Documentos de Académico
Documentos de Profesional
Documentos de Cultura
of vitiligo.
The work described in this thesis was carried out by me under the
Diploma
jJ17
R. D. A. R. Gunawardena. Date.
DECLARATION
We certify that the above statement made by the candidate is true and that
this thesis is suitable for submission to the University for the purpose of
evaluation.
Dr. C. D. J~
aweera.
* t 29 SEP2008 1
CONTENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
ACKNOWLEDGEMENTS
ABSTRACT xl"
1. Introduction
1.2 Vitiligo 6
EI
2.5.4 Psoralen concentration in different batches of
3.2.5 Corylin (I 1) 69
"U
3.5 Psoralen concentration in different "Bakuchi oil' samples 94
Process 101
4. Conclusion 109
5. References 110
6. Appendix 119
lv
LIST OF TABLES
Table page
for psoralen. 43
V
15. psoralen concentration in different batches of P. cory4foia fruits. 95
17 Solubility of psoralen MI
vi
LIST OF FIGURES
Figure page
of thymine in DNA. 8
4. Psoralea corylifolia-fruit. 12
vii
19. Gas chromatogram of steam distillate of 'Bakuchi oil. 106
viii
LIST OF ABBREVIATIONS
Hz: Hertz
ix
PEG: Poly Ethylene Glycol
encouragement and advice extended towards the successful completion of this work
through many hardships. I am indebted to him for directing me towards research while
for his keen supervision, commitment, invaluable suggestions and advice given me
Queensland, Australia for providing me NMR and LC-MS data while I would like to
University of Queensland. Australia for her assistance given in obtaining all NMR and
xi
I also wish to express my thanks to the Department of Chemistry, University of
The corporation of the academic, technical and nonacademic staff of the Department of
thank goes to Mr. Sri Lal Rangoda for his assistance given me in carrying out
I also take pleasure to thank Dr (Mrs.) U.G. Chandrika, Senior Lecturer, Department of
I wish to acknowledge National Science Foundation for partial funding this work.
xii
ABSTRACT
patches on the skin. Topical application of psoralen based drugs combined with
exposure to ultra violet radiation forms the major treatment method in both the modern
considered risk free. So their use in the treatment needs specialized medical expertise
"Bakuchi oil, prepared from Psoralea corvlifolia fruits is widely used in the treatment
of vitiligo in the Ayurvedic system of medicine in Sri Lanka. The fruit of Psoralea
abundant of which are psoralen and isopsoralen. Given the photo toxicity of psoralens,
it is essential that "Bakuchi oil" oil be standardized for its psoralen content.
Towards this end a method was developed for the quantification of psoralen in
"Bakuchi oil" by TLC-FD densitometry. The method was of acceptable precision and
accuracy with a Coefficient of Variation of a 4.3 % and a recovery rate of 103% when
concentrations from 0.038 to 0.226 mg/mi compared with the reference sample which
was prepared at the BMARI which gave a concentration of 0.083 mg/ml. This wide
variation in the concentration of psoralen indicates the need for standardization and
quality control of products used in Sri Lanka. Thin layer chromatographic profiles
(finger prints) which would be useful for this purpose were developed for "Bakuchi
oil'.
The chemical analysis of the oil required the development of a method to separate
Psorcileci corvlifolici secondary metabolites from fatty materials in the oil. A successful
plant extract and the medicinal oil. The six compounds are psoralen, isopsoralen,
were determined by UV, IIR, Mass, 'H NMR, DEPT, HSQC, HMBC, and '3C NMR
spectroscopic data.
A surprising fiiiding was the fact that bakuchiol which is a major secondary metabolite
though it is quite soluble in sesame oil and is stable at 140° C, the highest temperature
xlv
The rates of incorporation of psoralen in "Bakuchi oil" during the different stages of
manufacture were studied. It was found that the preparation process currently used for
"Bakuchi oil" at BMARI is wasteful in terms of both psoralen and energy, and that
about 90% of psoralen found in the fruit is thrown away. A modification of the drug
preparation process to eliminate the water extraction stage and to directly extract the
xv
1. Introduction
growing demand for traditional systems of medicine all over the world. Traditional
medicine meets the health care needs of a large percentage of the population in
for example in India 70% of the population uses traditional medicine for primary
health care, while in Canada 70% of the population are occasional users of traditional
medicine.'
Among the various systems of traditional medicine, such as Ayurveda, Sidda, Unani
and Homeopathy, Ayurveda stands out not only as a system of great antiquity but also
an organized system with distinct aims and objectives.2 It is believed to be one of the
oldest healing sciences in existence.3 The word Ayurveda has been formed by the
combination of two words !IayulI (life) and "veda" (knowledge) meaning thereby the
knowledge of life.4
In Ayurvedic therapeutics, only natural ingredients like plants, animal products and
minerals are used while plants constitute the major source of drugs. There are many
different types of Ayurvedic formulations.5 Some of the more important ones are given
below.
Arishta and Asava: fermented aqueous medicinal preparations containing a self-
Kalka: pastes, prepared by grinding dry or fresh herbs on a stone with a mallet and
with the addition of water when necessary. These are also mixed sometimes with
honey and sugar and cooked or boiled in ghee until they are reduced to a certain
consistency.
Churna: powders, prepared by pounding dry substances in a mortar with a pestle and
Thila: medicated oils, prepared by heating vegetable oils with fresh plants and or the
During the last decade, the use of traditional and Ayurvedic medicine has expanded
globally and has gained popularity. In olden days these Ayurvedic and other traditional
the patients and the responsibility for the quality of the drug lay with the physician.
Today the commercial value of herbal medicines is high and is increasing rapidly. The
manufacture of herbal drugs and the safety and efficacy as well as quality control of
2
herbal medicines have become important concerns for both health authorities and the
public.
The WHO has emphasized the need for proper standardization of plant products using
Standardization should be done at three different levels, namely, raw materials, process
The Sri Lankan Ayurvedic pharmacopoeia and specially the Indian Ayurvedic
pharmacopoeia have provided quality parameters and standards for many plant
materials which enable the identification and quality control of them. But neither of
them gives concrete methodology for the quality control of finished drugs.
Unlike modern pharmaceuticals which contain one or at the most two active
ingredients, Ayurvedic drugs can easily contain around 100 plant secondary
metabolites, all contributing in some way according to Ayurvedic theory towards the
applied for the standardization of Ayurvedic drugs and a different approach is needed.
for evaluating traditional medicines, simple tests like foreign matter, macroscopy,
3
microscopy, ash content, extractive values in various solvents, loss on drying, volatile
components, bitterness value, tannins, swelling and foaming indexes can be used as
basic parameters in the standardization of those drugs. These tests can provide
information on the purity and overall quality of the drug, but cannot provide evidence
on the identity and potency of the drug. Only the application of chromatographic
The use of marker compounds in the standardization of herbal drugs is now well
for that plant in a given formulation. Marker compounds can be used as identity tests to
establish the presence or absence of a plant in a formulation. When used as a test for
potency, a marker compound is chosen where its concentration is related to the activity
of the drug even though it may not be the active principle' of the drug. It is assumed
that the presence of the marker compound at the specified level, indicates that the other
plant metabolites including the 'active principle have been incorporated in the
formulation. For example, formulations of St. John's Wort, which is widely used in
mild to moderate depression are usually standardized to 0.3% hypericin, even though
of detection with high separation ability. However it is not useful for the finger printing
of polar non volatile compounds. These compounds are best finger printed using HPLC
and TLC. While HPLC provides very good separations, the usual UV detectors cannot
detect non chromophoric compounds. Further, it is not possible to inject crudc extracts
without significantly lowering column life times. These problems are not present in
TLC, where a multitude of spray reagents are available for visualization of any type of
more than one sample at the same time, unlike in HPLC. Our work concentrated on
TLC as the major chrornatographic tool for finger printing and identification of
While the comparison of standard finger prints with test results is usually done
visually, computer based comparison of complex finger prints for 'sameness' and
61
1.2 Vitiligo
melanocytes.'
While vitiligo is a world wide disease and affects all the races equally, it is a
particularly troubling social problem for persons whose normal skin colour is brown or
black. The world incidence of vitiligo is generally accepted to be 1-2 %•10 The exact
cause for this disfiguring disease remains unknown. But there are several theories
use of broad spectrum antibiotics, trauma, constant pressure on the skin, dietary
Even though treatment of vitiligo is not completely satisfactory, several therapies are
used in the treatment. The most widely used method is repigmentation therapy in
which the pigment cells are restored and gains the normal colour of the skin. The most
derivatives. In the depigmentation therapy the remaining cells are also destroyed to
match the already white areas. In some cases surgical transplantation of melanocytes
rel
1.2.2 Psoralen Drugs
Psoralens are a group of photosensitizing agents which sensitize the skin to ultra violet
radiation. They are furanocoumarin compounds and originally obtained from the plants
Psoralea car litbiia (Babchi) in India and Amini ma/us in Egypt, but now are available
as synthetic products in the form of tablets for oral administration and ointments or
solutions for topical application.12 The combination of psoralens and UVA (Ultra
Violet A) radiation exposure which is commonly called PUVA (Psoralen Ultra Violet
A) therapy is widely used for the treatment of several skin diseases such as vitiligo and
psoriasis. 13
In the presence of ultra violet radiation, these psoralen type compounds can stimulate
the repigmentation process. However the exact mechanisms of the therapy are not
known. In the treatment of vitiligo it has been suggested that melanocytes in the hair
follicle are stimulated to move up the follicle and to repopulate the epidermis. '
Psoralen (1) and isopsoralen (2) (angelicin) are the main furanocoumarins found in
Psoralea corvlifolia.
Absorption of light by psoralen can lead to nir* or it- excited states. The fluorescent
and phosphorescent excited states (S1 and T1 ) have been assigned to the it- it excited
states. Psoralens show efficient intersystem crossing. The quantum yield for
fiLlorescence is small (0.0 1-0.02) and fluorescence life time is short (1 to 5 ns). Triplet
7
Photoexcited psoralens can undergo photobinding reactions, the most important
reaction being [712s + 6 cycloaddition reaction with the 5, 6 double bond of thymine
in DNA. Psoralen reacts at the 3, 4 and / or the 4, 5 it-bond of the pyrone and furan
moieties respectively.
0
HNCH3 0
0 0
40
ON 6
4
DNA CH3
DNA'
I 0
hv
\ v
0 0
0 0
O
oQj
HN
HN
NNH
0 DNA
0 DNA DNA 1(
0
Fig.! Cycloaddition reaction of psoralen with the 5, 6 double bond of thymine in DNA.
Reaction at one of the psoralen sites gives DNA mono adducts whi!e reaction at both
sites gives rise to cross !inked DNA strands. In reaction with DNA, psoralen
intercalates with DNA in the dark, and irradiation at 400 nm leads to furan side mono
adduct formation, while irradiation at 350 nm leads to cross linking 17 (see Fig.1)
While the photo induced cyclization reaction can be expected to proceed rapidly via
the singlet excited state, there is evidence that the triplet state is the reaction precursor
fatty acids and lecithins in lipid membranes. They can also induce DNA-protein cross
links.'9 The exact role played by psoralen in the repigmentation process is not clear.
However, DNA cross linking in the presence of ultra violet A radiation appears to be
significance that only the linear furanocoumarins such as psoralen, Xanthotoxin (8-
methoxy psoralen) and Bergapton (5- methoxy psoralen) form interstrand cross
diseases it is a treatment that is still under investigation and there are serious concerns
regarding its potential long term hazards of cataract formation, accelerated aging of the
skin and development of skin cancers. Apart from that psoralen drugs are known to be
furanocoumarin containing herbs cannot he considered risk free. So their use in the
"Bakuchi oil" which is prepared from the fruits of Psoralea corylifolia and sesame oil
is widely used in the treatment of vitiligo in the Ayurvedic system of medicine in Sri
Lanka. Preparation method of this medicinal oil has been introduced to Bandaranayake
medicinal oil has been used to treat vitiligo at the hospital of the BMARI and other
Ayurvedic hospitals in Sri Lanka. In the treatment this oil is applied on the affected
area of the skin with a cotton bud and the patient is then exposed to mild sun for 1 hour
with sesame oil until the water is evaporated completely. Powdered fruits are added
Normally stainless steel vessels are used for the preparation of the oil and the heat is
supplied by gas cookers. (Fig.2) After adding the powder of fruits to the oil, it is heated
mildly while stirring manually. (Fig.2). Generally the maximum temperature achieved
during this process lies between 135° C and 145° C. Finally the oil is filtered through a
10
cloth. Typically at the pharmacy of the BMARI 100 bottles batches are prepaied and
hr
&!'kt.
