Genetic Diversity
TERI University-Ph.D. Thesis, 2006
Genetic diversity studies in Swertia
chirayita
3.1 Introduction
Swertia chirayita is a critically endangered (Anonymous 1997) medicinal herb
of temperate Himalayas. Owing to its medicinal importance, the plant has been
harvested from the wild for a long time and is now considered endangered in its
natural habitats. This requires immediate action for conservation and
sustainable management of the herb. Keeping in mind the diverse eco-
geographical zones occupied by the plant species, it is important to assess the
existing diversity among populations occupying different niches. In order to
devise pragmatic conservation strategies for endangered plant species such as S.
chirayita, identification of intra-population and inter-population diversity is a
preliminary step. This enables identification of regions representing maximum
genetic variations and therefore population from these areas can be conserved at
utmost priority (Shanker and Ganeshaiah 1997, Rao and Hodgkin 2002).
Further, as for most of the medicinal herbs, in case of S. chirayita also,
authentic material is not cultivated and collected, leading to substitution. This is
due to sheer ignorance on the part of the collector or unavailability of the
genuine germplasm. In such cases, diagnostic tools become essential for proper
identification of authentic samples. Enormous variations at morphological and
biochemical levels have been observed among different species of Swertia
(Karan et al. 1997 and Bhatia et al. 2003). However, similar studies have not
been extended to identify intra- species diversity in S. chirayita populations.
Also, the use of morphological markers as was done in the above-mentioned
study is not appropriate for identification of dried samples (the manner in
which, product is sold in the market), as the plant‟s floral and other identifying
characteristics cannot be recognised. Verma and Kumar (2001) used fresh apical
leaves and fibrous roots for isozyme analysis in order to differentiate the various
species belonging to the genus Swertia. The exact demarcation was not possible
during the investigation due to limited isozyme markers, germplasm collection
and poor separation techniques.
DNA based molecular markers can overcome the above mentioned
limitations associated with conventional markers (Weising et al. 1995) and have
been employed for genetic diversity studies in plants (Sangwan et al. 1999, Singh
3
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
et al. 1999, Negi et al. 2000, Lata et al. 2002, Fu et al. 2003). Among the
different molecular markers available, Inter-Simple Sequence Repeat (ISSR)
markers are dominant in nature. These make use of specific microsatellite
sequence anchored either at the 3‟ or 5‟ end to amplify scorable DNA bands
(Gupta et al. 1994, Zeitkiewicz et al. 1994). The ISSR based PCR makes use of
longer primers characterised by higher annealing temperatures than Random
Amplified Polymorphic DNA (RAPD) markers (Pomper et al. 2003), thus
allowing for more stringency. The ISSR markers produce high levels of
polymorphism, are reproducible, do not require prior knowledge of the target
sequences and can amplify even crude DNA preparations available in nanogram
amounts. These advantages have made ISSR the alternate tool for diversity
studies in many plant systems (Tsumura et al. 1996, Martins et al. 2004) and
were therefore employed in the present investigation.
The study was undertaken with the objective of fingerprinting S. chirayita
genotypes. Authentic samples of S. chirayita were collected from different
regions as detailed in the next section. Other species of Swertia genus were also
identified and collected with similar objective. Proper identification of the
authentic samples was the most crucial step prior to fingerprinting and hence
has been described in section 3.2.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
3.2. Morphological description of Indian species of chiretta and collection of authentic germplasm
Swertia Linn. (Fam: Gentianaceae), a large genus of herbs, is distributed in the
mountainous regions of tropical Asia, Europe, America and Africa. Of the thirty-
two species of Swertia recorded in Indian Himalayan region (Garg 1987), eleven
are considered medicinally valuable of which, S. chirayita is considered the
most important for medicinal purposes.
In order to collect and identify authentic S. chirayita, different species of
Swertia were collected from Chakrata, Jageshwar and Mandal regions in
Uttaranchal, India in the year 2004. These were morphologically identified and
authenticated based on their taxonomical characteristics described by Clarke
(1885). Distinguishing morphological characters of S. chirayita, S. cordata, S.
paniculata and S. purpurascens are illustrated in Figure 3.1 and summarised in
Table 3.1. Another common substitute of chiretta was also found growing
abundantly in the hills and was included in the collection and is commonly
called chiretta or pahari chiretta by the locals. However, it was identified as
Halenia elliptica D. Don belonging to fam: Gentianaceae, tribe: Gentianeae and
subtribe: Swertiinae. H. elliptica grows upto 15-60 cm, has tetramerous flowers
with blue to purple corolla. The corolla forms a distinct spur and hence it is also
referred as “the spurred gentian”.
S. chirayita grows to a height of about two feet during the flowering season
when it can be easily differentiated from other species through its floral
morphology. The flowers are small and stalked, green-yellow, tinged with purple
colour, rotate and tetramerous. The corolla is twice as long as the calyx and
divided near the base into 4 ovate-lanceolate segments. Upper surface of petal
has a pair of nectaries covered with oblong scales and ending as fringes. Fruit is
a small, one celled capsule with a transparent yellowish pericarp. It dehisces
from above, septicidally into two valves. Seeds are numerous, minute, many
sided and angular. Figure 1.2 illustrates the pictorial representation of S.
chirayita according to the Bentley and Trimen (1880).
In comparison, S. cordata grows up to 20-90 cm height, has pentamerous
flowers and each corolla lobe has a distinct viscous yellow gland, which is not
covered with any scales. S. paniculata grows to a height of 30-50 cm only and is
readily distinguished by its pentamerous flowers, which have twisted white
corolla lobes each having a crescent shaped blue-coloured gland at the base. S.
purpurascens gains a height of 30-80 cm and again has pentamerous flowers.
The corolla is reddish purple, and at the base of each corolla lobe are present
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
Table 3.1 Morphological differentiation of Swertia chirayita from few of its adulterants
Distinguishing
characters
S. chirayita S. cordata S. paniculata S. purpurascens Halenia
elliptica
Plant height 100-125 cm 20-90 cm 30-50 cm 30-80 cm 15-90 cm
Stem Round, 4-angled upwards. Orange
brown or purplish in colour
Quadrangular, winged Thin cylindrical, purple in colour Round, with two parallel violet
ridges. Greenish yellow in colour
which changes to violet on
maturity
erect,
subquadrangular,
striate, simple or
branched from base
and/or above base.
Leaf Sessile, acute, 3-7 nerved Sessile, cordate, acute, 5-7 nerved Sub-sessile, linear, tip bluntly acute,
1-3 nerved
Sub-sessile, oblong or lanceolate,
3-nerved
Basal leaves petiole
flattened, leaf blade
spatulate, elliptic, or
sometimes
suborbicular, base
narrowed to cuneate,
apex acute to
rounded, veins 3-5.
