ArticlePDF Available

The design, preparation and evaluation of Artemisia Afra and placebos in tea bag dosage form suitable for use in clinical trials.

Authors:

Abstract

Artemisia Afra, a popular South African traditional herbal medicine is commonly administered as a tea infusion of the leaves. However, clinical trials proving it safety and efficacy are lacking mainly due to the absence of good quality dosage forms and credible placebos for the plant. The objectives of this study were to prepare a standardized preparation of the plant leaves and freeze-dried aqueous extract powder of the leaves, in a tea bag dosage form and to design and prepare credible placebos for these plant materials.
THE DESIGN, PREPARATION AND EVALUATION OF ARTEMISIA
AFRA AND PLACEBOS IN TEA BAG DOSAGE FORM SUITABLE
FOR USE IN CLINICAL TRIALS
A. DUBE
A thesis submitted in partial fulfilment of the requirements for the degree of
Magister Pharmaceuticiae in the School of Pharmacy, University of the
Western Cape, Bellville, South Africa.
Supervisor: Prof. James A. Syce
June 2006
i
THE DESIGN, PREPARATION AND EVALUATION OF ARTEMISIA
AFRA AND PLACEBOS IN TEA BAG DOSAGE FORM SUITABLE
FOR USE IN CLINICAL TRIALS
A. DUBE
KEY WORDS
Artemisia afra
Standardized plant material
Plant freeze-dried aqueous extract
Tea bag dosage form
Infusion from tea bag
Luteolin
HPLC
Herbal stability
Herbal placebo
Clinical trials
ii
Summary
The design, preparation and evaluation of Artemisia afra and placebos in tea bag dosage
form suitable for use in clinical trials.
A. Dube
M. Pharm. Thesis: School of Pharmacy, University of the Western Cape, Bellville, South
Africa.
Artemisia afra, a popular South African traditional herbal medicine is commonly
administered as a tea infusion of the leaves. However, clinical trials proving its safety and
efficacy are lacking mainly due to the absence of good quality dosage forms and credible
placebos for the plant.
The objectives of this study were to prepare a standardized preparation of the plant leaves
and freeze-dried aqueous extract powder of the leaves, in a tea bag dosage form and to
design and prepare credible placebos for these plant materials. It was hypothesised that
the intra-batch variation in the plant flavonoid constituents would be lower in the
standardized A. afra leaves and freeze-dried aqueous extract powders compared to the
traditionally used non-standardized leaves, that the tea bag would be a pharmaceutically
suitable dosage form for the traditionally used plant leaves, that the infusion profiles of
luteolin (the flavonoid marker), from the tea bag would be similar to that of the loose
leaves traditional form and that credible placebos for the plant materials devoid of
pharmacological activity could be prepared.
To realise the objectives, different batches of plant materials were blended and an
aqueous decoction and freeze-drying method used to prepare the standardized leaves and
freeze-dried aqueous extract powder of the plant leaves, respectively. These plant
materials were packed in 36cm2 tea bags which were pharmaceutically evaluated using
the European Pharmacopoeia criteria, subjected to stability testing and their infusion
profiles determined using the British Pharmacopoeia dissolution apparatus I (basket) and
a modification of the BP apparatus II (paddle) incorporating a holding cell, for the loose
leaves and the tea bag preparations, respectively, and the infusion profiles compared
using the f1 and f2 mathematical method. The A. afra leaves were exhaustively solvent
iii
extracted and inert in-organic salts blended to prepare the placebos for the leaves and
freeze-dried aqueous extract powder, respectively. Finally, the placebos were evaluated
for lack of pharmacological activity using the isolated guinea pig tracheal muscle
preparation.
The A. afra standardized dried leaves and freeze-dried aqueous extract powder, contained
a total luteolin content of 2.065 ± 0.2347 and 13.870 ± 1.2460µg/mg, respectively, with
an intra-batch variation (% R.S.D) of 11.36% and 6.70%, respectively, reduced from an
initial %R.S.D of 21.24%, 30.00% and 16.77%, for the three separately collected plant
batches used. The tea bag of the A. afra standardized dried leaves was stable under room
temperature and humidity conditions for 6 months, while the freeze-dried aqueous extract
tea bag was not and both tea bag preparations were unstable at conditions of 40˚C/ 75%
RH. The f1 and f2 values for the infusion of luteolin from the leaves in tea bag compared
to the loose leaves were 73.52% and 13.85%, respectively, indicating that the profiles
were not similar. Finally, the placebo materials prepared closely resembled the respective
plant materials and the placebo of the A. afra leaves possessed only slight muscle relaxant
activity while the placebo for the extract powder was pharmacologically inert.
In summary, the results showed that the tea bag was a suitable dosage form for the A. afra
standardized dried leaves, but not the freeze-dried aqueous extract powder and that the
tea bag preparation did not have similar infusion profiles to the loose leaves, but could
still be used in clinical trials provided that adjustments in the dose preparation and
administration methods are made. Finally, credible placebos for the plant materials
suitable for use in clinical trials were prepared.
iv
DECLARATION
I declare that the thesis The design preparation and evaluation of Artemisia afra and
placebos in tea bag dosage form suitable for use in clinical trials is my own work, that it
has not been submitted before for any degree or examination in any other University and
that all the sources I have used or quoted have been indicated and acknowledged by
complete references.
A. Dube June 2006
Signed UWC, Bellville.
v
ACKNOWLEDGEMENTS
I would like to gratefully acknowledge the following;
Prof. James A. Syce my supervisor, for his guidance, support and the role he has played
in my personal development.
The National Research Foundation of South Africa (NRF) for providing the funding for
this project.
Coffee, Tea and Chocolate Company (Pvt) Ltd for donating the tea bag paper used in this
study.
A.R. Agencies for donation of the colourant chemical used in this study.
Mr B. Mouers and Mr F. Weitz for assistance with the identification and preparation of
voucher specimens of the plant material.
Mr V. Jeaven for his friendship and assistance with laboratory equipment and materials.
Mr. L. Cyster for assistance with the freeze-drying techniques.
Mr Y. Alexander for assistance with laboratory equipment.
Ms. Lorraine, Tanya Mapfumo, my girlfriend and wife to be, for her unending love and
support and assistance especially with the proof reading of this thesis.
Ms L. Manthata for assistance with the organ bath experiments.
vi
My colleagues and friends; Hai Qiu Ma, Max Ojangole, Sizwe Mjiqiza, Caroline Kinyua,
James Mukinda, Joseph Kabatende, Kenechukwu Obikeze, Emmanuel Molosiwa, Doris
Tshambuluka, Ian Backman, Denis Chopera, Chido Ruwodo, Natasha Katunga, for your
invaluable friendship and support that made the completion of this thesis possible.