Fig.2 Final stage in the large scale preparation of "Bakuchi oil'.
The fruits of Psoialea corylifb/ia which are used to prepare "Bakuchi oil" contains
photoactive. As there are serious concerns regarding the potential long term hazards of
these compounds it is important to standardize this medicinal oil specially for its
the overall compostion (and therefore the efficacy of the drug), as well as an "active
_e
.1
4
kftr
.4
India.2 It is an erect annual herb, bearing one seeded fruits which have a dark brown,
sticky, resinous pericarp (Fig.3and 4). Since ancient times, these fruits have been used
China.27 These fruits have been widely used for the cure of gynaecological bleeding,
29, 30.
leprosy and specially in certain skin diseases such as vitiligo and psoriasis.28'
Some therapeutic properties of these fruits such as laxative, diuretic, and anthelmintic
12
1.3.1 Chemical constituents of Psoralea corylifolia.
Very little work on Psoralea car lifolia had been done until a careful examination of
seeds and some clinical trials were carried out in the Calcutta school of Tropical
Medicine in 1923. As the results of its local application in leucoderma was satisfactory,
a considerable amount of interest in the further study of this plant was aroused.32 Since
then the fruits have been examined from time to time and isolation of variety of
flavanones, isoflavones and terpenoids have been reported.33 Additionally, a fixed oil
and an essential oil have also been isolated from the fruits.3 '
Compounds reported from the fruits of P.corylifolia and the relevant references are
Compound References
Coumarins
Psoralcn (1) 34, 35, 36, 15, 32, 38, 39.
Isopsoralen (2) 34, 35, 36, 15, 38.
Bakuchicin (3) 27
Neopsoralen (4) 40
Coumestans
Psoralidin (5) 35
Corylidin (6) 41
Bavacoumestan A (7) 42
Bavacoumestan B (8) 42
13
Table 1 cont.....
Compound References
Sophorocoumestan A (9) 42
Psoralidin oxide (10) 43
Flavonoids
Corylin (11) 28, 33, 35, 44.
Psoralenol (12) 45
Neobavaisoflavone (13) 28, 46, 47, 48.
Corylinal (14) 49
Bavachin (15) 29, 44, 46, 50.
Isobavachin (16) 47, 51
Bavachinin (7-0-methylbavachin) (17) 28, 29, 33, 51.
4'-methoxy flavone (18) 26
Corilifols A (19) 28.
Corilifols C (20) 28
Isoneohavaisoflavone (21) 28,46
Erythrin A (22) 28
8- prenyldaidzein (23) 28
Daidzein (24) 27
7, 8-dihydro-8-(4-hydroxyphenyl)-2,2- 28
dimethyl-2H, 6H-benzo[ 1 ,2-b:5,4-b']dipyran-
6-one. (25)
Chalcones
Corilifols B (26) 28
Brosimacutin 0 (27) 28
1 -[2,4-dihydroxy-3-(2-hydroxy-3-methyl-3 '- 28
butenyl)phenyl]-3-(4-hydroxyphenyl)-2-
propen- 1-one. (28)
Bakuchalcone (29) 28, 50
Isobavachalcone (30) 28, 29, 46, 47, 51.
14
Table 1 cont.... .
5'-formyl-2', 4-dihydroxy-4'-
methoxychalcone(31)
Isobavachromene (32) 46.
Bavachromene (33) 46.
Isoneobavachalcone (34)
Bavachromanol (35)
Bavachalcone (36) 28,51
4'0 methylbavachalcone (37) 46.
Terpenoids
Bakuchiol (38) 27, 29, 30, 35, 55, 56, 57, 58, 59, 60.
2,3- Epoxybakuchiol (39) 6 1,49.
Coumarins
cax CI u
0 0 0
9
0n~'~
Bakuchicin (3) Neopsoralen (4)
15
Coumestans
HO 0 0 >OOO
HO
OH
H I]
Psoralidin (5) Corylidin (6)
-
JH
Flavonoids
I
I 10~ 1 0
H
OH c 0
Neobavaisoflavone (13) Corylinal (14)
16
OH
HO 0 HOZyO1
Al
Corilifols A (19) Corilifols C (20)
HO 0
OH
Isoneobavaisoflavone (21) Erythrin A (22)
HO
HO O
OH
o OH
8- Prenyldaidzein (23) Daidzein (24)
17
OH
0
7,8-dihydro-8-(4-hydroxyphenyl)-2,2-
dimethyl-2H, 6H-benzo[ 1 ,2-b:5,4-
b]dipyran-6-one. (25)
Chalcones
OH A
OH OI f JIIIITIII OH
0
Corilifols B (26) Brosimacutin G (27)
. OH
1- [2,4-dihydroxy-3 -(2-hydroxy-3-
methyl-3' -butenyl)phenyl ]-3 -(4-
B akuchalcone (29)
hydroxyphenyl)-2-propen- 1-one. (28)
0(7)ZIIJ1
OH
OCH3)
OHC
0
5'-formyl-2', 4-dihydroxy-4'-
Isobavachalcone (30)
methoxychalcone (31)
OH
çOH
OH
HO)OH
OHC ::
0 101
Isoneobavachalcone (34) Bavachromanol (35)
OMe
Terpenoids
HOX<" HO
Bakuchiol (38) 2,3- Epoxybakuchiol (39)
/1 OH
X
HO < OH
HO
1 ,3-Hydroxybakuchiol (40) 3,2- hydroxybakuchiol (41)
19
1.3.2 Biological activities of Psoralea corylifolia.
A great variety of biological activities have been reported for the fruit extract of
Psoralea corvlifolia and for some of the compounds isolated from the fruits. (In the
literature, very often reference is made to the seed of Psoralea corvlifolia. It is a single
seeded fruit of which the pericarp is tightly attached to the seed coat and it is most
likely that the whole fruit has been used in these studies).
due to their ability to cross link the strands of DNA on photo activation. As the double
bond in the furan ring in the angular furanocoumarin, isopsoralen, is less effective in
forming such cross links, it shows comparatively less phototoxicity. Further psoralens
are reported to be insect antifeedant. It has been shown that the antifeedant activity of
furanocoumarin) and the reported EC 50 values were 170 and the 616 ppm
respectively.38
The use of psoralen in the treatment of AIDS and cancer has also been investigated.62
The phenolic compound bakuchiol (38) is another major bioactive compound found in
Porphvroinonas gin givalis (with MIC, ranging from 1 to 4 .ig/mI and sterilizing
concentrations for 15 min ranging from 5 to 20 tg/ml) It has also been effective
against adherent cells of Streptococcus inutans in water soluble glucan in the presence
of sucrose and also capable of reducing the pH in the broth. Therefore it has been
antibacterial agents against oral pathogens in humans and has a great potential for use
in mouthwash preparations for preventing and treating dental caries. Further, it could
be a useful antibacterial agent to be used as a food additive. (for candy and chewing
gum).58
I-Iyun-ock Pac et al. has shown that bakuchiol can inhibit the expression of inducible
bakuchiol would be a potential drug to treat inflammatory diseases due to iNOS gene
over expression.57
Bakuchiol (38) has also been reported to possess a variety of other activities such as
21
Bapat et al. has shown that the cytotoxicity of bakuchiol (38) could enhanced upon
radioiodination. Further they showed this radioiodinated bakuchiol could augment its
toxicity to the tumor cells and they suggest the 1251 bakuchiol could be studied with
various tumor cell lines of human origin to test its efficiency as a therapeutic agent.59
Although bakuchiol has also been listed as skin whitening agent in a Japanese patent
whole cell bioassay specific for inhibitors of DNA replication enzymes. Further studies
led to the isolation and identification of two compounds, namely, bakuchiol (38) and
DNA polymerase inhibitor found in Psoralea corylifolia. Apart from that daidzeiri was
topoisomerase II inhibitor.27
seed powder and its extracts have been examined by Gue et al. and shown that it has
for food. Six compounds were isolated and their antioxidant activities have been
studied, Psoralidin (5), bakuchiol (38), and corylin (11) had strong antioxidant
properties while psoralen and isopsoralen had no antioxidant properties. Further, the
antioxidant activity of psoralidin (5) was found to be even stronger than BHT
22
Hiroyule Ct al. has found that Psoralea seed extracts showed potent inhibition of
yielded bakuchiol (38), bavachin (15), isobavachin (16) and isobavachalcone (30). Out
microsomes. It also showed bakuchiol protected human red blood cells against
Psoralea corv/ifb/ia has shown remarkable antibacterial activities against two major
extract was fractionated sixteen compounds could be isolated and out of them nine
The aqueous extract of Psoralea corvlifolia has been found to have hepatoprotective
properties. An experiment which led to isolate the active compounds, yielded one
23
compounds psoralen (1) and bakuchicin (3)30 This has to be investigated further,
In a study of naturally occurring antiplatelet agents, it has been shown the methanol
extract of the seeds of Psoralea corylifolia had inhibited the aggregation of rabbit
platelets induced by arachidonic acid, collegen and platelet activating factor. Further
Psoralea corv/iftilia seed extract has been suggested to be useful as a remedy for bone
fracture, osteomalacia, and osteophorosis.28 Wang et al. have reported that the fruit
extract exhibited osteoblastic proliferation stimulating activity in UMR 106 cell line
cultured in vitro, and corylin (11) and bavachin (15) have been isolated as the active
principles. Further they suggest the fruit extracts, corylin (11) and bavachin (15) might
Prasad et al. showed that Psora/eci corvlifolia is a promising antifungal species and 4'-
methoxy flavone (18) was isolated as the principle antidermatophitic compound from
the methanol extract of fruits.26 According to Gupta et al. seeds of Psoralea corvliftlia
mentagrophyres and Candida albica11s.26 Further the essential oil of the fruits has been
24
As reported above, Psoralea corylfolia Contains a number of compounds which
exhibit different types of biological activities and are rich with therapeutic properties.
Further studies on these compounds and their properties may help to improve the
utility of Psoralea corylfo1ia as a therapeutic agent and that knowledge can be used as
OR
1.4 Scope of the work described in this thesis.
Chemical characterization and standardization of the drug, 'Bakuchi oil" and the effect
from Psoralea corylifolia fruits that were incorporated in it. This information was used
quantitatively by setting limits to its psoralen content. The manufacturing process was
studied to determine how it influenced the composition of the drug with respect to
2.1 Materials
Plant materials.
Dried fruits of Psoralea corylifolia were purchased from a local drug store and their
BMARL
Sesame oil was purchased from a local drug store and tested for the common
65
adulterant, cotton seed oil, before use.
Bakuchi oil.
A sample of Bakuchi oil prepared at the BMARI (BMARI-I) was used as the reference
sample.
H NMR and 13 C NMR spectra were measured at 600 MHz and 150 MHz respectively
(equipped with Ez-oinic software). Ultra Violet spectra were recorded on a Thermo
27
Spectronic HeXIOS a double beam spectrophotometer. El mass spectra were measured
column was DB wax, 30 in, 0.25 mm, O.SMm). Electrospray ionization mass spectra
(ES[MS), using both flow injection analysis (FIA) and liquid chromatography-diode
separations module quaternary (G131 1A) pump (G1312A) with vacuum degasser
Carrier gas: He
Oven temperature programming -1: initial temp.: 30° C, held for 3 mm., and raised
up to 255° C at a rate of 2°C per mm. and kept for 15 mm. at that constant temperature
(255° C)
Oven temperature programming -2: initial temp.: 60° C, raised up to 255° C at a rate
of 10°C per mm. and kept for 17 mm. at that constant temperature (255° C)
Densitornetric studies were carried out on a Desaga CD 60 HPTLC densitometer using
MERCK silica gel 60 precoated plates. All the densitometric scannings were carried
Melting points were recorded on a Reichert Thermover hot stage melting block
Precoated (Merck) analytical plates with and without fluorescent indicator (silica gel
60 F254 and silica gel 60) were used for thin layer chromatography (TLC). Preparative
plates were prepared in the laboratory using Merck silica gel PF251. Layer thickness of
Normal gravity column chromatography was carried out using silica (70-230 mesh)
and flash column chromatography was carried out with silica gel 60 (Fluka 60738).
