.670.93
KajliMorus indicaGangetic plains, eastern and central plateausLobed37.00131.000.63
S1Morus albaGangetic plains, eastern and central plateaus, hilly regionsUn lobed66.00183.002.38
BC259Morus indica×M. latifoliaHilly regionsUn lobed20.0138.001.06
C776M. nigra×M.multicaulisIsland regions and dry areas of Gangetic plainsUn lobed42.00163.672.16
S-36M. indicaSouthern plateau, west coast plains and Ghat regions, western dry regionsUn lobed34.00135.002.65
RFS-175M. indicaRain-fed regions of Southern plateau, West Coast plains and Ghat regionsUn lobed19.00142.001.50
Victory-1(M. nigra×M. multicaulis)×M. indicaIrrigated regions of southern and central plateaus, west coast plains, western Ghat regions and Trans gangetic plainsUn lobed11.83185.652.50

DNA extraction

Total genomic DNA was isolated from the fresh leaf using modified CTAB (cetyltrimethyl amonium bromide) protocols of Murray and Thompson (1980). About 500 mg of the leaf lamina was ground in liquid nitrogen using mortar and pestle. The leaf powder was transferred to a 50-ml polypropylene tube and incubated at 65oC for 30 min after adding 1.5 ml of 15% SDS and 5 ml extraction buffer containing 1% CTAB, 1.5% PVP (poly vinyl pyrollidone), 1.4 M NaCl, 50 mM EDTA, 50 mM Tris-HCl, 0.1% β-mercaptoethanol. After incubation, equal volume of phenol:chloroform (1:1) was added to the incubation mixture and after thorough mixing, it was centrifuged at 10 000 rpm in a Sorval centrifuge at 25oC for 15 min. To the supernatant, equal volume of phenol:chloroform: isoamyl alcohol (24:24:1) was added and centrifuged again at 10 000 rpm for 15 min. The supernatant was then transferred to a fresh tube, equal volume of chloroform was added and mixed thoroughly by inverting the tube gently for a number of time. The mixture was centrifuged at 10 000 rpm for 15 min. The DNA from the supernatant was precipitated by adding one-tenth volume of 3 M sodium acetate and two volumes of double distilled ethyl alcohol. The precipitated DNA was washed with 70% ethanol and air dried at 37oC. The DNA was later re-dissolved in TE Buffer (10 mM Tris-HCl [pH 8.0] and 1 mM EDTA). Contamination of the RNA was removed by incubating the dissolved DNA with DNAase free bovine pancreatic RNAase (10 mg ml−1) at 37oC for 1 h. The DNA was, then re-extracted following the steps stated above. Quantification of the DNA was performed by electrophoresis on 0.8% agarose gel (1×TBE) stained with ethidium bromide (0.5 μg ml−1) and using uncut λ DNA (10 ng μl−1) as standard. The PCR amplification of the genomic DNA with different primers was carried out on an MJ Research Thermal-Cycler, PTC 200 (MJ Research Inc. Watertown, Massachusetts, USA) following different cycles as follows.

PCR analysis with RAPD primers

A total of 17-selected decamer primers (OPW-1, 4, 5, 6, 7, 8, 11, 13, 17, 18, 20 and OPY-1, 6, 7, 8, 11, 13), designed by the Operon Technologies Inc USA, were used in this study. The PCR was conducted according to Williams et al. (1990) using 20 μl of reaction mixture containing 2.0 μl of 10×PCR Buffer (100 mM Tris-HCl pH 8.8; 500 mM KCl; 15 mM MgCl2; 0.1% gelatin; 0.05% Tween 20 and 0.05% NP-40), 2 mM dNTP, 2 mM MgCl2; 0.1 mM primer; 20 ng genomic DNA and 1 unit of Taq polymerase enzyme (Genetaq, Genetix, Singapore). The PCR schedule followed was 93oC for 2 min followed by 40 cycles of 93oC for 1 min, 35.5oC for 1 min, 72oC for 2 min and a final extension of 10 min at 72oC. The PCR product was separated on 1.5% agarose gel in 1×Tris boric acid buffer containing 0.5 μg ml−1 ethidium bromide as stain.

