Chemistry And Standardization Of Bakuchi Oil An Ayurwedic Medicial Oil Used Trasditionally In Thee Treatment Of Vitiligo.pdf

Chemistry and Standardization of "Bakuchi oil" an Ayurvedic medicinal oil used traditionally in the treatment of vitiligo.

Ranamuka Devage Aj antha Rohinie Gunawardena

Thesis submitted to the University of Sri Jayewardenepura for the award of the Degree of Master of Philosophy in chemistry

- I-..-------'-.

-

---.-

-

- -'-- -- - -

-.

-

-

-

I dedicate this thesis to my parents

DECLARATION The work described in this thesis was carried out by me under the supervision of Prof. A. M. Abeysekera (Department of Chemistry, University of Sri Jayewardenepura), Prof. G. M. K. B. Gunaherath (Department of Chemistry, The Open University of Sri Lanka.) and Dr. C. D Jayaweera (Department of Chemistry, University of Sri Jayewardenepura) and a report on this has not been submitted in whole or in part to any University or any other institution for other Degree / 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.

.....14 ...........

...... ........... .........

................

Prof. A.Ayvsekera. Mt

Prof. G. M. K.B. Gunaherath.

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

1.1

Traditional systems of medicine and Ayurveda

1.1.1

Types of Ayurvedic drugs

112

Standardization ofAyurvedic drugs

2

1.2

Vitiligo

6

1.2.1

Treatment methods

6

1.2.2

Psoralen Drugs

7

1.2.3

"Bakuchi oil"

10

1.2.3.1 Preparation of "Bakuchi oil"

10

1.2.3.2 Importance of standardization of "Bakuchi oil"

11

1.3

Psoralea corylfoiia L

12

I .3.1

Chemical constituents of Psoralea coryl(folia

13

1 .3.2

Biological activities of Psoralea coryIfolia

20

1.4

Scope of the work described in this thesis

26

2.0

Materials and Method

27

2.1

Materials

27

2.2

General procedures

27

2.3

Isolation and chemical characterization of compounds from the fruits of Psoralea corylifolia

30

2.3.1

Psoralen (1) and isopsoralen (2)

30

2.3.1.1

Separation of the two isomers

31

2.3.2

Bakuchiol (38)

33

2.3.3

Fixed oil

34

2.3.4

Dehydroisopsoralidin (42), corylin (11), psoralidin(5) and isobavachalcone (30)

2.3.5

34

Extraction and GC finger printing of essential oil of the fruits of Psoralea corylifolia

37

2.3.5.1

GC finger print of the essential oil

38

2.4

Analysis of 'Bakuchi oil"

38

2.4.1

Preparation of "Bakuchi oil' extracts for TLC experiments

38

2.4.2

Gas Chromatographic experiments to study the fate of bakuchiol

39

2.5

Standard i zation of "Bakuchi oil'

41

2.5.1

TLC finger prints

41

2.5.2

Determination of the concentration of Psoralen in "Bakuchi oil"

42

2.5.2.1

Standard curve

42

2.5.2.2

Quantitative extraction of psoralen from "Bakuchi oil"

43

2.5.2.3

Method validation

44

2.5.3

Psoralen concentration in different samples of "Bakuchi oil"

45

EI

2.5.4

Psoralen concentration in different batches of Psoralea corylfo1ia fruits

2.5.5

46

Psoralen concentration in sesame oil extract of seed powder and the extraction efficiency

47

2.5.6

Solubility of psoralen in sesame oil

48

2.5.7

Psoralen concentration in water extract of Psoralea corylifolia fruits and the extraction efficiency by water

48

2.5.8

Solubility of psoralen in water

49

3.0

Results and Discussion

51

3.1

Studies on the chemical composition of P.coryIfoiia and "Bakuchi oil'

3.2

51

Isolation and identification of compounds from the fruits of P.corylifo!ia

52

3.2.1

Psoralen (1) and isopsoralen (2)

53

3.2.2

Bakuchiol (38)

57

3.2.3

Fixed oil

63

3.2.4.

Dehydroisopsoralidin (42)

64

3.2.5

Corylin (I 1)

69

3.2.6

Psoralidin (5)

72

3.2.7

Isobavachalcone (30)

75

3.3

Analysis of "Bakuchi oil"

80

3.4

Standardization of "Bakuchi oil"

84

3.4.1

Chromatographic finger prints

84

3.4.2

Quantification of psoralen in "Bakuchi oil"

89

"U

3.5

Psoralen concentration in different "Bakuchi oil' samples

94

3.6

Psoralen concentration in different batches of P.corylsfolia fruits

95

3.7

Effect of processing parameters on the composition and the quality of the drug.

96

3.7.1

Preparation of 'Bakuchi oil"

96

3.7.2

Rate of incorporation of psoralen

98

3.7.3

Study of the fate of 'bakuchiol" during the manufacturing Process

101

4.

Conclusion

109

5.

References

110

6.

Appendix

119

Appendix i List of publications

120

Appendix ii: Spectral data.

121

lv

LIST OF TABLES

Table

page

1. Compounds reported from P. coryljfolia.

13

2. Densitometric readings obtained for the standard curve for psoralen.

43

3. Densitometric readings obtained for the psoralen spot in six replicates of the reference oil sample.

44

4. Densitometric readings obtained for the psoralen spot in the addition recovery experiment.

45

5. Densitometric readings obtained for the psoralen spot in different samples of 'Bakuchi oil.

46

6. Densitometric readings obtained for the psoralen spot in different batches of Psoralea corylifolia fruits.

47

7. Densitometric readings obtained for psoralen spot in experiments described in sections 2.5.5 - 2.5.8.

50

8. Compounds isolated from the fruits of P.cor'Iifoiea. and some of their physical properties.

53

9. NMR data for bakuchiol.

60

10. NMR (d6-DMSO, 600 MHz) data for dehydroisopsoralidin.

67

II. NMR data for corylin.

71

Coniparision of 'H NMR data of psoralidin with reported data.

74

NMR (d6-DMSO, 600 MHz) data for isobavachalcone.

77

Psoralen concentration in different samples of Bakuchi oil'.

95

V

15. psoralen concentration in different batches of P. cory4foia fruits.

95

16. Comparision of psoralen in "Bakuchi oil" and different extracts of P. corylifolia. 17 Solubility of psoralen

MI

vi

LIST OF FIGURES Figure

page

I. Cycloaddition reaction of psoralen with the 5, 6 double bond of thymine in DNA.

8

2. Final stage in the large scale preparation of the medicinal oil.

11

3. Psoralea corylifolia- plant.

12

4. Psoralea corylifolia-fruit.

12

5. Some HMB data of bakuchiol.

59

6. ö 4.9 - 6.0 ppm region of the 'H NMR spectrum of bakuchiol.

61

7. TLC of methanol extract of fruits of P. cory/ifolia.

62

8. Some key HMBC data for the coumestan fragments.

68

9

Appearance of 'H NMR pattern corresponding to H-b and H-c protons, and the doublet corresponding to 1-1-2' and 11-6'.

76

10 GC finger print of essential oil of the fruits of P.corylifolia.

79

11 TLC of "Bakuchi oil'.

82

12 TLC of fruits of P.corylifolia and "Bakuchi oil".

83

13 TLC finger prints of "Bakuchi oil".

86

14 Densitograms of thin layer chromatograms shown in Fig. 12.

87

15 Densitograms of the 'Bakuchi oil sample collected from Ayurvedic Drugs Corporation.

88

16 Standard curve for psoralen.

93

17 GC of an equal weight mixture of psoralen and bakuchiol

104

18 Gas chromatogram of 80% methanol soluble fraction of the hexane extract of Psoralea corylifolia fruits.

vii

105

19. Gas chromatogram of steam distillate of 'Bakuchi oil.

106

20. Gas chromatogram of steam distillate of of sesame oil extract of powdered fruits.

107

21. Gas chromatogram of steam distillate of seed residue after extracting with sesame oil.

108

22. 'H NMR spectrum of isopsoralen

122

23. 'H NMR spectrum of psoralen

123

24. 'H NMR spectrum of bakuchiol

124

25. 13C NMR spectrum of bakuchiol

125

26. 'H NMR spectrum of dehydroisopsoralidin

126

27. HMB correlations of dehydroisopsoralidin

127

28. HMB correlations of dehydroisopsoralidin

128

29. 13C NMR spectrum of dehydroisopsoralidin

129

30. HSQC spectrum of dehydroisopsoralidin

130

31. 'H NMR spectrum of corylin

131

32. '3C NMR spectrum ofcorylin

132

33. 'H NMR spectrum of psoralidin

133

34. 'H NMR spectrum of isobavachalcone

134

35. 13C NMR spectrum of isobavachalcone

135

viii

LIST OF ABBREVIATIONS AIDS: Acquired Immune Deficiency Syndrome BMARI: B andaranayake Memorial Ayurved ic Research Institute. BHT: Butylated Hydroxy Toluene DAD: Diode Array Detector DEPT: Distortionless Enhancement by Polarization Transfer DMSO: Dirnethyl Sulfoxide DNA: Deoxyribo Nucleic Acid EC 50: Half maximal Effective Concentration El: Electron Impact ESIMS: Electro Spray Ionization Mass Spectrometry FID: Flame Ionization Detector FT-IR: Fourier Transform- Infra Red GC: Gas Chromatography GC-MS: Gas Chromatography- Mass Spectrometry HMBC: Heteronuclear Multiple Bond Correlation HPLC: High Performance Liquid Chromatography HPTLC: High Performance Thin Layer Chromatography HSQC: 1-leteronuclear Single Quantum Coherence Hz: Hertz LCMS: Liquid Chromatography- Mass Spectrometry MIC: Minimum Inhibitory Concentration NMR: Nuclear Magenetic Resonance PDA: Photo Diode Array

ix

PEG: Poly Ethylene Glycol PUVA: Psoralen - Ultra Violet A TLC: Thin Layer Chromatography TLC-FD: Thin Layer Chromatography - Fluorescence Detection UV: Ultra Violet WHO: World Health Organization

Acknowledgements I take this opportunity to express my warmest gratitude to my supervisor, Prof. A. M. Abeysekera, Professor of Chemistry and Dean, Faculty of Applied Sciences, University of Sri Jayewardenepura, for his invaluable guidance, keen supervision, 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 working as the Director at Bandaranayake Memorial Ayurvedic Research Institute.