Plates for the chromatotrone (radial chromatography) were prepared with silica gel 60
F254 (Merck).
29
All solvents used were either analytical grade or were general purpose reagent grade
Psoralea corylifolia.
168
Procedure
800 g of dried whole fruits were extracted in a Soxhlet apparatus with petroleum ether
(60-80° C) for 12 hours, the seeds were then coarsely ground and the powder was
soaked in water for 8 days, dried in the sun and extracted in a Soxhlet apparatus with
petroleum ether for 6 hours. This extract was concentrated to a small volume and kept
suction filtration and the precipitate was washed with a little petroleum ether to remove
adhering coloured impurities to obtain a mixture of the two isomers as a creamy white
Procedure II
Powdered fruits (225.0 g) were extracted with hexane in a Soxhlet apparatus for 12
hours, concentrated to a small volume and kept in the cold. The precipitated
precipitate was washed with a little hexane to remove adhering coloured impurities to
30
obtain a mixture of the two isomers as a creamy white coloured amorphous mass (1.51
g. 0.67 %).
The mixture of the two isomers (0.893 g) were chromatographed over a silica gel
ethyl acetate in hexane, followed by psoralen. The elution of psoralen was completed
when the concentration of ethyl acetate was 8 %. There was a slight overlap of the two
bands. Therefore the middle fractions which contained both compounds were
discarded. The eluting fractions containing only isopsoralen yielded 32 % (364 mg) of
the compound, the late fractions containing only psoralen gave 35 % (402 mg) of the
The mixture of the two isomers (86 mg) were also separated by radial chromatography
using a gradient of ethyl acetate in hexane. Isopsoralen (20 mg, 23 %) eluted first at 4
% ethyl acetate in hexane followed by psoralen (28 mg, 33%) eluted at 5-6 % ethyl
acetate in hexane.
The two compounds were detected on TLC plates (silica gel 60254) as yellow and
yellowish green fluorescent spots respectively at 365 nm after spraying with 10%
methanolic potassium hydroxide. Psoralen and isopsoralen had R1 values of 0.55 and
0.58 respectively, when the solvent was hexane: ethyl acetate (4:1) (and developed
without presaturating the tank). Isopsoralen was recrystallized from hexane and
31
compounds were checked using different TLC systems, (hexane: ethyl acetate, 4: 1L,
toluene: ethyl acetate: hexane, 1: 3: 9, Toluene: ethyl acetate, 5: 2), melting points and
GC-MS.
2Lniax(CHC13): 250, 290, 333; JR vrnax(KBr)cm1: 3165, 3120, 3060, 2920, 2855, 1722,
1679, 1634, 1577, 1542, 1449, 1390, 1285, 1160, 1135, 1103, 1022, 925, 895, 848,
840, 824, 750, 761, 750, 602, 550, El: mlz: 186[MI, 158, 187, 102, 130, 63, 'HNMR
(CDCh 300 MHz) 8 : 6.37 (lH, d, J = 9.3 Hz, H-3), 6.83 (1H, dd, J = 2.3 Hz, 1.0 Hz,
H-2'), 7.45 (1H. m, H-8), 7.68 (1H, bs, H-5), 7.69 (1H, d, J= 2.3 Hz, H-3'), 7.78 (1H, d
of rn, J= 9.3 Hz, H-4), '3C NMR (CDC13 75 MHz) & 160,99 (C-2), 156.45 (C-7),
152.06 (C-9), 146.92 (C-12), 144.06 (C-4)1 124.89(C-6), 119.86 (C-5), 115.43 (C-10),
138° C ), UV Xniax(CHCI3) : 250, 299. IR viiia x(KBr)cm1: 3165. 3125, 3065, 2924, 2855,
1709, 1626, 1615, 1335, 1272, 1132, 1123, 1057, 1041, 832, 742, 470, El: mlz:
1 86IMI, 158, 130, 102, 49, 187, 'HNMR (CDC13 300 MHz ) : 6.39 (1H, d, J = 9.6
Hz, H-3), 7.13 (IH. dd, J = 2.3,1.0 Hz, H-3'), 7.37 (IH, d, J = 8.4 Hz, H-5), 7.43 (
1T-I,dd, J = 8.4 Hz,1.0 Hz, H-6), 7.69 ( 1H, d, J = 2.3 Hz, H-2'), 7.80 (1H, d, J = 9.6
Hz, H-4), 13C NMR (CDC13 75 MHz) ö: 160.80, 157.39, 145.89, 144.49, 123.83,
32
2.3.2 Bakuchiol (38)
using hexane and methanol. The solvents were evaporated under reduced pressure to
yield semisolids, (21.15 g and 60.62 g), 11.5 g of the hexane extract was repeatedly
stirred with 80 % MeOH (10 x 75 ml) for 12 hours. The 80% MeOH soluble fraction
of the hexane extract was partitioned with hexane. Evaporation of the solvent from the
hexane fraction resulted in a sticky residue (2.05 g). The residue (0.630 g) was
subjected to column chromatography over silica gel. Elution was carried out with a
gradient of ethyl acetate in hexane. The fraction eluting with 4% ethyl acetate in
hexane yielded bakuchiol with a red pigment as an impurity. The red pigment was
removed by subjecting the sample to preparative TLC (toluene: ethyl acetate: acetic
acid, 10: 1: 0.1, triple development), and pure bakuchiol (39 mg, 0.10 %) which
appeared as a dark blue spot under excitation at 254 nm on TLC, was obtained as a
Bakuchiol (38):
ESIMS: mlz: 255[M-H], 257[M+H], 'H NMR ( CDC13, 600 MHz) 6: 1.20 (3H, s, H-
16), 1.50 (2H, m, H-b), 1.59 (3H, s, 1-1-15 ), 1.69 (3H, s, H-14), 1.96 (2H, m, 7.8, H-
11), 5.02 (1H, d, 1=17.4 Hz, H-18a), 5.05 (1H, d, 1=10.5 Hz, H-18b), 5.11 (IH, t,
J=7.2 Hz, H- 12), 5.88 (1H, dd,J=17.4 Hz, 10.5 Hz, H-17), 6.07 (1H, d,J16.2 Hz, H-
8), 6.26 (1H, d, 1=16.2 Hz, H-7), 6.78 (2H, d, J8.4 Hz, H-3,5), 7.26 (2H, d, J= 8.4
Hz, H-2,6), 13C NMR ( CDC13, 150 MHz) 6: 154.58 ( C- 4),145.93 (C-17), 135.85 8)
33
(C-8), 131,30 (C-13), 130.89 (C-i), 127.35 (C-2,6 ), 126.44 (C-7), 124.78 (C-12 ),
115.34 (C-3,5), 111.86 (C-18), 42.50 (C-9), 41.27 (C-lU), 25.69 (C-14), 23.33 (C-16),
The 80% methanol insoluble fraction (1.930 g) of the hexane extract of powdered
fruits (5.27 g) (section 2.3.2) was subjected to column chromatography over silica gel
using a gradient of ethyl acetate in hexane to yield 17 fractions, and the fraction 8
eluting with 2-3% ethyl acetate in hexane yielded fixed oil. (1.13g. 2.5 %)
The fractions 10 and 11 eluted by 3 - 4 % ethyl acetate in hexane were combined and
further purified by preparative TLC (toluene: ethyl acetate, 12: 0.5) to yield a red
coloured pigment and spectral analysis showed that it was a mixture of compounds.
'H NMR: (CDC13 300 MHz) 6:0.89(m), 1.28 ( m). 1.60 (m), 2.03 (m), 2.30 (in), 2.78
(30).
Dried whole fruits (150 g) were ground to a powder and extracted successively with
hexane and methanol in a soxhlet apparatus and the solvents were evaporated under
reduce pressure to yield semisolids. The methanol extract (10 g) was subjected to silica
gel column chromatography using a gradient of ethyl acetate in hexane, which yielded
34
35 major fractions. TLC analysis (silica/ toluene: ethyl acetate: hexane, 1: 3: 9) of
fraction 14 (32 mg) which eluted with 10 % ethyl acetate in hexane, indicated the
presence of a purple fluorescent spot (0.41) overlapping with another blue fluorescent
spot, under 366 nm UV light. Development of the same plate 3 times with the same
solvent resulted in a separation of the two spots. The purple fluorescent spot was
pale yellow crystalline solid which was established as a new natural product,
dehydroisopsoralidin (42) (4 mg, 3x10 3 %). The analysis of fraction 13 revealed that it
of acetone. Fraction 17 which eluted with 15 % ethyl acetate in hexane yielded the
known compound corylin (11) as a white crystalline solid (21 mg, 1x10 2 %). The fine
was eluted with 20 % ethyl acetate in hexane was separated by centrifugat ion and
further washed with dichloromethane to obtain the pure compound. psoralidin (5) as a
white crystalline solid (14 mg, 9x10 3 %). The supernatant of fraction 23 and fraction
24 which eluted with 20% ethyl acetate in hexane were combined and further subjected
obtain more psoralidin (5) (9 mg, 6x10 3 %)) and isobavachalcone (30) as yellow
iit7 in. p. 292-294° C), ESIMS: mlz: 333[M-H]-, 335IM+H1, 'H NMR (d6-DMSO,
600 MHz) 3: 1.45 (6H. s), 5.91 (1H, d, J= 9.9 Hz), 6.63 (1H, d, J = 9.9 Hz), 6.96 (11-1,
35
dd, J= 8.4 & 2.0 Hz), 6.97 (1H, s), 7.17 (1H, d, J= 2.0 Hz), 7.71 (1H, d, J = 8.4
Hz), 7.77 (1H, s), '3C NMR (d,-DMSO, 150 MHz) 6: 157.2 (C-2), 102.9 (C-3), 159.0
(C-4), 118.5 (C-5), 118.7 (C-6), 155.9 (C-7), 104.2 (C-8), 153.9 (C-9), 105.6 (C-b),
114.5 (C-l'), 156.0 (C-2'), 98.6 (C-3'), 157.3 (C-4'), 114.1 (C-5'), 120.4 (C-6'), 120.8
33m.p.238-2390 C), JR vmax(KBr)cm: 3250, 2975, 1628, 1575, 1500, 1375, 1275,
1268, 1240, 1190, 1120, 960, 870, 830, 720, 560, ESIMS: m/z: 319[M-H],
321[M+H], 'H NMR (d6-DMSO, 600 MHz) 6: 1.39 (6H, s, H-6"Me), 5.78 (IH, d, J
9.0 Hz, H-5"), 6.43 (1H, d, J= 9.0 Hz, H-4"), 6.78 (1H, d, J 8.1Hz, H-5'), 6.86 (11-1,
d, J= 2.2 Hz, H-8), 6.93 (111, dd, 1=8.7 & 2.2 Hz, H-6), 7.28 (1H, d, 1=2.0Hz, H-
2'), 7.29 (1H, dd,J= 8.1 & 2.0 Hz, H-6'), 7.96 (1H, d, 1=8.7Hz, H-5), 8.34 (IH, s,
H-2), 13 C NMR(d6-DMSO, 150 MHz) 6: 174.5 (C-4), 162.7 (C-7), 157.44 (C-9), 153.1
(C-4'), 152.1 (C-2), 131.2 (C-5"), 129.6 (C-6'), 127.2 (C-5), 126.9 (C-2'), 124.4(C-i'),
123.09 (C-3), 121.7 (C-4"), 120.5 (C-3'), 116.4 (C-b), 115.5 (C-5'), 115.2 (C-6),
Psoralidin (5): White crystalline solid; JR vjnax(KBr)cni': 3450, 2930, 1720, 1635,
1600, 1580, 1510, 1420, 1370, 1260,1100, ESIMS: mlz: 335[M-H], 337[M+H], 'H
NMR (d(,-DMSO,600 MHz) 6: 1.71 (3H, s), 1.74 (3H, s), 3.32 (2H, d, 1=7.2Hz), 5.35
36
(IH, t,1 = 7.2 Hz), 6.92 (1H, s), 6.93 (11-I, dd,J= 8.4 Hz& 2.0 Hz), 7.16(1H, d,J