Flower Tetramerous, occur in large panicles Pentamerous Pentamerous Pentamerous Tetramerous
Calyx 4, gamosepalous, smaller than corolla 5, sepals almost the size of corolla,
lanceolate, green and 3-nerved
5, gamopetalous, equal or larger
than the corolla, granular surface
5, fused at the base, lanceolate,
green, 1-nerved
Calyx lobes elliptic to
ovate, apex
acuminate.
Corolla 4, gamopetalous, greenish yellow
lobes, which are purplish in the
centre; each corolla lobe has a pair of
nectaries at the base which are
covered with hair
5, gamopetalous, fused near the base,
white corolla lobes, with streaks of
purple near the margin. At the base of
each lobe is present a large viscous
orbicular greenish yellow gland
5, gamopetalous, caudate. White
petals. Single crescent shaped blue-
coloured gland at the base of each
corolla lobe
5, gamopetalous, the corolla lobes
reflexed, purple in colour, at the
base of each corolla lobe two dark
purple coloured spots, which
merge to from a complete ring. At
the base of each corolla lobe
single horse-shoe shaped gland
Corolla blue to
purple, campanulate,
1-2.5 cm, basal spurs
5-10(-14) mm; lobes
elliptic to ovate, apex
obtuse and apiculate.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
Distinguishing
characters
S. chirayita S. cordata S. paniculata S. purpurascens Halenia
elliptica
Androecium 4, versatile, filaments and anther
purple in colour. Base of filament
slightly dilated
5, versatile, purple hastate shaped
anthers with white filaments
spreading out from the gynoecium
5, versatile, violet anthers, which
encircle the stigma. Filaments purple
in colour each are slightly dilated at
the base
5, versatile, purple filaments with
dilated base
Filaments 3-5 mm
anthers ovoid, ca. 1
mm
Gynoecium Ovary one-celled Ovary one-celled Ovary one-celled Ovary one-celled Ovary 1-celled. Style
very short. Stigma 2-
lobed
Compiled from Wealth of India (Anonymous, 1982) and www.efloras.com
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
two dark purple spots, which merge to form a complete ring. Also present at the
base of each lobe is a horse shoe-shaped green coloured gland. The pictorial
representation of the whole plant and the flowers in Figure 3.1 helps to easily
differentiate the four species of Swertia and H. elliptica collected during the
flowering season. The leaves of the plants collected during this study were
lyophilised and employed for genetic diversity studies.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
3.3 Results
3.3.1 ISSR analysis for diversity studies
ISSR markers have proved ideal for fingerprinting various plant taxa due to their
abundance and ubiquitous nature (Gupta et al. 1994, Zeitkeiwcz et al. 1994).
Zeitkeiwcz et al. (1994) showed that anchoring of simple sequence repeats (SSR)
based primers at the 3‟ or 5‟ end assures annealing of primers and achieves
specificity for target sequences. 5‟ anchored primers display broader specificity
and also gives a clearer pattern as compared to 3‟ anchored primers (Huang and
Sun 2000). However 3‟ anchored primers are less prone to stuttering due to
strand slippage and hence produce more scorable bands when compared to 5‟
anchored primers (Blair et al. 1999). In the present study 3‟ anchored primers
were employed for the ISSR assay. An initial screening of 33 primers enabled
selection of 23 primers, which produced amplifications in S. chirayita
genotypes. Table 3.2 enlists the primers used for the preliminary screening.
Table 3.2 List of different UBC primers screened for amplification with four Swertia samples chosen at random from the lot of sixty samples
ISSR
Primer Primer sequence PCR amplification*
UBC 801 ATA TAT ATA TAT ATA TT N
UBC 802 ATA TAT ATA TAT ATA TG N
UBC 803 ATA TAT ATA TAT ATA TC N
UBC 804 TAT ATA TAT ATA TAT AA N
UBC 805 TAT ATA TAT ATA TAT AC N
UBC 806 TAT ATA TAT ATA TAT AG N
UBC 807 AGA GAG AGA GAG AGA GT Y
UBC 808 AGA GAG AGA GAG AGA GC Y
UBC 809 AGA GAG AGA GAG AGA GG Y
UBC 810 GAG AGA GAG AGA GAG AT Y
UBC 811 GAG AGA GAG AGA GAG AC Y
UBC 812 GAG AGA GAG AGA GAG AA Y
UBC 813 CTC TCT CTC TCT CTC TT N
UBC 814 CTC TCT CTC TCT CTC TA Y
UBC 815 CTC TCT CTC TCT CTC TG N
UBC 816 CAC ACA CAC ACA CAC AT Y
UBC 817 CAC ACA CAC ACA CAC AA Y
UBC 818 CAC ACA CAC ACA CAC AG Y
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
ISSR
Primer Primer sequence PCR amplification*
UBC 819 GTG TGT GTG TGT GTG TA Y
UBC 820 GTG TGT GTG TGT GTG TC Y
UBC 830 TGT GTG TGT GTG TGT GG Y
UBC 836 AGA GAG AGA GAG AGA GYA Y
UBC 838 TAT ATA TAT ATA TAT ARC N
UBC 841 GAG AGA GAG AGA GAG AYC Y
UBC 842 GAG AGA GAG AGA GAG AYC Y
UBC 843 CTC TCT CTC TCT CTC TRA N
UBC 848 CAC ACA CAC ACA CAC ARG Y
UBC 850 GTG TGT GTG TGT GTG TYC Y
UBC 852 TCT CTC TCT CTC TCT CRA N
UBC 857 ACA CAC ACA CAC ACA CYG Y
UBC 860 TGT GTG TGT GTG TGT GRA Y
UBC 868 GAA GAA GAA GAA GAA GAA Y
S2 C (TC) 4 (AC) 4 A Y
*Y: amplification obtained; N: no amplification
3.3.2 Genetic diversity in chiretta population from Mandal (Data Set I)
In the present study nineteen UBC primers were used after initial screening of
33 primers to assess genetic diversity in population of S. chirayita collected
from Mandal region. Table 3.3 gives details of the samples used in Data set I.
