Review of Literature

The word ‘species’ literally means outward or visible form. It comprises groups of actually or potentially interbreeding natural populations which are reproductively isolated from other such groups (Mayr, 1940). The Linnaean concept of species as relatively constant unit with most of the variations occurring among them is different from Darwin’s theory of evolution by gradual change, which states that the variations between species must be generated from variation within species. The historical review of the species and their varieties continued to the end of 19th century and it was during this period Mendel's concept of heredity made its appearance. Later, as the integration of Mendelian genetics and Darwin's evolutionary theory, often termed as the 'neo-Darwinism' was making its impact on biology, the significance of genetic variations within species invited wide attention. Studies on intraspecific (genetic) variations were initiated and further elaborated by many workers. During 1920s, Vavilov based on his epoch making observations on genetic diversity of cultivated plants and their wild relatives described what are known today as the geographical centers of genetic diversity/origin mostly in the tropical belt. Though these earlier workers described vast wealth of previously unknown genetic variations, little attention was paid to the necessity to preserve these reservoirs of genetic diversity or 'natural gene pools' (Dodds, 1991).
2.1 Origin and maintenance of intraspecific variations in plants

The spontaneous origin of new variations in an organism is termed as mutation. De Vries (1905) used this term for new phenotypes that arose abruptly in a stock of plants. At the most basal level, mutations may affect single locus (point mutations) generating new genes able to produce new individuals (Buss, 1987; Smith, 1989). Later there can be rearrangements of genes by inversions or translocations or even by recombination at the time of sexual reproduction giving rise to novel multi-locus gene combinations.

Approximately a gene consists of thousand to several thousand nucleotide pairs (Hartman and Suskind, 1965; Mays, 1981). This complexity may be one of the reasons for generating multiple alleles, as against the Mendelian concept of two alternative states of the gene. The multiple alleles can be visualised at isozyme or DNA level and are actually different variants in base sequence of DNA. According to Smith (1989), basic mutation rates in the order of 10-9-10-10 per base position in the DNA molecules per cell cycle result in mutation rates of 10-6-10-7 per locus (about 100-1000 bases) and even in higher rates for quantitative characters affected by genes at several loci. In highly inbred plants, the accumulation of mutant alleles can be as high as 0.5 per generation per individual (Charlesworth et al., 1990).

It was demonstrated that any gene has a low but measurable ‘mutation rate’ by which the particular ‘wild type’ allele can change to a mutant allele (Briggs and Walters, 1984). The extent of this change can be different in various situations. The causative factors may be external (eg. high temperature, chemicals or high-energy radiation) or internal environment (eg. other genes (Bachmann and Low, 1980), genome instability (Schmid, 1982; Kenton et al., 1988) or other physiological instabilities (eg. seed age (D’Amato and Hoffmann-Ostenhof, 1956; Meins, 1983; Levin, 1990). Spontaneous and induced mutations are also producing deleterious mutants, but without any use in the improvement of populations as they cannot compete with the normal wild type genotype.

The genetic structure of a population depends on allele frequencies at different loci and on their effects. Generation of genetic variation within a species by mutations is a continuous process and it may affect overall genetic structure of the species. But there are some other factors that affect allele frequencies in a random mating population. They are random genetic drift, gene flow and natural selection. Natural populations are often finite and small in number, and random genetic drift refers to the chance fluctuations in allele frequencies in each population. In such populations, the probability of extinction for any mutant allele is particularly high (Spiess, 1989). In many species, the population is a network of sub-populations with intermittent gene flow (migration) between adjacent sub-populations. Hence overall genetic variation in a species/population is primarily due to the dynamic balance between genetic drift and gene flow. Another important factor affecting allele frequency especially in the case of fitness-related characters is natural selection (Endler, 1986). Selection can stabilize phenotypic characters in such a way that the genetic variation within a group is reduced and among groups it is increased. Thus, any genetic variation that arises as a result of mutation changing allele frequencies will further be under the positive or negative control of factors such as drift, gene flow and selection, which contribute to the overall evolutionary processes.
2.2. Analysis of genetic variations of plants below species level

According to Frankel (1983), an essential prerequisite for a species to survive against environmental pressures is the availability of a pool of genetic diversity and in the absence of that extinction would appear inevitable. Determining how much genetic diversity exists in a species and explaining this diversity in terms of its origin, organization and maintenance are thus of fundamental significance in the application of genetic principles to conservation. Moreover, while assessing genetic diversity it is essential to have a quantitative measure of the hierarchy of organisms as genes, genotypes, populations and species. This is often based on characterization of amount and distribution of genetic diversity in the hierarchy, i.e., the population genetic structure, which is the most fundamental piece of information for species that require genetic management (Brown, 1978).

Biodiversity consists of the variety of morphology, behavior, physiology, and biochemistry in living things. Underlying this phenotypic diversity is a diversity of genetic blueprints, nucleic acids that specify phenotypes and direct their development. Although some argue that the gene is the only true unit of selection (Dawkins, 1982) and that the species is the only ‘real’ taxon (Ghiselin, 1975), all levels of classification such as genes, chromosomes, cells, organisms, local populations, races, species, and higher taxa are important in evolution (Mallet, 1996). Genetic diversity can be measured at any of these functional classes and also is possible to measure it at any level from genotype to phenotype; but as a rule variation arising at the most genetic level is strictly heritable and do not alter under cultivated conditions. Therefore in general, a great deal of information on morphology, phytochemistry, physiology and genetics is necessary before an observed pattern of diversity is interpreted.