2.0 Hz), 7.62 (11-1, s), 7.68 (IH, d,.J= 8.4 Hz), 10.09 (1H, s), 10.75 (1H, s).
C) JR viiiax(KBr)cm': 3415, 2930, 2850, 1630, 1600, 1560, 1540, 1508, 1460, 1380,
1320, 1295, 1240, 1170, 1008, 1040, 830, 670, 625, ESIMS: m/z: 323[M-H]-,
325[M+H], il_I NMR (d6-DMSO,600 MHz) ö: 1.62 (3H, s, H-4"), 1.72 (3H, s, H-5"),
3.22 (2H, d, 1=7.2 Hz, H-i"), 5.17 (1H, t, 1=7.2Hz, H-2"), 6.46 (1H, d, J = 8.9 Hz,
H-5), 6.835 (2H, d, J = 8.6 Hz, H-5',3'), 7.718 (IH, d, J=15.6 Hz, H-b), 7.77 (11-I, d,
1=15.6 Hz, H-c), 7.749 (21-I, d, 1= 8.6 Hz, H-6',2'), 8.03 (1H, d, J = 8.9 Hz), '3C NMR
(d(,-DMSO, 150 MHz) & 174,5 (C-4), 162.7 (C-7), 157.4 (C-9),153.1 (C-4'), 152.1 (C-
2), 131.2 (C-5"), 129.6 (C-6'), 127.2 (C-5), 126.9 (C-2'), 124.4 (C-il), 123.09 (C-3),
121.7 (C-4"), 120.5 (C-3'), 116.4 (C- 10), 115.5 (C-5'), 115.2 (C-6), 102.1 (C-8), 76.2
2.3.5 Extraction and GC finger printing of essential oil of the fruits of Psoralea
cory1folia.
Whole fruits (75 g) were hydro distilled using a Clavenger apparatus for 12 hours.
The oil layer was separated out and dissolved in 10 ml of dichloromethane, it was dried
over anhydrous sodium sulfate and the solvent was evaporated to yield 55 mg (0.074
%) of essential oil.
Powdered fruits (75 g) were hydro distilled using a Clavenger apparatus for 12
hours. The oil layer was separated out and dissolved in 10 ml of dichioromethane, it
37
pq
was dried over anhydrous sodium sulfate and the solvent was evaporated to yield 61
(3) 100 g of powdered fruits was steam distillated for 10 hours. The distillate (600 ml)
was collected and extracted with dichloromethane. The dichioromethane extract was
dried over anhydrous sodium sulfate and the solvent was evaporated to yield 82 mg
3.0 mg of essential oil was dissolved in 5.00 ml of dichlorometharie and 1.00 j.tl was
The medicinal oil 'Bakuchi oil" (30 ml) was extracted with methanol (8 x 60 ml),
solvent was evaporated under vacuum and the methanol extract of the oil (4.21 g) was
obtained. The methanol extract (3.8 g) was partitioned between 80% methanol and
hexane, The 80% methanol soluble fraction was separated out and the solvent was
evaporated under reduced pressure. This extract was analyzed for the presence /
methods.
I. The 80% methanol soluble fraction of the methanol extract of the "Bakuchi oil"
(850 mg) was subjected to colunm chromatography over silica gel using hexane
FACULTY OF APPLIED SCIENCE
Dean
Faculty of Graduate Studies
University of Sri Jayewardenepura
The above thesis had three supervisors, myself, Prof. Karnal Gunaherath
and Dr. Champa Jayaweera.
Although Prof. Kamal Gunaherath and I have each received a copy of the
thesis and I have been informed that Dr. Champa Jayaweera has not
received her copy.
compounds in the Bakuchi oil" co-TLC were carried out on the same plate)
IL The 80% methanol soluble fraction of the methanol extract of "Bakuchi oil"
III. The 80% methanol soluble fraction of the methanol extract of the "Bakuchi oil"
Psoralea corylif b/ia fruits using the solvent system toluene: ethyl acetate:
potassium hydroxide, at 254 nm without spraying and in the visible light after
(1) Bakuchiol (1 mg) was dissolved in 1.00 ml of dichlorornethane and 1.00 p1 of the
39
Psoralen (1 mg) was dissolved in 1.00 ml of dichloromethane and 1.00 jil of the
Hexane extract of powdered fruits (30 mg) was partitioned between 80% methanol
and hexane. The 80 % methanol fraction was separated and the solvent was removed
dichloromethane and 1.00 tl of the solution was injected into GC. (Temperature
programming 1)
Powdered fruits (100 g) were steam distilled forl8 hours, steam distillate was
sodium sulfate and the solvent was evaporated to obtain 82 mg of essential oil, the
essential oil (3 mg) was dissolved in 5.00 ml of dichloromethane and 1.00 t1 of the
"Bakuchi oil" (60.00 ml) was steam distilled for 4 hours, steam distillate was
sodium sulfate and the solvent was evaporated to obtain 3 mg of a volatile oil, 2 mg of
the volatile oil was dissolved in 2.00 ml of dichloromethane and 1.00 p] of the solution
Powdered fruits (8.00 g) were heated with 120 ml of sesame oil (140-145° C) for
three hours, filtered through muslin cloth, sesame oil extract (60 ml) was steam
40
dichloromethane fraction was dried over anhydrous sodium sulfate and the solvent was
evaporated to obtain 8 mg of a volatile oil. Volatile oil (2 mg) was dissolved in 1.00 ml
of dichloromethane and 1.00 p.1 was injected into GC. (Temperature programming 1)
The seed residue was steam distilled for 4 hours to obtain 7 mg of essential oil, 2 mg of
the essential oil was dissolved in1.00 ml of dichloromethane and 1.00 p.1 was injected
Bakuchiol (10.0 mg) was mixed well with 1.5 ml of sesame oil and 0.1 ml of the
solution was withdrawn and dissolved in 1.50 ml of dichloromethane, 1.00 p.1 was
injected into GC, remaining sesame oil fraction was heated around 145° C for three
hours, 0.1 ml was withdrawn and dissolved in 1.50 ml of dichloromethane 1.00 p.1 was
psoralen was dissolved in 5.00 ml of dichloromethane. Equal volumes (1.00 ml) from
each were mixed well and 1.00 p.1 of the solution was injected into GC. (Temperature
programming 2)
"Bakuchi oil" (15 ml) from the reference sample (BMARI-1) was extracted with methanol
(15 ml x 5), sovlent was evaporated under reduced pressure, and the residue was
partitioned between 80% methanol and hexane. The 80% methanol soluble fraction was
evaporated under reduced pressure, and the residue dissolved in methanol and spotted oil
41
precoated TLC plates (silica gel 60), developed in three different solvent systems (toluene:
ethyl acetate, 5: 2, hexane: ethyl acetate, 4: 1, toluene: ethyl acetate: hexane, 1: 3: 9) and
observed under Ultra Violet light (335 mn) after spraying with 10% methanolic potassium
hydroxide.
The plates were subjected to densitometric scannings. Measurements were carried out in
the remission mode and the fluorescence was measured at 366nm using a filter of 470
nm.
Five other oil samples of different production batches, collected from different sources
Ayurvedic Drugs Corporation) were analyzed in the same way and compared with the
Pure psoralen isolated from the fruits of Psoralea cory/ift/ia was used to prepare the
standard solutions. Stock solution was prepared by dissolving 5.00 rng of psoralen in
25.00 ml of acetone. Standard solutions containing 40, 80, 120, 160 and 200 ng/sl
were prepared by diluting the stock solution. From each solution 5.00 1iJ was applied
on precoated TLC plate (MERCK, silica gel 60 G) in duplicate with a Hamilton micro
syringe. The plate was developed in hexane: ethyl acetate, 4: 1 (solvent front 8.0 cm)
after saturating the chamber for two hours. The plate was dried with a stream of hot air
and psoralen was observed at R1 = 0.62 as a blue fluorescent spot under excitation at
366 nm. Densitometric scanning was carried out with excitation at 240 nm and the
fluorescence was measured using a filter of 370 nm. (Table-2). The standard curve (Y=
Table 2: Densitometric readings obtained for the standard curve for psoralen.
1 200 3069.402
2 400 5207.474
3 600 6336.803
4 800 7896.882
5 1000 9009.066
The reference sample of "Bakuchi oil" (4.00 ml) was dissolved and mixed well with 50
ml of acetone and 30 ml of methanol was added. The mixture was cooled at -10 - 15
C° for three days for precipitation of fats. The supernatant was decanted and the
precipitate was washed with cold methanol. The washings and the supernatants were
combined and the solvent was evaporated completely under vacuum. The residue was
dissolved in acetone and made up to 5.00 ml and 5.00 pJ of the solution was applied on
the TLC plate in duplicate along with a standard solution of psoralen (standard.3.120
ng/J). The plates were developed and scanned as described for the standard curve. The
43
2.5.2.3 Method validation
Precision.
Six replicates of the reference sample were prepared separately as described in the
section 2.5.2.2 and 5.00 il from each final solution was applied in duplicate on the
same plate, the plates were developed and densitometric scanning was carried out as
described for the standard curve (2.5.2.1.). The coefficient of variation was calculated
from the readings obtained for the six replicates (Table 3).
Table 3: Densitometric readings obtained for the psoralen spot in six replicates of the
2 3958.106
3 3767.977
4 3702.435
5 3823.687
6 3829.000.
Accuracy (Addition-Recovery)
A sample of the reference oil was analyzed for its psoralen content as described in
section 2.5.2.2 and the densitornetric reading obtained is given in the Table 4.
5.0 rng of psoralen was weighed accurately, dissolved in the reference oil and made
Psoralen concentrations of the reference oil sample before and after addition of
psoralen were determined using the standard curve and the percentage recovery was
calculated.
Table 4: Densitornetric readings obtained for the psoralen spot in the addition recovery
experiment.
Each oil sample (4.00 ml) was dissolved in 50 ml of acetone and proceeded as
described in section 2.5.2.2, 5.00 il from each was applied on the TLC plate in
45
duplicate along with a standard solution of psoralen (std.3), plate was developed and
scanned as described for the standard curve and the desitometric readings obtained for
Table 5: Densitometric readings obtained for the psoralen spot in different samples of
"Bakuchi oil".
Psoralen concentration of each oil sample was determined using the standard curve.
1.00 g of powdered fruits from five different batches were Soxhlet extracted separately
with chloroform (50 ml) for 10 hours, chloroform was evaporated completely under
reduced pressure, residue was dissolved in acetone and made up to 25.00 ml, 5.00 Ml
from each was applied on a precoated TLC plate along with a standard solution of
psoralen (std.3) The plate was developed and scanned as described for the standard
curve and the densitometric readings obtained for each batch of fruits are given in the
table 6.
Table 6: Densitometric readings obtained for the psoralen spot in different batches of
1 6347.45
2 5976.66
3 4594.25
4 7435.38
5 6128.97
Psoralen concentration of each batch of Psoralea fruits was determined using the
standard curve.
2.5.5 Psoralen concentration in sesame oil extract of seed powder and the
extraction efficiency.
10.00 g of powdered fruits (batch-1) were extracted with 150 ml of sesame oil in a
beaker heated over a bunsen flame (140-150°C) for 3 hours, filtered and the sesame oil
extract was obtained. The oil extract (4.00 ml) was dissolved and mixed well in 50 ml
of acetone and proceeded as described in the section 2.5.2.2. Plates were developed
and scanned as described for the standard curve. Densitometric reading obtained is
47
Psoralen concentration of the sesame oil extract of seed powder was determined using
Psoralen was added little by little into 6 ml of Sesame oil and stirred well with a
been added an excess of undissolved psoralen could he observed. Then the addition
was stopped and stirring was continued for 48 hours. The solution was centrifuged and
decanted carefully to obtain a sample of Sesame oil saturated with psoralen , 4.00 ml of
that was dissolved and mixed well in 50 ml of acetone, and proceeded as in the section
2.5.2.2, after evaporating the solvent from the combined supematants residue was
dissolved in acetone and made up to 100.00 ml and 5.00 j.il was applied on the TLC
plate in duplicate along with a standard solution of psoralen, The plates were
developed and scanned as described for the standard curve. Densitometric reading
Psoralen concentration of Sesame oil saturated with psoralen was determined using the
standard curve.