The data set comprised 13 genotypes of S. chirayita collected from Mandal
region of Uttaranchal along with 2 genotypes each of S. cordata, S. paniculata
and S. purpurascens, which were used as outliers in the study.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
Table 3.3 Description of plant material used in Data set I, revealing the codes, the species name, the region of collection and other identifying remarks
Data Set I
Code Species Location Background
A S. chirayita Mandal Flowering, wild collection
B S. chirayita Mandal Flowering, wild collection
C S. chirayita Mandal Flowering, wild collection
D S. chirayita Mandal Flowering, wild collection
E S. chirayita Mandal Flowering, wild collection
F S. chirayita Mandal Flowering, wild collection
G S. chirayita Mandal Flowering, wild collection
H S. chirayita Mandal Flowering, wild collection
I S. chirayita Mandal Vegetative, wild collection
J S. chirayita Mandal Vegetative, wild collection
K S. chirayita Mandal Vegetative, wild collection
L S. chirayita Mandal Samples from previous yr, wild
collection
M S. chirayita Mandal Samples from previous yr, wild
collection
N S. cordata Jageshwar Flowering, wild collection
O S. cordata Jageshwar Flowering, wild collection
P S. paniculata Jageshwar Flowering, wild collection
Q S. paniculata Jageshwar Flowering, wild collection
R S. purpurascens Chakrata Flowering, herbal garden
S S. purpurascens Chakrata Flowering, herbal garden
A total of 315 amplification products were scored averaging to 16.57 bands
per primer. Table 3.4 gives details regarding the amplification products obtained
by each primer used. The number of bands produced ranged between 21 (UBC
840, 850, 857 primers), and 7 (UBC 848 and 860 primers). The UBC primers
produced a high percentage of polymorphic bands (98.73 %) among the
genotypes assayed. However, on exclusion of the bands produced within the
outliers, percent polymorphism was reduced to 42.59 (Table 3.4). This shows
that maximum polymorphism was generated due to the outliers. Primer UBC
820 generated the highest polymorphism of 80% among the S. chirayita
genotypes. The primers UBC 842 and UBC 860 could not detect any
polymorphism. Representative gels illustrating the amplification profiles
produced by the ISSR maker assay by employing primers UBC 809, 810, 818
and 840 are shown in Figures 3.2, 3.3, 3.4, and 3.5, respectively.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
Primer UBC 809 amplified a total of 15 scorable bands within the molecular
range of 300 bp to 1.6-Kb. Of the seven bands amplified within Mandal
genotypes A to M, four bands were monomorphic. The rest of the bands were
polymorphic within the Mandal population and also among the outliers. The
amplification profile generated by primer UBC 810 is represented in Figure 3.3.
A total of 14 bands were scored within the size range of 500 bp to 2.0-Kb. In all,
four monomorphic bands were scored within the Mandal population.
Primer UBC 818 enabled scoring of 19 bands in total (Figure 3.4). Eleven
bands were scored for the S. chirayita genotypes A to M within size range of 500
bp to 1.6-Kb; of these seven were monomorphic. A strikingly variable pattern
was scored for the allied species of chiretta and also within the two genotypes of
the same species. Primer UBC 840 revealed a total of 21 scorable bands of which
12 were exclusively amplified within the S. chirayita population (Figure 3.5). Of
these seven were monomorphic and the rest five were polymorphic.
3.3.2.1 Statistical analysis of the data
The binary data obtained by ISSR analysis was used to estimate genetic
similarity by Jaccard‟s coefficient. The Jaccard‟s similarity matrix revealed high
genetic diversity in S. chirayita; the values of similarity coefficient ranged
between 0.97 and 0.67 in the population of S. chirayita. Highest similarity
coefficient value of 0.97 was scored between genotypes C and D; followed by
0.93 between A and B; J and K; F and G each. Within the chiretta population the
lowest similarity value of 0.67 was observed between genotypes coded F and M.
On inclusion of the outliers (samples coded N to S), the similarity coefficients
decreased considerably. Different clustering methods are available for revealing
the genotype association during genetic diversity studies. We employed
complete linkage (CL), single linkage (SL), Unweighted Pair Group Method of
Arithmetic Averages (UPGMA) and Weighted Pair Group Method of Arithmetic
Table 3.4 Information obtained in terms of percent polymorphism detected on employing the enlisted set of primers on samples belonging to data set I
Data Set I
Primer
Number of amplicons
Including outliers Excluding outliers
Total Polymorphic Total Polymorphic
UBC 807 14 13 3 1
UBC 809 15 15 7 3
UBC 810 14 14 12 8
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
UBC 811 16 16 12 7
UBC 812 18 18 9 6
UBC 817 19 19 7 1
UBC 818 19 19 11 4
UBC 819 13 13 6 4
UBC 820 16 16 5 4
UBC 830 16 16 7 3
UBC 836 20 19 16 5
UBC 840 21 21 10 2
UBC 841 18 17 13 7
UBC 842 13 13 3 0
UBC 848 12 12 7 4
UBC 850 21 20 11 5
UBC 857 21 21 12 2
UBC 860 12 12 6 0
UBC 868 17 17 5 3
Total 315 311 162 69
Percent Polymorphism 98.73 42.59
Averages (WPGMA) methods of cluster analysis to deduce the associations
among the S. chirayita genotypes. Similar results were obtained by each of these
clustering methods except for minor changes in placement of some genotypes
(Figures 3.6 and 3.7).
The dendrogram obtained by subjecting the similarity matrix to UPGMA
method of cluster analysis is shown in Figure 3.6. It consisted of two major
clusters. Cluster-I labelled I in Figure 3.6, consisted mainly of the Swertia
chirayita genotypes, and cluster-II (labelled II) included the allied species of the
Swertia genus, which were used in the study as outliers. Within cluster I, three
sub-clusters were formed. Interestingly, the genotypes which were in flowering
stage during the collection, grouped together as sub-cluster Ia with the
exception of H and those at the vegetative phase formed sub-cluster Ib. The
genotypes representing one-year-old population (L and M) grouped separately
as subcluster Ic.
In cluster II, the three species of Swertia genus formed three separate
clusters. The two S. cordata individuals grouped together at 0.75 similarity
coefficients, those of S. purpurascens grouped at 0.86 and the two genotypes of
S. paniculata formed a cluster at 0.62 value of similarity coefficient. Considering
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
that the genotypes of each species were collected from same localities the
diversity observed among the species is very high.
Bootstrap analysis was performed to estimate the statistical support to the
internal nodes of the UPGMA tree. The bootstrap P values obtained after 1000
replications are represented in Figure 3.8. According to Hillis and Bull (1993),
the internal branches showing > 70% bootstrap values are likely to be accurate
upto 90% level of significance. Our results were interpreted on the basis of this
rule. The clusters, which grouped genotypes of S. cordata, S. paniculata, S.
purpurascens were supported by 100% bootstrap values. The major cluster I,
which consisted of all the genotypes of S. chirayita, was also supported by 100%
bootstrap P- value. Within this cluster, a few nodes were supported by low
bootstrap values (32.6%, 34.9% and 35.9%). However, for most of the other
nodes this value ranged between 69.2 to 95.9%. These values suggest overall
robustness of the UPGMA tree.
