Assessment of genetic diversity in Berberis lycium Royle complex using RAPD markers

Genomic diversity among fifty accessions of Berberis lycium Royle complex was studied, using random amplified polymorphic DNA (RAPD) markers. Out of 80 RAPD primers, 28 were polymorphic and showed reproducible results. Total of 11,683 amplicon generated 50 accessions of B. lycium complex with 28 primers. 332 amplification products scored, 284 (85%) were polymorphic and 48 monomorphic. Maximum numbers of 21 amplification products were obtained with primers OPAP-3 and 20 products with OPB-4. Average number of 11.5 bands obtained per primer and amplificon size ranged from 100 to 4, 500 bp and after study, no primer gave single band among all accessions. Polymorphic Information


INTRODUCTION
Berberis lycium, economically and medicinally important genera (Bhakuni et al., 1968), belongs to the family Berberidaceae.B. lycium (2n = 28) Royle complex is one of the important species of this genera and interesting for molecular study due to its misclassification (Rao et al., 1998).B. lycium is being abundantly distributed in Western Ghat Himalaya, West Pakistan and Nepal.In India, it is frequently distributed in Himachal Pradesh and Uttrakhand.It is extensively used for the treatment of several human diseases (Watt, 1893;Kirtikar et al., 1933;Anonymous, 1988;Khan, 2001;Chand et al., 2007;Lahiri et al., 1967).It is used as a single plant remedy or in polyhedral formulation in organized medicine such as Ayurveda, Siddha and Unani (Khare, 2004).The plant contains major alkaloid known as berberine (Khosla, 1992;Rastogi et al., 1993), an isoquinoline alkaloid, known for its activity against cholera (Rabbani, 1996), severe diarrhea (Yamamoto et al., 1993), amoebiasis, *Corresponding author.E-mail: genevivek@rediffmail.com.and latent malaria (Ghosh et al., 1985).In the British pharmacopoeia patented, a drug made from Berberine is Orisol.
Population studies and genetic diversity studies in the family is almost non-existent.Inspite of three major revisions (Rao et al., 1998), the taxonomy of B. lycium Royle complex still remains, and utter confusion, perhaps due to difficulty in their correct identification.Due to the great variation among them, the taxonomic identification is difficult.In the present study, an attempt has been done to clarify the existing confusion and efforts have also been made clear to rearrange their taxonomic position.Ahrendt (1941) surveyed the Berberis spp.and published a detailed revision of Berberis.Recently, Rao and Hajra (1993) while treating the family for the flora of India, included 54 species of Berberis in Indian region.The conclusion of most of the mentioned workers are primarily based on previous herbarium collections and often on solitary collections scattered in different herbaria.Study of the live plants in the natural habitat is rarely attempted.
Molecular genetic diversity studies of this family were not attempted before, which raveled the neglect status and the extent of gap in the knowledge of family.Rao et al. (1998) solved the identification and taxonomy of B. lycium complex based on morphological basis.These identifications were based on extensive field studies as well as herbarium specimens.This helped extremely in solving the taxonomic problem of several speciescomplexes.It is important to use DNA based markers to study genetic diversity in the species as they are expected to reveal results that are less affected by environmental factors.
The objective of the current research is to deploy RAPD-DNA marker to study the genetic diversity of fifty accessions of B. lycium complex.

Plant materials
In the present study fifty accessions of Berberis lycium complex were collected from different parts of Uttrakand, and Himachal Pradesh, India (Table 1).Different accessions of B. lycium complex were identified on morphological basis.The gross morphological attributes were taken into consideration for the identification of taxa which include terete or sulcate nature of stem, colour of bark, colour of leaf surface, nature of inflorescence, etc.The morphologicallyclosely related plants have been identified and categorized in to four major groups; B. lycium var.lycium, B. lycium var.simlensis, B. lycium var.subfascicularis and B. lycium var.subvirescens.Some species were collected from mixed wild populations and some were maintained by local people in India.

Total genomic DNA extraction
Total genomic DNA was extracted from young leaves of each variety using CTAB method.The leaves were first ground into a fine powder in liquid nitrogen, using mortar and pestle, and then, following the steps of the protocol of Doyle and Doyle (1987) with some minor modification, DNA was extracted.A fluorometer (Hoefer DyNA Quant 200 pharmacia Biotech, USA) was used to determine the quantity and quality of the DNA.The stock DNA samples were diluted with sterile TE buffer to make a working solution of 5 ng µl -1 for use in PCR analysis.