2.5.7 Psoralen concentration in water extract of Psoralea coryhfo!ia fruits and the
35.00 g of whole fruits were extracted with water (467m1), in a beaker heated over a
Bunsen flame and boiled down to 1/8, the decoction was filtered and centrifuged to
remove fine particles, filtrate was partitioned with chloroform (60 ml x 4)(a
INZ
troublesome emulsion formed during the partitioning was broken by centrifugation),
solvent was evaporated completely, residue was dissolved in acetone and made up to
50.00 ml, applied 5.00 tl on the TLC plate along with a standard solution of psoralen
(std.3). The plate was developed and scanned as described for the standard curve and
Psoralen concentration was determined using the standard curve and the extraction
Psoralen was added little by little into 150 ml of water at room temperature and the
psoralen had been added an excess of undissolved psoralen could be observed. Then
the addition was stopped and stirring was continued for 24 hours. The saturated
solution was filtered, 50.00 ml of the filterate was extracted with dichiorornethane (50
ml x 3), it was dried over anhydrous sodium sulfate, solvent was evoporated under
reduced pressure, and diluted up to 100.00 ml with acetone, 5 tl was applied on the
TLC plate, plates was developed and scanned as described for the standard curve, and
psoralen concentration of the saturated solution was determined using the standard
curve.
Me
Table 7: Densitometric readings obtained for psoralen spot in experiments described in
sections 2.5.5 - 2.5.8.
2.5.5 7522.63
2.5.6 7073.57
2.5.7 7495.63
2.5.8 5494.50
50
3.0 Results and Discussion
Plant materials used in the preparation of herbal drugs can be considered to be complex
herbal drugs, all the compounds present in the plants do not get incorporated to the
drug at the same rate. They are incorporated into the drug depending on their physico-
chemical properties such as solubility and polarity, and the processing parameters such
as temperature and time of heating which are applied during the manufacturing
process.
In the analysis of herbal drugs, one of the aims is to find out which plant components
have been incorporated into the drug. To answer this question, two approaches can be
made.
(ii) Isolation and identification of compounds from plants which are used in the
preparation of the drug and analysis of the drug to check whether those compounds are
For the investigation of a particular plant constituent in the drug, thin layer
particular compound on the chromatogram and for the confirmation of their presence
or absence, this technique can be adjusted in different ways. For examples, fine tuning
of the solvent gradient, applying multiple developments, the use of specific spray
51
reagents, carrying out co-TLC and 2D-TLC would be helpful in the identification of a.
particular compound.
Method (i) gives quick answers to the absence or presence of major plant constituents
whereas method (ii) has the advantage that is possible to determine the absence or
presence of even minor constituents using it. In the present study method (ii) was
Seven compounds (see Table 8) including one new natural product were isolated from
and they were identified by spectral analysis. (UV, IR, 'H NMR, 13 C NMR, HSQC,
HMBC, Mass) During these studies a fixed oil and the essential oil from the fruits of
52
Table 8: Compounds isolated from the fruits of P. corylifolia and some of their
physical properties.
The two major compounds, psoralen (1) and isopsoralen (2) were isolated as a mixture
using two methods as described in 2.3.1. In procedure-I 68, whole fruits were extracted
first with petroleum ether to remove petroleum ether soluble materials present in the
pericarp and they were ground to a powder. Then the powder was soaked in water and
extracted again with petroleum ether. The concentrated extract deposited psoralens on
Apart from this a much more convenient method was developed which involved direct
extraction with hexane (procedure-lI in section 2.3.1) and gave an equally pure
53
Larger amounts of pure compounds, psoralen and isopsoralen were obtained by
Small amounts (less than 100 mg) of pure compounds were obtained by subjecting the
Due to the fact that these two isomeric compounds have very similar R1 values, it was a
challenge to isolate these two compounds in the pure form using coloumn
carried out to sharpen the bands and increase resolution. Recently, a new method using
supercritical fluid extraction and high speed Counter current chromatography has been
reported, for the isolation, separation and purification of the two cornpounds.34' 5
The identification of the two compounds, psoralen (1) and isopsoralen (2) were
2 3
66 ~X3
Isopsoralen- (2)
The compound which started eluting at 5 % ethyl acetate in hexane had a Rf of 0.68 on
TLC (silica gel 60 254 / toluene: ethyl acetate: hexane, 1:3:9). It had a melting point of
54
14001430 C and was identified as isopsoralen. It showed a molecular ion peak at 186
in the mass spectrum. IR absorption at Vniax 1709 cm' revealed the presence of an ct, 13
unsaturated 6 lactone carbonyl, 'H NMR spectrum of (2) showed two characteristic
doublets at 6 6.39 (IH, d, J= 9.6 Hz) and 6 7.80 (11-1, d, J= 9.3 Hz) for H-3 and 11-4
proton signals of cournarin ring. Apart from the doublet at 6 7.37 (1 H. d, J= 8.4 Hz, H-
5) which is ortho coupled to the double doublet at 6 7.43 (1 H, dd, J= 8.4 Hz, 1.0 Hz,
H-6) indicates the presence of two adjacent aromatic protons, Presence of two couples
of proton doublets supports the angular arrangement of the molecule. Further the
doublet at 6 7.69 (11-1, d, J= 2.3 Hz, H-2') and the double doublet at 6 7.13 (1 H, dd, 1=
2.3 Hz, 1.0 Hz, H-3') which are coupled to each other can be assigned to the two furan
1 020
2QJ
3 5 4
psoralen- (1)
The second compound which eluted after isopsoralen had a Rf of 0.59 on TLC (silica
gel 60 254 / toluene: ethyl acetate: hexane, 1: 3: 9) and a melting point of 1600 163° C.
It was identified as psoralen. It also showed a molecular ion peak in the mass spectrum
at 186. IR absorption for the a, P unsaturated lactone carbonyl was observed at Viiax
1722 cm'. (It was noted that the IR absorption for the a, 13 unsaturated lactone
carbonyl was lower in isopsoralen than in psoralen.) LH NMR spectrum of (1) showed
two characteristic doublets at 6 6.37 (111, d, J= 9.3 Hz) and 6 7.78 (iT-I, d of m, J= 9.3
55
Hz) for H-3 and H-4 proton signals respectively of the coumarin system. The doublet
at 6 7.69 (lH, d, J= 2.3 Hz, H-3') and the double doublet at 6 6.83 (1H, dd, J= 2.3 Hz,
1.0 Hz, H-2') which are coupled to each other can be assigned to the two furan
hydrogens. Further psoralen gave a singlet at 6 7.68 (1H, bs, H-5) which indicates the
presence of an isolated aromatic hydrogen in the central aromatic ring, revealing the
linear arrangement of ring system. The remaining isolated aromatic hydrogen, H-8
(IH, in. H-8) appears as a multilpiet at 6 7.45 due to long range couplings. ('H NMR
These spectral data of psoralen and isopsoralen were compared with reported data and
it was found that they were in good agreement with reported values.34 (However in the
13
C NMR spectrum of isopsoralen the signal at 6 148.5 which is reported for C-9
(toluene: ethyl acetate: hexane, 1: 3: 9 / silica gel 60 F 254) and observed at 365 nm, the
two compounds could be detected as fluorescent blue spots at Rf 0.59 and 0.67 and
they gave intense yellow and yellowish green colours respectively when sprayed with
10% methanolic potassium hydroxide. Further they could be observed as dark blue
spots when illuminated at 254 nrn without spraying any reagent. When a thin layer
56
methanolic potassium hydroxide, these two compounds can be easily detected as the
Column chromatography of the hexane extract of fruits yielded bakuchiol (38). It was
eluted with 4 % ethyl acetate in hexane and was further purified by preparative TLC
obtained after sequential extraction of fruits with methanol and hexane, it was observed
that the column separation was disturbed by the high content of fixed oil present in the
hexane extract. Therefore instead of subjecting the total hexane extract to column
chromatography, the solvent was evaporated from the hexane extract and the residue
was re extracted with 80 % aqueous methanol leaving behind most of the fixed oil. The
aqueous methanol phase was partitioned with hexane and the hexane fraction which
was rich in bakuchiol was evaporated to remove solvent and chromatographed over
silica.
It was found that bakuchiol eluted from column was contaminated with a red coloured
bakuchiol in the pure form by column chromatography alone. Therefore the column
fractions containing bakuchiol as the major compound were pooled together and
further subjected to preparative TLC to remove the red pigment and to yield pure
bakuchiol as a colourless oil. Multiple development was carried out in order to achieve
57
It was possible to identify this compound as the phenolic terpenoid, bakuchiol (38) by
means of spectral analysis. Bakuchiol has been reported to have a number of biological
Hb
l>__Ha 15
16
)::2
HO 4
Bakuchiol (38)
Mass spectrum (ESIMS: mlz: 255[M-H], 257[M+H]) of the compound showed that
the molecular weight of the compound was 256. 'H NMR spectrum clearly indicated
6.78 ppm (21-1, d, J= 8.4 Hz, H-3, H-5), and 6 7.26 ppm (2H, d, J= 8.4 Hz, H-2, H-6),
The presence of a trans double bond (C-7 - C-8) is supported by the two doublets at 6
6.27 (lH, d, J=16.2 Hz,H-7) and 6 6.07. (11-1, d, J16.2 Hz, H-8). The vinyl group is
indicated by double doublet at 6 5.88 (11-1, dd, J=17.7 Hz, 10.5 Hz, H-17) and two
doublets at 6 5.02 (11-1, d, J=17.7 Hz, H-18a) and at 6 5.05 (11-1, d, J=10.5 Hz, H-18b).
Two singlets each integrating to three hydrogens at 6 1.59 (31-1, s, H-IS) and 6 1.69
(31-1, s, H-14) can be assigned to geminal dimethyl group while the other singlet at 6
1.20 (31-1. s, H-16) can be assigned to tertiary methyl group. (see appendix ii. Fig.24
coupled to C-10 protons. The remaining multiplets at 6 1.50 (211, m) 6 5.11 (1H, m)
were assigned to the proton signals at C-10 and C-12 respectively with the aid of
The spectral data obtained for bakuchiol agrees well with reported values.56 However
HSQC and HMBC experiments revealed that the reported 13C NMR assignments for
C-8 / C-7, C-i / C-13 and the 'H NMR assignments for H-10 / H-I 1 need to be
14
Based on 2D NMR correlations, 'H NMR and 13C NMR shifts of bakuchiol could now
Table 9 also gives the previous assignments of 'H and 13C NMR assignments of
bakuchiol.56
Table 9: NMR data for bakuchiol (38)
4 154.59 154.5
9 42.50 42.5
13 131.3 131.4'
14 17.63 1.59 (3H, s) C-14 1.56(s)
Reported chemical shift values for C-7 and C-8(*). C-I and C13(t)and H- 10 and I-I-I 1(t)should be
interchanged as evident by the HSQC and HMBC experiments.
M
OO
It is also interesting to note that, the 111 NMR signals corresponding to the terminal
protons of vinyl group (H-18a and H-18b) are well resolved and can be identified as
two doublets at 8 5.02 (11-1, d, J=17.7 Hz, H-18a) and 6 5.05 (11-1, d, J=10.5 Hz, H-
18b). The corresponding double doublet (IH, dd, 17.7, J=10.5 Hz, H-17) can be
observed at 6 5.88.( Fig. 6) But in the literature these signals are not well resolved and
IIII
.. .......'-I-. ,-.-. ..
5.9 5.8 5.7 5.6 5.5 5.4 5.3 5.2 5.1 ppm
Fig 6: ö 4.9 - 6.0 ppm region of the 'H NMR spectrum of bakuchiol.