Principal component analysis was performed on the correlation matrix of the
original data set. The PC analysis revealed four distinct clusters of the four
Swertia species. Figure 3.9 and 3.10 represents the 2-D and 3-D PCA scatter
plot, respectively. The S. chirayita population formed a tight cluster along the
PC-1 axis and the other three species of Swertia separated along the second
principal axis. The first two PC axes accounted for 69.3% of the variance, with
the first axis explaining 58.9% variance (eigenvalue=11.20) and the second axis
accounting for 10.3% (eigenvalue = 1.96) of variance in the correlation matrix.
These results corroborate the clustering pattern obtained by the UPGMA cluster
analysis.
Genetic Diversity
TERI University-Ph.D. Thesis, 2006
3.3.3 Variation among Swertia chirayita collected from different regions
and among related species of S. chirayita (Data Set II)
The objective of the present study was to demarcate authentic chiretta from
allied species of S. chirayita and to validate the authenticity of commercially
available samples with reference to S. chirayita genotypes collected from
different geographical locations. These were morphologically identified and
authenticated based on their taxonomic characteristics described by Clarke
(1885), as shown in Figure 1.2.
The commercial samples were procured from herbal markets and
commercial plantations. The samples collected from the wild by local collectors
were also included in this study. However, these samples from herbal markets
and commercial plantations could not be authenticated either because they were
collected during the vegetative phase or were obtained in the dried form.
Halenia elliptica was included as an outlier in the study (Figure 3.1 e). Table 3.5
provides details of the genotypes used in the data set. In all, Data set II consisted
of ten genotypes of S. chirayita representing different locales; four market
samples and five genotypes of allied species belonging to the genus Swertia.
A total of 246 bands were produced by amplification with 18 UBC primers
giving an average of 13.6 scorable amplicons per primer. Primer UBC 818
amplified the highest of 21 scorable bands and only 7 bands were scored with
primer UBC 807, thus, representing the lowest amplifications produced. The
amplified bands were 100% polymorphic and on eliminating the profiles
generated for the outliers (the allied species of S. chirayita) percent
polymorphism was reduced to 97.84% only (Table 3.6).
Figures 3.11, 3.12, 3.13 and 3.14 represent the amplification profiles produced
by primers UBC 816,818, 841 and 857, respectively. Primer UBC 816 amplified a
total of 19 polymorphic bands within the size range of 300 bp to 2.0-Kb. The
ISSR profile generated by primer UBC 818 is shown in Figure 3.11. The primer
enabled scoring of 21 bands of which 15 were exclusively produced in the S.
chirayita genotypes. These bands ranged from the molecular sizes of 300 bp to
2.0-Kb. Primer UBC 841 amplified bands from 200-bp size to 2.0-Kb size. A
total of 21 bands were scored among all the genotypes and on excluding the
outliers, 12 bands were scored for the S. chirayita genotypes. Primer UBC 857
amplified a total of 10 bands, which was comparatively lesser than those
amplified by UBC 816, 818 and 841. However, all these bands were polymorphic
and ranged within the molecular size of 500 bp to 1.6-Kb. Table 3.6 provides the
details of the amplification products scored by different primers in the whole
data set as well as exclusively in S. chirayita genotypes.
Bibliography
TERI University-Ph.D. Thesis, 2006
Table 3.5 List of plant material, providing the codes used in Data set II, the species name, the region of collection and identifying remarks
Data Set II
Code Species Location Background
A H. elliptica Mandal Identified, wild collection
B S. angustifolia Mandal Unidentified, wild collection
C S. angustifolia Mandal Unidentified, wild collection
D S. chirayita Chakrata Identified, Wild collection
E S. chirayita Chakrata Identified, Wild collection
F S. chirayita Devariyataal Unidentified, wild collection
G S. chirayita Devariyataal Unidentified, wild collection
H S. chirayita Jageshwar Identified, herbal garden
I S. chirayita Jageshwar Identified, herbal garden
J S. chirayita Joshimath Unidentified, commercial
plantation
K S. chirayita Khaljuni Unidentified, commercial
plantation
L S. chirayita Khaljuni Unidentified, commercial plantation
M S. chirayita Khari Baoli Unidentified, dried sample
from herbal market
N S. chirayita Sikkim Identified; in vitro
maintained
O S. chirayita Sikkim Identified; in vitro maintained
P S. chirayita Mandal Identified, wild collection
Q S. cordata Chakrata Identified, wild collection
R S. paniculata Chakrata Identified, wild collection
S S. purpurascens Chakrata Identified, wild collection
Bibliography
TERI University-Ph.D. Thesis, 2006
Table 3.6 Fingerprint details in terms of polymorphism generated for data set II, as obtained with the enlisted primers
Data Set II
ISSR Primer
Number of amplicons
Including outliers Excluding outliers
Total Polymorphic Total Polymorphic
UBC 807 7 7 5 5
UBC 809 13 13 9 9
UBC 810 12 12 9 8
UBC 811 15 15 13 13
UBC 812 11 11 8 8
UBC 816 19 19 19 19
UBC 817 15 15 13 13
UBC 818 21 21 15 15
UBC 819 10 10 6 6
UBC 820 11 11 9 9
UBC 830 14 14 10 10
UBC 836 16 16 13 13
UBC 840 16 16 14 14
UBC 841 19 19 12 12
UBC 842 11 11 5 5
UBC 848 15 15 10 9
UBC 857 10 10 8 8
UBC 860 11 11 8 8
Total 246 246 186 182
Percent Polymorphism 100 97.84
3.3.3.1 Statistical analysis of the data
The binary data was employed to deduce the genetic similarity by Jaccard‟s
coefficients. Highest similarity coefficient value of 0.86 was observed between
the two genotypes, N and O of S. chirayita from Sikkim, followed by 0.84
between genotypes coded B and C (S. angustifolia from Mandal); and also
between chiretta sample coded M (from Khari Baoli) and sample K (from
commercial plantation at Khaljuni). The Devariyataal genotypes grouped at a
similarity coefficient of 0.66. As expected, the lowest similarity coefficient value
of 0.24 was observed between S. chirayita genotype (E) and S. purpurascens
(S), both collected from Chakrata.
Bibliography
TERI University-Ph.D. Thesis, 2006
Jaccard‟s similarity coefficient was subjected to CL, SL, UPGMA and
WPGMA methods of cluster analysis. The clustering patterns obtained by the
four methods were similar (Figure 3.15 and 3.16). Cophenetic correlation
coefficients of these clustering methods were calculated to test goodness of fit
between similarity matrix and the cophenetic matrix. Phenetic dendrograms
obtained were a true representation of the similarity matrix as indicated by their
high cophenetic correlation values. The complete linkage and single linkage had
correlation value of 0.94, whereas the UPGMA and WPGMA methods had
correlation value of 0.96 each. UPGMA clustering method is the most preferred
method of cluster analysis in terms of consistency in grouping of biological
materials. Hence, we employed UPGMA based dendrogram for further analysis
and interpretation (Figure 3.15).