I wish to express my sincere gratitude to my supervisor, Prof. G. M. K. B. Gunaherath, Professor of Chemistry, Department of Chemistry, The Open University of Sri Lanka for his keen supervision, commitment, invaluable suggestions and advice given me through out my research work.

I would also like to express my sincere thanks to my supervisor, Dr (Mrs.) C. D. Jayaweera, Senior Lecturer, Department of Chemistry, University of Sri Jayewardenepura for her encouragement, advice and invaluable suggestions I am grateful to Prof. R. J. Capon. Institute of Molecular Bioscience, University of Queensland, Australia for providing me NMR and LC-MS data while I would like to extend my sincere thanks to Dr. Ranjala Ratnayake, Institute of Molecular Bioscience. University of Queensland. Australia for her assistance given in obtaining all NMR and Mass spectra and for her contribution to the structure elucidation.

xi

I also wish to express my thanks to the Department of Chemistry, University of Peradeniya for providing me NMR and GC-MS facilities.

I take this opportunity to express my thanks to the Department of Chemistry, University of Sri Jayewardenepura for providing facilities to carry out my research. The corporation of the academic, technical and nonacademic staff of the Department of Chemistry given me through out the work is gratefully acknowledged. My special thank goes to Mr. Sri Lal Rangoda for his assistance given me in carrying out laboratory work through out my research.

I acknowledge the support given by the Department of Ayurveda and Bandaranayake Memorial Ayurvedic Research Institute at the initial stage of the work.

I also take pleasure to thank Dr (Mrs.) U.G. Chandrika, Senior Lecturer, Department of Biochemistry, University of Sri Jayewardenepura and my colleague Mrs. Chandani Ranasinghe for their support given me in numerous ways.

I wish to acknowledge National Science Foundation for partial funding this work.

Finally I would like to express my deepest gratitude to my mother for her encouragement and kindness extended to me to achieve my goals.

xii

ABSTRACT Chemistry and Standardization of "Bakuchi oil" an Ayurvedic medicinal oil used traditionally in the treatment of Vitiligo. Ranamuka Devage Ajantha Rohinie Gunawardena Vitiligo is a disfiguring disease, which is characterized by the appearance of white 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 and Ayurvedic systems of medicine. However, according to available data, any synthetic form of psoralen or extracts from furanocoumarin containing herbs cannot be considered risk free. So their use in the treatment needs specialized medical expertise and standardized products. "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 corylifolia contains a considerable amount of psoralen type compounds, the most 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 30 % of psoralen was added.

"Bakuchi oil" samples collected from different manufacturers gave a range of 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 method which involved selective solubility in a mixture of acetone and methanol at low temperatures was developed. Six of the compounds present in Psoralea cory'lfolia that were incorporated in "Bakuchi oil" were identified by detailed comparative chromatographic analysis of the plant extract and the medicinal oil. The six compounds are psoralen, isopsoralen, dehydroisopsoralidin, corylin, isobavachalcone, and psoralidin. Of these, dehydroisopsoralidin is a new natural product. Structures of the isolated compounds 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 found in the fruits of Psoralea coryliiia is not incorporated in "Bakuchi oil"even though it is quite soluble in sesame oil and is stable at 140° C, the highest temperature reached during processing.

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 fruits with sesame oil is suggested as being worthy of further study.

xv

1. Introduction

1.1 Traditional systems of medicine and Ayurveda. In recent years traditional systems of medicine have become a topic of global importance. In spite of the phenomenal progress of modern medicine there is a 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 developing countries, while it is increasingly becoming popular in developed countries, 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

1.1.1 Types of Ayurvedic drugs 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 selfgenerated alcohol content. 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 passing through a cloth. Thila: medicated oils, prepared by heating vegetable oils with fresh plants and or the decoction orjuice of plants. Swarasa: expressed juice of fresh herbs. Kvatha: decoctions of medicinal plants.

1.1.2 Standardization of Ayurvedic drugs. 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 medicinal formulations were prepared by the physicians themselves and delivered to 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 tremendous expansion in the use of traditional medicine has resulted in industrial 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 modern techniques by applying suitable standards.

Standardization should be done at three different levels, namely, raw materials, process and finished drugs.

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.

The standardization of Ayurvedic preparations is very complex because many ingredients of variable composition are used in the manufacture of a single preparation. 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 action of the drug. Therefore modern pharmaceutical standards cannot be readily applied for the standardization of Ayurvedic drugs and a different approach is needed.

Although there is a lack of common standards, understanding and appropriate methods 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 techniques such as Thin Layer Chromatography (TLC), High Performance Liquid Chromatography (HPLC), Gas chromatography (GC) for chromatographic finger printing and quantification of specific compounds (marker compounds) can provide information on the identity and potency of complex herbal preparations.

The use of marker compounds in the standardization of herbal drugs is now well established. Marker compounds are compounds characteristic of a plant and specific 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 the "active principle" is now considered to be hyperforin 6

Chromatographic finger prints provide a better overall view of the composition of the drug, than a single marker compound. Of the instrumental methods available GC is very important in finger printing of volatile components in a formulation. The extraction of volatile components can be conveniently standardized and individual components in the profile can be identified by coupling the gas chromatograph to a mass spectrophotometer. (GC-MS analysis). The GC profile of a popular Ayurvedic drug. "Dasamoolarishtaya" has been reported recently.7 GC combines high sensitivity 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 compound. Furthermore, sample preparation is simple. It is also possible to analyze 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 compounds, sLipported by GC for steam volatile compounds.

While the comparison of standard finger prints with test results is usually done visually, computer based comparison of complex finger prints for 'sameness' and "difference" in a meaningful manner is now drawing the attention of researchers.8

61

1.2 Vitiligo Vitiligo is a common acquired idiopathic pigmentory disorder which is characterized by well demarcated patches of depigmented skin due to loss of cutaneous 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 suggesting metabolic errors, autoimmunity, genetic tendencies and hormonal connection. In some cases vitiligo is believed to be caused by mental stress, frequent use of broad spectrum antibiotics, trauma, constant pressure on the skin, dietary deficiency and low serum copper levels."

1.2.1 Treatment methods 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 common type of compounds used in repigmentation therapy are the psoralen derivatives. In the depigmentation therapy the remaining cells are also destroyed to match the already white areas. In some cases surgical transplantation of melanocytes and the use of cosmetics may also be helpful.

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.

Photo chemistry of psoralens 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 state life times are significantly longer (iMs to 1 s) 16

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 6

ON DNA

4

I

0 0

40

CH3

DNA'

hv

0

\ v 0 0

oQj

0 0 O HN

HN 0 DNA 0 DNA

NNH DNA

1( 0

mono-adducts

cross linked di-adducts

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 for mono adduct formation.18

Photoactivated psoralens can bind to other cellular components such as unsaturated 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 important in photosensitizing activity of these compounds. It is of biological significance that only the linear furanocoumarins such as psoralen, Xanthotoxin (8methoxy psoralen) and Bergapton (5- methoxy psoralen) form interstrand cross linkages and are thus able to induce cutaneous photosensitization; Angular furariocournarins such as isopsoralen (angelicin), which form single strand covalent linkages, lack skin photosensitizing ability.20

Although psoralen photocheinotherapy is extremely effective in the treatment of skin 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 potentially hepatotoxic. Among the short-term hazards, burning or blistering of the skin, gastric discomfort, nausea, nervousness, depression can be seen. 72223 24

According to available data, any synthetic form of psoralen or extracts from furanocoumarin containing herbs cannot he considered risk free. So their use in the treatment needs specialized medical expertise and standardized products.

1.2.3 "Bakuchi oil' "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 Memorial Ayurvedic Research Institute by its former Assistant Director, Dr. 25 H.I.Chandrasekera in 1968 and was documented at the BMARI. Since then this 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 in the morning. (between 8.00 a.m. and 9.00 am.).

1.2.3.1 Preparation of "Bakuchi oil". "Bakuchi oil" is prepared by heating a decoction of whole fruits of Psoralea corylifàlia with sesame oil until the water is evaporated completely. Powdered fruits are added directly to the oil, towards the end of this process.

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 then the whole process takes about two weeks.

hr

T

&!'kt. Fig.2 Final stage in the large scale preparation of "Bakuchi oil'.

1.2.3.2 Importance of standardization of "Bakuchi oil". The fruits of Psoialea corylifb/ia which are used to prepare "Bakuchi oil" contains considerable amount of psoralen type compounds which are known to be highly 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 psoralen content. Psoralen can be considered to be both a marker compound reflecting the overall compostion (and therefore the efficacy of the drug), as well as an "active principle" of the drug.

1.3 Psoralea coryilfolia L

.1

kftr

_e

4

.4

Fig. 3 Psoralec: coiy/ifo/ici- plant

Fig. 4 J'sora/ea coiyli/b/ia-fruits

Psoraleci cory/ifo/ici Lin. (fabaceae) is a commonly occurring medicinal plant in South 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 in traditional medicine to treat a wide variety of diseases, specially in India and China.27 These fruits have been widely used for the cure of gynaecological bleeding, leprosy and specially in certain skin diseases such as vitiligo and psoriasis.28'

29, 30.

Some therapeutic properties of these fruits such as laxative, diuretic, and anthelmintic have also been reported.3 '

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 compounds which belong to different chemical groups such as coumarins, coumestans, 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 given in the Table 1.

Table 1: Compounds reported from P.cor''1iJ61ici 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)-2,2dimethyl-2H, 6H-benzo[ 1 ,2-b:5,4-b']dipyran6-one. (25)

28

Chalcones Corilifols B (26)

28

Brosimacutin 0 (27)

28

1 -[2,4-dihydroxy-3-(2-hydroxy-3-methyl-3 'butenyl)phenyl]-3-(4-hydroxyphenyl)-2propen- 1-one. (28) Bakuchalcone (29)

28

Isobavachalcone (30)

28, 29, 46, 47, 51.