When a thin layer chromatogram (toluene: ethyl acetate: hexane, 1: 3: 9 / silica gel 60
F254) of the plant extract of methanol or hexane is observed under 254 nm irradiation,
in addition to the two psoralen spots, bakuchiol can be seen as a dark blue spot at a
relatively higher Rf value (0.82). When the plate was sprayed with anisaldehide
sulphuric acid reagent and heated that could be visualized as a very prominent bluish
61
During this isolation process the red coloured pigment mentioned above also could be
isolated in an impure state. E-Iowever time did not permit the purification and structural
elucidation.
bakuchiol
Isopsoralen
psoralen
EI
62
3.2.3 Fixed oil
80% methanol insoluble fraction of the hexane extract of P.corylifolia was subjected to
column chromatography over silica gel using a gradient of ethyl acetate in hexane to
The fixed oil of P.corylifolia has not been subjected to analysis previously and no
of the 'H NMR spectral data with those of the known triacyiglycerols and
The 'H NMR spectrum shows a multiplet at about 6 5.3 which can be collectively
assigned to the olefinic protons in the polyunsaturated fatty acid chain and the methine
proton in the glycerol backbone of the triacyiglycerol. The multiplet between 6 4.40
and 6 4.1 corresponds to the A2 B2 section of the A2B2X system of the glycerol
backbone where the two carboxylic acid residues attached in the 1 and 3 positions of
the glycerol molecule are same. AB, AX, and BX coupling constants calculated from
the 'H NMR spectrum are 12 Hz, 4.2 Hz and 6.0 Hz. They are in agreement with the
values quoted in the literature for similar systems.7° The signal at 6 1.28 with relatively
a very higher integration value can be assigned to the hydrogens in the CH2 groups,
attached to the saturated carbon atoms of fatty acid chains. The multiplet at 6 0.89 can
be explained as overlapping triplets due to terminal CH3 groups. While the hydrogens
in the CH2 groups a the carbonyl groups are appeared at 6 2.30, hydrogens in the CH2
63
groups to the carbonyl groups are shown by the signal at 1.60. The multiplet at
2.03 can be assigned to the allylic protons. The signal at 6 2.78 can be assigned to
doubly allylic methylene protons indicating that at least one of the fatty acids is poly
unsaturated.
Fixed oil could be seen on TLC plates (toluene: ethyl acetate: hexane, 1:3:9 / silica gel
60 F254) at a relatively very higher R1 value 0.87 as a large ilTegular reddish purple spot
The methanol extractive obtained from the sequential extraction of fruits with hexane
and methanol was subjected to column chromatography and 35 major fractions were
silica gel and eluted with acetone: hexane, 1: 20 which resulted in the isolation of
that this compound also could be precipitated by addition of acetone to the column
fraction 13.
Mass spectrum of the compound gave pseudo molecular ions at mlz 335[M+HI and
mass and 'H NMR data with those of the compounds reported from Psoralea sp.
suggested the compound to be a cournestan. The 'H NMR spectrum showed signals for
14 protons including a phenolic OH, while the ' 3C NMR gave signals for 20 carbons.
Based on DEPT and HSQC NMR experiments these were assigned for two methyl
carbons and 7 methine carbons with the rest being accounted for quaternary carbons. A
The singlet at 61.45 (6H) in the 'H NMR indicated the presence of a gem- dimethyl
group. Further, the downfield proton signals observed at ö 6.97 (c 104.2) and 7.77 (Sc
118.5) for a 1,2,4,5 tetra substituted benzene ring and a ineta coupled doublet at 8 7.17
(J = 2.0 Hz), a ortho coupled doublet at ö 7.71 (J = 8.4 Hz) and a double doublet at
6.96 (J = 8.4 & 2.0 Hz) for a 1, 2, 4-tri substituted benzene ring were all in agreement
with a coumestan class of compound. The HMBC correlations from both CH3 groups
at C-6" to C-5" and C-6" confirmed the position of the two methyl groups at C-6"
while the downfield shift of C-6" (6c 77.8) favored the pyran oxygen attachment.
Further J2 and J3 HMBC correlations from H-4" and H-5" protons to C-7 and C-6
65
confirmed the pyran ring system while the de shielded NMR signals for C-4" (c
120.7) and C-5" (öc 131.8) olefinic positions in comparison to the already known
from H-S to C-7, H-8 to C-6 and H-4" to C-7, H-5" to C-6 confirmed the colmectivity
correlations from H-S to deshielded carbons C-9 (c 153.9) and C-4 (c 159.0)
indicated they were adjacent to two oxygen atoms, which accounted for the ring
oxygens of the coumarin fragment and the furari in the system. A downfield proton
signal at 10.2 accounting for a phenolic 01-1 was placed at C-4', based on 3-bond
correlations from both H-5' (dd) and H-3' (d) to C-15. The downfield shift of C-2' (6c
156.0) and the aromatic carbon shift of C-l' (c 114.5) were found to be characteristic
for a furan / pyran substituted aromatic system (Fig. 8). From these interpretations it
was possible to account for a fragment of C18 H1404. Since an HMBC correlation was
not observed from H-6' to C-3, but given that the molecular formLlla accounts for two
more quaternary carbons, one of which has to be a carbonyl as supported by both the
coumestan structure and chemical shift evidence (6c 157.2) the unaccounted carbons
were assigned as C-2 and C-3. ('H NMR, ' C NMR, HSQC and HMBC spectra of
LIM
Table 10: NMR (d6-DMSO, 600 MHz) data for dehydroisopsoralidin (42) *
2 157.2a
3 102.9
4 159.0
6 118.7
7 155.9
9 153.9
10 105.6
I' 114.5
156.0'
77.8
* Chemical shifts are given in 5ppm and referenced to the standard deuterated solvent.
LChemical shift values can be interchanged between these two carbons.
b
Chemical shift values can be interchanged between these two carbons
67
0
A~ 0
1 W 0- C',
-V~/~
0 k7
OH
On the basis of the above spectroscopic evidence this compound was established as
dehydroisopsoralidin. (42)
Dehydroisopsoralidin has been synthesized from psoralidin (5) and first reported by
Gupta ci. al. (1977).69 It is noteworthy that the compound is being reported here for the
first time as a natural product while complete NMR assignments are also being
reported for the first time. Based on 2D NMR correlations the 'H NMR shifts should
When the methanol extract of fruits was chrornatograph on TLC plates (toluene: ethyl
acetate: hexane, 1: 3: 9 / silica gel 60 F254) and observed at 365 nm, this compound
can be seen as a fluorescent purple spot (R1 0.41) which is almost completely
be separated from the overlapping blue fluorescent spot by developing the same plate
three times in the same solvent. When the plate is treated with 10 % methanolic
W.
3.2.5 Corylin (11)
The compound was isolated as a white crystalline solid (1 x 102 %) from the methanol
Molecular weight of the compound was given as 320 amu by the mass spectrum.
Presence of the hydroxyl group is supported by the broad peak at 3230 in the IR
HO 7 gO
a 0
Corylin (11)
The sharp singlet at 1.39 which integrates to 6H in the 'H NMR spectrum suggests
the presence of a geminal dimethyl group. The two ortho coupled doublets, two meta
coupled doublets and two ortho and ineta coupled double doublets in the 'H NMR
spectrum indicate the presence of two 1,2,4- trisubstitiuted benzene rings. Double
doublet centered at S 6.94 (1H, dd, J=8.7 Hz, 2.0 Hz) can be assigned to H-6 which is
art/ia coupled to H-S (7.96, 1H, d, J=8.7 Hz) and meta coupled to H-8 (6.86, 1H, d,
J=2.2 Hz). The other double doublet at 5 7.29 (IH, dd, J=8.7 Hz, 2.0 Hz) can be
assigned to H-6' which is ortho coupled to H-5' (6.78, 111, d, J=8.7 Hz) and ineta
coupled to H-2' (7.28, lH, d, J=2.0 Hz). The two doublets centered at 8 5.78 (11-1, d,
M
.
J=9.0 Hz) and 6 6.44 (1H, d, J=9.0 Hz) can be assigned to two pyran protons. The
singlet at 6 8.34 indicates the isolated hydrogen of C-2. The chemical shifts of C-2 (6c
152,1), C-3 (6c 123.0) and C-4 (6c 174.5) agrees with their location in an an a, 13-
field carbon shift at 6 162.7. (see appendix ii, Fig.3 1 and 32 for the'H NMR and 13 C
A comparision of 'H NMR and 13C NMR data of the compound with reported data 31
and HMBC correlations for the compound are given in Table 11.
70
Table 11: NMR data for corylin (11)
No. '3C (d6- 'H (multiplicity and gHMBC '3 C(CDCI3,150 'H multiplicity
DMSO, J(Hz) MHz)
150 (d6-DMSO,600MHz)
MHz)
2 152.1 8.34(IH, s) C-3, C-4, C-9, c-I' 8.22(1H,$)
3 123.0 122.6
4 174.5 175.4
9 157.4 158.4
10 116.4 117.6
124.4 124.2
120.5
153.1 153.3
71
These comparison shows that the spectral data obtained for corylin very closely
This compound can be observed on TLC plates as a yellowish green spot when
illuminated with 366 nm at 0.35 (toluene: ethyl acetate, 5: 2) and can be intensified by
spraying with natural product reagent. It gives a fluorescent blue colour when sprayed
Psoralidin was isolated as a colourless crystalline solid (9x1ft3%) from the methanol
Psoralidin (5)
Initial comparision of the spectral data (Mass, IR, 'H NMR) with those of the known
compounds of Psoralea corylifolia suggested that the compound was a coumestane and
section.
72
Mass spectrum of psoralidin (5) gave the molecular ion peak at mlz 337[M+H] and
appeared at 1720 cm 1 indicating the presence of a carbonyl group. Broad peak at 3450
cm' suggests the presence of hydroxyl groups. 1.2,4,5 tetra substituted benzene ring is
evident from the two down field singlets at 6 7.62 (1H, s, H-i) and 6 6.92 (114, s, H-4)
6.94 (1H, dd, J=8.4 Hz, 2.0 Hz, H-8), ortho coupled doublet at 6 7.69 (1H, d, J=8.4
Hz, H-7) and meta coupled doublet at 6 7.16 ( IH, d, J=2.0 Hz, H-10).The two singlets
at 61.71 (3H, s,) and 61.74 (3H, s) each integrating to three hydrogens, triplet at 6 5.36
(IH, 1, J=7.2 Hz, H-2') and doublet at 63.33 (2H, d, J=7.3 Hz, H-i') indicate the
presence of a prenyl group. Further the 'H NMR broad peaks at 610.09 and 610.25
indicate the presence of two hydroxyl groups in the molecule. 'H NMR spectrum of
The spectral data of the compound (UV, IR, 'H NMR and Mass) agreed well with
reported data of psoralidin confirming its identity. 'H NMR data of the compound is
73
Table 12: Comparision of 'H NMR data of psoralidin (5) with reported data.
6.92(1H,$) 6.93(1H,$)
7.16(1H,d,2.0) 7.18(1H,d)
0.26 (toluene: ethyl acetate, 5: 2) and turns to yellow when sprayed with 10%
methanolic hydroxide.
74
3.2.7 Isobavachalcone (30).
Isobavachalcone was isolated as bright yellow needles from the methanol extract of
Isobavachalcone (30)
Initial comparision of the spectral data with those reported in literature suggested this
compound to be a chalcone and the detailed spectral analysis (Mass, 'H NMR, 13C
Mass spectrum of the compound gave pseudo molecular ions at rnlz 325IM+H] and
mlz 323 IM-H[1 corresponding to a MW of 324 amu. It gave two sharp singlets at 61.62
(31-1. s, H-4") and 81.72 (3H, s, H-5") together with a doublet at ö 3.22 (21-1, d, J=7.2
Hz. 1-1-1") and a triplet at 6 5.17 (1H, t, J=7.2 Hz, H-2") in its 'H NMR spectrum
indicating the presence of a prenyl group. The two ortho coupled doublets centered at
6.84 (21-1, d, J=8.6 Hz) and 6 7.75 (21-1, d, J=8.6 Hz) indicated the presence of the pam
disubstituted benzene ring. The two overlapping proton doublets of C-b (6 7.72, 1H, d.