PCR amplification with ISSR primers

A total of 11 selected primers UBC-807, 808, 810, 812, 820, 825, 828, 830, 834, 861 and 881 (University of British Columbia Biotechnology Laboratory, Vancouver; Canada; Primer set 9) were used for the PCR using 20 μl of reaction mixture containing 2.0 μl of 10×PCR buffer of Genetaq, Genetix (100 mM Tris-HCl pH 8.8; 500 mM KCl; 15 mM MgCl2; 0.1% gelatin; 0.05% Tween 20 and 0.05% NP-40), 2 mM dNTP, 2 mM MgCl2; 0.1 mM primer; 50 ng genomic DNA and 1 unit of Taq polymerase enzyme (Genetaq, Genetix, Singapore). The PCR schedule followed was 94oC for 2 min followed by 35 cycles of 94oC for 30 sec, 50oC for 30 sec, 72oC for 2 min and a final extension of 10 min at 72oC. The PCR product was resolved on 2.0% agarose gel as mentioned above.

Data collection and statistical analysis

Amplification with each primer was repeated three times and clearly resolvable and reproducible fragments were considered for analysis. Each fragment was treated as a unit character and was scored as 1 (present) or 0 (absent). Pair wise genetic similarity was calculated as 2Nij/(Ni+Nj), where Nij is the number of common bands in i and j varieties and Ni+Nj is the total number of bands produced by the varieties ith and jth (Nei and Li 1979). Cluster analysis of the data was carried out using UPGMA (Sneath and Sokal 1973) method version 3.572c (PHYLIP, Felsenstein 1993) as because this method does not assume rate constancy of evolution among the varieties. The bootstrap values were calculated using WINBOOT (Yap and Nelson 1996).

The efficiency of UPGMA in estimating the genetic relationship among the varieties was tested by calculating the cophenetic values for each variety and measuring the cophenetic correlation between similarity matrix and the cophenetic values using Mantel's Z-statistics (Mantel 1967). The Mantel test was carried out using the free software developed and distributed by Liedloff (1999). Estimates of the differences between the dendrograms based on RAPD and ISSR markers were also assessed by computing the cophenetic values and constructing the cophenetic matrices for each primer set. These cophenetic matrices were compared using Mantel's test for matrix correspondence (Mantel 1967; Liedloff 1999).


It is evident from the morphological traits and the regions of cultivation that the varieties selected in this study showed considerable variations (Table 1). Varieties like Mysore local, Bilidevalaya and Kajli have highly lobed leaves while Victory-1, C776, S1, BC259, and RFS-175 have unlobed leaves. The branching pattern of these varieties also showed variations as S1, Kajli, C776 and S-36 produce profuse branching while Victory-1, Mysore local, Bilidevalaya have fewer number of branches. The yield potentials of Victory-1 and S36 were many folds higher than that of the old varieties like Bilidevalaya, Kajli. This indicates that a quantum jump has been achieved in recent times on leaf yield potential of mulberry. The adaptability of these varieties also showed considerable variations as BC259 was found suitable for hilly regions of the eastern India while C776 was a better for both salt and drought stress regions (Chakraborti et al. 2000). Similarly, Victory-1 is the best variety for irrigated plains of the Indian subcontinent (Sarkar et al. 2000) while RFS-175 and S1 are suitable for rain fed regions of both southern and eastern regions of India. Thus, considerable genetic diversity was observed on morphological characters of these varieties. The molecular analysis also showed significant DNA polymorphism among these varieties (Fig. 1) as the 17 RAPD primers generated a total of 164 markers, of which 123 were polymorphic, generating 75.00% polymorphism among the 11 varieties. Out of the total 71 ISSR markers produced by the 11 ISSR primers, 57 were polymorphic giving 80.28% polymorphism. Thus, it could be seen that both types of primers could generate a high amount (>75%) of polymorphism among the varieties. Among the ISSR primers, UBC-812 gave the maximum number of polymorphic bands, similarly among the RAPD primers, OPY-9 produced 12 bands of which 11 were polymorphic. However, the genetic similarity coefficients estimated from the RAPD markers varied from 0.877 between Kanva-2 and Blidevalaya to 0.645 between Kajli and Victory-1 with an average similarity index of 0.768. The genetic similarity from the ISSR markers varied from 0.873 between Kanva-2 and BC259 to 0.600 between Kajli and Victory-1 (Table 2 - upper half) with an average genetic similarity of 0.750. It is obvious from the genetic similarity matrices that Victory-1 and Kajli were genetically most unrelated while Kanva-2, Bilidevalaya and BC259 were the most closely related varieties being cultivated in India.