14

28, 50

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.

1 ,3-Hydroxybakuchiol (40)

61, 49.

3,2-Hydroxybakuchio1 (41)

61.

Coumarins

cax

CI u

Psoralen (1)

Isopsoralen (2)

0

0

0

I 9 0n~'~ Bakuchicin (3)

Neopsoralen (4)

15

Coumestans

HO

0 0

>OOO HO OH H

I]

Psoralidin (5)

Corylidin (6)

-

JH

Bavacoumestan A (7)

Bavacoumestan B (8)

Sophorocoumestan A (9)

Psoralidin oxide (10)

Flavonoids

Corylin (11)

Psoralenol (12)

I I

10 ~ 1

OH

c

Neobavaisoflavone (13)

Corylinal (14)

16

0

0 H

OH HO 0

HOZyO1

Bavachin (15)

Isobavachin (16) OH

OCH3

B avachinin (7-0-methylbavachin) (17)

4'-methoxy flavone (18)

OH

Al Corilifols A (19) HO

Corilifols C (20)

0

OH

Isoneobavaisoflavone (21)

Erythrin A (22)

HO HO

O

o

OH

OH

8- Prenyldaidzein (23)

Daidzein (24)

17

OH

0

7,8-dihydro-8-(4-hydroxyphenyl)-2,2dimethyl-2H, 6H-benzo[ 1 ,2-b:5,4b]dipyran-6-one. (25)

Chalcones

OH

OH

OI f

A

JIIIITIII OH

0

Corilifols B (26)

Brosimacutin G (27)

OH

.

1- [2,4-dihydroxy-3 -(2-hydroxy-3methyl-3' -butenyl)phenyl ]-3 -(4hydroxyphenyl)-2-propen- 1-one. (28)

B akuchalcone (29)

0(7)ZIIJ1

OH

OCH3) OHC

0

Isobavachalcone (30)

5'-formyl-2', 4-dihydroxy-4'methoxychalcone (31)

OH çOH

0cr

>O

LJ

(OH<

0

0

Isobavachromene (32)

Bavachromene (33) OH

HO) OH OHC :: 0

101

Isoneobavachalcone (34)

Bavachromanol (35) OMe

Bavachalcone (36)

4'-0- methylbavachalcone (37)

Terpenoids

HOX<"

HO

Bakuchiol (38)

2,3- Epoxybakuchiol (39)

/1

X

HO

<

OH

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).

Psoralen is considered to be the major photoactive compound of Psoralea coryIift1ia Derivatives of Psoralen are widely used as phototherapeutic agents in the treatment of certain skin diseases such as vitiligo and psoriasis. 34

These compounds are toxic to a variety of organisms. Toxicity of these compounds is 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 psoralen (linear furanocoumarin) was more than the isopsoralen (angular 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 Psoralea coryiiftlia. In an in vitro experiment, antimicrobial activities of bakuchiol

against some oral bacteria have been evaluated. It showed bactericidal effects of bakuchiol against Streptococcus inutans, Streptococcus san guis. Streptococcus salivarius. Streptococcus sobrinus, Enterococcus faecaiis, Enterococcus faeciu,n. Lactohacillus acidophilus, Laclobacillus plan tarurn, L. casei, Actinoinyces viscosus, 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 suggested that bakuchiol would be a useful compound for the development of 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 NO (nitric oxide) synthase (iNOS) gene through the inactivation of nuclear transcription factor-kB (NF-kB), hence inhibit NO production. They suggest that 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 antimutagenic, antiinfiammatory, insect juvenile hormone, anti-helmenth ic, and cytotoxic activities. 29, 58, 59

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 no scientific work has been reported on its activity.63

Ethanol extract of Psoralea corylifolia caused strong DNA polymerase inhibition in a 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 neobavaisoflavone (13) as DNA inhibitors. Bakuchiol (38) is reported as the strongest DNA polymerase inhibitor found in Psoralea corylifolia. Apart from that daidzeiri was identified as a DNA polymerase and topoisomerase II inhibitor, and bakuchicin (3) as a topoisomerase II inhibitor.27

During the study of antioxidative materials in botanicals sources, Psoralea cory/ifrilia seed powder and its extracts have been examined by Gue et al. and shown that it has strong antioxidant activities. It has been suggested as a potential antioxidant resource 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 (butylated hydroxyl toluene).35

22

Hiroyule Ct al. has found that Psoralea seed extracts showed potent inhibition of mitochondrial microsomal lipid peroxidation. The fractionation of active extracts had yielded bakuchiol (38), bavachin (15), isobavachin (16) and isobavachalcone (30). Out of those compounds, bakuchiol was found to be the most potent antioxidant in microsomes. It also showed bakuchiol protected human red blood cells against oxidative haemolysis. Further these phenolic compounds in psoralen were shown to be effective in protecting biological membranes against various oxidative stresses.29

During an antibacterial screening programme, the ethanol extract of the seeds of Psoralea corv/ifb/ia has shown remarkable antibacterial activities against two major pathogenic bacteria, Staphylococcus aureus and Staphylococcus epidertnidis. When the extract was fractionated sixteen compounds could be isolated and out of them nine compounds showed in vitro antibacterial activity against Staphylococcus aureus and Staphylococcus epidermidis at the level of MIC 0.009-0.073 mM. These compounds were corylifol B (26), neobavaisoflavone (13), isobavachalcone (30), 7, 8-dihydro-8(4-hydroxyphenyl)-2,2-dimethyl-2H,6H-benzo[ 1,2-6:5,4-6' ]dipyran-6-one

(25),

isoneobavaisoflavone (21), bavachalone (36), bavachin (15) and bavachinin (17). Antibacterial activities of isobavachalcone (30), havachinin (17) and erythrin A (22) were found to be higher than bakuchiol and magno!ol.28

The aqueous extract of Psoralea corvlifolia has been found to have hepatoprotective properties. An experiment which led to isolate the active compounds, yielded one hepatoprotective compound, bakuchiol (38) and two other moderately active

23

compounds psoralen (1) and bakuchicin (3)30 This has to be investigated further, considering that psoralen has been reported as hepatotoxic in some studies.22

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 fractionation had led to isolate three flavonoids, isobavachalcone (30), neobavaisoflavone (13) and bavachin (15) as active compounds.47

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 stimulate hone formation or have potential activity against osteophorosis.44

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 showed antifungal activities against several fungi such as Aspergillus niger, Aspergilius Jwnigatus, Aspergillus candidus, Tricophyton rubrum, Tricophyton mentagrophyres and Candida albica11s.26 Further the essential oil of the fruits has been reported to be active against Streptococci and paramecium.64

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 a powerful tool in modern drug development.

OR

1.4 Scope of the work described in this thesis. The work described in this thesis covers the following areas. Chemical characterization and standardization of the drug, 'Bakuchi oil" and the effect of processing on the composition and quality of the drug.

We attempted to characterize the drug "Bakuchi oil" by identifying the compounds from Psoralea corylifolia fruits that were incorporated in it. This information was used to standardize the drug qualitatively by establishing chromatographic finger prints, and 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 psoralen and bakuchiol.

2.0 Materials and Methods.

2.1 Materials

Plant materials. Dried fruits of Psoralea corylifolia were purchased from a local drug store and their authenticity was confirmed by comparing them with the herbarium specimens at BMARL

Sesame oil (Gingelly oil). Sesame oil was purchased from a local drug store and tested for the common adulterant, cotton seed oil, before use.

65

Bakuchi oil. A sample of Bakuchi oil prepared at the BMARI (BMARI-I) was used as the reference sample.

2.2 General procedures H NMR and 13 C NMR spectra were measured at 600 MHz and 150 MHz respectively on a Bruker avance 600 spectrophotometer or at 300 MHz and 75 MHz respectively on a Varian 300 spectrophotometer in d6-DMSO or CDC13 with or without tetramethylsilane as an internal standard. Chemical shifts are given in 8 (ppm). IR spectra were recorded on a Thermo Nicolet AVATAR 320 FT-JR spectrometer (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 on a Varian Saturn 2000 mass spectrophotometer coupled with Varian 3800 GC ( GC 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 array-mass spectrometry (HPLC-DAD-MS), were measured on a Agilent 1100 Series separations module quaternary (G131 1A) pump (G1312A) with vacuum degasser (G1379A), diode array detector (DAD) (G1315B), colunrn compartment (G1316A) and quadrupole mass detector (G1946D). GC studies were carried out on Agilent 689 N GC system with a FID.

Conditions for GC analysis: Sample size: 1.00 t1 Injector temperature: 250° C Detector (FID) temperature: 270° C Carrier gas: He Flow rate: 2 ml mind 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 out within 30 minutes after the development.

Melting points were recorded on a Reichert Thermover hot stage melting block apparatus. All melting points recorded are uncorrected. 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 analytical and preparative plates were 0.25mm and 0.5 mm respectively.

Chromatographic spray reagents 10% Methanolic KOH reagent. Anisaldehyde sulphuric acid reagent. 66 0.5 ml anisaldehyde was mixed with 10 ml of glacial acetic acid, followed by 85 ml of methanol and 5 ml of concentrated sulphuric acid, in that order. Natural Products- polyethyleneglycol reagent. (NP/PEG) 67

The plate was sprayed with 1 % methanolic diphenylboric acid-/3-ethylamino ester (NP), followed by 5 % ethanolic polyethyleneglycol-4000(PEG). 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 purified in the laboratory by fractional distillation.

2.3 Isolation and chemical characterization of compounds from the fruits of Psoralea corylifolia.

2.3.1 Psoralen (1) and isopsoralen (2) Procedure

168

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 in the cold. The precipitated psoralenhisopsoralen mixture was then separated by 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 coloured amorphous mass (6.3 g, 0.79 %).

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 psoralen/isopsoralen mixture was then separated by suction filtration and the 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 %).

2.3.1.1 Separation of the two isomers The mixture of the two isomers (0.893 g) were chromatographed over a silica gel column, using a gradient of ethyl acetate in hexane. Isopsoralen started eluting at 5 % 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 compound. Both were crystalline solids.