Morphological variation

Even though morphological variation may be related to genetic variation, the two do not correlate perfectly. Much morphological variation depends on seasonal or developmental changes that affect many individuals in a population regardless of genotype. For example, domesticated populations of black caraway (Bunium persicum) showed a large heterozygocity in the population for seed yield and tuber weight per plant apparently due to the differential performance of plants belonging to different age groups (Kapila et al., 1997). Interestingly, there are ‘plastic’ phenotypes in which growth form depends on the environment (Stearns, 1989; Schmid, 1992). On the other hand, the extensive variation exhibited by many Piper species in their leaves, flowering habit and fruiting spikes are not formed by environmental influence but through genetic divergence (Ravindran et al, 1990).

The phenotypic correlation between any two characters is the net result of both genetic and environmental correlation between these characters (Falconer, 1989). Changes in phenotypic correlation between characters will result when the change in environment produces different types of plastic responses by characters. The manner in which changes in correlation structure across environments affect fitness, and alter the intensity of and response to selection could have a significant impact on the evolutionary potential of populations (Schlichting, 1989).

Genetically determined morphological diversity can be continuous or discrete; and may be coded by many or few genes together with some input from the environment (Mallet, 1996). Genes affecting morphology are among the most interesting because morphology is often under strong selection. In many real cases of adaptation, small number of loci have major effects on traits such as resistance to chemicals, diseases, parasites etc. (Macnair, 1991), although there are good theoretical reasons for expecting that most adaptation should be gradual and polygenic (Fisher, 1958; Barton and Turelli, 1989).

Complex or continuous traits cannot be studied as easily as discrete differences inherited at single genes. The method of quantitative genetics allows a partitioning of the total phenotypic variance of a quantitative trait consisting of additive genetic variance (VG) plus environmental and non-additive genetic variance (VE), i.e. VP=VG+VE, to give an estimate of heritability, h2= VG/VP, which varies between 0 and 1. Heritabilities can be estimated from phenotypic correlation between relatives (Falconer, 1989). Heritability is useful because it predicts the approximate success of selection on any continuous trait. Provided that the environment (VE) remains relatively constant, and that many independent genes, each of individually small effect, act on the selected trait.

Variation in secondary metabolites

Variation in medicinal plants is often noticed at chemical level which is due to the synthesis and accumulation of a wide variety of biochemicals that are often plant-specific. These compounds collectively grouped as secondary metabolites are ‘high-value low–volume’ compounds biosynthetically derived from primary metabolism which help to defend, tolerate, adapt and adjust themselves against abiotic and biotic stresses including insect pests and fungal and other pathogenic diseases. Some of these agents can also act within the human body against microorganisms and other causes of disease, and represent an important source of natural drugs (Van-Seters, 1997). These phytochemical constituents may also reflect the genetic diversity or epigenetic responses or both.

In many experiments in the early 1960s, there were fruitful attempts to identify the secondary chemical compounds separated by various techniques and inferences were made to explore plant variation. Stebbins et al. (1963) found chromatographic evidence in support of the view that the tetraploid Viola quercetorum is a polyploid hybrid derivative of the cross between the diploid taxa V. purpurea sub-sp. purpurea and V. aurea sub-sp. mohavensis. Bose and Frost (1967), by thin layer chromatography, detected the variation in phenolic compounds in Galeopsis pubescens, G. speciosa and their presumed polyploid derivative G. tetrahit. In these studies, no attempts were made to locate secondary plant compounds that have medicinal rather than systematic importance.

It was interesting to note that the same medicinally active principle can be met within two species belonging to different genera. During recent phytochemical studies a small tree known as Nothapodytes foetida (Wight) Sleumer (Icacinaceae), growing in the evergreen forest areas of Western Ghats was found to contain the antiduodenal cancer compound Camptothecin. Earlier this compound was found to occur in Camptotheca acuminata Decne. (Nyssaceae), a plant native to China and Tibet, but only up to 0.0005%, whereas N. foetida contains about 0.1% and has become the promising source, for the treatment of intestinal cancer (Subramanian and Sasidharan, 1997).

Genetically determined phytochemical variations would not change under cultivated conditions. There are numerous examples. Out of the 28 genotypes of Senna (Cassia angustifolia) screened for genetic variation and strain selection, one recorded highest accumulation of sennosides A and B in the leaves (3.51% ~ 4 times more) compared to 0.89% in the control (Singh et al., 1998). The essential oils that are commercially harvested in a number of broad-leaved forest tree species such as eucalypts (Eucalyptus sp.) and melaleucas (Melaleuca alternifolia and M. linariifolia) are terpenes (Weiss, 1997). Extracting from leaves, the desired trait for medicinal oils in eucalypts is high concentration of 1,8-cineole whereas in melaleucas it is high concentration of terpin-4-ol levels. There is good evidence that variation in oil qualitative traits is genetically determined in melaleucas (Butcher et al., 1996) and in eucalypts (Doran, 1988). Within and among populations of species of Eucalyptus and Melaleuca there can be chemotypes or chemoraces which differ in proportions of terpenes (Butcher et al., 1994). To some extent these could be thought of as genetic markers but clearly a number of genes are involved.

Phytochemical variation is also possible among populations, local races etc. and among different organs of the same plant. The germplasm collection of Cymbopogon flexuous organised at Tropical Botanic Garden and Research Institute is reported to be composed of a number of chemical variants showing significant differences in essential oil content and quality (Anonymous, 1996a). Bajpai et al. (1998) reported an interesting study in which percentage of morphine in leaves and peduncles of opium poppy was about 8 to 12 fold lower than that in capsule. Instances of environment-induced variations in the curative properties of a medicinal plant and preferential collection of such drugs only from certain localities are many. It is known for years that Glycirrhyza glabra cultivated in Spain and Afghanistan has better liquorice content than that cultivated in the northern belts of India (Seeni et al., 1998). An array of chemo-variants is also reported in Savi or the Lesser Calamint (Calamintha nepeta L.). A profile of essential oils revealed the presence of chemotypes rich in piperitone oxide, mixed pulegone-piperitone oxide and menthone in this species (Thoppil, 1997). In a significant study, analysis of the major active constituents of Atractylodes lancea from China showed that not only plants growing in different geographical areas with different morphological characteristics could have different chemical constituents but also plants with similar morphological features and growing on the same site may have different contents of chemical constituents (He and Sheng, 1997).