Finally, the entire school of pharmacy staff for making my study pleasant.
Thank you all.
vii
DEDICATION
I dedicate this master’s thesis to my parents Mr and Mrs Dube for their love,
support and encouragement that has got me where I am today.
viii
Contents
Title page i
Summary ii
Declaration iv
Acknowledgements v
Dedication vii
List of tables xiii
List of figures xiv
List of graphs xvi
List of appendices xvi
Chapter 1: Introduction 1
Chapter 2: Literature review 3
2.1 Introduction 3
2.2 Artemisia afra – An important indigenous medicinal plant 3
2.2.1 Vernacular names 3
2.2.2 Botanical classification, and morphology of Artemisia afra 4
2.2.3 Geographical distribution of Artemisia afra 4
2.2.4 Traditional uses and dosage forms of Artemisia afra 5
2.2.5 Shortcomings of the traditional dosage formulations 6
2.3 The need for safety and efficacy data 8
2.3.1 Clinical trials 9
2.3.1.1 Placebos 10
2.4 Teas 11
2.4.1 Instant teas 12
2.4.2 Tea bags 13
2.4.2.1 Infusion from tea bags 13
2.4.2.2 Advantages of tea bags as a dosage form 15
ix
2.5 Phytochemical constituents of Artemisia afra 16
2.6 The flavonoids 16
2.7 Quality considerations for herbal materials 19
2.7.1 Standardization of herbal materials 19
2.7.2 Chemical fingerprints of herbal materials 20
2.8 Quality assessment of herbal materials 20
2.8.1 Quality assessment of herbal starting materials 21
2.9 Quality assessment of herbal preparations 21
2.9.1 Dissolution / infusion testing 22
2.9.2 Dissolution / Infusion profile comparison methods 23
Chapter 3: Work Plan 26
3.1 Study objectives 26
3.2 Hypothesis 26
3.3 Study approach 27
3.3.1 Why Artemisia afra? 27
3.3.2 Why a tea bag dosage form? 27
3.3.3 Why luteolin? 28
Chapter 4: Preparation and evaluation of Artemisia afra plant material and design
of placebos 29
4.1 Introduction 29
4.2 Equipment and materials 29
4.3 Methods 30
4.3.1 Collection and preparation of the plant material 30
4.3.2 Standardization of the Artemisia afra dried leaves 31
4.3.3 Preparation of the freeze-dried aqueous extract powder 31
4.3.4 Irradiation of the plant materials 32
4.3.5 Determination of the plant material organoleptic characteristics 32
4.3.5.1 Determination of plant material particle size and shape 33
x
4.3.5.1.1 The sieve method 33
4.3.5.1.2 Determination of the plant material particle shape 34
4.3.5.2 Determination of the plant material colour 34
4.3.5.3 Determination of the plant material odour 34
4.3.5.4 Determination of the plant material bitterness value 35
4.3.6 Determination of the plant material ash values 36
4.3.7 Determination of the plant material moisture content 37
4.3.8 Determination of microbial contamination of the materials 37
4.3.9 Development and validation of the HPLC assay 38
4.3.10 Quantitation of luteolin in the plant materials 39
4.3.11 Design of the plant material placebos 41
4.3.11.1 Design of placebo for the standardized dried leaves 41
4.3.11.2 Design of placebo for the freeze-dried aqueous extract powder 43
4.3.11.3 Matching of odour between the placebos and the plant materials 44
4.3.11.4 Matching of taste between the placebos and the plant materials 46
4.4 Results and discussion 47
4.4.1 Preparation of the plant materials 47
4.4.2 Organoleptic properties of the A. afra plant materials 48
4.4.2.1 Particle size and shape of the A. afra plant materials 49
4.4.2.2 Bitterness value of the A. afra plant materials 49
4.4.3 Ash values of the A. afra plant materials 49
4.4.4 Moisture content of the A. afra plant materials 50
4.4.5 Microbial contamination of the plant materials 51
4.4.6 Development and validation of the HPLC assay 52
4.4.7 Quantitation of luteolin in the A. afra plant materials 54
4.4.8 Design of the placebos for the plant materials 57
4.4.8.1 Design of placebo for the A. afra standardized dried leaves 57
4.4.8.2 Matching of odour of the placebo and plant materials 60
4.4.8.3 Matching of taste of the placebo and plant materials 60
4.4.8.4 Design of placebo for the freeze-dried aqueous extract powder 60
4.5 Conclusions 61
xi
Chapter 5: The preparation and evaluation of tea bags and placebos of
Artemisia afra 63
5.1 Introduction 63
5.2 Equipment and materials and animals 63
5.3 Methods 64
5.3.1 Preparation of the tea bags 64
5.3.1.1 Determination of the amount of plant material for the tea bags 65
5.3.1.2 Determination of the appropriate tea bag size 65
5.3.1.3 Selection of the appropriate tea bag paper 66
5.3.1.4 Preparation of the Artemisia afra and placebo tea bag dosage forms 66
5.3.2 Pharmaceutical evaluation of the tea bag dosage forms 67
5.3.2.1 Determination of uniformity of mass of the tea bag dosage forms 67
5.3.2.2 Determination of the infusion profile of the A. afra dosage forms 67
5.3.2.2.1 Determination of the infusion profile of the A. afra loose leaves 68
5.3.2.2.2 Determination of the infusion profile of the A. afra tea bag dosage
forms 69
5.3.2.2.4 Sample preparation, analysis and comparison of the infusion profiles 71
5.3.2.3 Determination of the stability of the dosage forms upon storage 72
5.3.3.4 Comparison of the A. afra tea bag dosage forms and their placebos 73
5.3.4 Pharmacological evaluation of the placebo for the plant leaves 73
5.3.5.1 Preparation of solutions used in the tracheal muscle experiments 74
5.3.5.2 Procedure for evaluation of the effect of the A. afra and placebo
materials on the isolated guinea pig tracheal muscle 72
5.7 Results and discussion 76
5.7.1 Preparation of tea bags containing the A. afra plant material 76
5.7.2 Uniformity of mass of the dosage forms 77
5.7.3 Infusion profile of the A. afra loose leaves 78
5.7.4 Infusion profile of the tea bags containing the A. afra leaves 82
5.7.5 Infusion profile of the tea bags containing the freeze-dried aqueous
extract of A. afra 84
xii
5.7.6 Comparison of the infusion profiles of the A. afra loose leaves and tea
bags containing the leaves and freeze-dried aqueous extract powder 85
5.7.7 Results of the stability studies conducted on the dosage forms 87
5.7.7.1 Organoleptic and chromatographic evaluation: Conditions at site A 87
5.7.7.2 Organoleptic and chromatographic evaluation: Conditions at site B 90
5.7.7.3 Organoleptic and chromatographic evaluation: Conditions at site C & D 91
5.7.7.4 Infusion profile characteristics of the dried leaves tea bag stored under
the conditions at site A and B 92
5.7.7.5 Infusion profile characteristics of the freeze-dried aqueous extract
powder in tea bag stored under conditions at site C and D 94
5.7.8 Comparison of tea bags of the A. afra plant material and their placebos 95
5.7.9 Results of test for absence of activity of the placebo for the A. afra
dried leaves 97
5.8 Conclusions 99
Chapter 6: Conclusion 101
References 103
Appendices 116
Vitae 136
xiii
List of the Tables
Table 4.1 Various excipients and their proportions used in different formulations
in the design of the placebo for the A. afra freeze-dried aqueous