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 psoralen was recrystallzed from hexane/ethanol as white needles. Purity of the

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.

Psoralen (1) : White needles; m. p. 161-163° C (lit. 35m. p. 162-163° C ), UV 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), 114.66 (C-3), 106.42 (C-il), 99.84 (C-8).

Isopsoralen (2) : White needles; mp.140-143°C (lit.46 m. p. 143-144° C, lit.35 m.p.137138° 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, 116.94, 114.17, 113.53, 108.80, 104.12.

32

2.3.2 Bakuchiol (38) Powdered fruits (225.0 g) were subjected to sequential extraction in a soxhiet apperatus 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 colourless oily compound.

Bakuchiol (38): ESIMS: mlz: 255[M-H], 257[M+H], 'H NMR ( CDC13, 600 MHz) 6: 1.20 (3H, s, H16), 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, H11), 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, H8), 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), 23.21 (C-li), 17.63 (C-15).

2.3.3 Fixed oil. 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.

Fixed oil: Pale yellowish oil 'H NMR: (CDC13 300 MHz) 6:0.89(m), 1.28 ( m). 1.60 (m), 2.03 (m), 2.30 (in), 2.78 (m), 4.14 (dd). 4.29 (dd). 5.34 (m).

2.3.4 Dehydroisopsoralidin (42), corylin (11), psoralidin (5) and isobavachalcone (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 separated by flash column chromatography (ethyl acetate: hexane, 1: 20) to obtain a 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 too contained dehydroisopsoralidin, which could be precipitated (2 mg) by the addition 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 white precipitate deposited on addition of dichloromethane to the fraction 23 which 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 to preparative TLC (toluene: ethyl acetate: acetone, 11: 1: 1, triple development ) to obtain more psoralidin (5) (9 mg, 6x10 3 %)) and isobavachalcone (30) as yellow needles (8 mg, 5x10 3 %).

Dehydroisopsoralidin (42) Pale yellowish crystalline solid m. p. 290° C (decomp.) ( 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 (C-4"), 131.8 (C-5"), 77.8 (C-6"), 27.9 (C-6"-Me).

Corylin (11): White crystalline solid; m. p. 242-243° C (lit.31m.p.2282300 C, lit. 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, H2'), 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), 102.1 (C-8), 76.24 (C-6"), 27.7 (C-6"-Me).

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).

Isobavachalcone (30): Bright yellow needles; m. p. 153-154° C (lit.51 M. p. 154-156° 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 (C2), 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 (C- 6"), 27.71 (C-6"-Me).

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 mg (0.081 %) of essential oil. (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 (0.082 %, w/w) of essential oil.

2.3.5.1 GC finger print of the essential oil. 3.0 mg of essential oil was dissolved in 5.00 ml of dichlorometharie and 1.00 j.tl was injected into GC. (temperature programming 1)

2.4 Analysis of "Bakuchi oil".

2.4.1 Preparation of "Bakuchi oil" extracts for TLC experiments. 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 / absence of compounds isolated from Psoralea cory1ftilia by the following three 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 18111 January 2010

Dean Faculty of Graduate Studies University of Sri Jayewardenepura

M.PHJL THESIS OF MS. R.D.A.R. AJANTHA GUNAWARDENA CHEMISTRY A ND STANDARDIZA TION OF 'BAKUCHI' OIL A N AYURVEDIC MEDICIAL OIL USED TRADITIONALLY IN THE TREA TMENT OF VITILIGO

-

-

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. Please be good enough to issue a copy to me so that I can hand it over to Ii er.

~~"

A'sekera Prof. A.M. t Dean/Faculty of Applied Science

/ ethyl acetate and ethyl acetate / methanol to yield 13 fractions. Each fraction was chromatographed on TLC plates along with isolated compounds of Psoralea corylifolia and observed at 366 nm after spraying with 10% methanolic potassium hydroxide (In order to confirm the presence of isolated 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" was directly chromatographed on TLC plates alongside isolated compounds of Psoralea cory/itolia with triple development using toluene: ethyl acetate: hexane, 1: 3: 9 as the eluvant. The plates were observed at 366 nm after spraying with 10% methanolic potassium hydroxide and at 254 nm without spraying. (in order to confirm the presence of isolated compounds in the "Bakuchi oil" co-TLC were carried out)

III.

The 80% methanol soluble fraction of the methanol extract of the "Bakuchi oil" was chromatographed on TLC plates alongside the total methanol extract of Psoralea corylif b/ia fruits using the solvent system toluene: ethyl acetate: hexane. 1: 3: 9, and observed at 366 nm after spraying with 10 % methanolic potassium hydroxide, at 254 nm without spraying and in the visible light after spraying with anisaldehyde sulphuric reagent.

2.4.2 Gas Chromatographic experiments to study the fate of bakuchiol. (1) Bakuchiol (1 mg) was dissolved in 1.00 ml of dichlorornethane and 1.00 p1 of the solution was injected into GC. (Temperature programming 1 and 2)

39

Psoralen (1 mg) was dissolved in 1.00 ml of dichloromethane and 1.00 jil of the solution was injected into GC.(Temperature programming land 2) Isopsoralen (1 mg) was dissolved in 1.00 ml of dichloromethane and 1.00 ti of the solution was injected into GC. (Temperature programming 1) 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 under reduced pressure. The extract (2 mg) was dissolved in 2.00 ml of 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 partitioned with dichloromethane, dichloromethane fraction was dried over anhydrous 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 solution was injected into GC. (Temperature programming 1) "Bakuchi oil" (60.00 ml) was steam distilled for 4 hours, steam distillate was partitioned with dichloromethane, dichloromethane fraction was dried over anhydrous 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 was injected into GC. (Temperature programming 1) 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 distilled for 4 hours, steam distillate was partitioned with dichloromethane,

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 into GC. (Temperature programming 1) 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 injected into GC. (Temperature programming 1) Bakuchiol (6.0 rng) was dissolved in 5.00 ml of dichloromethane and 6.0 mg of 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)

2.5 Standardization of "Bakuchi oil".

2.5.1 TLC finger prints "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 (BMARI-II, BMARI-III, Beliatta Ayurvedic hospital, Kurunegala Ayurvedic hospital, Ayurvedic Drugs Corporation) were analyzed in the same way and compared with the reference oil sample.

2.5.2 Determination of the concentration of Psoralen in "Bakuchi oil".

2.5.2.1 Standard curve 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= 7.2844 x + 1933.3040) was linear with a correlation coefficient of 0.9931.( Fig. 16 ).

Table 2: Densitometric readings obtained for the standard curve for psoralen.

Standard

Weight of psoralen I spot

Densitometric reading

(ng) 1

200

3069.402

2

400

5207.474

3

600

6336.803

4

800

7896.882

5

1000

9009.066

2.5.2.2 Quantitative extraction of psoralen from "Bakuchi oil". 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 psoralen concentration was determined using the standard curve.

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 reference oil sample. Replicate

Densitometric reading

1

4164.210

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 up to 200.00 ml in a volumetric flask. An aliquot (4.00 ml) of the solution was dissolved in 50 ml of acetone and proceeded as described in the section 2.5.2.2. The densitometric reading obtained is given in the Table 4 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. Sample

Densitometric reading

Reference sample

4341.780

Sample after addition of psoralen

5092.499

2.5.3 Psoralen concentration in different samples of "Bakuchi oil". Psoralen concentrations of six different oil samples (BMARI-I, BMAR-II. Ayurvedic Drugs Corporation, Beliatta Ayurvedic hospital, Kurunegala Ayurvedic hospital, BMARI-III) were determined as described bellow.

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 each sample are given in the Table 5. Table 5: Densitometric readings obtained for the psoralen spot in different samples of "Bakuchi oil". "Bakuchi oil sample

Densitometric reading

BMARI - I (reference)

4341.78

BMARI —1I

8511.05

Ayurvedic Drugs corporation

4460.64

Ayurvedic hospital - Kurunegala

3028.73

Ayurvedic hospital - Beliatta

4924.83

BMARI - 111*

7559.60

* p repa red from fruit batch - I Psoralen concentration of each oil sample was determined using the standard curve.

2.5.4 Psoralen concentration in different batches of Psoralea corylifolia fruits. 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 Psoralea corylifoha fruits. Batch of P.coryhfolia fruits

Densitometric reading

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 given in the table 7.

47

Psoralen concentration of the sesame oil extract of seed powder was determined using the standard curve, and the extraction efficiency was calculated.

2.5.6 Solubility of psoralen in sesame oil. Psoralen was added little by little into 6 ml of Sesame oil and stirred well with a magnetic stirrer at room temperature, When approximately 28.9 mg of psoralen had 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 obtained is given in the table 7.

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 extraction efficiency by water. 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 the densitometric reading obtained is given in the table 7.

Psoralen concentration was determined using the standard curve and the extraction efficiency was calculated.

2.5.8 Solubility of psoralen in water. Psoralen was added little by little into 150 ml of water at room temperature and the mixture was stirred well with a magnetic stirrer. When approximately 45 mg of 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 the densitometric reading obtained is given in the table 7.

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. Section

Densitometric reading

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

3.1 Studies on the chemical composition of P.corylifolia and 'Bakuchi oil' Plant materials used in the preparation of herbal drugs can be considered to be complex mixtures of many different types of compounds. During the manufacturing process of 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 physicochemical 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. (I) Isolation and identification of compounds directly from the drug. (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 present or absent in the drug.

For the investigation of a particular plant constituent in the drug, thin layer chromatography can be used as an important tool. For the clear observation of a 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 followed. Accordingly, isolation and identification of major constituents present in the fruits of P.corylifolia was attempted initially.

3.2 Isolation and identification of compounds from the fruits of P.corylifolia. Seven compounds (see Table 8) including one new natural product were isolated from the fruits of P.corylifolia by means of various chromatographic techniques such as TLC, co-TLC, coloumn chromatography, preparative TLC and radial chromatography, 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 P.cory1ijlia was also isolated and analyzed.