The cluster analysis revealed presence of a broad genetic base of the species.
Two major clusters (CI and CII) were identified; Cluster-I consisted of Swertia
chirayita genotypes collected from Mandal (P), Devariyataal (F, G), Chakrata
(D, E), Jageshwar (H, I) and Sikkim (N, O) regions, along with B and C
genotypes of S. angustifolia. Genotype D of S. chirayita from Chakrata and H
from Jageshwar clustered together at a similarity value of 0.58. Genotype D of S.
chirayita from Chakrata region clustered with S. chirayita genotype P from
Mandal region at similarity coefficient value of 0.48.
Cluster-II included the chiretta samples collected from Khari Baoli (M) and
from commercial plantations at Joshimath (J), and Khaljuni (K, L), and the
genotypes Q, R and S of Swertia genus, which included S. cordata, and S.
purpurascens and H. elliptica, as outliers. The genotypes K and L collected from
commercial plantations of Khaljuni clustered together with dried samples of
chiretta from Khari Baoli (M) at 0.84 and 0.69 values of similarity coefficients.
Bootstrap analysis was carried out to find the statistical support for the internal
branches of the UPGMA tree. Bootstrap re-sampling technique makes use of the
data itself to assess its own utility in statistical analysis. Figure 3.17 represents
the bootstrap estimates obtained after 1000 replications. The S. chirayita
representatives from Sikkim (N, O), Devariyataal (F, G) and S. angustifolia
genotypes F and G from Mandal clustered together in all the 1000 replications
during bootstrap analysis, showing a very strong geographical association. The
node clustering the S. chirayita collections from Mandal (P), Chakrata (D and
E), Jageshwar (H and I) and Sikkim (N and O) was supported by 94.7% of
bootstrap analysis. High bootstrap values for the internal nodes of this cluster
confirmed its robustness.
Bibliography
TERI University-Ph.D. Thesis, 2006
However, in UPGMA based dendrogram (Figure 3.15), S. angustifolia
genotypes were found to cluster with S. chirayita samples from Sikkim, leading
to difficulty in deciphering the true identity. In absence of any identified
material of S. angustifolia, its grouping with the rest of the genotypes of genuine
chiretta samples raise speculations about its authenticity. These samples were
collected prior to flowering, and their identification was based solely on
morphological characteristics. In order to clarify this discrepancy, principal
component analysis was carried out because it leads to reduction of data so as to
clarify the relationships between two or more characters and to divide the total
variance of original characters into a limited number of uncorrelated new
variables (Wiley 1981).
As expected, the PCO analysis revealed a clearer picture (Figure 3.18) and it
segregated the S. angustifolia from S. chirayita. In fact, the PCO analysis also
segregated the chiretta collected from Devariyataal, and hence it can be
speculated that the collection made solely on the morphological basis can be
misleading and therefore molecular characterisation is essential to establish the
authentic identity of the collected material. This is at least essential if
germplasm collections are done with specific aims of forming core collection,
conservation, and for establishing herbal gardens and commercial plantations.
In the present analysis the first two PC axes explained 41% of the variance in
the correlation matrix. The first axes accounted for 22% of variance with
eigenvalue of 4.28 and the second PC explained 19% of the variance (eigenvalue
= 3.62). PC with eigenvalues > 1.0 are considered as inherently more
informative. In our analysis the first six PCs had eigenvalues greater than 1 and
explained 72% of the variance, suggesting a high heterogeneity in the data set.
The PC scatter plot revealed a clear-cut separation of authentic S. chirayita
genotypes (identified on the basis of their floral morphology) and the
commercial samples along the first PC axes (please see Figure 3.18). A 3-D view
of the same is illustrated in Figure 3.19.
Bibliography
TERI University-Ph.D. Thesis, 2006
3.4 Discussions
In the present study we employed ISSR markers for assessing the levels of
genetic diversity within S. chirayita genotypes. Use of molecular markers is
preferred over conventional morphological and biochemical markers for genetic
diversity studies as they are completely devoid of environmental effects and the
developmental stages of the experimental material. Among the three widely used
PCR based molecular markers, namely RAPDs (Willaims et al. 1990), AFLP (Vos
et al. 1995) and SSR or microsatellites (Tautz 1989), the SSRs owing to their
ubiquity, hypervariability, abundance and genome wide distribution are
considered very powerful genetic markers. However, their widespread use is
impeded due to requirement of sequence information from flanking regions for
designing primers.
On the other hand ISSR, a modification of SSR based marker system,
circumvents the requirement of the prior sequence information and hence finds
a widespread application. The analysis involves PCR amplification of genomic
DNA using a single primer that targets the microsatellite repeats. Further, the
amplifications result in multilocus and highly polymorphic patterns. The
effectiveness of ISSR markers in genome analysis has been reported is many
plant species (Kantety et al. 1995, Nagaoka and Ogihara 1997, Blair et al. 1999,
Huang and Sun 2000, McGregor et al. 2000, Ruas et al. 2003). Each amplified
band corresponds to a DNA sequence delimited by two inverted microsatellites.
In the present study, ISSR marker assay enabled us to quantify the levels of
genetic diversity within the population of S. chirayita from Mandal region (Data
set I); establish relationship among chiretta from different geographical
locations and finally helped in distinguishing the unauthentic chiretta samples
from the authentic ones (Data set II).
3.4.1 Data set I
The present study was conducted to gain an insight into the diversity levels
existing within S. chirayita genotypes collected from Mandal region in
Uttaranchal. A total of 315 bands were amplified by using 19 ISSR primers
across the thirteen genotypes of S. chirayita and six genotypes of its allied
species. This is a sufficiently good number to explore genetic diversity of Swertia
species, as is also supported by other studies in the literature. Nagaoka and
Ogihara (1997) scored a total of 260, 300 and 350 bands in diploid, tetraploid
and hexaploid wheat germplasm respectively, whereas Gilbert et al. (1999)
scored 137 ISSR markers in 37 accessions of lupin germplasm. Ruas et al. (2003)
Bibliography
TERI University-Ph.D. Thesis, 2006
observed that 200 RAPD markers were sufficient for dendrogram stability while
assessing genetic diversity of Coffea species.