extract powder. 44
Table 4.2 Organoleptic characteristics of the A. afra standardized dried leaves and
freeze-dried aqueous extract powder. 48
Table 4.3 Residual moisture content of the A. afra standardized dried leaves 50
Table 4.4 Residual moisture content of the A. afra freeze-dried aqueous extract
powder 51
Table 4.5 Accuracy and recovery of the luteolin standard 54
Table 4.6 Intra-day and inter-day precision of the assay 54
Table 4.7 The series of solvent extractions employed to produce the placebo material
of the A. afra leaves 58
Table 4.8 Excipients and quantities used in the formulation of the placebo for the
A. afra freeze-dried aqueous extract powder 61
Table 5.1 Uniformity of mass of the prepared tea bags of the A. afra standardized
dried leaves 78
Table 5.2 Uniformity of mass of the prepared tea bags of the freeze-dried aqueous
extract powder of A. afra 78
Table 5.3 Mean percentage release values (n = 6) of luteolin from the A. afra
loose leaves and leaves in tea bag 86
Table 5.4 Comparison of the infusion profiles of luteolin from the A. afra leaves
in tea bag (test) against that from the loose leaves (reference) 86
Table 5.5 Loss on drying of the A. afra dried leaves after 180 days stability testing at
Site A 89
Table 5.6 Loss on drying of the A. afra dried leaves after 90 days stability testing at
Site B 90
Table 5.7 Percentage released by the tea bags of the A. afra dried leaves
at 180 minutes from tea bags stored at storage sites A and B 93
Table 5.8 Luteolin content (µg/ml) infused after 5 minutes infusion of the tea
xiv
bags of the freeze-dried aqueous extract powder stored under
conditions at site C and D 94
Table 5.9 Physicochemical comparison of tea bags of the A. afra leaves and
placebo for the leaves 96
Table 5.10 Physicochemical comparison of tea bags of the A.afra freeze-dried
aqueous extract and placebo for the extract powder 96
Table 5.11 Percentage relaxation produced by cumulative concentrations of the
A. afra standardized dried leaves and placebo of the leaves on the
methacholine-induced contraction of the isolated guinea pig tracheal
muscle 97
List of Figures
Figure 2.1 Artemisia afra plant in its natural habitat at the Montagu museum 4
Figure 2.2 Geographical distribution of Artemisia afra in South Africa 5
Figure 2.3 Schematic diagram showing the biosynthesis of the flavonoids 17
Figure 2.4 The flavonoid nuclear structure and the structure of luteolin (3,4,5,7-
tetrahydroxyflavone) 17
Figure 4.1 Sampling positions used in the quantitation of luteolin in the A. afra plant
materials 40
Figure 4.2 Chemical structure of linalool 46
Figure 4.3 The standardized dried leaves of A. afra plant and the freeze-dried
aqueous extract powder of the A. afra standardized dried leaves 47
Figure 4.4 Retention times and peak heights of morin and luteolin
standards injected for generation of the standard curve 53
Figure 4.6 Chromatogram showing the retention of morin and luteolin within the
plant matrix 54
Figure 4.7 Concentrations of the different forms of luteolin at various sampled
positions in the A. afra plant materials 57
Figure 4.8 Placebo leaves produced by repeated solvent extractions of the A. afra
xv
dried leaves 56
Figure 4.9 A. afra standardized dried leaves chromatographic fingerprint after first
solvent extraction 59
Figure 4.10 A. afra leaves chromatographic fingerprint after final solvent extraction 60
Figure 5.1 The holding cell devised to contain the A. afra tea bags during the infusion
process and a schematic diagram of the modified dissolution apparatus II
used for the determination of the infusion profile of the tea bags
containing the A. afra plant materials 70
Figure 5.2 The organ bath system used for evaluation of the activity of the plant
materials 75
Figure 5.3 A close up picture of a single organ bath 75
Figure 5.4 Prepared tea bag containing 4g of the standardized dried leaves of
A. afra and that containing 0.55g of the A. afra freeze-dried aqueous
extract powder 77
Figure 5.10 A tea bag of the A. afra dried leaves at day 0 and at day 180, stored
under conditions of normal room temperature and humidity 89
Figure 5.11 A tea bag containing the A. afra freeze-dried aqueous extract
powder at day 0 and at day 28, stored under conditions of normal
room temperature and humidity 89
Figure 5.12 A tea bag of the A. afra leaves at day 0 and after 90 days storage
at conditions of 40ºC/75% relative humidity 90
Figure 5.13 The tea bag of the A. afra freeze-dried aqueous extract powder at
day 0, after day 90 stored under conditions at site C and after 90 days
storage at the conditions at site D 91
Figure 5.14 The A. afra standardized dried leaves in tea bag and the A. afra dried
leaves placebo in tea bag 96
Figure 5.15 The freeze-dried aqueous extract powder of A. afra in tea bag, the
placebo of the freeze-dried aqueous extract powder of A. afra in
tea bag and the colours of the teas produced by A. afra placebo
and A. afra freeze-dried extract powder 97
Figure 5.16 Percentage relaxation versus log concentration of the A.afra
xvi
standardized dried leaves and placebo effect on the isolated guinea pig
tracheal muscle 98
List of Graphs
Figure 4.5 The standard curve and linear regression line values used in the
quantitation of luteolin 53
Figure 5.5 The infusion profiles of the A. afra loose leaves and the percentage release
profile of the A. afra loose leaves determined using UV spectroscopic
analysis at 220 and 349nm 79
Figure 5.6 The infusion profiles of luteolin release and the percentage release profiles
of luteolin from the A. afra loose leaves 79
Figure 5.7 The infusion profiles of the A. afra and the percentage release profiles of
the tea bags containing the A. afra leaves determined using UV
spectroscopic analysis at 220 and 349nm 82
Figure 5.8 The infusion profiles of luteolin release from the A. afra tea bag and the
percentage release profiles from the tea bags containing the A. afra
leaves 82
Figure 5.9 The infusion profiles and the percentage release profiles of the tea bags
containing the A. afra freeze-dried aqueous extract, determined using UV
spectroscopy analysis at 220 and 349nm 84
List of Appendices
Appendix A1 Table showing results of the quantitative determination of percentage of
leaves, flowers and stems which constitute Artemisia afra plant and the
average loss on drying of the A. afra leaves after dyring in an oven for 3
days at 30ºC 116
Appendix A2 Figure of certificate of irradiation for the Artemisia afra dried leaves and
freeze-dried aqueous extract powder 117
Appendix A3 Table of the extractions conducted and resultant yields obtained in
xvii
the preparation of the freeze-dried aqueous extract of Artemisia afra 118
Appendix A4 Tables of the results of the particle size classification of the A. afra plant
materials 118