52

Table 8: Compounds isolated from the fruits of P. corylifolia and some of their physical properties. Compound Psoralen (1) Isopsoralen (2) Bakuchiol (38) Dehydroisopsoralidin (42) Corylin (11) Psoralidin (5) Isobavachalcone (30)

Physical description White needles 'White needles Colourless oil Pale yellow needles White needles White needles Bright yellow needles

Melting point 160-163° C 140-142° C

290° C (decomp.) 241-243° C

153-154° C

Rf value 0.59(toluene: ethyl acetate: hexane, 1:3:9) 0.67(toluene: ethyl acetate: hexane, 1:3:9) 0.82(toluene: ethyl acetate: hexane, 1:3:9 0.41(toluene: ethyl acetate: hexane, 1:3:9) 0.35(toluene: ethyl acetate, 5: 2) 0.26(toluene: ethyl acetate, 5: 2) 0.21(toluene: ethyl acetate, 5: 2)

3.2.1 Psoralen (1) and isopsoralen (2). 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 cooling. This yielded 0.79 % of psoralen and isopsoralen as a mixture.

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 material in a yield of 0.67 %.

53

Larger amounts of pure compounds, psoralen and isopsoralen were obtained by subjecting the mixture to colunm chromatography. (0.36 % and 0.32 % respectively). Small amounts (less than 100 mg) of pure compounds were obtained by subjecting the mixture to radial chromatography. (33% and 23 % respectively for psoralen and isopsoralen). (section 2.3.1.1 )

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 chromatography. However careful adjustment of solvent gradient using ethylacetate/ hexane mixture enabled the separation of the two compounds by column chromatography. (section 2.3.1.1). In radial chromatography multiple development was 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 achieved by the analysis of spectral data.

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, H5) 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 hydrogens. ('H NMR spectrum of isopsoralen is given in appendix ii, Fig.22)

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 spectrum of psoralen is given in appendix ii, Fig.23)

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

carbon34 could not be clearly observed in our spectrum due to noise).

When a mixture of psoralen and isopsoralen was chromatographed on TLC plates (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 chromatogram (toluene: ethyl acetate: hexane, 1: 3: 9 / silica gel 60 F 254) of a plant extract of methanol or hexane is observed at 365 nm after spraying with 10%

56

methanolic potassium hydroxide, these two compounds can be easily detected as the most prominent spots. (Fig: 7-A) 3.2.2 Bakuchiol (38). 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 and obtained as a colourless oil in a yield of 0.10 %. When attempting the column chrornatographic separation of the hexane extract 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 pigment which is soluble in bakuchiol. Hence it was found to be difficult to isolate 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 a better resolution. (2.3.2)

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 activities.27 29, 57. 58, 59.

Hb l>__Ha )::2

HO

15

16

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 the presence of a paradisubstituted benzene ring by the appearance of two doublets at 6 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 and 25 for the 'H NMR and 13 C NMR spectrum of bakuchiol)

As supported by HSQC and HMBC experiments, multiplet at 6 1.9 (2H, in, H-il) can be explained as two overlapping triplets corresponding to each C- il proton which is 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 HSQC and HMBC correlations.

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 interchanged. The HMB correlations are shown in Fig.5.

14

Fig. 5 Some HMBC data of bakuchiol.

Based on 2D NMR correlations, 'H NMR and 13C NMR shifts of bakuchiol could now be reassigned unambiguously to their respective positions as depicted in the Table 9. Table 9 also gives the previous assignments of 'H and 13C NMR assignments of bakuchiol.56

Table 9: NMR data for bakuchiol (38) Experimental data Position No

Reported data"'

'3C (CDCI,, 150 MHz) 130.89

'H (multiplicity, J(Hz) (CDCI3, 600 MHz)

2

127.35

7.26(2H, d, 8.4)

C-5, C-4, C-7

127.4

7.22(d, 7.2)

3

115.34

6.78(2H, d, 8.4)

C-i, C-4, C-5

115.4

6.74(d, 7.2)

4

154.59

5

115.34

6.78(2H. d, 8.4)

C-I, C-3, C-4

115.4

6.74(d, 7.2)

6

127.35

7.26(2H, d, 8.4)

C-4, C-S, C-7

127.4

7.22(d, 7.2)

7

126.44

6.27(IH,d, 16.2)

C-1,C-2,C-6.C-9

135.9*

6.23(d, 16.2)

8

135.85

6.07(IH,d, 16.2)

C-1,C-17,C-10

126.5*

6.03(d, 16.2)

9

42.50

10

41.27

1.50(2H,m)

C-8, C-9, C- 12, C- 17

41.3

1.93(m)

11

23.212

1 .96(2H. m)

C-10, C-12, C-13

23.3

1.47(m) r

12

124.78

5.11(IH,m)

C-b, C-Il, C-14, C15

124.8

5.09(bt)

13

131.3

14

17.63

1.59 (3H, s)

C-14

1.56(s)

15

25.69

1.69(3H, s)

C-12, C-13, C-15

1.65(s)

16

23.32

1.20(3H, s)

C-8, C-9, C-10, C-17

23.4

1.17(s)

17

145.93

5.88(1H. dd, 17.7, 10.5)

C-9, C-b, C-16

146.0

5.86(dd,10.9,17.3)

18a

111.86

5.02(1H, d, 17.7)

C-9, C-17

111.9

4.98(m)

18b

111.86

5.05(IH,d, 10.5)

C-9.C-17

111.9

gHMBC correlation

13C, (CDCI3 , 75.4 MHz)

'H multiplicity (CDCI3 . 300 MHz)

131.7

154.5

42.5

131.4'

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, H18b). 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 have been collectively assigned as a multiplet.

1I-18a (d. 17.7 I-li)

H-I8h (d. 10.5 Hz) -

14-17 (dd, 17.7, 10.5 Hz) 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 green spot under visible light. (Fig.7-B).

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

0 EI Fig. 7: TLC of methanol extract of fruits of P. co,yifoiia. Solvent system- toluene: ethyl acetate: hexane, 1:3:9 detection under 366 nm I 10% methanolic KOH. detection under visible light I anysaldehyde sulphuric reagent.

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 obtain a fixed oil in a yield of 2.5 %.

The fixed oil of P.corylifolia has not been subjected to analysis previously and no literature available regarding its structure.

This was identified as a triacyiglycerol with polyunsaturated fatty acids by comparision of the 'H NMR spectral data with those of the known triacyiglycerols and polyunsaturated fatty acids 70 and studying simulated spectra of triacylglycerols.7 '

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 when treated with anisaldehide sulphuric acid reagent.

3.2.4. Dehydroisopsoralidin. (42) The methanol extractive obtained from the sequential extraction of fruits with hexane and methanol was subjected to column chromatography and 35 major fractions were collected. Fraction 14 was further subjected to flash column chromatography over silica gel and eluted with acetone: hexane, 1: 20 which resulted in the isolation of dehydroisopsoralidiri as a pale yellowish solid (3 x 10 %). (section 2.3.4). Apart from that this compound also could be precipitated by addition of acetone to the column fraction 13.

This compound was elucidated as dehydroisopsoralidin (42) by means of detailed analysis of spectral data. (UV, IR, 'H NMR, 13C NMR, HSQC, HMBC) as follows.

Dehydroisopsoralidin (42) Mass spectrum of the compound gave pseudo molecular ions at mlz 335[M+HI and mlz 333[M-H] corresponding to a MW of 334 amu. Initial comparisons of the UV, 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 molecular formula of C20H140 was evident from these interpretations.

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 cournestan, corylidin4 ' indicated this to be a dihydropyran system. HMBC correlations 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 of the tetra-substituted aromatic ring attachment to the dihydropyran system. Strong 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 dehydroisopsoralidin are given in appendix ii, Fig.26-30)

NMR data for dehydroisopsoralidin is given in Table 10.

LIM

Table 10: NMR (d6-DMSO, 600 MHz) data for dehydroisopsoralidin (42) * No.

'3C

2

157.2a

3

102.9

4

159.0

5

118.5

6

118.7

7

155.9

8

104.2

9

153.9

10

105.6

I'

114.5

'H (multiplicity and J(Hz))

gHMBC

7.77(s)

C-4", C-7, C-9, C-4

6.97(s)

C-6, C-7, C-9, C-b

98.6

7.17(d, J = 2.0)

C-i', C-5', C-4', C-2'

114.1

6.96 (dd, J = 8.4 & 2.0)

c-i', C-3'

120.4

7.71 (d,J=8.4)

C-4',C-2'

120.8

6.63 (d, J = 9.9)

c-6", c-7, C-5

131.8

5.91(d, J = 9,9)

C-6", C-4", C-6

156.0'

77.8 6"-Me

27.9

1.45(s)

C-6"-Me, C-6", C-5"

6"-Me

27.9

1.45(s)

C-6"-Me, C-6", C-5"

* 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

A~ 0

0

W 0-

1

C',

~/~ 0 k7

-V

OH

Fig. 8: Some key HMBC data for the coumestan fragments

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 be corrected from the earlier report as given in Table 10.

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 overlapping with another blue fluorescent spot. However dehydroisopsoralidin could 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 potassium hydroxide this fluorescent purple spot turns to yellowish blue.

W.

3.2.5 Corylin (11) The compound was isolated as a white crystalline solid (1 x 102 %) from the methanol extract of fruits by column chromatography (section 2.3.4) and identified as corylin based on spectral analysis and comparision with reported data.

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 spectrum. JR peak at 1628 indicates the presence of a carbonyl group.

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 coupled doublets and two

ortho

and

ineta

meta

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, 13unsaturated carbonyl system. Hydroxyl attachment at C-7 is supported by the down field carbon shift at 6 162.7. (see appendix ii, Fig.3 1 and 32 for the'H NMR and 13 C NMR spectra of corylin)

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) Reported data35

Experimeatal data '3C

'H (multiplicity and J(Hz) (d6-DMSO,600MHz)

gHMBC

8.34(IH, s)

C-3, C-4, C-9, c-I'

'3 C(CDCI3,150 MHz)

'H multiplicity

2

(d6DMSO, 150 MHz) 152.1

3

123.0

122.6

4

174.5

175.4

5

127.2

7.96(1H, d, 8.7)

C-4, C-7, C-9

127.8

8.00(IH,d)

6

115.2

6,94(IH, dd, 8.7, 2.2)

C-8, C- 10

115.9

6.99(1H,dd)

8

102.1

6.86(1 H, d, 2.2)

C-6, C-9

102.9

6.90(1H,dd)

9

157.4

158.4

10

116.4

117.6

124.4

124.2

No.