15.6 Hz) and C-c (6 7.77, IH, d, 15.6 Hz) carbons of the trans double bond appears as
a single peak at 6 7.745. (Fig 9 ). Two other ortho coupled doublets appeared at ö 6.46
75
(111, d, J=8.9 Hz) and 6 8.03 (111, d, J=8.9Hz) could be assigned to H-5 and H-6
protons of the tetrasubstituted benzene ring. NMR assignments along with 11MB
correlations for the compound are given in the Table 13. (see 'H NMR and ' C NMR
Fig 9: Appearance of 111 NMR pattern corresponding to H-b and H-c protons, and the
76
Table 13: NMR (d6-DMSO, 600 MHz) data for isobavachalcone (30)
2 163.5
3 114.4
4 162.3
1 112.6
A 191.7
1' 125.75
4' 160.2
131.16
it is illuminated at 366 nm, as a dark spot. When the plate was sprayed with
77
anisaldehyde sulphuric acid reagent and heated this compound can be observed as a
Essential oil: An essential oil was extracted from the fruits. Hydrodistillation of whole
fruits yielded 0.074 % (w/w) essential oil, while powdered fruits yielded 0.081 %
essential oil. Steam distillation of powdered fruits yielded 0.082 % (w/w) essential oil.
A GC finger print was developed for the essential oil (Fig: 10) and it can be used to
The major compounds identified in the essential oil by comparision of retention times
were to be isopsoralen (peak no.1), psoralen (peak no.2) and bakuchiol (peak no.3).
i1
1'
r
r
0
0
r
0
0
N a
0)
0
0
a
co
-0
0
0
a
0)
0
0
0
to
0
0
0
0
0
0
N
0
0
a
N
0
0
0
r
79
3.3 Analysis of "Bakuchi oil'.
The analysis of the 'Bakuchi oil" was based on the information obtained from the
analysis of the plant. The oil was subjected to careful chromatographic analysis with
the view of confirming either the presence or absence of the previously identified plant
components.
Direct thin layer chromatography of a solution of the oil in chloroform was not
successful due to the fatty material from sesame oil. The plant components could be
separated partially from the fatty material by extracting the oil with methanol.
However, even this extract was found to be unsuitable for effective chromatographic
analysis. Partitioning the methanol extract between 80% aqueous methanol and hexane
resulted in removal of the fatty material into the hexane phase, and the resulting
aqueous methanol fraction was used for subsequent analysis in following ways.
ethyl acetate as the solvent. Sixteen fractions were collected and examined by TLC.
While the fraction-3 and 4 contained the two major compounds psoralen and
Corylin (11). psoralidin (5) and isobavachalcone (30) were found in the fraction-7.
The 80% aqueous methanol soluble fraction of the methanol extract of the was
examined by TLC in order to detect the isolated compounds. A TLC system was
developed for this purpose and the presence of all the isolated compounds in a single
TLC could be demonstrated by using the solvent system toluene: ethyl acetate: hexane,
possible to identify all the compounds isolated from P.cory1ij1ea, namely, psoralen
(1). isopsoralen (2), dehydroisopsoralidin (42), corylin (11), psoralidin (5) and
iii) Using the same solvent system, the 80 % methanol soluble fraction of the methanol
extract of the "Bakuchi oil' and the total methanol extract of the fruits were compared.
Several other minor compounds besides the compounds already isolated could be
observed in the oil. However most of them could be observed in the plant extract too.
(Fig 12-A) In order to get a total picture of the composition of the "Bakuchi oil" these
It was noted that bakuchiol was not found in "Bakuchi oil". It was further noted that
the red pigment eluting with bakuchiol was also not incorporated in "Bakuchi oil".
The non detection of bakuchiol (38) in "Bakuchi oil" was surprising as it is a major
component of the fruit. (Fig 12-B, 12-C, The faint spot observed at the R1 value for
bakuchiol has been shown by control experiments to arise from sesame oil). In order to
test whether bakuchiol was remaining in the sesame oil without getting extracting into
methanol, the oil sample left after the extraction by methanol was saponified and then
thin layer chromatograms. In order to test whether bakuchiol has been extracted into
hexane during the partitioning of the methanol extract of "Bakuchi oil" between 80%
81
methanol and hexane, the hexane extract was chromatographed. But bakuchiol (38)
could not be detected on TLC. These results indicate quiet definitely that bakuchiol
(38) is not incorporated into "Bakuchi oil". The absence of bakuchiol in "Bakuchi oil"
do
1 2 3 4 5 6 7 1 2 3 4 5 6 7 8
F,,
A
123 123
A C
Fig 12: TLC of fruits of P.corylfolia and "Bakuchi oil".
83
3.4 Standardization of "Bakuchi oil"
Chromatographic finger printing is one of the most important methods used in the
The ideal chromatographic finger print will contain a well separated peak for each
compound found in the mixture. While this is almost impossible to achieve in practice,
a chroinatogram which contains well separated peaks for selected marker compounds
will contain much information relevant to the quality control of the herbal drug.
Furthermore, the complex patterns of peaks contain information that can be unravelled
TLC has the advantage over other chromatographic methods in terms of simplicity,
In these studies on "Bakuchi oil', chromatographic finger prints were developed based
development of TLC finger prints the extracts of "Bakuchi oil" were prepared as
described in the section ( 2.4 ) Plates were developed in three different solvent
systems, namely, hexane: ethyl acetate, 4: 1, toluene: ethyl acetate, 5: 2, toluene: ethyl
acetate: hexane. 1: 3: 9 and evaluated visually under irradiation by UV light at 366 nrn
as most of the compounds isolated from the plant that were shown to he present in the
oil, exhibit fluorescence at this wave length. Spraying the chromatograms with 10%
13-A, 13-13, 13-C) Densitometric scanning with a filter of 420 nm, was used to obtain
chromatographic profiles. (Fig. 14-A, 14-13, 14-C) The initial solvent systems used
hexane: ethyl acetate (4: 1) and toluene: ethyl acetate: hexane (1: 3: 9) while separating
clearly the low polar compounds did not resolve the highly polar compounds. A better
profile representing all the compounds was obtained by toluene: ethyl acetate (5: 2),
although the resolution of the low polar compounds were not as clear as with the two
previous solvents.
Samples from different production batches of 'Bakuchi oil from the BMART as well
sample from Ayurvedic Drugs Corporation were analyzed using the TLC systems
densitograms did not show any marked difference in the pattern from that of the
Psoralen and isopsoralen are the most important marker compounds found in these
finger prints.
85
A B C
Solvent systems:
86
A
Ifl
C
Fig. 14. Densitograms of thin layer chromatograms shown in Fig. 12.
Solvent systems:
EM
A
90.0 mm
Solvent systems:
The major group of photoactive compounds in Psoralea corvlifolia fruits are the
'psoralens", the most abundant of which are psoralen (1) and isopsoralen (2). The
parameter in the standardardization of "Bakuchi oil". Towards this end a method was
densitometry.
The major problem to he overcome in the development of the method was to obtain a
total extract of psoralen free from the fixed oil which would interfere with the TLC
separation. Thus, even a dilute solution of the oil in chloroform, would not yield an
extract of the oil gave better chromatograms. However, a certain proportion of the fatty
was the fact that even repeated extraction by methanol was not successful in obtaining
It was clear that an initial removal of the fatty material from the oil was necessary
"Bakuchi oil" in acetone and precipitating the fatty material by adding the protic
solvent methanol to the solution and cooling it. A series of controlled experiments
yielded the optimum ratios of oil to solvents, and the optimum temperature to achieve a
near-total separation of fatty material from the organic extract containing psoralen. (see
2.5.2.2). TLC experiments confirmed that the fatty material separated did not contain
any detectable amount of psoralen. The extract of psoralen so obtained was used for
densitometric studies.
Although the chromatographic behaviour of psoralen and isopsoralen are similar, they
could be well separated from each other on TLC by using hexane: ethyl acetate (7: 1)
could not be carried out to measure psoralen present in "Bakuchi oil' as visual
indicator, showed that the psoralen spot was partially overlapped by another spot
which quenches fluorescence. Observation of TLC of the plant extract and sesame oil
extract, showed that this non fluorescent spot arose from sesame oil. Therefore in order
to avoid the interference of this spot, densitometry was carried out in the fluorescence
Isopsoralen present in "Bakuchi oil' could not be quantified using this system as it
Chromatographic investigation showed that this interfering spot also arose from
sesame oil.
00
(b) Standard curve
Pure psoralen isolated from P.corvlifclia fruits was used to prepare the standards, and
the standard curve was determined as described in the section 2.5.2.1. The standard
Each time the standard curve was used for measurements its validity was checked by a
single point calibration at its mid point. The calibration range for confidence interval, p
= 0.05 was found to be 5860-6750. This was calculated using following equations. 72
(2) If y0 is the value for a given x0 , obtained from the regression line, then
- )2
S2 = + 1S2
_)2] 5 where = average of x values, S = variance of
[fl (x1
estimated y value.
2 - - b y - m x1 y
-
s2 - where 5' = best estimate for y
n-2 -
value.
91
This standard curve was used to determine psoralen concentration in several Bakuchi
92
11 '
IMI
4- -
j I
93
( c ) Method validation
trial(VCR I) was selected as the reference sample. The precision of the TLC-FD
six replicates of the reference sample and the Coefficient of Variation (CV) was found
recovery rate of 103 %. These results show the method developed is of acceptable
Psoralen concentration in six different "Bakuchi oil" samples were determined using
the above method and the results are given in the tablel4. These results show that there
is a wide range of concentrations (0.038 mg/mI to 0.226 mg/ml) of psoralen in the oils
tested. Further more, the widely different results obtained for the two BMARI oil
samples (BMARI-I and BMARI-lI) prepared using the same process, indicates the
It was found that the level of psoralen found in "Bakuchi oil" is much less than that
found in the topical applications used in modern medicine (0.2 - 1%).73 However total
BMARI - II 0.226
using the same method and the results are given in table 15.
2 2.78
3 1.3
4 3.78
5 2.88
These results show that there is a considerable variation in the psoralen concentration
among different batches of fruits (over a factor of 2 from the lowest value) indicating
the need to standardize the raw material used in the manufacture of the oil. This could
3.7 Effect of processing parameters on the composition and the quality of the
drug.
The large scale preparation of "Bakuchi oil" (scheme-i) follows the traditional
procedure of obtaining a water extract of the plant material (stage I), and heating the
extract with a vegetable oil until the water is evaporated (stage II). During the latter
part of stage II, where the aqueous extract is heated with oil, finely powdered plant
material (churna) is added. In the case of "Bakuchi oil", the vegetable oil used is
sesame oil and the "churna" consists of powdered Psoralea corylifolia fruits.
gel
Scheme -1 Process for the preparation of Bakuchi oil' at the BMARI.
1
Stage I
Boiling down to 75 L
Filtration
Stage II
Addition of powdered fruits (5 kg)
prior to total evaporation of water ( 90 % complete)
'Jr
'Jr
range of polarities in the final product than if only an oil extraction was used. The
aqueous extract would contain high polar and medium polar compounds. Low polar
compounds found suspended in the aqueous extract as well as medium and some of the
97
high polar compounds found in the aqueous solution can be expected to be
incorporated into the product during the stage II. However most of the low polar
compounds found in the product would be extracted into the oil from the powdered
fruits during stage II. Some of the low polar compounds may be lost by steam
volatilization during stage I and II. Direct extraction of the plant material by the oil
would not extract the medium and high polar compounds to the same extent.
In order to study the factors affecting the rate of incorporation of psoralen into
prepared from that sample were determined. (section 2.5.3 and 2.5.4) Table-16
A sample from the same batch of fruits was extracted by water (section2.5.7) and
the psoralen concentration of the water extract and the extraction efficiency was
determined. ) TabIe16
The same batch of fruits was extracted by sesame oil and the psoralen
concentration of sesame oil extract and the extraction efficiency was determined.