Physiological changes

Many examples of physiological plasticity are also reported in plants. A particular barley variant produced albino phenotypes out of doors, yet the same genotype grown at higher temperatures in a glasshouse has normal foliage (Collins, 1927). In Lotus corniculatus certain plants known to possess alleles responsible for the production of hydrocyanic acid (HCN) when their foliage is crushed, do not in fact produce the cyanogenic reaction in every circumstance (Dawson, 1941).

Many plants, especially facultative CAM (Crassulacean acid metabolism) plants (Osmond, 1978) fix atmospheric CO2 through different pathways, depending on the availability of light in the environment. The plasticity is attributed to possible difference in the intercellular expression of principal photosynthetic enzymes. This is met within a number of species including Eleocharis vivipara, the only non-succulent species discovered which also has a habitat-dependent shift in photosynthetic pathways (Ueno et al., 1988) and Flaveria browni, a terrestrial species exhibiting C4 (or Hatch-Slack) pathway (Cheng et al., 1989).

Perusal of literature also reveals that leaf photosynthetic characteristics are dependent upon the incident light received during growth (Boardman, 1977). Studies with leaves of maize and amaranthus indicate that high growth irradiance increases the activities of several key C4 enzymes, and changes both the maximum photosynthetic rates and the light-saturation characteristics of leaves (Hatch et al., 1969).

As Briggs and Walters (1984) put it, different plants are capable of different responses to the same environment and show different degrees of ‘stability’ in different environments. This is particularly evident from descriptive experiments in which genotypes are reared under several controlled conditions. For example, Dwivedi and Mishra (1997) found out that different varieties of wheat responded with changes in zinc and iron composition on chlorophyll content and enzyme activities to different degrees. Maliwal (1997) reported differential tolerance of wheat varieties to the salinity and their ions absorption. This study holds some significance since about 12 million hectares of land in India are affected by saline and alkaline conditions (Yadav and Gupta, 1984).

There can be underlying reasons for most of the visible or other kind of plant variations. For example, immediate and dramatic changes in several aspects related to photosynthesis have been noticed in many crop plants due to polyploidy particularly of interrelated effects on structural, behavioural and physiological elements (Warner et al., 1987; Warner and Edwards, 1988). Neena and Srivastava (1994) reported the biochemical aspect of photosynthesis in Brassica. A considerable enhancement in CO2 fixation and enzyme activity was noticed when expression was made on cellular basis from seedling till flowering stages. Recently, Neena and Srivastava (1997) estimated the electron transport activity of diploids and allotetraploids of Brassica by measurements of the photosystem-I (PS-I) and photosystem-II (PS-II) activity in isolated thylakoid preparations from the leaves as a function of their chlorophyll content. For both diploids and allotetraploids of Brassica, the oxygen evolution rates increased with age up to first anthesis, thereafter it declined at 50% flowering in all the cases. Highest value for PS-I activity on chlorophyll basis was reported for
B. juncea, while B. oleracea vars' capita and botrytis displayed the poorest rate of electron transport activity. But there appeared to be no trend or indication of increasing or decreasing rates of electron transport as chromosome number changed. On cellular basis, PS-I activity increased with increase in ploidy. B. juncea had PS-I activity more than the sum of the activities of its parents indicating that ploidy has more than additive effect. Bansal and Abrol (1990) reported that in wheat also there is a rise in electron transport activity with an increase in ploidy from diploid to tetra or hexaploid.

Karyotypic differences

Botanists have been studying the karyotype of plants for more than 70 years and the results of their work have been published in a number of books and scientific journals (e.g. Flowering plants in general: Darlington and Wylie, 1961; Arctic flora: Love and Love, 1975; Flowering plants of Indian subcontinent: Kumar and Subramanyam, 1985; Variation and Biology of Chromosomes: Kew Chromosome Conference – Brangham (Ed), 1988). An examination of the available information makes possible some important generalizations. Chromosome number variations, especially polyploidy or having more than a pair of each type of chromosome is rare in animals because it would disrupt the mechanism of dosage compensation that normally inactivates one X chromosome in females (Orr, 1990). However, there is a recent report on polyploidy in animals. The genetic evidence indicates that the red viscacha rat, Tympanoctomys barrerae (Octodontidae) is tetraploid (Gallardo et al., 1999). Polyploidy is also very rare in fungi and gymnosperms, but is recorded in algae and many bryophytes, and is particularly common in ferns and higher plants (White, 1978; Lewis, 1980).

Chromosome research has made extensive contributions particularly in the elucidation of the systematic relationships of many closely related species (Bocher et al., 1955; Rahn, 1957; Cartier, 1971, 1973; Zemskova, 1977; Roy, 1988), karyomorphology of various plants (Geitler and Ischermak-woess, 1962; Govindarajan and Subramanian, 1983; Roy, 1988) and cytology of species of different phytogeographic regions (Gregor, 1939; Runemark, 1969; Briggs, 1973; Sen and Sharma, 1990; Pramanik and Raychaudhuri, 1997).

Cytological investigations of Plantago sp. from different phytogeographical regions (Sen and Sharma, 1990) have revealed relative uniformity among the collections. Despite the habitat differences, the populations of one species did not show any difference in karyotype, even though they differed to some extent from the karyology reported earlier (Bocher et al., 1955; Badr and El-Kholy, 1987). Evidently such a constant karyology from different geographical regions may have certain adaptive value. Some morphologically high-contrasting individuals of P. laneolata have been observed in the Kashmir Himalayas, but no specific differences in the karyotypes of different variants were found. According to Jain (1978), the diversity distinguishing the variants has been produced by simple genetic differences but not at affecting ploidy.