126.9

7.28(IH, d, 2.0)

C-3, C-4', C-6', C-4"

8.22(1H,$)

125.4

7.32(IH,d)

120.5 153.3

153.1 115.5

6.79(IH, d, 8.1)

C-I', C-3', C-4'

116.3

6.77(1H,d)

129.6

7.29(1H, dd, 8.1, 2.0)

C-2', C-4'

130.4

7.35(IH,dd)

121.7

6.44(IH, d, 9)

C-6"

121.4

6.40(IH,d)

131.2

5.78(IH, d, 9)

C-3', C-6", C-6"Me

131.8

5.78(IH, dd)

76.9

6"

76.2

6"-Me

27.7

1.39(6H, s)

C-5", C-6", C-6"Me

28.2

1.40(6H,$)

6"-Me

27.7

1 .39(6H, s)

C-5", C-6", C-6"Me

28.2

1

71

.40(6H,$)

These comparison shows that the spectral data obtained for corylin very closely resemble those reported for corylin.35

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 with 10% methanolic potassium hydroxide.

3.2.6 Psoralidin (5) Psoralidin was isolated as a colourless crystalline solid (9x1ft3%) from the methanol extract of fruits by column chromatography and preparative TLC. (section 2.3.3)

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 further studies confirmed the compound to be psoralidin, as described in the following section.

72

Mass spectrum of psoralidin (5) gave the molecular ion peak at mlz 337[M+H] and mlz 335[M-H] corresponding to a MW of 336 amu. In the JR spectrum a sharp peak 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) in its 'H NMR spectrum.

Presence of a 1,2,4 trisubstituted benzene ring is supported by the double doublet at 6 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 psoralidin is given in appendix ii, Fig.33)

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 given in the Table12 along with reported data.

73

Table 12: Comparision of 'H NMR data of psoralidin (5) with reported data. Experimental (DMSO, 600 MHz)

Reported data 5 (400 MHz, TMS as internal standard)

1.71 (3H, s)

1.77 (3H, s)

1.24 (3H, s)

1.75 (3H, s)

3,33 (2H, d, 7.2)

3.38 (3H, d)

5.36 (1H, t, 7.2)

5.40(111, t)

6.92(1H,$)

6.93(1H,$)

6.94 (11-1, dd, 8.4, 2.0)

7.00 (1H,dd)

7.16(1H,d,2.0)

7.18(1H,d)

7.62 (1H, s)

7.67 (1H, d)

7.68 ( IH, d, 8.4)

7.72 (1H, d)

10.09

9.97 (1H, s)

10.25

10.70 (1H,$)

This compound appears as a strong pinkish purple fluorescent spot at 365 nm at Rf 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 fruits of P.cory1i11ia by column chromatography and preparative TLC. (section 2.3.4).

H

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 NMR, HSQC, HMBC) revealed that the compound to be isobavachalcone.

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 spectra in appendix ii, Fig.34 and 35)

H-6', H-2' (6 7.75, 2H, d, J=8.6 Hz)

\4

I

Two overlapping proton doublets H-c (ö 7.718, 1H, d, 15.6 Hz) and H-b ( 7.77, IH, d. 15,6 Hz)

Fig 9: Appearance of 111 NMR pattern corresponding to H-b and H-c protons, and the doublet corresponding to H-2' and H-6'.

76

Table 13: NMR (d6-DMSO, 600 MHz) data for isobavachalcone (30) No.

'C

2

163.5

3

114.4

4

162.3

5

'H (multiplicity and J(Hz))

gHMBC

107.3

6,46(111, d, 8.9)

C-3, C-I

6

129.77

8.03(111, d, 8.9)

C-4, C-a

1

112.6

A

191.7

b

117.3

7.745(IH,d. 15.6)?

C-I'

C

144.0

7.745(1H,d, 15.6)?

C-a,C-6'

1'

125.75

6'

130.4

7.749(IH, d, 8.6)

C-c, C-4', 2'

5'

115.81

6.835(1H, d, 8.6)

C-I', C-3', C-4'

4'

160.2

3'

115.81

6.835(111, d, 8.6)

C-i', C-S', C-4'

2'

130.4

7.749(111, d, 8.6)

C-c, C-4'

21.24

3.22(2H, d, 7.2)

C-3,C-4,C-2"

122.3

5.17(1H, t, 7.2)

C-5", C-4"

17.69

1 .62(3H, s)

C-5", C-2"

25.46

1 .72(3H, s)

C-5", C-2"

131.16

Isobavachalcone could be observed on TLC (toluene: ethyl acetate, 5: 2, R1 0.21) when 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 yellow spot under visible light.

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 check the quality of Psoralea corylifolia fruits.

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

0000 0000 00 000 0000 000 0 0 0 0 000000000 000000000000000 0000000000 00000000000000 0000000000 00000000000000 0000000000 00000000000000 N 0 0 N0 0 0 1 N0 0 0 1 N0 0 I 1 N

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.

This extract was chromatographed on a silica column using a gradient of hexane / 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 isopsoralen, fraction-5 contained the new natural product dehydroisopsoralidin (42). Corylin (11). psoralidin (5) and isobavachalcone (30) were found in the fraction-7. (Presence of those compounds were confirmed by carrying out co-TLC)

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, 1: 3: 9 as the TLC eluant and developing it three times on a 10 cm plate. It was possible to identify all the compounds isolated from P.cory1ij1ea, namely, psoralen (1). isopsoralen (2), dehydroisopsoralidin (42), corylin (11), psoralidin (5) and isobavachalcone (30) except bakuchiol (38). (See Fig 11 A and B)

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 minor compounds should also be isolated and identified.

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 it was chromatographed on TLC plates. However bakuchiol could not be detected on 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" was further confirmed by gas chromatographic analysis, (see section 3.7.2)

do

1 2 3 4 5 6 7

1 2 3 4 5 6 7 8

A Fig 11: TLC of "Bakuchi oil" Track 1: 80% methanol soluble fraction of the methanol extract. Track 2: isobavachalcone (30). Track 3: psoralidin (5). Track 4: corylin (11). Track 5: dehydroisopsoralidin (42). Track 6: psoralen (1). Track 7: isopsoralen (2). Track 8: bakuchiol (38). Solvent system: toluene: ethyl acetate: hexane, 1: 3: 9 ( x 3) detection under 366 nm / methanolic KOH. detection under 254 nm.

F,,

-

0

123

A

Fig 12: TLC of fruits of P.corylfolia and "Bakuchi oil".

123 C

Track 1: methanol extract of fruits of P. coryljfolia. Track 2: 80% methanol soluble fraction of the methanol extract of "Bakuchi oil". Track 3: bakuchiol. Solvent system: toluene: ethyl acetate: hexane, 1: 3: 9 detection under 366 nm / 10 % methanolic KOH. detection under 254 nm. detection under visible light / anysaldehyde sulphuric reagent.

83

3.4 Standardization of "Bakuchi oil"

3.4.1 Chromatographic finger prints Chromatographic finger printing is one of the most important methods used in the quality control of herbal medicines which consist of complex mixtures of compounds. 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 using chromatographic - spectroscopic hyphenated techniques such as HPLC-PDA and LC-MS. The readily availability of computers has resulted in the development of computer aided comparison of chromatograms for "sameness" and "differences".8 TLC has the advantage over other chromatographic methods in terms of simplicity, versatility, speed, sample preparation and economy.

In these studies on "Bakuchi oil', chromatographic finger prints were developed based on the thin layer chromatography-fluorescence detection (TLC-FD). For the 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% methanolic potassium hydroxide resulted in an intensification of fluorescence. (Fig. 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 as samples prepared at other Ayurvedic hospitals at Beliatta and Kurunegala and a sample from Ayurvedic Drugs Corporation were analyzed using the TLC systems described above. Visual examinations of their thin layer chromatograrns and densitograms did not show any marked difference in the pattern from that of the reference sample. (Densitograms of the 'Bakuchi oil" sample collected from Ayurvedic Drugs Corporation is given in the Fig.15.)

However quantitative measurements of psoralen and isopsoralen concentration show that there is a substantial variation of active components present in different samples. (Section 3.5, Table 15)

Psoralen and isopsoralen are the most important marker compounds found in these finger prints.

85

A

B

C

Fig. 13. TLC finger prints of "Bakuchi oil". Detection: under 365nm / methanolic KOH. Solvent systems: A - toluene: ethyl acetate, 5: 2 B - hexane: ethyl acetate, 4: 1 C - toluene: ethyl acetate: hexane, 1: 3: 9

86

A

Ifl

C Fig. 14. Densitograms of thin layer chromatograms shown in Fig. 12.

Densitometric scanning: excitation at 365nm / fluorescence detection Solvent systems: A - toluene: ethyl acetate, 5: 2 B - hexane: ethyl acetate, 4: 1 C - toluene: ethyl acetate: hexane, 1: 3: 9

EM

A

90.0

mm

C Fig.15. Densitograms of the "Bakuchi oil" sample collected from Ayurvedic Drugs Corporation.

Densitometric scanning: excitation at 365nm / fluorescence detection. Solvent systems: A - toluene: ethyl acetate, 5: 2 B - hexane: ethyl acetate, 4: 1 C - toluene: ethyl acetate: hexane, 1: 3: 9

3.4.2 Quantification of psoralen in "Bakuchi oil". 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 concentration of the major photoactive compound, psoralen would be an important parameter in the standardardization of "Bakuchi oil". Towards this end a method was developed for the quantification of psoralen in "Bakuchi oil", using TLC-FD densitometry.

(a) Method development 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 acceptable TLC separation due to the interference of fatty materials. A methanol extract of the oil gave better chromatograms. However, a certain proportion of the fatty material dissolved in the methanol causing interference. A more significant problem was the fact that even repeated extraction by methanol was not successful in obtaining a total extraction of psoralen from the oil.