Figure 1.

Polymorphism generated by the primers (A) ISSR-825 (B) OPY-15. The arrows indicate the bands showing polymorphism. The labels are M- Marker, 1-Mysore local, 2-Bombaipiasbari, 3-Kanva-2, 4-Bilidevalaya, 5-Kajli, 6-S1, 7-BC259, 8-C776, 9-S36, 10-RFS-175, 11-Victory-1.

Table 2.  Similarity matrices among 11 mulberry varieties realized from ISSR (upper half) and RAPD markers.
VarietyMysore localB. piasbariKanva-2Bilide valayaKajliS1BC259C776S-36RFS-175Victory-1
Mysore local 0.8570.8000.8330.7930.7670.7640.8000.6400.7200.731
Bombay piasbari0.855 0.8250.8250.8200.8570.7930.7170.6790.6790.691
Kanva-20.8140.782 0.8670.6900.7330.8730.7200.7600.7600.808
Bilidevalaya0.8610.8150.877 0.7240.7670.8360.7200.6800.7600.769
Kajli0.8070.8020.7930.836 0.7930.7170.6670.6250.6250.600
S10.8110.7870.7870.8220.817 0.7640.7200.6800.6800.654
BC2590.7960.7720.8100.8440.8020.826 0.7560.8440.6670.809
C7760.7160.6950.7570.7460.7030.7500.754 0.7500.8000.810
S-360.7320.7130.7510.7220.6910.7340.8020.716 0.6500.762
RFS-1750.7190.6770.7490.7490.6960.7420.7230.8030.683 0.810

The dendrogram realized from RAPD markers through UPGMA clustered the 11 varieties into three groups and one outlier (Figs. 1 and 2A). The first group contains two varieties (Mysore local and Bombay piasbari) both are very old varieties, cultivated in the traditional sericulture zones of Karnataka and West Bengal, though the leaf yield potential of these varieties is very low. The second group comprised of five varieties. This group is a mixture of local varieties and old evolved varieties with comparatively less yield potential. The third group consists of three varieties, which were evolved recently (Sastry 1984; Sarkar et al. 1999; Chakraborti et al. 2000) and have very higher leaf yield potential. Thus, the dendrogram generated from the RAPD markers could discriminate the varieties according to their yield potential and also on the basis of their evolution as the old varieties were discriminated from the newly developed varieties. The position of S36 as an outlier with a bootstrap value of 38 indicates that this variety is genetically different from others. The highest bootstrap value (69) was present at the node of the varieties S1 and BC259 and lowest (35) was at the node where C776 joins with others. Out of the total nine nodes only four nodes were showing a value higher than 50%, which in turn indicates that the structure of the dendrogram is not very solid. This is further substantiated by the correlation coefficient (r=0.58; p=0.001), obtained between the cophenetic correlation matrix and the similarity matrix, as revealed by Mantel test (1967). However, the grouping pattern observed in this study would definitely help the breeder to select genetically divergent parents for breeding purpose.

Figure 2.

The phylogenetic relationships among 11 mulberry varieties from India analyses through UPGMA-dendrogram on the basis of (A) RAPD and (B) ISSR markers; Numbers at the nodes indicate bootstrap value (%).

The dendrogram obtained from ISSR markers also grouped the 11 mulberry varieties into 3 groups and one outlier. The first group included four traditional varieties, i.e. Mysore local, Bombay piasbari, Bilidevalaya and Kajli. In the second group varieties, with less leaf yield potential, such as S1, BC259 and Kanva-2 were clustered together. The third group contained the three recently evolved varieties with higher leaf yield potential. In this case also S36 stood as an outlier. The bootstrap values varied from 37 at the node of cluster one with cluster two to 83 at the node where C776 joins with Victory. These comparatively higher bootstrap values observed at the nodes indicate that the dendrogram obtained from ISSR markers is more reliable than the one obtained from RAPD markers. This better reproducibility of the dendrogram was further substantiated by the higher correlation coefficient value (r=0.72, p=0.001), observed between the cophenetic correlation matrix and the similarity matrix, as reveled by Mantel test (1967).