In cases where chromosomes are large, an examination of karyotypic differences is often of great value in understanding the nature of plant variations particularly from the level of population to genus. Karyotypes also show differences in absolute chromosome size indicating changes in nuclear DNA in evolution (Govindarajan and Subramanian, 1983; Roy, 1988). At the same time, it is well understood that the karyotype is often under strong selection. But this is not usually because of direct effects on the phenotype but because rearrangements affect the mechanism of meiosis and recombination (Dobzhansky, 1937; White, 1978; King, 1993). These changes often lead to unbalanced gametes and reduced fertility, which have great evolutionary significances (Mallet, 1996).

Isozyme variation

The advent of isozymes as genetic markers in the early 1970’s heralded a great advance for genetic studies of plant populations, since only morphological and in some cases cytological markers were available up to that time. Suddenly 20-30 loci per species were available that enabled estimation of genetic diversity and mating system parameters to be made for populations of many species. Even today isozyme markers are the best tools to answer many research questions in analysis of genetic variations.

Markert and Moller coined the term ‘isozyme’ in 1959 to describe the multiple molecular forms of enzymes that exhibit the same enzymatic specificity. In one of the earlier studies, Mallette and Dawson (1949) obtained five purified tyrosinase preparations from the common mushroom, Psalliota campestris. Additional examples were obtained when paper chromatography of brewers yeast invertase (Cabib, 1952) and rice amylase (Giri et al., 1952) gave results consistent with multiple forms.

Among the methods available to isolate and distinguish isozymes are electrophoresis, chromatography, gel filtration, catalysis, immuno-techniques, and sedimentation. Out of these, electrophoresis is often treated as the best investigative tool that has been applied successfully in many botanical disciplines, but for which comparisons among laboratories are limited by individualized and varied approaches (Carr and Johnson, 1980). Originating in the 1930’s (Tiselius, 1937), electrophoresis, coupled with the zymogram technique (Hunter and Markert, 1957), has been the tool of choice for studies of heritable variations by geneticists, systematists, and population biologists (Gottlieb, 1977; Brown, 1979; Hamrick et al., 1979; Conkle et al., 1982; Moran and Bell, 1983; Crawford, 1985; Liengsiri et al., 1990; Adams et al., 1991).

The application of isozyme markers in monitoring genetic diversity has been successfully tested and documented (Simpson and Wither, 1986; Brown and Weir, 1983; Mehta and Ali, 1996; Ranade and Sane, 1996; Dunham and Minckly, 1998). In Brassica, it was possible to distinguish 'base genetic groups' with single enzyme (esterase) alone (Arunachalam et al., 1996). In another interesting study on Phyllanthus emblica, four spatially separated populations showed considerable genetic variations as evident from diversity index values (Uma Shankar and Ganeshaiah, 1997). The concurrence of genetic diversity changes between pre-harvest and post-harvest gene pools of two adjacent virgin, old-growth (~250 years) stands of Pinus strobes in Canada analysed at 54 isozyme loci suggests that real and repeatable genetic erosion occurred as a result of harvesting (Buchert et al., 1997).

Preservation of biological diversity at the most basal level entails maintenance of genetic variation (Frankel and Soule, 1981; Schoenewald–Cox et al., 1983). Differing patterns in magnitude and distribution of genetic variation among taxa demand different conservation measures (Templeton et al., 1991). For example, Delphinium bolosii C.Blanche and Molero (Ranunculaceae), a rhizomatous, tall perennial larkspur is known from few populations only. Studies of Bosch et al. (1998) resulted in the identification of low levels of genetic variation (FST (Nei, 1978) 0.026, mean alleles per locus with order of 1.6 - 1.7 and polymorphic loci ranging from 33.3 to 50.0%) and low heterozygocity, together with some disturbed pollinator activity and declining population size neatly confirming the endangered character of D. bolosii. Wickneswari and Norwati (1993) analysed 18 populations of Acacia auriculiformis A. Cunn. Ex Benth and detected 39.8% polymorphism inferred from 12 loci. Average and effective numbers of alleles per locus were 1.5 and 1.1 respectively. Mean expected heterozygocity was 0.081 and genetic differentiation between populations was high (GST=0.270). The study demonstrated that both intra- and interpopulation genetic variations are important in A. auriculiformis improvement programmes or conservation and management of the maternal populations. Similar studies were conducted in a number of taxa that revealed threatened or otherwise significant populations for conservation or management of genetic resources (Waller et al., 1987; Lesica et al., 1988; Schaal et al., 1991; Soltis and Soltis, 1991; Soltis et al., 1992; Godt et al., 1995; Ge et al., 1997; Petit et al., 1998; van Wyk et al., 2001).

Correlation between geographic and genetic distances was revealed by allozyme electrophoresis in a number of organisms. Peterson and Heanery (1993) found significant correlation between genetic and historical geographic distances for two species of Philippine fruit bats. Schmitt et al. (1995) found significant correlation between genetic distances and geographic distances among the Indonesian island populations of Cynopterus nusatenggara. The genetic analysis of Egyptian Rousette (Rousettus egyptiacus) suggests that geographical isolation has played a role in allozyme differentiation among the populations (Juste et al., 1996). Studies of Wickneswari and Norwati (1993) suggested three distinct clusters of populations corresponding to the geographic distribution of Acacia auriculiformis in the Northern Territory and Queensland, Australia and Papua New Guinea using Nei’s unbiased genetic distance (Nei, 1978) between populations grouped by UPGMA cluster analysis.