Each primer used in the present study uncovered polymorphism within the
complete data set. However, on excluding the outlier species, primers UBC 842
and UBC 860 failed to detect any polymorphism among the S. chirayita
genotypes. A 98.73% polymorphism was detected in the complete data set and
on excluding the outliers, S. chirayita population scored 42.7% of polymorphic
bands.
In the present study, the high genetic diversity revealed within the
population of S. chirayita, may be attributed to the mating system of the species.
A close correlation has been demonstrated between the breeding system of a
species and the levels of genetic variations displayed. In fact, the breeding
system directly influences the pattern of genetic variation by regulating the
inheritance and transmission of genes from parents to the offspring. DNA
fingerprints of cross-pollinated species are individual specific and therefore
display high levels of genetic variation. For instance, Perera et al. (1998) studied
the level of genetic variation in two coconut varieties using AFLP markers. It was
revealed that the cross-pollinated varieties harbour more variations than the
dwarf selfing ones. Similarly, Das et al. (1999b) have displayed higher diversity
in self - incompatible varieties of Brassica campestris as compared to self-
compatible ones. Singh et al. (2001) demonstrated high levels of genetic
variation even at progeny levels in neem, which was explained by high rates of
out crossing in the species. Since each individual is derived through genetic
recombination, it is considered to be genetically unique and thus contributes to
high levels of polymorphism within a population (Weising et al. 1995). S.
chirayita as explained by its floral morphology, is also a cross pollinator; bees
being the predominant pollinator of the species (Khoshoo and Tandon 1963).
This helps in explaining the broad genetic base of the plant.
A part of the polymorphism generated can also be ascribed to the type of
molecular marker employed because different markers uncover different levels
of polymorphism. ISSR markers are known to detect substantial levels of
polymorphism and have been employed in many plant systems. For instance, in
Trigonella foenum-graecum, 14 ISSR primers uncovered 72% polymorphism in
17 accessions and 93.64% polymorphism was detected in 9 accessions of T.
caerulea (Dangi et al. 2004). Similarly, Wu et al. (2004) observed 64%
polymorphism among 14 populations of Oryza granulata and 26 and 22%
polymorphism within two distinct populations by employing ISSR markers. In
fact, Godwin et al. (1997) indicated that the polymorphism deduced in their
Bibliography
TERI University-Ph.D. Thesis, 2006
studies was not a result of greater polymorphism genetically, but rather due to
technical reasons related to detection methodology used for ISSR analysis. In
the present study as well, use of ISSR markers could likely be one of the reasons
for the high polymorphism detected in S. chirayita genotypes.
It is a prevalent view that rare species maintain low levels of genetic
variation. Though S. chirayita has a rare distribution in its natural habitat, high
levels of genetic diversity were revealed in our study, contrary to this prevalent
view. The ruthless exploitation of the species and destruction of its habitat,
which actually account for rare occurrence of the species, can likely explain these
high levels of genetic variations in the study. High genetic variation has also
been reported for other endangered plants such as in Cathaya agrophylla 32%
(Wang et al. 1996); Changium smyrnioides 69% (Fu et al. 2003); Dacydium
pierrei 33.3% (Su et al. 1999); and Lactoris fernandeziana 24.5% (Brauner et al.
1992). In such cases it is prudent to save the natural habitat of the endangered
species as a conservation measure.
Veritably, a firm grasp of the genetic structure and the diversity grade in rare
and endangered species is required to determine efficacious measures and
strategies for protecting them. Shanker and Ganeshaiah (1997) related the utility
of intra-population diversity evaluation to development of effective conservation
strategies, as was done in the case of Phyllanthus. The genetic base of the
species determines the number of accessions to be maintained for conservation
purpose. Fewer plants would be required for conservation in case the germplasm
has genetically uniform accessions. If the stocks were genetically more variable,
the number of accessions to be maintained would increase.
Various marker techniques have been employed to achieve the above
objective. Fu et al. (2003) reported genetic diversity of Changium smyrniodes,
an endangered medicinal plant of China. Gaudel et al. (2000) used AFLP
markers to infer the genetic diversity of Eryngium alpinum, another endangered
medicinal plant belonging to European Alps. Singh et al. (2004a) reported high
genetic diversity within the population of Podophyllum hexandrum, an
endangered medicinal herb of north-western Himalayas, by molecular analysis.
Molecular marker data is most widely employed to generate a binary matrix
for absence and presence of alleles for statistical analysis and further analysis is
done by different measures of genetic distance/similarity. Among the most
prevalent are Nei and Li (1979) coefficient, Jaccard‟s coefficient (1908), Simple
Matching (GDsm) (Sokal and Michener 1958) and modified Roger‟s distance. In
Bibliography
TERI University-Ph.D. Thesis, 2006
the present study, we employed Jaccard‟s similarity coefficient to estimate the
genetic distance, as this is most suitable for dominant markers such as ISSR.
The Jaccard‟s similarity matrix was used for cluster analysis based on CL, SL,
UPGMA and WPGMA methods. Different methods of cluster analysis were used,
as a single clustering method might not always be effective in revealing the most
authentic genetic association. In our study, the dendrograms obtained by these
methods were found to be more or less similar. Kantety et al. (1995), similarly,
compared CL, SL, UPGMA, UPGMC and ward‟s method of cluster analysis while
assessing genetic diversity of dent and popcorn maize line by ISSR method and
reported consistent results by all the five methods used.
The UPGMA cluster analysis of ISSR data separated the S. chirayita
genotypes of Mandal region as a separate cluster (I) and the genotypes of S.
cordata, S. paniculata and S. purpurascens were segregated into distinct
groups. It was interesting to note that the sub-grouping within cluster I
correlated with the phenological differences displayed by the genotypes,
especially in relation to flowering behaviour. The samples collected for the
present study comprised of plants at the vegetative phase as well as at the
flowering stage, which could very likely be due to the rapid cycling and normal
variants, growing as a mixed population and individuals with different flowering
times growing side by side. Our results from the ISSR assay indicated that there
is sufficient natural variation present within these S. chirayita genotypes. The
distinct flowering habit may be the outcome of adaptation to distinct
environment and eco-geographical regimes. The level of gene flow between
populations is known to influence the common gene pool shared and the
amount of genetic subdivisions within individuals of a species (Mayr 1970),
which ultimately leads to adaptation of individuals according to their local
environment (Gustafsson and Lonn 2003). However, further studies with larger
sample size and different molecular makers would help us validate this
hypothesis.
Consequently, the present study highlighted the genetic variations available
within the population of S. chirayita collected from Mandal and segregated the
allied species according to their taxonomic groupings. The sub-groupings within
the major cluster of S. chirayita according to genotypes‟ flowering phenology
helped to hypothesise the existence of flowering variants. These results give a
very interesting direction for future research, considering that population
genetics of S. chirayita is a totally unexplored research area. Lastly, considering
the fact that S. chirayita is an endangered and rare species, there is an urgent
need to conserve the maximum diversity available within the species.