Solubility of psoralen in water and in sesame oil were determined. (section2.5.8 and
WI
Table-16. Comparision of psoralen in "Bakuchi oil" and different extracts of P.
cor'ciifriia.
The solubility of psoralen in water and sesame oil at room temperature determined and
found to be 0.1956 mg/mi and 3.528 mg/ml respectively showing that the weakly polar
psoralen was more soluble in sesame oil than in water. The psoralen concentration of a
batch of oil prepared from fruits containing 3.029 rng/g of psoralen was determined to
be 0.207 mg/mI, showing that only 10.23 % of the total psoralen content is
Then a water extract prepared as in stage I, contains only 0.136 mg/mi of psoraien,
amounting to only extracting 7.2 % of the psoralen found in the fruits. Assuming that
ARE
all the psoralen in the water extract is incorporated in the oil, it would account for
65.84 % of the psoralen present in the oil which means that at least 34.15 % of
psoralen is incorporated from the fruits in stage II. However, extraction of P.corylifolia
powder with sesame oil under the conditions of stage 11(5kg of powder in 75 ml of oil)
indicates that 95.3 % of the psoralen content of the fruits can be extracted. If stage I in
the preparation process is omitted, and P.corylifolia powder is extracted directly with
sesame oil as in stage 11, a medicinal oil containing 92.6 % of the psoralen content
expected from the normal process (i.e. stage I followed by stage II) may be expected.
These results suggest that it may be economical to modify the procedure for the
preparation of "Bakuchi oil" by eliminating the water extraction step and to directly
The thin layer chromatographic comparisons (Fig 12) show that the chemical
compositions of "Bakuchi oil" and the sesame oil extract of the fruits are qualitatively
bearing on the composition of the oil as regards minor components. Therefore a careful
100
analysis and clinical trials are necessary before a recommendation to change the
In order to study the fate of bakuchiol, following GC experiments were carried out.
(section 2.4.2)
The steam distillate of the "Bakuchi oil' was analyzed by gas chromatographiy.
(section 2.4.2/ 6)
The steam distillate of the sesame oil extract of the fruits of P.corvlifoiia was
The steam distillate of the seed residue left behind after extracting with sesame
2.4.2/7)
2.4.2/8)
101
The ratio of the GC response factor of psoralen (1) to bakuchiol (38) were first
determined and it was found to be 5: 2. (Fig. 17) The relative amounts of psoralen to
bakuchiol in the fruits were determined by analyzing the 80% methanol soluble
fraction of an exhaustive hexane extract of the fruits. (It has been shown previously
psoralen and bakuchiol into the methanol phase from hexane, leaving behind the fixed
oil in the hexane layer.) This was found to be 3: 10 and bakuchiol could be detected as
the major compound found in the GC chromatogram. (Fig. 18) Examination of the
steam distillate of the fruits showed that it contained psoralen to bakuchiol in a ratio of
2: 15 indicating that the relative amount of bakuchiol has increased when compared to
the hexane extract. (Fig.10- GC finger print of essential oil of the fruits of
P.corvlifhlia.) Thus, bakuchiol is more steam volatile than psoralen. A steam distillate
of the "Bakuchi oil' did not give a peak for bakuchiol while a small peak was seen for
psoralen. (Fig. 19) This finding confirms the results from TLC that the oil does not
contain bakuchiol.
lost by steam volatilization. However, in stage II some of the bakuchiol present in the
powdered fruits added should get incorporated, and one would expect to find sonic
bakuchiol present in . Therefore, the total absence of bakuchiol in Bakuchi oil" cannot
be explained by steam volatilization alone. In order to study this further, the following
102
Powdered fruits were extracted by sesame oil at 135-145° C which is the temperature
applied during the manufacturing process, and the steam distillate of that extract was
(Fig.20). Given the fact that bakuchiol is more steam volatile than psoralen, it is clear
that the sesame oil extract contain more psoralen than bakuchiol, even though the fruits
contain more bakuchiol than psoralen. In order to check whether bakuchiol was not
being extracted into sesame oil, the fruit residue was analyzed by GC (Fig.21) and only
a trace amount of bakuchiol could be detected. The final experiment was to test
whether bakuchiol was being extracted and then decomposing under the preparation
conditions. A solution of bakuchiol in sesame oil was heated to 145° C and the change
in bakuchiol content was measured after 2 hours by GC. There was no significant
change in the concentration of bakuchiol in the sample after heating showing that no
decomposition had taken place at 145° C which would be the maximum temperature
A possible explanation of these findings is that bakuchiol reacts with other compounds
present in Psoralea corvliftlia at 145° C in the presence of sesame oil. This hypothesis
needs to be investigated.
103
N
0
0•
H
10
0000000000000000000
++++++++++000000000
0 e0 ø0 06 000000000
rrrrrrrrr
104
en
000000000000000000000000 '
000000000000000000000000
000000000000000000000000
000000000000000000000000
105
N —:
106
a 0 aaaaaaa aaóóa
I OOODOOOOOOOOOOOOOOOOOOOO,
I 000000000000000000000000
000000000000000000000000
C 000 000000000000000000000 E
o
a
I
I
L
107
IN
C-'
108
4 Conclusion
bakuchiol are incorporated in "Bakuchi oil'. The hypothesis that bakuchiol could be
reacting with other plant components during the manufacturing process needs to be
investigated. All the compounds can be observed in the simple TLC system developed
profile can be used as a finger print in the quality control of the oil. Psoralen
concentration in the oil can be determined using TLC-FD densitometry, and the
material and good process control is needed to obtain a standardized oil. The
preparation process currently used for "Bakuchi oil" is wasteful in terms of psoralen
and energy, and the possibility of preparing it by direct extraction of the fruits by
sesame oil should be explored. Careful chemical analysis and clinical trials are needed
The method developed in this study for separating plant secondary metabolites from
the matrix in medicinal oils by precipitating the fatty materials from a solution of the
oil, by varying the polarity of the medium and the temperature, could be used in the
109
References
hyperforin for the clinical efficacy, Pharmacopsychiatry, vol. 31, issue SUPPL. 1,
p.54-59, 1998)
Yi-Zeng Liang, Peishan Xie, Chan, K.. Quality control of herbal medicines.
1996, p.843.
Marks, R., Roxburgh's common skin diseases, 171h ed.. London, 2003, p.298.
Medicine. Ministry of health and family welfare, New Delhi, 1986, p.20-2 1.
110
Pasricha, J.S., Treatment of Skin Disease, 4th ed., Raju Primlani, New Delhi,
1992, p.174.
Lowe, N.J., Practical psoriasis theraphy, 2' ed., Mosby year book Inc., 1986,
p.276.
Renmin Liu, Aifeng Li, Sun, A., Kong, L., Preparative isolation and purification
Andreoni, A., Cubeddu, R., DalI'Acqua, F., Knox, C.N. and Truscott. T.G.,
Llano, J., Raber, J., Eriksson, L.A., Theoretical study of phototoxic reactions of
235-243., 2003.
111
Abel, E.A., Farber, E.M., Photochemistry (chapter 11), in Recent advances in
dermatology, Rook, A. and Savin, J., Churchill Livingstone, New york, 1980,
p.263.
dermatology, Rook, A. and Savin, J., Churchill Livingstone, New york, 1980,
pp.274-276.
The Merk Index of Chemicals and Drugs, 71h ed., Merk and Co. Inc.. Rahway,
p.1 179.
24, Tripathi, K.D., Essentials of medical pharmacology, 4t) ed., Jaypee Brothers,
Sun, J.N., Woo, S.H., Cassady, J.M., Snapka, R.M., DNA Polyrnerase and
362-366.
112
Yin, S., Cheng-Qi Fan et al., Antibacterial prenyiflavone derivatives from
Cho, H., Jung-Yang Jun, Eun-Kyoung Song, Ki-Hong Kang ci a?., Bakuchiol: A
Chakravarti, K.K., Bose, A.K., Siddiqui, S., Chemical Examination of the seeds
Jam, A. C., Gupta, G.K., Rao, P.R., Isolation and constitution of corylin: a new
Wang, X., Wang, Y., Yuan, J., Sun, Q. etcii., An efficient new method for
113
Innocent, G, Cappelletti, EM., Caporale, G., Furocoumarin contents in the
No.4,311-316, 1991.
BE
38, Chjntalwar, G.J., Ramakrishnan, V., Luthria, D.L and Baneiji, A., Insect
Gouping Peng, Panhua Wu, Dongjun Chen, Hongyang Li, Chem.Abstr., 126:
303679.
Gupta, G.K., Dhar, K.L. and Atal, C.K., Isolation and constitution of corylidin: a
403-404, 1977.
Gupta, S., Jha, B.N,, Gupta, G.K. et al., Coumestans from seeds of Psoralea
Gupta, B.K., Gupta, G.K., Dhar, K.L., Atal, C.K., Psoralidin oxide, a coumestan
114
Wang, D., Famei Li, Jiang, Z., Osteoblastic proliferation stimulating activity of
Psoralea corylifolia extracts and two of its flavonoids, Planta Medica, 67, 748-
749, 2001.
Sun, J.L., Gupta, G.K., Dhar, K.L. and Atal, C.K., Psoralenol: a new isoflavone
Bajwa, B.S., Pyare La! Khanna and Seshadri, T.R., Components of different
15-19, 1974.
Wei-Jern Tsai, Wen-Chi Hsin, Chien-Chih Chen, Ant iplatelet flevonoids from
Bajwa, B.S., Pyare Lal Khanna and Seshadri, T.R., A new isoflavone, Neo-bava--
isoflavone, from the seeds of Psoralea corylifolia, Current Science, 41, No:24,
882-883, 1972.
Gupta, G.K., Dhar, K.L. and Atal, C.K., Corylina!, A new isoflavone from seeds
Gupta, G.K., Sun, J.L., Gupta, B.K. and Dhar, K.L., Bakuchalcone, a
Bhalla, V.K., Kayak, U.R., Sukh Dev, Some new flavonoids from Psoralea
115
Gupta, S.R., Seshadri, T.R., Sood, G.R., A new formylated chalkone, 5'-formyl-
Gupta, B.K., Gupta, G.K., Dhar, K.L., Atal, C.K., A C-formylated chalcone from
Sun, J.L., Gupta, G.K., Dhar, K.L. and Atal, Bavachromanol: A new chalcone
Labbe, C., Faini, F., Coll, J., Bakuchiol derivatives from the leaves of Psoralea
Hyun-Ock Pae, Hyun Cho, Gi-Su Oh et al, Bakuchiol from Psoralea corvlifrlia
inhibits the expression of inducible nitric oxide synthase gene via the
Bapat, K., Chintaiwar, G, J., Pandey, U. and et al, Preparation and in vitro
116
Shuki Araki and Yasuo Bustugan, Short synthesis of ( )-bakuchiol via a
Shah, Chandrakant C., Bhalla, V.K., Sukh Dev, Chem.Abstr., 129: 81856.
Ogawa, S., Tamura, K., Oda, M., Inomoto, T., Kataoka, S., Japanese patent
Kokate, C.K., Purohith, A.P., Gokhale, S.D., Pharmacognosy. 12t1 ed., Nirali
Wagner. H., Bladt, S., Zagainski, E. M., Plant Drug Analysis, Springer-Verlag.
Wagner, H., Bladt, S., Zagainski, E. M., Plant Drug Analysis, Springer-Verlag.
657-658, 1977.
(chapter 9), p. 401, in Analysis of Oils and fats, Hamilton, R. J., Rossell, J. B.,
117
ACD 50 H NMR predictor
1982.
118
Appendix
119
Appendix i List of Publications / Communications
other Natural Products; Trends and Advances, Colombo: Programme and Abstracts P.
41.
120
Appendix ii: Spectral data of
Isopsoralen
Psoralen
Bakuchiol
Dehydroisopsoralidin
Corylin
Psoralidin
Isobavachalcone
121
E
cx
I0
122
.1
"I
123
N
124
125
9*-
11
ET
J
126
127
128
129
E 0 0 0 0 0 0 0 0 0 0 0 0 0 0
N 0 - N N fl
130
P
131
132
fl0(ft
133
U." 0
A
411
11•iu/
t91
9•S
134
69Lt
,Ht
135