Variation at DNA level

The ability to investigate DNA sequences directly became available to population biologists only during the late 1970s. Currently, three major DNA-based techniques have been widely used for analysing the genetic diversity in natural populations. These include (i) restriction fragment length polymorphism (RFLP; Botstein et al., 1980), (ii) polymerase chain reaction (PCR; Mullis and Faloona, 1987), and its derivatives, termed random amplified polymorphic DNA (RAPD) (Williams et al., 1990); AP-PCR (Welsh and McClelland, 1990) and (iii) a hybrid of both the above techniques named amplification fragment length polymorphism (AFLP; Vos et al., 1995).

RFLP markers

RFLPs were the original DNA markers developed in the late 1970s (Botstein et al, 1980). Their development was facilitated by the discovery of restriction enzymes which cut DNA at specific sequences giving rise to restriction fragments.

In the case of plants, RFLPs have been widely employed in determining linkage maps, in identifying genes linked to characters of agronomic importance and in mapping the quantitative trait loci (QTL) (Young et al., 1988; Paterson et al. 1988, 1990). The RFLPs are generally not very informative in identification of genotypes within random mating or non-pedigreed populations, since the RFLP probes are invariably specific to a single or few genetic loci. Hence it is of limited application in population genetics and biodiversity studies. However, in a few instances, RFLP studies have been employed to determine geographical or even ecological basis for genetic variation. In 1993, King et al. estimated genetic diversity among ecotypes of Arabidopsis thaliana. The mean genetic similarity between the 28 ecotypes was 0.69 ranging from 0.32 to 1.00, while no relationship was observed between genetic similarity and geographical origin of the ecotypes.

The percentage of nuclear clones that produce RFLPs in plants varies widely depending on the source of the probes, the range of diversity encompassed by the accessions studied, and the number of enzyme-probe combinations used (Landry et al., 1987; McCouch et al., 1988; Miller and Tanksley, 1990; Nodari et al., 1992). In a study conducted by Brubaker and Wendel (1994), 51% of probes revealed polymorphisms in Gossypium hirsutum. This is roughly equivalent to percentages obtained in Phaseolus vulgaris (Nodari et al., 1992) under similar conditions.

RAPD markers

RAPD markers are a modification of PCR contrived in the late 1980’s (Wiliams et al., 1990). PCR provides a means by which billions of copies of a particular target DNA fragment can be made from a complex mixture of genomic DNA. Now it is becoming more powerful with the introduction of user-friendly and fully automated techniques (Kreader et al., 2001).

The technical ease of RAPD markers and the facility of their application to new species has led their employment in many organisms including forest trees, crop as well as medicinal plants and lower plants for genetic linkage mapping (e.g., Carlson et al., 1991; Tulsieram et al., 1992; Grattapaglia et al., 1994; Nelson et al., 1994), phylogeny and systematics (Caetano-Anolles et al., 1991; Kaemmer et al., 1992; Wilde et al., 1992; Joshi and Nguyen, 1993; Samec, 1993; Caetano-Anolles, 1994) and population genetics (e.g., Heusden and Bachmann, 1992; Chalmers et al., 1992; Kazan et al., 1993; Mosseler et al., 1992; Chalmers et al., 1994; Isabel et al., 1995; Bucci and Menozzi, 1995; Nesbitt et al., 1995; Ranade, 1995; Tuskan et al., 1996; Schierenbeck et al., 1997; Inglis et al., 2001). The technique is one of the best available DNA-based tools for scoring variations between cultivars within species (Lakshmikumaran and Bhatia, 1998). One probable disadvantage, however, is the degree of reproducibility of these markers which can sometimes be low (e.g., Muralidharan and Wakeland, 1993; Ellsworth et al., 1993; Skroch and Nienhis, 1995) particularly between laboratories (Penner et al., 1993; Jones et al., 1997). This is due to the sensitivity of RAPD banding patterns to reaction conditions, and the difficulty in exactly replicating reaction conditions across laboratories, where different brands of thermocyclers may be used.

RAPD analysis has been widely employed for assessing the genetic diversity and relationships in many plants. In an attempt to identify pigeonpea (Cajanus cajan L) cultivars and their wild relatives, Ratnaparkhe et al (1995) successfully used this technique. In coffee (Coffea arabica) the differences in morphology and geographical origin of the genotypes was reflected in the RAPD patterns (Orozco-Castillo et al., 1994). The procedure was also used to establish genetic diversity of potato genotypes including siblings and genotypes with no immediate relationship (Demeke et al., 1996). The genetic relationships between members of Fagaceae family was assessed by RAPDs which ascertained the taxonomic studies of a particular population of Fagus sylvatica, the ‘tortuosa’ variety (Gallois et al., 1998). Rajasegar et al. (1997) have also demonstrated usefulness of RAPD analysis for cultivar development in Ixora, a group of tropical ornamental plants. In another study, clone identification and fingerprint analysis was demonstrated for Picea abies by Scheepers et al. (1997) and a single primer was used to distinguish all the 100 different clones. On the other hand, the study in case of Eucalyptus genotypes with RAPD markers revealed the power of the technique in resolving ambiguities in sampling and genotype identification (Keil and Griffin, 1994). The technique can also effectively be used to develop strain-specific SCAR (sequence-characterised amplified region) marker (Hermosa et al., 2001). The potential use of RAPD technique to study the genetic diversity in Brassica juncea and its relationship to heterosis was successfully demonstrated by Jain et al. (1994). Their studies showed that genetic diversity forms a very useful guide for the selection of parents for heterotic hybrid combinations.