Appendix A5 Method and results obtained for the determination of the bitterness
value of A. afra dried leaves and freeze-dried extract powder 119
Appendix A6 Results of the ash value determinations on the A. afra plant materials 121
Appendix A7 The microbiological testing methods used and the results obtained
in the determination of the microbial contamination on the
A. afra standardized dried leaves and freeze-dried aqueous
extract powders before and after irradiation of the plant materials 123
Appendix B The data and results of the determinations of the luteolin
content and variation in the plant material batches 126
Appendix C The specifications for the DynaporeTM 117/7/0 tea bag
paper used for manufacture of the A. afra plant material tea bags 128
Appendix D1 The results of the organoleptic evaluation conducted on the tea bags
of the A. afra leaves subjected to different storage conditions 129
Appendix D2 Chromatographic fingerprint and UV spectral overlays of the
samples from the A. afra plant materials in tea bag after storage at the
various conditions 131
Appendix E The pen recordings (data) from the pharmacological evaluation
of the placebo material using the isolated guinea pig trachea model 135
1
Chapter 1
Introduction
Artemisia afra is a popular traditional herbal medicine widely used in South Africa for
the treatment of a variety of ailments (Roberts, 1990; Van Wyk & Gerike, 2000).
Infusions of the plant leaves are commonly prepared as teas for treatments ranging from
asthma, malaria to diabetes, and the steam from such infusions may also be inhaled for
headaches and colds (Hutchings et al., 1996; Thring & Weitz, 2006).
Despite the popularity and widespread use of this herb, clinical trials demonstrating its
safety and efficacy in humans are lacking. For traditional herbal medicines in general,
several reasons have been given to account for this state of affairs and include the
variability in the preparation methods and poor quality of the traditional dosage forms,
the variation in the phytochemical composition of the plant materials used and the lack of
credible placebos for use in the clinical trials (Wolsko et al., 2005). For example, in the
Artemisia afra traditional preparations, it is commonly directed that a quarter cup or a
double handful of the wet or dried leaves be infused as a tea (Roberts, 1990) and from
this it may be anticipated that such non-specific directions to measurement may lead to
variations in dose each time a treatment is prepared. In addition, the use of wet leaves in
the dosage preparations may result in poor product quality, as the presence of moisture
may encourage microbial growth. The A. afra plant materials used may also be subject to
factors that may affect their phytochemical composition such as variations in the
geographic source of the materials, time of harvest, drying processes and storage
conditions and as a result, variations in the observed therapeutic effects may be expected
(Graven et al., 1990; Zidorn et al., 2005). Collectively, the above issues, together with the
fact that there is currently no placebo material for A. afra plant, makes it difficult to
conduct good quality clinical trials on the herb.
To address some of the above issues and therefore enable clinical trials on the herb, it was
proposed to standardize the plant leaves and prepare them in a tea bag dosage form. It
was anticipated that such a preparation would be of suitable pharmacopoeial quality and
2
would have release profiles of the actives that were comparable to those observed in the
traditional infusions (and therefore could be used in place of the traditional preparation in
clinical trials). In addition, the process of standardization would reduce the inherent
variability in the phytochemical constituents of the plant materials, guarantee
reproducible clinical effects and therefore enable good quality safety and efficacy data for
the herb to be generated. Furthermore, in order to obtain an even more uniform and
constant material composition, a freeze-dried aqueous extract powder of the plant leaves
was to be prepared in the tea bag dosage form, for administration as an instant tea, with
the added advantage of rapid preparation for the patients (Wichtl, 1994). Finally, it was
intended to design placebo materials which looked, smelt and tasted similar to the A. afra
plant materials and were devoid of pharmacological activity.
Suitable analytical methods such as HPLC were used to characterise the quality of the
plant materials and of the dosage preparations. Methods were designed for the
preparation of the placebo materials and their pharmacological activity was evaluated
using the isolated guinea pig tracheal muscle preparation.
3
Chapter 2
Literature Review
2.1 Introduction
In the present chapter, an overview of Artemisia afra plant is given and this includes its
description, uses and traditional methods of preparation. The shortcomings of the
traditional preparations shall also be highlighted and the proposed methods of addressing
them discussed. The need for safety and efficacy data, including the general aspects of
clinical trials and quality aspects, for herbal products shall also be discussed. Finally, a
review of the methods used for dissolution profile comparisons shall be presented.
2.2 Artemisia afra – An important indigenous medicinal plant
2.2.1 Vernacular names
Artemisia afra Jacq. Ex Willd, is one of the oldest known and most widely used of
medicinal plants in South Africa (Roberts, 1990; Watt & Breyer-Brandwijik, 1962).
Several of the indigenous ethnic groups of South Africa have a long tradition of use of
this plant for various ailments, and as such different names relate to it some of which are
given below (Watt & Breyer-Brandwijik, 1962):
Xhosa : Umhlonyane
Zulu : Mhlonyane
Sotho : Lanyana
Tswana : Lengana
English : African Wormwood
Afrikaans : Wildeals
4
2.2.2 Botanical classification and morphology of Artemisia afra
Artemisia afra belongs to the;
Division : Magnoliphyta
Class : Magnoliopsida
Sub class : Asteridae
Order : Asterales
Family : Asteraceae
Genus : Artemisia
Species : A. afra
Artemisia afra is an erect growing, shrubby, woody, perennial plant growing up to 2 m
tall with a leafy and hairy stem (Van Wyk et al., 1997). Its leaves grow up to 8 cm long
and 4 cm wide. The leaf shape is narrowly ovate, bi-pinnatipartite, feathery and finely
divided. The ultimate segments which are linear in shape with an acute tip and smooth or
toothed margin, grow up to 10 mm long and 2 mm wide. It has a pectinated midrib with
similar lobes, a smooth or glandular-punctated upper surface and a canescent (grayish)
lower surface. The petiole is up to 2 cm long, dilated at the base with a pair of simple or
divided leaf-like stipules. The inflorescence is subglobose with nodding, leafy, terminal,
racemose panicles up to 40 cm long (Hilliard, 1977).