Through this study we have documented the genetic relationships among 11 mulberry varieties, which were evolved at different periods but are being cultivated at present in different parts of India. Further, this study revealed that considerable differences could be observed between results from molecular markers and those from morphological traits. In earlier studies, where only morphological traits were used for assessing the genetic relationships, most of the varieties used in this study have grouped under one or two clusters along with a few other varieties without showing any particular pattern of grouping (Mala et al. 1997; Vijayan et al. 1999). However, in this study, using molecular markers, these varieties were clustered into small groups according to their yield potential. From the dendrograms, obtained from both RAPD and ISSR markers, it is clear that the old traditional varieties with very low leaf yield were grouped together in one cluster. For example, Mysore local and Bombay piasbari, both with very low leaf yield but grown in different geographical areas were grouped together with a bootstrap value higher than 60%. Likewise, varieties with moderate leaf yield potentials (S1, BC259 and Kanva-2) were also grouped together. Similarly, the recently developed high yielding varieties like C776, RFS-175 and Victory-1 were clustered into a group with more than 50% bootstrap values. This grouping of the varieties according to the leaf yield potentials, is of much importance in mulberry breeding as it is clear from the results that through breeding with conventional methods like hybridization and selection considerable amount of genetic diversity has been created among the mulberry varieties, presently being cultivated in different parts of India. However, the poor adaptability of these high yielding varieties to the varying climatic conditions is a major point of concern, as the high yielding varieties developed in the south do not perform well in the east and vice versa. This lack of wider adaptability among the high yielding mulberry varieties could be due to tailoring down of the genetic pool of these varieties to make them better yielding. However, it is not possible for the sericulture farmers, as most of them belong to the economically weaker section, to grow different varieties for different seasons, as mulberry is a perennial plant and establishment of mulberry gardens entails huge investments. Hence, the adaptability of these varieties needs to be tackled urgently. The higher genetic variability observed between the traditional varieties like Mysore local, Bombay piasbari, Kajli, Bilidevalaya and higher leaf yielding varieties like RFS-175, C776, Victory-1 points to the possibility of transferring the adaptive nature of these traditional varieties into the high yielders through introgressive breeding.

Another important point observed in this study is the lack of relationships between geographic origin and grouping of the varieties. Varieties that originated in the eastern region joined with those originated in the south. For instance, the variety Bombay piasbari grouped with Mysore local with significantly high bootstrap value. Similarly, Kanva-2 joined with either Bilidevalaya or S1 from the eastern India. Sharma et al. (2000) and Vijayan and Chatterjee (2003) also could not observe any relationship between geographic origin and grouping among mulberry varieties.

Another point that needs special attention is the positions of S36 in the dendrograms as in both dendrograms it remained as an isolate from the rest, though with medium bootstrap values. Cytologically S36 is an aneuploid containing a somatic chromosome number of 2n=30. This variety was developed from a local mulberry variety, originated from west Bengal, through chemical mutagenesis (Sastry 1984). This variety is known to have a number of good agronomic characters like better adaptability to different climatic conditions, good leaf yield, high response to fertilizers. Since this variety is having a number of desirable agronomic characters and is genetically different from most of the presently cultivated varieties, it can be considered as a good candidate for the genetic studies in mulberry as crossing between this and other mulberry varieties like Kajli and Bilidevalaya may lead to the evolution of plants with varied chromosome numbers, which would help to find out the effect of the extra chromosome on certain characters. Another interesting finding from this study is the grouping of C776, RFS-175 and Victory-1 into one group as RFS-175 and C776 drought resistant varieties (Sastry 1984; Rahman et al. 1999; Sarkar et al. 2000). Though, we could not identify any specific marker association with only these varieties. The possibility of identifying such marker for these varieties cannot be ruled out, unless a larger number of primers is tested. Such marker identification would be of much use to the breeders to identify and select drought tolerant hybrids in mulberry at an early stage of development.


The authors express their sincere thanks to Dr. K. Thangavelu, Director, CSGRC, Hosur, Tamil Nadu, for providing the plant materials and necessary information.