AFLP markers

These fairly recent genetic markers born after combining positive qualities of both RFLP and PCR techniques are named as amplification fragment length polymorphism (AFLPs – Vos et al., 1995). AFLPs have shown high degree of reproducibility in contrast to RAPDs (Akerman et al., 1996), and were shown, again in contrast to RAPDs, to be highly reproducible across laboratories (Jones et al., 1997). Due to these advantages, AFLP is now increasingly used in a number of species including many wild plant species (Cervera et al., 1996; Akerman et al., 1996; Beismann et al., 1997; Gaiotto et al., 1997; Remington et al., 1988). Escaravage et al. (1998) have demonstrated utility of AFLP markers in determining genotypes and clonal diversity in Rhododendron ferrugineum L. (Ericaceae). They are also successfully used to characterize 67 different grapevine accessions in Spain (Cervera et al., 1998). In India, AFLP markers have been used in the assessment of genetic diversity in 37 neem accessions from different eco-geographic regions of India and four exotic lines from Thailand (Singh et al, 1999). The study indicated that neem germplasm within India constitutes a broad genetic base.

Other DNA markers

All DNA markers other than RFLPs are based in some way or other upon the PCR. After the advent of cycle sequencing methodology (Murray, 1989), direct sequencing of PCR products became a routine matter at least in organelle DNA loci or repetitive nuclear DNA such as ribosomal DNAs (Savard et al., 1994). This technology is considered to be one of the most powerful method for phylogenetic studies (Nei, 1987). In another method, designated as PCR-RFLP, the PCR products are digested with restriction enzymes usually having 4 base recognition sequences. The resultant products are analysed by agarose or polyacrylamide gel electrophoresis. This method is used successfully in many forest tree species (see Neale and Harry, 1994; Tsumura et al., 1997, 1998).

There are a few cheaper options available compared to DNA sequencing for the detection of nearly any sequence differences in PCR products. For example, SSCP (Single Strand Confirmation Polymorphism) analysis (Hayashi, 1992) allows, with minimal manipulation, the detection of greater than 90% of the possible single-nucleotide substitutions within PCR products (or restriction fragments of PCR products) of ?200 bases in length. There are several reports in plants employing PCR-SSCP analysis (Watano et al., 1995; Dumolin-Lapegne et al., 1996; Bodenes et al., 1996; Bodenes et al., 1997; Quijado et al., 1997; Ujino et al., 1998). Watano et al. (1995) has reported utility of this technique to detect chloroplast variation within an intergenic spacer region as compared to restriction analysis in species and hybrid identification in the Pinus pumila – P. parviflora complex. DGGE (Denaturing Gradient Gel Electrophoresis) provides yet another means by which sequence differences in PCR products can be detected (Myers et al., 1987; Sheffield et al., 1989). PCR-DGGE, though technically more demanding to perform than PCR-SSCP, has shown promise in its first application in forest trees (Temesgen et al., 1988).

Another powerful derivation of PCR technology is microsatellite markers. Microsatellites or simple sequence repeats (SSRs) are short (1-5 bp long) tandemly repeated DNA sequences. These were first detected in humans (Weber and May, 1989) and were later found to be abundant in plants as well (Morgante, 1993). The high degree of polymorphism and codominance of microsatellites makes them extremely informative from a genetic point of view. Hence, they find wide application in DNA sequencing especially in paternity disputes, analysis of mating systems, intraspecific variations and patterns of gene flow in plants.

Microsatellites have recently been developed in many plants including radiata pine (Smith and Devey, 1994), eastern white pine (Echt et al., 1996), oaks (Dow et al., 1995), Norway spruce (Pfeiffer et al., 1997) and several tropical trees (Teranchi, 1994; Chase et al., 1996; White and Powell, 1997; Dawson et al., 1997; Steinke Ilner et al., 1997). A study conducted by Sefc et al. (1998) determined the genetic profiles of 52 grapevine cultivars using 32 microsatellite markers. It was possible to define the complex genetic relationships among nine European grapevine cultivars. Huang et al. (1998) have isolated and characterised microsatellite DNA markers from Actinidia chinensis and proved high degree of polymorphism in this taxa. Ramakrishna et al. (1994) demonstrated use of microsatellite- derived DNA fingerprint patterns in identification of rice genotypes. This was one of the initial attempts for a systematic study with reference to the inter- and intraspecific variability in plants using simple oligonucleotide probes.

Minisatellites, on the other hand are relatively short DNA sequences (15-35 bp) repeated in tandem, dispersed throughout the genome, and associated with extensive allelic variations based on the number of core units they contain (Jeffreys et al., 1985). Many studies have demonstrated the polymorphic nature of minisatellites, which allows their use as highly informative genetic markers in individual identification, parentage testing, genetic relatedness and linkage mapping (Wolff et al., 1994; Nybom et al.,1989; Santoni et al., 1992; Poulsen et al., 1993; Rongwen et al., 1995). These multiple loci revealed by DNA fingerprinting represent a set of polymorphic markers that are usually dispersed throughout the genome (Epplen et al., 1991).

Data analyses and estimation of genetic variation

Several statistics have been used in the literature to summarize data on genetic variation, which come under two major concepts (Brown and Weir, 1983). First, there is allelic richness, or the number of distinct kinds of alleles encountered in a sample of particular size. The second component is evenness, which is related to the distribution of allelic frequencies. One of the most common measure of genetic diversity has been the equivalent of Simpson’s (1949) index of ecological diversity. This measure, or simple transforms of it, has received various names such as effective number of alleles (Kimura and Crow, 1964), heterozygocity expected under panmixia (Hubby and Lewontin, 1966), polymorphic index (Marshall and Jain, 1969) and gene diversity (Nei, 1973; Nei, 1978). While all these labels have their merits and demerits, it seems that Nei’s gene diversity is the one commonly used (Brown and Weir, 1983). It does not rely on knowledge of genotype frequencies and can be estimated from allele frequencies in terms of the expected heterozygocities within and between populations.