Polymerase chain reaction (PCR) analysis
A total of eighty decamer random primers from kits AP, B, C and U from Operon Technologies (Alameda, Calif.) were used for amplification of template DNA.A standard 20 µl reaction contained 50 ng template DNA, 1.5 U Taq DNA polymerase (Bangalore genei, India) , 2 × PCR reaction buffer containing 1.5 mM MgCl2, 10 picomoles primer and 100 µmoles of each dNTPs.DNA amplification was performed in Perkin Elmer DNA thermal cycler 9700 according to Williams et al. (1990).The following thermal cycling protocol was used: (1) One cycle for 2 min at 94°C; (2) 44 cycles of 94°C for 1 min, 36°C for 1.30 min and 72°C for 1.30 min; (3) one cycle for 5 min at 72°C, followed by a soaking at 4°C.The RAPD products were separated by electrophoresis according to their molecular weight on 1.4% (w/w) agarose gels submerged in 0.5 × TBE buffer and then stained with ethidium bromide (100 µg ml -1 ) solution for 15 min.The DNAs were visualized on a UV-transilluminator and documented using the gel documentation system of AlphaImager (System and Control, India).The λ DNA digested by EcoRI and Hind III was used (Banlalore genei, India) on the gel as standard size marker.

Data analysis
PCR of each sample was repeated three times.Only reproducible and unambiguous fragment were scored as (1) for presence or (0) for absence of a band after electrophoresis.A fragment was considered polymorphic if both the presence and absence of that fragment were observed in the same species and monomorphic if it was present among all individual within a species.To reduce the possibility of comparing non-homologous bands, a positive control (an individual possessing the band to be scored) was included on each agarose-gel electrophoresis.Analysis of RAPD markers was based on the following three assumptions: (1) each RAPD marker represented a single locus comprising two alleles, a marker allele (amplified product) and a non-marker alleles (null allele); (2) RAPD marker is inherited in a dominant fashion with the marker allele dominant to the non-marker allele; (3) co-migrating bands from different populations present homologous amplified products (Allan et al., 1997;Hadrys et al., 1992).
The genetic associations among B. lycium were evaluated by calculating the jaccard similarity coefficient for pair-wise comparisons based on the proportion of shared bands (alleles) produced by primer.Similarity matrices were generated using 'Simqual subprogram, similarity coefficients were used for cluster analysis of accessions performed using the 'SHAN' sub program, dendrogram were built by the un-weighted Pair Group Method with Arithmetic average (UPGMA) Figure 3.The computer program NTSYS-pc Version 2.02 was used (Rohlf, 2000).
The polymorphic information content (PIC) was calculated by applying the formula given by Powell et al. (1996) and Smith et al. (1997): Where f i is the frequency of the i th alleles and the summation extends over n alleles.