It was clear that an initial removal of the fatty material from the oil was necessary before extracting the psoralens. The method developed consisted of dissolving "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) as the solvent, with presaturation of the development tank.

Reflectance spectroscopy of a sample of psoralen (isolated from the plant) on a TLC plate indicated maximum absorption at 240 nm. However, absorption densitometry could not be carried out to measure psoralen present in "Bakuchi oil' as visual examination of a chrornatogram developed on a plate containing a fluorescent 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 mode with excitation at 240 nm and a 370 urn filter.

Isopsoralen present in "Bakuchi oil' could not be quantified using this system as it overlapped with another fluorescent spot even in different mobile phases. 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 curve was linear with a correlation coefficient of 0.9931. (Fig .16.)

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

For the regression equation, y = mx + h, confidence limit at any x0 is given by y0

(2) If

y0

± ts1

where t = Students t value, S = variance.

is the value for a given x0 , obtained from the regression line, then

- )2 S2 =

+ [fl

(x1

1S2 _)2] 5

where = average of x values, S = variance of

estimated y value.

Variance of estimated y value,

s2

-

n-2

2 -

- b y - m x1 y

-

value.

91

where 5' = best estimate for y

This standard curve was used to determine psoralen concentration in several Bakuchi oil samples collected from different places.

92

11

'

I

IMI

4-

j

I

93

( c ) Method validation

Precision and accuracy. In this experiment a sample of "Bakuchi oil' prepared at BMARI for a clinical trial(VCR I) was selected as the reference sample. The precision of the TLC-FD densitometric method was determined by measuring the concentration of psoralen in six replicates of the reference sample and the Coefficient of Variation (CV) was found to be 4.3 %. A sample with an approximately 30 % addition of psoralen gave a recovery rate of 103 %. These results show the method developed is of acceptable precision and accuracy. (section 2.5.2.3)

3.5 Psoralen concentration in different "Bakuchi oil" samples 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 need for better process control in the preparation of the oil.

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 psoralens present in "Bakuchi oil" shows a synergistic action.

Table 14: Psoralen concentration in different samples of 'Bakuchi oil" "Bakuchi oil" sample

Psoralen concentration (mg/mi)

BMARI - I (reference)

0.083

BMARI - II

0.226

Ayurvedic Drugs corporation

0.087

Ayurvedic hospital - Kurunegala

0.038

Ayurvedic hospital - Beliatta

0.103

BMARI - 111*

0.207

* p repared from fruit

3.6 Psoralen concentration in different batches of P.corylifolia fruits Psoralen concentration in different batches of P.corylijolia fruits were determined using the same method and the results are given in table 15.

Table 15: Poralen concentration in different batches of P.corylifolia fruits. Batch number

Poralen concentration (mg/g)

1

3.03

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 be achieved by mixing in appropriate ratios of different batches of fruits, to obtain a master batch containing a desired level of psoralen.

3.7 Effect of processing parameters on the composition and the quality of the drug.

3.7.1 Preparation of "Bakuchi oil' 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.

Whole seeds + Water (45kg) (600L)

1 Boiling down to 75 L

Stage I

Filtration

Addition of 75 L of sesame oil to the filtrate.

Heating to evaporate water Stage II Addition of powdered fruits (5 kg) prior to total evaporation of water ( 90 % complete) 'Jr

Total evaporation of water 'Jr

Sedimentation at room temperature (2 days)

Filtration

Stage III

Storage in dark bottles.

Generally this type of procedure results in incorporating compounds with a wider 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.

3.7.2 Rate of incorporation of psoralen. In order to study the factors affecting the rate of incorporation of psoralen into 'Bakuchi oil' the following experiments were carried out. The psoralen content in a batch of P. corylifolia fruits and in "Bakuchi oil" 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. (section 2.5.5) ) Table-16

Solubility of psoralen in water and in sesame oil were determined. (section2.5.8 and 2.5.6) The results are given in the Table 17.

WI

Table-16. Comparision of psoralen in "Bakuchi oil" and different extracts of P. cor'ciifriia. Sample

Psoralen concentration

Extraction efficiency

medicinal oil

0.207 mg/mi

10.23 %

Sesame oil extract

0.192 mg/mi

95.36 %

water extract

0.136 mg/mi

7.2%

Concentration qfpsoralen in fruits used in above experiments- 3.029 mg/g.

Table -17. Solubility of psoralen Medium

Psoralen concentration

Water

0.195 mg/ ml

Sesame oil

3.528 mgI ml

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 incorporated in the oil.

The initial extraction of fruits by water is wasteful as regards extraction of psoralen. 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 extract the powdered fruits with sesame oil.

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 the same, as far as the major compounds are concerned.

However according to Ayurveda theory, the action of a drug caimot be attributed to a single active component.

Further in stage I, processes not identified so far such, hydrolysis of compounds, evaporation of compounds by steam volatilization may be taking place, which have a 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 traditional preparation process can be made.

3.7.3 Study of the fate of "bakuchiol" during the manufacturing process. In order to study the fate of bakuchiol, following GC experiments were carried out. (section 2.4.2) The ratio of the GC response factors of psoralen and bakuchiol were determined.( section 2.4.2/ 9) The 80% methanol soluble fraction of a hexane extract of fruits of P.coryliftlia was analyzed by gas chrornatographiy. (section 2.4.2/ 4) The steam distillate of the fruits of P.corylfo1ia was analyzed by gas chrornatographiy. ( section 2.4.2/ 5) 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 analyzed by gas chromatographiy. ( section 2.4.2 / 7) The steam distillate of the seed residue left behind after extracting with sesame oil in the experiment 5 above was analyzed by gas chromatographiy. ( section 2.4.2/7) Heat stability of bakuchiol in sesame oil at 145° C was measured. ( section 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 that partitioning of hexane extract with 80 % methanol results in total extraction of 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.

In stage I (scheme-fl of the preparation process, it is likely that most of bakuchiol is 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 experiments were carried out.

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 obtained and analyzed by GC. Psoralen to bakuchiol could be detected in a ratio of 8: 5 (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 applied during the manufacturing process.

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

0 0• r

H 10

o

0000000000000000000 ++++++++++000000000 0 e0 ø0 06 000000000 rrrrrrrrr

104

en

000000000000000000000000 000000000000000000000000 000000000000000000000000 000000000000000000000000

'

o

105

N —:

a

106

I

I C

o a

a

0

aaaaaaa aaóóa

OOODOOOOOOOOOOOOOOOOOOOO, 000000000000000000000000 000000000000000000000000 000 000000000000000000000 E

I I

L

107

IN C-'

I

108

4 Conclusion Of the major secondary metabolits of Psoralea corylfolia fruits, psoralen, isopsoralen, psoralidin, corylin, isobavachalcoe, dehydroisopsoralidin and bakuchiol, all except 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 ( toluene: ethyl acetate: hexane, 1: 3: 9, triple develpoment) and the chromatographic 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 psoralen level can be used as a quantitative standard. Standardization of the plant 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 before implementing a change in the manufacturing process.

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 analyzing of other oils with suitable modifications.

109

References

WHO Policy Perspectives on Medicines, No.2, p., 2002. Vinaya Kumar, A., Principles of Ayurvedic Therapeutics, Sri Satguru publications, India, 1995, p.3. The Ayurveda encyclopedia, Sri satguru publications, India, p.13 Dash, V. B., Fundamentals of Ayurvedic Medicine, Sri Satguru publications, India, 1999, p.2. Ayurveda Pharmacopoea, vol.1, part I, Dept. of Ayurveda, Sri Lanka, 1976 Loakmann, G., St.John's Wort in mild to moderate depression: The relevance of hyperforin for the clinical efficacy, Pharmacopsychiatry, vol. 31, issue SUPPL. 1, p.54-59, 1998) Sameera, W.M.C., Aheysekera, A.M., Sarnarasinghe, S., Gas chromatographic studies of volatile components in Dasamoolarishtaya(DMA), Chemistry in Sri Lanka, vol.23, No.2, 27-28, 2006. Yi-Zeng Liang, Peishan Xie, Chan, K.. Quality control of herbal medicines. Journal of chromatography B. vol.8 12, issue 1-2, p.2-4, 2004. Grimes, P.E., Diseases of Hypopigmentation (chapter 72). in Principals and practice of dermatology, Mitchell Sams (ed), 2 d ed., Churchill Livingstone Inc.. 1996, p.843. Marks, R., Roxburgh's common skin diseases, 171h ed.. London, 2003, p.298. Clinical studies on Bars (Vitiligo), Central Council for Research in Unani 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. MIMS annual, Sri Lanka, 1999, p.477. Renmin Liu, Aifeng Li, Sun, A., Kong, L., Preparative isolation and purification of psoralen and isopsoralen from Psoralea corylifolia by high-speed countel-current chromatography, Journal of Chromatography-A, 1057, 225-228, 2004. Andreoni, A., Cubeddu, R., DalI'Acqua, F., Knox, C.N. and Truscott. T.G., Fluorescence lifetimes of furocoumarins. Psoralens, Chemical Physics Letters, vol.114, issue 3, p.329-333, 1985, Llano, J., Raber, J., Eriksson, L.A., Theoretical study of phototoxic reactions of psoralens, Journal of Photochemistry and Photobiology A, 1.154, issue 2-3, pp. 235-243., 2003. Kitamura, N., Kohtani, S., Nakagaki, R., Molecular aspects of furocoumarin reactions: Photophysics, photochemistry, photobiology, and structural analysis, Journal of Photochemistry and Photobiology C, pp.168-185, 2005. Bordin, F., Carlassare, F., Busulini, L., Baccichetti, F., Furocoumarin sensitization induces DNA-protein cross-links, Journal of Photochemistry and Photobiology, vol.58, issue 1, pp.133-136, 1993.