The plant is easily identifiable by its characteristic aromatic odour. It flowers between
January and June, producing yellow coloured flowers. The fruits are about 1mm long,
somewhat 3-angled and slightly curved with a silvery-white coating. In winter, the
branches die back but rapidly regenerate from the base (Hilliard, 1977; Van Wyk et al.,
1997).
2.2.3 Geographical distribution of Artemisia afra
Artemisia afra is found growing abundantly in Namibia, Zimbabwe, and in the mountain
regions of Kenya, Tanzania and Uganda. It is also found as far up as Ethiopia, in the east
Fig. 2.1: Artemisia afra plant.
5
and south of tropical Africa and down to the eastern part of South Africa including
Swaziland and Lesotho (Van Wyk et al., 1997; Hilliard, 1977).
In South Africa, the plant can be found growing in regions from Stellenbosch, the
Western Cape, Aliwal North and the Graaff Reinet Mountains in the North. It is also
widespread in Natal from the coast to the Drakensberg (Hilliard, 1977). Its geographical
distribution in South Africa is shown in the Figure 2.2 below.
2.2.4 Traditional uses and dosage forms of Artemisia afra
Artemisia afra is a herb used for the treatment of several ailments (Van Wyk et al., 1997).
It is a well-known treatment for coughs, colds, croup, whooping cough, colic, heartburn,
flatulence and gout (Roberts, 1990; Thring & Weitz, 2006). It is also commonly used for
the treatment of asthma, acute bronchitis, hay fever, bladder and kidney disorders,
convulsions, diabetes, fever, headache, inflammation, rheumatism, stomach disorders and
worms (Felhaber, 1997; Thring & Weitz, 2006). The herb is also used for the treatment of
malaria (Watt & Breyer-Brandwijik, 1962).
In traditional Zulu medicinal use, leaf infusions are taken as teas or administered as
enemas and the steam from the infusions is commonly inhaled for the treatment of
Figure 2.2: Geo
g
raphical distribution of
Artemisia afra in South Africa
(SATMERG, 1999).
6
headaches and colds (Hutchings et al., 1996). Enemas made from ground plants
suspended in water or milk are administered for constipation or intestinal worms in
children. Decoctions are also taken as blood purifiers for acne, boils, measles and
smallpox (Bryant, 1966). In a survey of medicinal plant use in the Bredasdorp/ Elim
region of the Southern Overberg in the Western Cape Province of South Africa,
Artemisia afra was found to be the plant with the greatest use value amongst the
respondents. In other words, Artemisia afra was used the most for the treatment of
various ailments more than any other traditional plant medicine among the people
surveyed (Thring & Weitz, 2006).
The usual dosage preparation of the plant is in the form of a tea or a decoction (Roberts,
1990; Felhaber, 1997; Van Wyk et al., 1997). Commonly, a quarter cup of fresh leaves to
one cup of boiling water is allowed to stand and steep for 10 minutes, then strained and
sweetened with honey before drinking (Roberts, 1990). This preparation may be
administered orally for the relief of most of the ailments previously mentioned. An
infusion may also be prepared, by pouring 2 litres of boiling water over 1 cup of fresh
leaves and stems and allowed to draw for an hour before being strained and this may be
used as a bath for measles, wounds, bites and stings (Felhaber, 1997). A strong brew of
the herb may be prepared using a half-cup of leaves to 11/2 or 2 cups of boiling water,
allowed to draw for 10 minutes and then strained. This can be used as a mouthwash for
gumboils, mouth ulcers or for the relief of earache (Bryant, 1966). Apart from these
orally administered dosage forms, poultices, vapours for inhalation and enemas are
prepared for the treatment of a variety of ailments (Roberts, 1990; Felhaber, 1997).
2.2.5 Shortcomings of the traditional dosage formulations
As stated above, the common method of dosage preparation is that of a decoction which
involves the use of wet leaves immersed in boiling water to make a tea. The traditional
method of preparation has several disadvantages attributed to it. Firstly, the use of wet
leaves is undesirable. This is because the presence of moisture in the leaves may promote
microbial growth and accelerate degradation of the product (McCutcheon, 2002). The
7
microbes may however be killed by use of boiling water during the decoction process, but
in some circumstances the microbial load may still be outside the acceptable limits as set
by the European Pharmacopoeia for herbal medicinal products (EP, 2002c).
Under the traditional method, the directions for dosage preparation are often obscure.
Often the directions are stated as ‘a double handful’, ‘cup of boiling water’ or ‘quarter
cup of leaves’ (Roberts, 1990). It is difficult to accurately quantify a double handful or
measure a quarter cup consistently. Furthermore the size of ‘cups’ varies and this may
lead to variations in the actual dose the patient receives.
It is well known that plants grown in different geographic locations may contain different
compositions of the active principles (Zidorn et al., 2005; Gilani & Atta-ur-Rahman,
2005). The chemical constituents may vary depending on time of harvest, plant origins,
drying processes, storage times and other factors (Liang et al., 2004; Yang et al., 2005).
Phenolic compounds such as flavonoids in plants have been shown to increase
proportionally with increases in altitude and this has been attributed mainly to their
function as UV-B protective agents (Zidorn et al., 2005). Graven et al (1990) have
reported significantly higher oil yields from Artemisia afra when the crop was harvested
during anthesis and early seed set as opposed to earlier or later harvesting in the
reproductive period. The aforementioned observations may be expected to affect the plant
materials used in the traditional setting, resulting in variations in the phytochemical
composition of the plant and consequently the amount of actives the patient receives in a
given dose.
In the traditional method of preparation, there is no mention of the expected degree of
comminution of the leaves used. In the extraction of constituents from plant matrices, i.e.
in the decoction or infusion process, the degree of comminution (leaf size) is of crucial
importance as the amount of constituents released into the tea per given time period
increases as the degree of comminution increases and the highest yields are often
obtained with smaller leaf sizes (Wichtl, 1994). In addition to the leaf size, the method of
8
preparation or drying/ storage of the plant material may affect the kinetics of release of
the active principles due to changes in the internal leaf structure (Jaganyi & Price, 1999).