RESULTS AND DISCUSSION
RAPD marker system has been used for the molecular characterization of B. lycium complex.A total of thirty-two accessions of B. lycium var.lycium, five accessions of B.s lycium var.simlensis, eleven accessions of B. lycium var.subfascicularis and two accessions of B. lycium var.subvirescens were included in this study.Eighty primers were used to study molecular genetic diversity.Most of the primers did not amplify with B.lycium DNA.Only 28 primers yielded scorable amplification pattern (Figure 1) and rest primers gave unreadable and smear band pattern.A total of 11,683 amplicon generated 50 accessions of B. lycium with 28 primers.332 amplification products were scored and 284 (85%) were polymorphic and the rest, monomorphic (Table 3).Maximum numbers of 21 amplification product were obtained with primer OPAP-3 and 20 products with OPB-4.Minimum numbers of 6 RAPD products were generated with primer OPAP-14.Average numbers of 11.8 bands were obtained per primer and amplification ranged from 100 bp to 4.5 kb, and after study, no primer gave single band among all 50 accessions (Table 4).Highest similarity (0.97%) was identified between BLL10 or BLL11 accessions and least (0.23%) genetic similarity in BLL10 or BLS6 (Table 2).Polymorphic information content (PIC) scores represented gene diversity for specific locus.PIC scores for the RAPD primers ranged from 0.013 to 0.52 with an average of 0.12.
The relationship amongst the B. lycium complex obtained by RAPD method differs from the mostly cited classification of B. The genetic relatedness among the B. lycium Royle complex, showed large diversity when confirmed by the principal component analysis (PCA).In Figure 2, the B. lycium cultivars were dispersed on PCA graph, which is a reflection of narrow genetic base of this genus.The results of PCA show a clear-cut separation of 50 B. lycium cultivars into five clusters.The present study concluded that molecular evidence is not supporting previous taxonomic classification of B. lycium Royle complex (Figure 3).
The evolution of varieties in distinct climatic location demonstrates significant levels of variation in response to the selection pressure in the location (Millan et al., 1996).
It is therefore not surprising to find significant levels of polymorphism among 50 accessions of B. lycium and related varieties in RAPD (85%).The success of our study in identifying polymorphism is due to the use of a number of randomly selected, prescreened, and highly informative primers.Geographically isolated population accumulates genetic differences as they adapt to different environment.All 50 accessions in this study revealed a unique profile with the 28 primers and thus can be used for the DNA-RAPD fingerprinting.Generally, a large chromosome size and more repetitive sequences provided greater chances for the primers to find homology and to give more and differently sized amplified fragment.
Several doubts have been raised regarding the suitability of RAPD for variety identification and diversity studies, the most important one being that co-migrating bands may not be allelic or composed of similar sequences (Swaboda and Bhalla, 1997).However, the homology of co migrating RAPD bands has been demonstrated in some species of Glycine and Allium (Williams et al., 1993;Wilkie et al., 1993).In addition, conformity of phylogenetic grouping, based on RAPD data to these, based on conventional approaches likemorphological and cytology analysis, is in itself indirect, but significant evidence in support of the allelism of co migrating RAPD bands.The use of a large number of polymorphic markers would minimize the skewing of result due to non-allilism.
Another problem often encountered in RAPD analysis is that of reproducibility of bands pattern between different PCR reactions.These aspects can be overcome by using a thoroughly optimized PCR protocol and by scoring only reproducible bands.The RAPD method has been employed in the past, successfully for varieties relationship in mango (Srivastava et al., 2006), cassia   (Whity et al., 1994), rice (Takeuchi, 1994), mustard (Lin et al., 1996) and soybean (FuJishiro and Sasakuma., 1994).We analyzed genetic relationship between genotypes of B. lycium varieties.It was apparent that RAPD marker is capable of differentiating between closely related varieties.Similar results have been found in case of peanut and blackgram where they could differentiate the species with RAPD (Raina et al., 2000;Souframanian et al., 2004).The present study showed that B. lycium is quite different from the rest of the varieties and did not show any close relationships.Four varieties of B. lycium did not arrange into four clusters.Within species, genetic diversity existed among the accessions but not in the case of B. lycium complex.Ahrendt (1941)  simlensis can be separated from other species by pubescent nature of the stem but this is not true as all the other varieties of B. lycium also have pubescent stem either at young or mature stage (Rao et al., 1998).This result is not supportive of previous taxonomic classification of B. lycium Royle complex proposed by Ahrendt (1941), Rao et al. (1998) and Uniyal and Rao (1993).
It was concluded from the present study that to obtain identification, tracing genetic relationships and characterization of the B. lycium accessions, the molecular approach based on RAPD profile is a powerful technique.RAPD markers amplified from few primers could identify the B. lycium cultivars.The information obtained will facilitate the choosing of appropriate breeding program, genome mapping, and tagging numerous traits of economic importance.It was also suggested that RAPD technique would be more useful for identification of cultivars and for estimating genetic relationship in B. lycium complex.The present study analyzed the genetic relationships at molecular level utilizing RAPD assay.RAPD markers have been used earlier to study taxonomic and phylogenetic relationships Demeke et al., 1992;Millan et al., 1996).Virk et al. (1995) have analyzed the germplasm collection of rice accessions by RAPD markers and classified the unclassified rice accessions as indica and japonica types.Similarly, Howell and Newburg (1994) have used RAPD for identifying and classifying Musa germplasm.Pipe et al. (1995) supported the separation of two groups of Opiostoma piceae into two species based on the clear-cut divergence revealed by RAPD.In another case, the genus Scaevola, which was initially misclassified by Linnaneus (1753), and further rearranged several times by other scientist (Bentham, 1868;Krauze, 1912;Carolin, (1992), has now been reclassified, resolving the previous confusions through RAPD analysis (Swoboda et al., 1997).Phylogenetic relationships investigated using RAPD analysis among the Rosa species accessions proved useful in assigning unclassified accessions to specific taxonomic groups (Millan et al., 1996).RAPD analysis of Tibetan wheat, common wheat and European spelt wheat supported the previous classification of Tibetan wheat as a subspecies of common wheat (Sun et al., 1998).Recently, Singh et al. (2004) classified Ocimum species using RAPD markers.
In day-to-day management of the germplasm collection, RAPD allow identification of redundancy and provide additional cultivars verification method.The genetic diversity analysis within the B. lycium germplasm collection may provide useful information for proper management and its future utilization in basic and applied studies.To our knowledge, this study is the first attempt in using molecular markers for the varietial identification and genetic diversity assessment of B. lycium accessions.This study will be helpful for the breeding, conservation and germplasm management of B. lycium.

AKNOWLEDGEMENTS
We give thanks to Dr. Anil Kumar, Dr. Bhaskar Dutta and Dr. K.N.Nair, NBRI, for their help during plant material collection and varieties identification.We also thank to Dr. Rakesh Tuli, Director NBRI, Lucknow, India, for providing necessary facilities related to this work and Department of Biotechnology, New Delhi, India, for financial support.

Figure 2 .
Figure 2. Principle co-ordinate analysis of RAPDs products from Berberis lycium complex.The accessions are separated into five groups.

Figure 3 .
Figure 3. Dendogram showing the relationship among different species of Berberis lycium complex based on UPGMA and sequential agglomerative hierarchical nested.

Table 1 .
Collected Berberis lycium Royle complex accessions, accession name and their location.
BSI*-botanical survey of India.
lycium Royle complex-reference.Cluster I included one accession of B. lycium and come out separately from rest of the species.Cluster II included 46 accessions of B. lycium var.lycium, B. lycium var.simlensis, B. lycium var.subfascicularis and B. lycium var.subvirescens.

Table 2 .
Genetic relationships among the Berberis lycium and varieties.

Table 3 .
Analysis of polymorphism among Berberis lycium complex obtained with random primers.The pair wise similarity indices RAPD band data among 50 accessions of Berberis lycium.

Table 4 .
Summary of detection of RAPD marker in Berberis lycium complex.