Genetic distances are designed to express the genetic differences between any two populations as a single number (Smith, 1977). Among the many estimates of genetic distances using allelic frequency differences among populations, Nei’s standard genetic distance, D (Nei, 1972) has been most frequently used. Based on isozyme data from a number of studies, Nei (1974) reported that species are characterized by genetic distances of 0.1 to 1.0, subspecies and varieties by 0.02 to 0.20 and races by 0.01 to 0.05.

The genetic structure of subdivided populations can be analysed by F-statistics using the correlation between uniting gametes (Wright, 1943, 1951). These provide an approximate measure of inbreeding in each subpopulation (FIS) and in the entire population (FIT), (Weir and Cockerham, 1984; Weir, 1990), and FST measures the degree of genetic differentiation among populations. The latter is always positive, between 0 and 1, and as such is used as a measure of genetic distance (Gregorius and Roberds, 1986; Long, 1986; Long et al., 1987; Weir, 1990). Nei’s distance is appropriate for long-term evolution when populations diverge because of drift and mutation. The distance is proportional to the time since divergence in the special case of the infinite alleles mutation model and equilibrium in the ancestral population. The FST distance (or Coancestry Distance) is the most appropriate for short-term evolution, for divergence due only to drift, and no assumptions need to be made about the ancestral population (Reynolds et al., 1983; Weir, 1990).

The genetic distance values can also be used for reconstruction of phylogenies that may shed light on the question, whether inferred relationships of species suggest anything about the nature of evolutionary forces? There are many reports and reviews regarding construction of phylogenies (e.g. Nei, 1987; Li and Graur, 1991; Swofford et al., 1996; Hillis et al., 1996). Phylogenetic trees may be based on distance matrices between all pairs of operational taxonomic units (OTUs). Using these distances to group OTUs in a phenetic context may employ clustering. There is a class of strategies used for finding clusters, called sequential, agglomerative, hierarchical, and non-overlapping by Sneath and Sokal (1973). The most widely used is unweighted pair group method with arithmetic averages (UPGMA).

2.3.Conservation and utilization of intraspecific variations in medicinal plants

The importance of plants as valuable sources for chemical compounds of medicinal value is well known from time immemorial. But presently due to the depletion of forest areas and other wild lands particularly in the tropical countries, medicinal plant materials are becoming scarce (Subramanian and Sasidharan, 1997; Mali and Ved, 1999). Unless there is immediate action to salvage the remaining unprotected hotspot areas, the species losses will be more than double (Pimm and Raven, 2000). Apart from habitat degradation and loss, in medicinal plants in particular, injudicious collection is yet another important reason for genetic depletion and endangerment of species.

Conservation of medicinal plants

The basic principle governing the conservation of any species is the inclusion and maintenance of maximum genetic diversity (Frankel and Soule, 1981; Wickneswari and Norwati, 1993; Stewart and Porter, 1995; Dunham and Minckley, 1998; Ananthakrishnan, 2001). Incidently, genetic diversity in plant populations is structured in a way that it reflects the biological characteristics, distribution and ecology of the species examined (Nevo et al., 1988; Nevo and Beiles, 1989; Hamrick and Godt, 1990). There are numerous reports regarding the number of individuals required for long-term survival of the species without loss of any genetic variation. Many such reports based on genetic considerations suggest that effective population size (Ne) values may be in the range of 50-500 individuals (Frankel and Soule, 1981). However, a study conducted by FRLHT (Foundation for Revitalization of Local Health Traditions) in medicinal plants suggests population size of 800 would be required for long-term survival of the species (Mali and Ved, 1999). Though these numbers give a reasonably satisfactory level of confidence regarding genetic conservation of a given species, it is also vital to consider population dynamics, which has broad applications for biodiversity and conservation in general (Hochberg and Weis, 2001). A recent study carried out in this line delineates new ways of analyzing population dynamics of natural populations (Bjornstad et al., 2001).

Of the various methods, in situ conservation, which is regarded as conservation in any habitat where the germplasm naturally occurs ensures maintenance of original allele/gene frequencies whilst permitting continued evolution (Hayward and Hamilton, 1997). If the plant species is a major constituent of a climax forest ecosystem, which does not regenerate adequately after logging and is not readily cultivated by present knowledge, the in situ method protection of the ecosystem is recommended (Subramanyam and Sasidharan, 1997). Now, conservationists are enquiring about the practical possibilities of saving most species at the least cost. In this context, it is interesting to note that as many as 44% of all species of vascular plants and 35% of all species of vertebrate groups are confined to 25 hotspots comprising only 1.4% of the land surface of the Earth
(Myers et al., 2000). This opens the way for a 'silver bullet' strategy on the part of conservation planners, focusing on these hotspots in proportion to their share of the world's species at risk.

Ex situ conservation, which essentially involves storage of genetic resources outside their natural habitat after appropriate collection, requires firstly, that they should truly represent genetic profile of the original population from which they were sampled and secondly, that the genetic integrity of these samples is conserved (Hawkes, 1987). Different methods available for storage and conservation of genetic diversity ex situ include live plant collections in botanic gardens, seed banks, meristem banks or cryobanks (Ashton, 1987). While plant germplasm organised as ex situ living collection has got specific advantages in education, display and research, they have the relative disadvantage of high maintenance costs, including high spatial requirements especially in the case of trees. Thus the most preferred method of ex situ conservation as of now is through storage as seed (Gomez-Campo, 1985, 1987). The principal advantage of seed banks is their economy of space, the larger sample size and their low labour demands (Ashton, 1987). Tissue cultures, especially meristems, can also maintain genotypes unaltered over long periods (Henshaw, 1975; Wilkins and Dodds, 1983). They also provide an economic means of suspending, at least temporarily, changes in gene frequency. Though cryogenic storage of seeds may in future provide a solution for long-term preservation of natural patterns of genetic variation within population samples in vitro, DNA libraries are probably the most stable form in which genetic information can be stored (Ashton, 1987).
Utilisation of intraspecific variations in medicinal plants