2.3 The need for safety and efficacy data
According to the World Health Organization (WHO), about 65-80% of the world
population rely on herbal medicines for their primary health care needs (WHO, 1993). A
marked growth in worldwide use of phytotherapies has occurred over the last 20 years
and in Germany, it is estimated that about 80% of physicians prescribe herbs and sales of
herbal medicines worldwide are in the region of hundreds of billions of dollars a year
(Gilani & Atta-ur-Rahman, 2005). However, the boom in use and popularity has not been
followed by an increase in supporting scientific data and insufficient safety and efficacy
data exists for most plants to support their widespread use (Bombardelli, 2001; Calixto,
2000; Chang, 2000). This state of affairs has been attributed mainly to the lack of patent
protection and the diversity and relatively small-scale of the industry involved, which has
seen many of them unable to meet the financial demands of efficacy and safety studies
(Mills, 1998). In addition, quality issues such as lack of plant material standardization,
good quality dosage forms and thorough characterization of the traditional plant materials
has also hampered clinical trial research or where conducted, has led to poor quality
studies (Wolsko et al., 2005). Finally, the use of different dosages and variations in the
duration of treatment (as administered in the traditional setting) makes comparison and
analysis of the clinical data difficult (Kroes & Walker, 2004).
However, it is important that claims for safety and efficacy of herbal medicines be
scientifically validated. Once the herb has been scientifically validated, it can gain the
acceptance of conventional medicine (Calixto, 2000). Ispaghula, Garlic, Ginseng,
Gingko, St. John’s Wort and Saw Palmetto are a few examples of herbs that have gained
popularity and approval by physicians as a result of scientific validation through clinical
studies (Gilani & Atta-ur-Rahman, 2005). In addition, clinical studies help reveal the
toxic effects, risks and inherent side effects of the plants for safer and more effective use
of the medicines (Calixto, 2000).
9
Once scientifically validated, the herb may be approved by drug regulators and be
marketed for public use. Although preliminary assessments of efficacy can be obtained
through in vitro testing and experiments on animals, authorities licensing new medicines
for public use require evidence of their effect on human beings and clinical trials, with
minimum bias, are able to satisfy these requirements (Mills, 1998).
2.3.1 Clinical trials
A clinical trial is any systematic study of a medicinal product in human subjects whether
in patients or in non patient volunteers in order to discover or verify the effects of and/ or
identify any adverse reaction to the investigational product and/ or study their absorption,
distribution, metabolism and excretion in order to ascertain the efficacy and safety of the
product (EEC, 1997).
Clinical trials on herbal medicines may have one of two types of objectives. One is to
evaluate the safety and efficacy that is claimed for a traditional herbal medicine and the
other is to develop new herbal medicines or examine a new indication for an existing
herbal medicine or change of dose formulation, or route of administration. In some cases,
trials may be designed to test the clinical activity of a purified or semi purified compound
derived from herbal medicines (EEC, 1997; WHO, 1998).
Clinical trials are generally divided into 4 phases. Phase I trials are carried out on a small
number of healthy volunteers or patients suffering from the disease for which the
medicine is intended. The main purpose of this type of trial is to observe tolerance to the
medicine and therefore get an indication of the dose that may be used safely in
subsequent studies (EEC, 1997; WHO, 1998). Phase II studies are conducted on a limited
number of patients to determine clinical efficacy and further confirm safety. Such studies
are usually randomized, double-blind, controlled studies, using for control groups either
an existing alternative treatment or a placebo. The dosage schedules used in such studies
are then used for more extensive clinical studies (EEC, 1997; WHO, 1998). Phase III
studies are performed on larger patient groups usually situated at several study centres
10
using a randomized double blind design to validate the preliminary evidence of efficacy
obtained in earlier studies. Normally such studies are conducted in conditions which
mimic the anticipated or normal conditions of use as closely as possible (WHO, 1998).
Finally, phase IV studies are performed after the dosage form is available for public use
and the main purpose of such studies is to detect toxic events that may occur so rarely
that they are not detected in earlier studies (WHO, 1998).
Clinical trials are required to conform to the standards of Good Clinical Practice (GCP),
which essentially, is an international ethical and scientific quality standard for designing,
conducting, recording and reporting trials. Compliance with this standard provides public
assurance that the rights, safety and well-being of trial subjects are protected, consistent
with the principles in the Declaration of Helsinki, and that the clinical trial data is
credible (EEC, 1997).
2.3.1.1 Placebos
A placebo, also commonly referred to as a dummy pill, is in theory a pharmacologically
inert substance. The placebo may be defined as any therapy or component of therapy that
is deliberately used for its non-specific, psychological, or psychophysiological effect, or
that is used for its presumed specific effect, but is without specific activity for the
condition being treated (Emilien et al., 1998).
Historically, placebos began to be utilized in therapeutic research around the early
nineteenth century in response to accusations that the efficacy observed had to do with
just the appearance of a treatment, as opposed to any real effect from the therapy itself
(Kaptchuck et al., 2001). As a result, placebos are now included in clinical research in
order to function as a mechanism for controlling and ensuring that any observed
improvement in the condition under investigation is due to the effects of the treatment
alone. The requirements for a credible placebo are that factors such as the appearance, the
volume or quantity, the number of intakes, preparation and route of administration be
similar to the treatment under investigation (Emilien et al., 1998).
11
Herbal materials are known to present significant challenges in designing credible
placebos. This is mainly due to the nature of plant materials with respect to their
appearance, smell and taste, which is difficult to mimic using synthetic materials or
chemicals (Kaptchuck et al., 2001). In literature, few herbal clinical trials employing the
use of a placebo actually discuss the ingredients used in the formulation of the placebo
and from the studies found, none were identified as using a placebo of the plant parts
themselves, i.e. the leaves, bark, roots, etc. All studies found used extracts of the herbs
which were formulated as tablets or capsules. One example is that of a placebo of a
Chinese herbal preparation, that contained 78.2% calcium hydrogen phosphate, 19.6%
soy fibre, 0.3% cosmetic brown, 0.5% cosmetic yellow, 0.01% edicol blue 0.09%
identical liquorice dry flavour, and 0.03% bitter flavour (Bensoussan et al., 1998).
2.4 Teas
Tea was first processed as a beverage in China over 3000 years ago and today millions of
cups of tea are consumed everyday around the world making tea one of the most popular
beverages in the world (Peterson et al., 2004).
Teas vary by type (i.e. black, oolong, green and pu’er), variety (i.e. blended and
unblended), and processing (i.e. conventional, cut, tear, and curl). Unblended teas are
named by their country of origin (e.g. Assam, China, Darjeeling), while blended teas are
named generically (Earl Gray, Irish Breakfast), rather than by the teas they contain
(Peterson et al., 2004).