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. Abel, E.A., Farber, E.M., Photochemistry (chapter 11), in Recent advances in 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, New Jersey, U.S.A., 1960, P. 670. McIntyre, N., Jean-Pierre Benhamou, Bricher, J., Rizzetto, M., Rodes, J.(eds),Oxford Textbook of Clinical Hepatology, vol 2, Oxford University Press, p.1 179. 24, Tripathi, K.D., Essentials of medical pharmacology, 4t) ed., Jaypee Brothers, Medical Publishers, New Delhi, p.855. Dr. H.I. Chandrasekara, Former Director of Bandaranayke Ayurvcdic Research Institute, Personal communications. 2006. Prasad, N.R., Anandi, C., Balasubramanian, S., Pugalendi, K.V., Antidermatophytic activity of extracts from Psoralea corylifolia (fabaceae) correlated with the presence of a flavonoid compound, Journal of Ethnopharmacology, 2004, 91, 21-24. Sun, J.N., Woo, S.H., Cassady, J.M., Snapka, R.M., DNA Polyrnerase and Topoisomerase II Inhibitors from Psoralea corylifolia, J.Nat.Prod., 1998, 61, 362-366.

112

Yin, S., Cheng-Qi Fan et al., Antibacterial prenyiflavone derivatives from Psoralea corvlifolia, and their structure-activity relationship study, Bloorganic and Medicinal Chemistry, 12, 4387-4392, 2004. Haraguchi, H., Inoue, J. et.al., Antioxidative components of Psoralea corlitolia (Leguminosae), Phytotherapy Research, 16, 539-544, 2002. Cho, H., Jung-Yang Jun, Eun-Kyoung Song, Ki-Hong Kang ci a?., Bakuchiol: A hepatoprotective compound of

Psoralea corylitolia

on tacrine-induced

cytotoxicity in Hep G7 cells, Planta Medica, 67, 750-751, 2001. Chakravarti, K.K., Bose, A.K., Siddiqui, S., Chemical Examination of the seeds of Psoralea coryliftilia Linn. (Babchi), (Council Sci.Ind. Research, Delhi) J. Sci. Ind. research(India), 1948, 7B, No.2, 24-26. Mukeiji, B., Psoralea and

other indigenous drugs used in leucoderma,

J.Sci.Ind.Res.(India), iSA, No.5, supplement, p.1-12,1956. Jam, A. C., Gupta, G.K., Rao, P.R., Isolation and constitution of corylin: a new isoflavone from the fruits of Psoralea coryliftlia, Indian journal of chemistry, 12, 659-660, 1974. Wang, X., Wang, Y., Yuan, J., Sun, Q. etcii., An efficient new method for extraction and purification of psoralen and isopsoralen from Fructus Psoraleae by supercritical fluid extraction and high-speed counter-current chromatography. Journal of Chromatography A, 1055, 135-140, 2004. Jiangning. G., Xinchu., W., Wo Hou et cii., Antioxidents from a Chinese medicinal herb-Psoralea corli10/ia L., Food Chemistry, 91, 287-292, 2005.

113

Innocent, G, Cappelletti, EM., Caporale, G., Furocoumarin contents in the vegetative organs of cultivated Psoralea species, Tnt. J. Pharmacognosy, 29, No.4,311-316, 1991. Vo-Dinh, T., White, D.A., O'Malley, M.A. et al., Flurescence detection of phototoxic psoralens in vegetable products, J. Agric. Food Chem., 36, 333-337,

BE 38, Chjntalwar, G.J., Ramakrishnan, V., Luthria, D.L and Baneiji, A., Insect atifeedents from the roots of Psoralea con'lifolia Lirm.. Indian Journal of Experimental Biology, 30, 858-859, 1992. Stanley. W.L., and Vannier, S.H., Chemical composition of lemon oil. I. Isolation of a series of substituted coumarins, Journal of American Chemical Society, 79, 3488-349 1, 1957. 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 new cournestrol from the fruits of Psoralea corylifolia, Phytochemistry, 16, 403-404, 1977. Gupta, S., Jha, B.N,, Gupta, G.K. et al., Coumestans from seeds of Psoralea cor''lifo1ia, Phytochemistry, 29, No:7. 2371-2373, 1990. Gupta, B.K., Gupta, G.K., Dhar, K.L., Atal, C.K., Psoralidin oxide, a coumestan from the seeds of Psoralea corylifolia, Phytochemistry, 19, 2232-2233, 1980.

114

Wang, D., Famei Li, Jiang, Z., Osteoblastic proliferation stimulating activity of Psoralea corylifolia extracts and two of its flavonoids, Planta Medica, 67, 748749, 2001. Sun, J.L., Gupta, G.K., Dhar, K.L. and Atal, C.K., Psoralenol: a new isoflavone from the seeds of Psoralea corylifoiia, Phytochemistry, 17, 2046. Bajwa, B.S., Pyare La! Khanna and Seshadri, T.R., Components of different parts of seeds (fruits) of Psoralea corylfoiio, Indian journal of Chemistry, 12, 15-19, 1974. Wei-Jern Tsai, Wen-Chi Hsin, Chien-Chih Chen, Ant iplatelet flevonoids from seeds of Psoralea corylifolia, J.Nat.Prod., 59, 67 1-672, 1996. 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 of Psoralea corylifolia, Phytochemistry, 17, 164, 1978. Gupta, G.K., Sun, J.L., Gupta, B.K. and Dhar, K.L., Bakuchalcone, a dih ydrofuranochal cone from the seeds of Psoralea corvlifolia, Phytochemistry, vol.21, No.8, 2149-2151, 1982. Bhalla, V.K., Kayak, U.R., Sukh Dev, Some new flavonoids from Psoralea corylifolia, Tetrahedron Letters, No. 20, 2401-2406, 1968.

115

Gupta, S.R., Seshadri, T.R., Sood, G.R., A new formylated chalkone, 5'-formyl2', 4-dihydroxy-4'-methoxychalkone, from the seeds of Psoralea corvlifolia, Indian J. Chem., 13, 402-404, 1975. Gupta, B.K., Gupta, G.K., Dhar, K.L., Atal, C.K., A C-formylated chalcone from Psoralea corylifolia, Phytochemistry, vol.19, 2034-2035, 1980, Sun, J.L., Gupta, G.K., Dhar, K.L. and Atal, Bavachromanol: A new chalcone from the seeds of Psoralea corylifolia, Phytochemistry, vol.19, 336-337, 1980. Adhikari, S., Joshi, R., Patro, B.S. et a! , Antioxidant activity of bakuchiol: Experimental evidences and theoretical treatments on the possible involvement of the terpenoid chain, Chem. Res. Toxico!., 16, 1062-1069, 2003. Labbe, C., Faini, F., Coll, J., Bakuchiol derivatives from the leaves of Psoralea glandulosa, Phytochemistry, vol.42, No: 5, 1299-1303, 1996. 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 inactivation of nuclear transcription factor-k B in RAW 264.7 macrophages. International Immunopharmacology, 1, 1849-1855, 2001. Katsura, H., Ryo-Ichi Tsukiyama, Suzuki, A., Kobayashi, M., In vitro antimicrobial activities of bakuchiol againt oral microorganisms, Antimicrobial Agents and Chemotherapy, vol.45, No.11, p.3009-30113, 2001. Bapat, K., Chintaiwar, G, J., Pandey, U. and et al, Preparation and in vitro evaluation of radioiodinated bakuchiol as an anti tumor agent, Applied Radiation and Isotopes. 62, 389-393, 2005.

116

Shuki Araki and Yasuo Bustugan, Short synthesis of ( )-bakuchiol via a geranylindium reagent, J. Chem. Soc. Perkin Trans. 1, 2395-2397, 1991. Shah, Chandrakant C., Bhalla, V.K., Sukh Dev, Chem.Abstr., 129: 81856. Duke, A., Psoralea corylifoia L.(fabaceae), Econ.Bot., 41(4), 524-526, 1987 (Med. and Aromat. Plants Abstr., 10, No.5, 434, 1988.) Ogawa, S., Tamura, K., Oda, M., Inomoto, T., Kataoka, S., Japanese patent application: JP 97-2459 12 28 Aug 1997. The Wealth of India, vol.VI1I, Council of Scientific and Indrustial Reaserch, New Delhi, 1969. Kokate, C.K., Purohith, A.P., Gokhale, S.D., Pharmacognosy. 12t1 ed., Nirali Prakashan, India, p.229, 1999. Wagner. H., Bladt, S., Zagainski, E. M., Plant Drug Analysis, Springer-Verlag. Berlin, Heidelberg, New York, Tokyo, p.229, 1984. Wagner, H., Bladt, S., Zagainski, E. M., Plant Drug Analysis, Springer-Verlag. Berlin, Heidelberg, New York, Tokyo, p.303, 1984. Bhauacharji, S. and Moti, L. Dhar, Chem. Abstr., vol.53, 4460h, 1959. Gupta, B.K., Dhar, K.L., Atal, C.K., Cyclodehydrogenation of Psoralidin with Dichlorodicyano-henzoquinone (DDQ), Indian journal of Chemistry, vol.1513, 657-658, 1977. Pollard, M., Nuclear Magnetic Resonance Spectroscopy (High Resolution) (chapter 9), p. 401, in Analysis of Oils and fats, Hamilton, Elsevier Applied Science Publishers, London, 1986.

117

R. J.,

Rossell, J. B.,

ACD 50 H NMR predictor Connors, K.A., A text book of Pharmaceutical Analysis, 591608, John Wiley, 1982.

118

Appendix

Appendix i : List of Publications / Communications Appendix ii: Spectral data

119

Appendix i

List of Publications / Communications

Abeysekera, A. M., Gunawardena, R. D. A. R., Ratnayake,R. Capon, R.J., Gunaherath, G.M.K.B., A New Natural Coumestan from the Fruits of Psoralea Corylifolia. (2006) Chemistry in Sri Lanka, vol.23, No.2, p28.

Abeysekera, A. M., Gunaherath, G. M. K. B. and Gunawardena, R. D. A. R. (2005) Standardization of 'Bakuchi" oil, an Ayurvedic Medicinal oil used in the treatment of vitiligo, International Symposium on Herbal Medicines, Phytopharmaceuticals and 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*-

R 11

ET

J

126

127

128

129

E

0

0

0

0

0

0

0

130

0 N

0 0

0 -

0 N

0 N

0

0 fl

P

131

132

fl0(ft

133

U."

0 A

411 11•iu/

A

t91 9•S

134

69Lt

,Ht

135