Bibliography
TERI University-Ph.D. Thesis, 2006
3.4.2 Data set II
DNA based techniques are being widely used for authentication of plant species
of medicinal importance and are essentially useful for medicinal plants that are
frequently substituted or adulterated with other species or varieties that are
morphologically and phytochemically indistinguishable (Joshi et al. 2004). For
instance, RAPD technique was adopted to identify eight types of dried Coptis
rhizomes and one type of Picrorhiza rhizome, a substitute for the former in the
Chinese herbal market (Cheng et al. 1997). Similarly, characterisation of
Echinacea species and detection of possible adulteration was done using RAPD
technique (Wolf et al. 1999). Hon et al. (2003) employed RAPD for genetic
authentication of ginseng and other traditional Chinese medicines.
The chiretta available for trade is usually substituted with allied species of S.
chirayita either intentionally or due to ignorance, during collections from the
wild by the locals. Hence, the second part of this investigation involved assessing
the genetic diversity in authentic S. chirayita genotypes collected from different
geographical locations and commercial sources. The objective of the study was
to characterise S. chirayita genotypes from different geographical locations and
to establish the authenticity of the samples collected from different commercial
sources vis-à-vis the authentic germplasm.
The ISSR assay was conducted with 18 primers and a total of 246
polymorphic bands were scored. The polymorphism was reduced to 97.84%
after excluding the outliers. The high level of polymorphism revealed, could be
attributed to the heterogeneity of the samples constituting the experimental data
set, which included market samples, authentic chiretta from different
provenances as well as the outliers. The commercial samples were collected from
different trade sources as authentic chiretta, however our analysis segregated
them as a completely different species. Further, even within the samples
authenticated as true chiretta through their floral morphology, a high level of
genetic variation was revealed. Since S. chirayita is a critically endangered
medicinal plant, these results emphasise the importance of conserving
maximum populations possible. This would help in maintaining the maximal
biodiversity for conservation and reestablishment of the plant and also in
defining core collections for breeding programmes.
The UPGMA based cluster analysis of the similarity matrix, clearly brought
out the genetic similarity of all the genotypes of S. chirayita and segregated the
outliers and commercial samples of chiretta into separate clusters. The
dendrogram obtained by CL, SL and WPGMA also revealed similar genetic
alliance. In UPGMA based dendrogram, the S. angustifolia genotypes were
Bibliography
TERI University-Ph.D. Thesis, 2006
found to cluster with S. chirayita samples from Sikkim, leading to difficulty in
deciphering the true identity of the sample collected as S. angustifolia. In
absence of any identified material of S. angustifolia, its grouping with the rest of
the genotypes of genuine chiretta samples raised speculations about its
authenticity. These samples were collected prior to flowering, and their
identification was based solely on morphological characteristics. However, the
segregation on the basis of PCO analysis revealed a clearer picture. It segregated
the S. angustifolia as well as the chiretta collected from Devariyataal from
genuine chiretta samples. The analysis of the binary data by different clustering
methods and also by PCO analysis revealed a very consistent clustering pattern.
A strong geographical association was noticed among the S. chirayita
genotypes from Sikkim (bootstrap P value=100); among the unidentified
chiretta samples from Devariyataal; and S. angustifolia from Mandal region
(bootstrap values=100). Geographical association found in the above mentioned
cases were however missing among the genotypes collected from Jageshwar and
Chakrata regions of Uttaranchal.
Considering the different geographical zones and niches where Swertia
chirayita is found, this study can be considered a preliminary study since it
focuses only on samples collected from a few geographical locales and market
sources. Despite this, the study was decisive in suggesting the level of genetic
diversity that is present among the different genotypes of S. chirayita and also
among other related species within Swertia genus. This helps to conclude that
though the species has declined in numbers, the populations available represent
high genetic diversity, which needs to be conserved with utmost priority.
Figure 3.1 Pictorial representation of the four species of Swertia along with a local substitute of chiretta a) Swertia chirayita b) S. cordata c) S. paniculata d) S. purpurascens e) Halenia elliptica (local name pahari chiretta).
a b c d e
Figure 3.2 ISSR fingerprint generated on employing primer UBC 809. Lanes A to M represent different genotypes of Swertia chirayita and lanes N to S represent the genotypes of S. cordata, S. paniculata and S. purpurascens, which were used in the study as outliers. Lane X represents the 1-Kb ladder.
X A B C D E F G H I J K L M N O P Q R S
X A B C D E F G H I J K L M N O P Q R S
Figure 3.3 ISSR fingerprint profile of chiretta samples using primer UBC 810. Lanes A to M represent the Swertia chirayita population from Mandal region. The outliers are represented in lanes N to S; lane X represents the 1-Kb ladder.
X A B C D E F G H I J K L M N O P Q R S
Figure 3.4 A representative ISSR profile of Swertia chirayita genotypes generated by the UBC 818 primer. The lanes A to M are the genotypes of S. chirayita, and lanes N to S represent the controls S. cordata, S. paniculata and S. purpurascens used in the study. The lane X represents the 1-Kb ladder.
X A B C D E F G H I J K L M N O P Q R S
Figure 3.5 Amplification profile obtained on employing primer UBC 840 for ISSR assay of Swertia chirayita population collected from Mandal (Uttaranchal, India). Lane X represents the 1-Kb ladder, followed by genotypes of S. chirayita from lanes A to M. Lanes N to S represent the outliers.
Figure 3.6 Dendrogram representing the UPGMA based clustering obtained for the population of Swertia chirayita collected from Mandal (Uttaranchal, India). The similarity matrix was obtained using Jaccard’s coefficient using 315 ISSR marker bands.
Similarity Coefficient
0.10 0.33 0.55 0.78 1.00
A
B
C
D
E
F
G
I
J
K
H
L
M
N O R
P
Q
S CII
A
B
C
CI
A
Figure 3.7 Dendrograms generated by employing A) complete linkage; B) single linkage and C) WPGMA methods of cluster analysis. The clusters obtained on employing the three different clustering algorithms produced nearly similar clustering patterns, as is evident from the figures.