In brewing a tea, several factors are involved in influencing the characteristics of the final
brew. Apart from the quality of the tea leaf itself, the quality of the water used is
important in determining the brew (Jaganyi & Wheeler, 2003). Bottled or filtered water is
normally preferred and this has to do mostly with the influence on the resultant flavour of
the tea (Anonymous, 2006a). The temperature of the water used for the brew is also
important and it is recommended that black teas be brewed with boiling water, oolongs
with water just below boiling point and green teas with water around 80˚C. Again, the
12
temperature affects the flavour as well as the constituents extracted (Anonymous, 2006a).
The other important factor in brewing teas is the steeping time. Steeping times depend on
the type of leaf and its grade. Black teas are normally steeped from 3 - 7 minutes while
oolongs are steeped from 1 ½ - 4 minutes and greens from 2 - 3 minutes. Steeping time
affects the amount of constituent extracted, which ultimately affects the flavour of the tea.
Longer steeping times normally result in bitter teas (Anonymous, 2006a).
Flavonoids are known to be extracted from teas during the tea making process. The
resultant flavonoid content in a cup of tea may depend on one of the two factors, i.e. the
characteristics of the tea leaf itself and the brewing characteristics, as described above
(Peterson et al., 2005). Currently, there is much interest in the type and amount of
flavonoids extracted from teas and herbal infusions. This is largely due to the reported
health benefits of teas such as antioxidant, artherosclerotic and anti- carcinogenic benefits
and flavonoids are believed to be mainly responsible for these actions (Wang &
Helliwell, 2001; Atoui et al., 2005).
2.4.1 Instant teas
An instant tea is produced as a result of an exhaustive extraction of the tea leaves/herb
using water or in some instances water/ ethanol mixtures. The resultant powders are then
mixed with hot water for consumption as a tea (Wichtl, 1994). Generally, two types of
instant teas are available resulting from differences in the manufacturing process. Spray-
dried extracts are dry and hollow pellets produced as the extract is sprayed through a
nozzle and forms fine droplets in a current of warm air. The result is a low-density highly
soluble powder. However, these extracts are often hygroscopic with caking as a major
problem (Wichtl, 1994). Tea granules are granular or cylindrical aggregates produced
when the extract is sprayed onto a carrier (usually saccharose) and dried with heating.
These granular extracts are often very soluble in water and slightly hygroscopic.
However, they contain a small percentage of drug extract, as carrier and filler substances
occupy a large proportion of the granule (Wichtl, 1994).
13
Generally, spray- dried instant teas are considered the ideal for a tea for medicinal
purposes. They have the advantage of rapid preparation, which can be of convenience for
the patients. They also have the added advantage of a uniform and constant composition,
which is important in order to produce reproducible pharmacological effects (Wichtl,
1994; Bombardelli, 2001).
2.4.2 Tea bags
Millions of cups of tea are consumed everyday around the world and most of these are
brewed using tea bags (Jaganyi & Ndlovu, 2001). Different shapes and sizes exist on the
market, all with the aim of capturing the attention of the consumer. Apart from the
traditional square and rectangular shapes, round and pyramidal shaped tea bags are also
common (Jaganyi & Ndlovu, 2001). Tea bag papers used are commonly manufactured
from nonwoven fibres usually based on cellulose from the seeds of cotton or stem fibres
of hemp, jute or abaca trees (Anonymous, 2006b). These may be oxygen bleached or be
processed unbleached to form the tea bag paper. The fibres may also be processed in
combination with synthetic polymers to form papers that are heat sealable (Schoeller &
Hoesch, 2005). High strength at the sealing joints, good cut-ability, high wet strength,
good particle/dust retention and flavour infusion are some of the requirements for a good
quality tea bag paper (Schoeller & Hoesch, 2005).
2.4.2.1 Infusion from tea bags
It has been reported in literature that when infused in water, the tea leaf swells by a factor
of about 4.25 (Spiro & Price, 1985) and therefore, depending on the size of the tea bag,
some hindrance occurs to the swelling of the tea leaf and this affects the rate of infusion
of the constituents (Jaganyi & Mdletshe, 2000). In experiments conducted on the effect of
tea bag size and shape on the rate of caffeine extraction from Ceylon orange pekoe tea
(Jaganyi & Ndlovu, 2001), it was observed firstly, that the infusion of caffeine from loose
tea was much faster than from tea in the tea bag. Secondly, significant increases in the
rate of extraction were observed with increases in tea bag size. From a 16 to 36 cm2 tea
14
bag area, an increase of 25% was observed and smaller increases in the rate of extraction
were observed from tea bags larger than 36 cm2 up to an area of 64 cm2 (Jaganyi &
Ndlovu, 2001).
The above observations have been attributed to increases in the space available for the
swelling and movement of the tea leaf inside the tea bag. The theory behind this
explanation assumes that when the tea leaf comes into contact with water, it swells first
and then infusion follows. In the case of smaller sized tea bags, the tea leaves after
swelling a compacted together. This makes penetration of water to the leaf and to the
centre of the tea bag, more difficult. Moreover, the extracted constituents from the middle
of the tea leaf to the bulk of the solution have to follow a more tortuous passage within
the leaf particles; hence, a slower infusion process results (Spiro & Lam, 1995; Jaganyi &
Ndlovu, 2001).
The hindrance to infusion can also be explained from the point of view of Nernst
diffusion layers (Jaganyi & Ndlovu, 2001). A Nernst diffusion layer by definition is a
fictitious layer above the true concentration profile or diffusion layer (IUPAC, 1997).
This diffusion layer is present both inside and outside the tea bag membrane. The Nernst
diffusion layer on the inside decreases with increase in the tea bag size because of an
increase in free movement of the tea leaves and solution inside the bag. This explains the
observed increase in rate of infusion with tea bag size. It also explains why the infusion is
faster from loose leaves than from a tea bag as the Nernst diffusion layer in the former is
zero (Spiro & Jaganyi, 2000). It also follows that, any motion which decreases the
thickness of the inner and outer Nernst layers adjacent to the tea bag paper membrane,
will increase the rate of tea brewing and this explains empirical findings among tea
drinkers that stirring the brew around the tea bag, moving the tea bag up and down and
jiggling the tea bag all speed up the brewing of the tea (Spiro & Jaganyi, 2000).
Other factors which affect the rate of tea brewing in a tea bag include the transfer of the
constituents through the membrane, diffusion towards and away from the membrane plus
through the swollen leaf, method of manufacture/preparation and storage of the tea leaves
15
and composition and temperature of the extracting medium (Spiro & Jaganyi, 2000;
Jaganyi & Mdletshe, 2000).