The increasing awareness in the conservation and economic utilization of biological and genetic resources is closely associated with the enforcement of international agreements such as Trade Related Aspects of Intellectual Property Rights (TRIPS) of the World Trade Organisation (WTO), the Convention of Biological Diversity (CBD) and International Convention for the Protection of New Varieties of Plants (UPOU). India, being a rich repository of plant genetic resources has recently responded to the international movement by introducing three bills, the Protection of Plant Varieties and Farmer's Rights Bill, the Patents (Second Amendment) Bill and the Biological Diversity Bill. In the amended bills the local communities can enjoy some form of compensation determined by a central authority (Cullet, 2001).

The rapid loss of biological diversity, with the extinction of 30-300 species per day (Anonymous, 1995), has initiated a new attitude towards the exploration of natural resources. Costa Rica’s Instituto Nacional de Biodiversidad (the National Biodiversity Institute, INBio) has pioneered a new concept of bioprospecting that integrates product discovery with financial and intellectual returns to ‘nature’. This model finds application to other biodiversity rich countries for simultaneous development of economic and conservation activities (Sittenfeld and Villers, 1994). Since 1991, the INBio's programme has raised more than $8 billion of which $2.5 billion has been used to support public universities and conservation areas in Costa Rica (Tyler, 2001).

In India, Department of Biotechnology has launched a Bioprospecting and Molecular Taxonomy programme which opens avenues for sustainable utilisation and bioprospecting of medicinal plant genetic resources. This has resulted in the characterization of intraspecific genetic diversity of many medicinal and other plants (Symplocos laurina, S. racemosa, Gaultheria fragrantissims, Eurya nitida, Vitex negundo, Podophyllum hexandrum, Rhododendron nilgiricum etc); identification, isolation, sequencing and cloning of a gene tolerant to extreme cold temperature from a plant species of the Spiti Valley of Himachal Pradesh; isolation of a salt tolerant gene from a mangrove species (Avicennia marina); identification and characterization of HMG CoA reductase gene from Andrographis paniculata; GIS mapping of selected plants like Taxas wallichiana and development of a species database aimed at documenting various information related to the bioprospecting programme (Anonymous, 2000).

As a result of many bioprospecting projects carried out worldwide many therapeutic compounds have been discovered. But so far only 90,000 natural compounds have been well studied which represent about 40% of total possible new drugs (Tyler, 2001). However, the increasing need for phytoceuticals as a safe alternative or an adjunct to modern medicine is seriously felt particularly due to the widely perceived side effects of the latter. For example, the sale of phytomedicine in United Kingdom has recorded a 40% increase during 1991-'96 (Ernst et al, 2001) and in US it has increased by 380% between 1990 and 1997, paralleling a reported increase in the overall use of complementary and alternative medicine from 33.8% to 42.7% during the same period (Jackson and Kanmaz, 2001). India can play a leading role in this context as it has got rich heritage of medicinal plant combined with vast storehouse of medicinal plant germplasm.

Intraspecific genetic variations can also be utilised as a possible solution to the vulnerability of many monocultured plants to disease and pests. Both theory and observations indicate that genetic heterogeneity provides greater disease suppression when used over large areas, though experimental data are lacking particularly in medicinal plants. In a recent study conducted in China, disease-susceptible rice varieties planted in mixtures with resistant varieties had 89% greater yield and blast was 94% less severe than when they were grown in monoculture (Zhu et al, 2000). In addition to disease resistance, broadly, greater genetic diversity leads to greater productivity in plant communities, greater nutrient retention in ecosystems as well as greater ecosystem stability. For instance, grassland field experiments have shown that each halving of the number of plant species within a plot leads to a 10–20% loss of productivity (Tilman, 2000).

The increase in world human population combined with altering climatic conditions, forest fires and global warming are contributing towards extinction of many vulnerable species. It is predicted that of the identified 2,50,000 higher plant species, as many as 60,000 are in danger of extinction leading to gene erosion during the next 30-40 years (Anonymous, 1994). Though most of the varieties and their wild relatives of cultivated crops and garden plants are well represented and conserved in many botanic gardens of the western countries, until recently little attention has been accorded to the protection of medicinal plants. Paradoxically, industrial/pharmaceutical utilisation of medicinal plants is largely being done by the developed countries and over 60% of all prescription drugs in Eastern Europe are proved to consist of unmodified or slightly altered higher plant products (Anonymous, 1994). What ever may be the case, conservation and characterisation of medicinal plant genetic resources is essential to ensure sustainable utilization of these neglected resources in for years to come.

2.4.Studies on the intraspecific variations in A. paniculata

Even though number of studies have been conducted to evaluate and test the medicinal properties of A. paniculata (appropriate references cited under Chapter 1), genetic variations of the species particularly using molecular markers have been a subject of least investigation. Gupta and Srivastava (1995) have investigated the diversity of the genotypes with respect to the plant morphology and andrographolide content, while Roy and Datta (1988) have used plant tissue culture techniques to evolve new varieties of the species with enhanced andrographolide content. Recently Padmesh et al. (1998) from this laboratory have successfully used the RAPD technique to assess genetic variability in A. paniculata which is possibly the only available report on the application of PCR based technique in this species. Considering the origin, diversity and medicinal importance of the species, understanding of the intraspecific variations is all the more significant in order to identify superior genotype(s) for possible development as cultivar(s) and also for appropriate utilization/conservation of the diversity-rich natural populations.