S
B
C
D E F
G
H
I J K
L
M N
O
P
Q
R
0.10 0.33 0.55 0.78 1.00
0.10 0.33 0.55 0.78 1.00
0.10 0.33 0.55 0.78 1.00
S
B
C
D
E F
G
H
I
J
K
L
M
N
O
P Q
R
A
S
B
C
D
E
G
H
I
K L
M
N
O
P
R
A
F
J
Q
Similarity coefficient
C
A
B
+----------K
+-----------68.7
| +----------A
|
+-88.4 +------B
| | +-95.9
| | +-35.9 +-------C
| | | |
| +-34.9 +----------D
+-32.6 |
| | | ----------E
| | +------55.6
| | +----------F
| |
+-95.4 | +----------H
| | +-----------69.2
| | | +-----------I
| | +-86.3
+100.0 | +------J
| | |
| | +-----------------------------G
| |
| | +----------K
| +-------------------99.8
| +----------L
|
| +----------P
| +100.0
| | +----------Q
| +-83.8
| | | +-----------R
| | +100.0
+---------------------74.6 + --------S
|
| +----------N
+-----100.0
+----------O
Figure 3.8 Bootstrap consensus tree of Swertia chirayita genotypes, showing bootstrap value at each node of the tree after 1000 replications. Codes A to M represent S. chirayita genotypes representing Mandal population. Genotypes N to S represent the outliers used in the study.
Figure 3.9 Principal component analyses depicting a tight clustering among representative genotypes of Swertia chirayita collected from Mandal (Uttaranchal, India) based on ISSR assay data. The genotypes of S. cordata, S. paniculata, S. purpurascens also separate out as three distinct groups along PC-2.
PC-1 -0.25 0.06 0.38 0.69 1.00
PC-2
-0.50
-0.16
0.18
0.51
0.85
S. chirayita genotypes
S. cordata 1
S. cordata 2
S. purpurascens 1
S. purpurascens 2
S. paniculata 1
S. paniculata 2
S. paniculata 2
S. paniculata 1
S. purpurascens 2
S. purpurascens 1
S. cordata 2
S. cordata 1
S. chirayita genotypes
Figure 3. 10 A three dimensional view of the scatter plot obtained by the principal component analysis of the ISSR amplification data for Swertia chirayita genotypes collected from Mandal region.
Figure 3.12 Polymorphism generated by UBC 818 among the different collections of chiretta. Lane X represents the 1-Kb ladder. Lanes A to M represent the different collections of S. chirayita and lanes N to S represents the outliers used in the study.
X A B C D E F G H I J K L M N O P Q R S
X A B C D E F G H I J K L M N O P Q R S
Figure 3.11 A representative ISSR profile produced by UBC 816 in Swertia chirayita collected from different regions. Lanes A to M represents the different collections of S. chirayita and M to S represent allied species of S. chirayita, which were used in the study as outliers. Lane X represents 1-Kb ladder.
X A B C D E F G H I J K L M N O P Q R S
Figure 3.13 Amplification profile produced by UBC 841 in different collection of S. chirayita collected from different regions. Lane X is the 1-Kb ladder and lanes A to M represent the different genotypes of Swertia chirayita. The lanes N to S represent the outliers (allied species of S. chirayita).
Figure 3.14 Amplification profile produced by primer UBC 857. Lane X is the 1-Kb marker lane and lanes A to M represent the different collections of Swertia chirayita. The lanes N to S represent the allied species of S. chirayita.
X A B C D E F G H I J K L M N O P Q R S
Figure 3.15 Phenetic dendrogram generated by UPGMA based cluster analysis of similarity matrix generated from ISSR assay of 16 chiretta genotypes and three Swertia species procured from different commercial sources and geographical locations. Values on the X-axis represent Jaccard’s similarity coefficient values on a scale of 0 to 1.
0.00 0.25 0.50 0.75 1.00
D
E
H
I N O
B
C
F G R
K
M
L J
Q
A
S
P
CI
CII
Figure 3.16 The dendrograms obtained by using (A) complete linkage, (B) single linkage and C) WPGMA methods of cluster analysis.
0.00 0.25 0.50 0.75 1.00
1.00 0.00 0.25 0.50 0.75
0.00 0.25 0.50 0.75 1.00
S
B C
D
G
H I
L M
N O
P
R
A
F
J Q
E
K
S
B C
D
G
H
I
L
M
N
O
R
A
F
J Q
K
E
P
S
B
C
D
G
H
I
L
M
N
O
R
A
F
J
Q
K
E
P
Similarity coefficient
A
B
C
+-------------------P
+------71.5
| +-----------D
+-69.3
| | +-----------E
| | +-90.7
| +-72.8 +-------H
| |
+-94.7 +-----------I
| |
| | +-----------N
| | +100.0
| | | +-----------O
+------58.6 +------50.3
| | | +-----------B
| | +100.0
| | +-----------C
| |
| | +-----------F
| +---------------100.0
| +-----------G
|
| +-----------K
| +-99.4
| +-59.7 + -------M
| | |
| +-48.0 + -----------L
| | |
| +100.0 + --------------J
| | |
| +-44.4 + -------------------Q
| | |
+-33.4 + ----------------------A
|
| +-----------R
+---------------------37.1
+-----------S
Figure 3.17 A bootstrap tree for chiretta genotypes with bootstrap P values at each node of the cluster. The P values were obtained after 1000 replications
PC-1 -0.50 -0.17 0.15 0.48 0.80
PC-2
-0.30
0.00
0.30
0.60
0.90
Tungnath Chakrata1 Chakrata2 Jageshwar1
Jageshwar2
Kahljuni1 Kahljuni2
Kharibaowli
Devariyataal1 Devariyataal2
Sikkim1 Sikkim2
Joshimath
S. cordata
Halenia elliptica
S. purpurascens
S. paniculata
S. angustifolia 1
S. angustifolia 2
Figure 3.18 Principal component analysis of the chiretta genotypes collected from different commercial and geographical locations. The first PC axis clearly separates out the authentic S. chirayita genotypes collected from Chakrata, Jageshswar, Sikkim and Tungnath and the commercially available chiretta (collected from Khari baoli, Delhi and Kahljuni and Joshimath, Uttaranchal). The PC analysis clearly shows the distinction between authentic chiretta and the adulterated samples available in the market and also separates out the allied species distinctly.
S angustifolia 2
S angustifolia 1
S. paniculata
S. purpurascens
Halenia elliptica
S. cordata
Joshimath
Sikkim2 Sikkim1
Devariyataal2
Devariyataal1
Khari bawli
Kahljuni2
Kahljuni1
Jageshwar2
Jageshwar1 Chakrata2
Chakrata1
Tungnath
Figure 3.19 A three dimensional view of the scatter plot obtained by the principal component analysis of the chiretta genotypes. The first PC axis clearly separates out the authentic S. chirayita genotypes collected from forest areas of Chakrata, Jageshswar, Sikkim and Tungnath and the commercially available chiretta (collected from Kahri baoli, Kahljuni and Joshimath). The PC analysis clearly shows the distinction between authentic chiretta and the adulterated samples available in the market.