2.4.2.2 Advantages of tea bags as a dosage form
Tea bags offer several advantages compared to the traditional tea preparation. Firstly, the
fact that the correct amount (dose) is already accurately measured out offers both safety
(from over- or under- dosing) and convenience in preparation for the patient. This
advantage is more apparent in the case of mixed herbal teas, where the correct
proportions of the herbs (with similar particle size) are already measured into the tea bag
and therefore the patient does not have to scoop and measure each constituent herb as
would be necessary with loose teas (Wichtl, 1994). Secondly, the considerable degree of
comminution, which is required to pack into tea bags, also allows for a better extraction
of the constituents compared to the uncut herb. The amount of constituents in the tea has
been shown to increase as the degree of comminution increases and higher yields are
often obtained using powdered drugs (Wichtl, 1994).
Tea bag paper material has been shown to retard the rate of infusion by about 29% when
compared to loose tea (Jaganyi & Mdletshe, 2000) and this may offer an advantage if one
wishes to control the amount of actives the patient receives within a given time period.
However, in practice, because of the considerable degree of comminution these smaller
leaves infuse faster inside the tea bags such that the overall infusion rate is increased over
and above the larger-sized loose tea leaves (Jaganyi & Mdletshe, 2000).
Finally, tea bags also offer practical advantages. For example, the tea is easier to handle
and simpler and less messy to dispose of as there is no residue left in the cup (Jaganyi &
Mdletshe, 2000).
16
2.5 Phytochemical constituents of Artemisia afra
Literature sources contain some information on the phytochemical constituents of
Artemisia afra. The essential oil of A. afra is known to contain the compounds α-pinene,
γ-terpinene, camphene, ρ -cymene, 1.8-cineole, α-thujone, β- thujone, camphor, borneol,
Artemisia ketone and sesquiterpene- 1-3 (Graven et al., 1990; Van Wyk et al., 1997;
Piprek, 1982). In addition to these compounds found in the essential oil, other
compounds have been detected within the leaves and these include the tannins, saponins,
terpenoids of the eudesmadien and germacratien types, triterpenes α- and β- amyrin, the
friedelin alkanes ceryl cerotinate and n- nonacosane as well as the coumarins and
acetylenes (SATMERG, 1999; Van Wyk et al., 1997).
However, little is documented in the literature concerning the flavonoid constituents of A.
afra. Investigations in our laboratory by Waithaka (2004), Komperlla (2005) and
Mukinda (2006) on leaf infusions and aqueous extracts of the plant leaves have revealed
the presence of the flavonoids luteolin, kaempferol, apigenin and quercetin.
2.6 The flavonoids
Flavonoids are a class of natural compounds present in all vascular plants. They occur in
virtually all plant parts including the leaves, roots, wood, bark, pollen, flowers, berries
and seeds (Markham, 1982). They are the pigments responsible for the hues and various
colours observed in flowers and are also required for the normal growth, development
and defense mechanisms in plants (Di Carlo et al., 1999; Harborne & Williams, 2000).
Flavonoids are biosynthesized via a combination of the shikimic acid and
acylpolymalonate pathways. A cinnamic acid derivative synthesized from shikimic acid
acts as the starting compound in a polyketide synthesis in which additional acetate
residues are incorporated into the structure (Figure 2.3). This is followed by ring closure
and through subsequent hydroxylations and reductions, plants are then able to form
different classes of flavonoids (Markham, 1982; Di Carlo et al., 1999).
17
Fi
g
ure 2.3: Schematic dia
ram showin
the bios
nthesis of the flavonoids
Di Carlo et al.
,
1999
)
.
Structurally, flavonoids consist of a fifteen-carbon atom basic nucleus (C6-C3- C6
configuration) with several phenolic groups and, in some instances, attached sugars to
form glycosides. They possess two benzene rings namely A and B, which are connected
by an oxygen-containing heterocyclic ring C (Figure 2.4). Flavonoids are grouped
according to the presence of different substituents on the rings and by the degree of ring
saturation. Flavonoids containing a pyran ring i.e. a hydroxyl group in position C3 of the
C ring are classified as 3-hydroxyflavonoids (i.e. the flavonols, anthocyanidins and
catechins) and those lacking a hydroxyl group in position C3 as 3-desoxyflavonoids i.e.
the flavonones (e.g. hesperetin, naringenin) and flavones (e.g. luteolin, apigenin)
(Markham, 1982; Heim et al., 2002).
Figure 2.4: The flavonoid nuclear structure (on the left) and the structure of luteolin ( 3,4,5,7-
tetrahydroxyflavone) (on the right).
18
Flavonoids may occur as aglycones (consisting of a benzene ring condensed with a six
member ring which possesses a phenyl ring at the 2 position), glycosides (that carry one
or more sugar residues on the ring) or as their methylated derivatives. Linking to various
sugar residues increases the variation in structure and polymerization of the flavonoid.
For example, quercetin alone is known to have over 179 glycosides and of the 2 x 103
different flavonoids already identified, up to 2 x 106 are thought to exist (Molnar-Perl &
Fuzfai, 2005).
Flavonols and flavones are the most widely occurring flavonoids; of these luteolin (figure
2.4), quercetin, kaempferol, myricetin, chrysin and apigenin are widely distributed. The
flavanones, flavanols, dihydroflavones and dihydrochalcones, are considered minor
flavonoids because of their limited natural distribution (Di Carlo et al., 1999).
The starting step in the analysis of flavonoids, normally involves extraction or isolation
of the flavonoids. Extraction with methanol or solid phase extractions are the commonly
used methods. Optimum extracting conditions, however, vary and depend on the matrix
to be isolated from e.g. plasma, urine or herbal material. Chromatographic techniques
such as HPLC are the methods of choice in the separation, identification and quantitation
of flavonoids. (Molnar-Perl & Fuzfai, 2005; de Rijke et al., 2006). In the analysis
procedures, it is accepted practice to first acid hydrolyse the glycosides and then identify
or quantify the released aglycones. This is mainly due to the large number of flavonoid
conjugates present and the few numbers of commercially available reference compounds
for them (Crozier et al., 1997). The hydrolysis procedure may also be used as a step to
reduce the number of compounds to be determined, resulting in better resolution and
improved characterization of the flavonoid constituents (Molnar-Perl & Fuzfai, 2005).
A wealth of information exists in the scientific literature concerning the pharmacological
properties of flavonoids. Antioxidant, anti-mutagenic, anticancer, anti-hypertensive, anti-
inflammatory, anti-diabetic, anti-allergy, anti-asthma, anti-ischaemic, anti-ulcer,
antibacterial, antiviral and immune stimulating properties (Havsteen, 2002). Some of the
19
reported pharmacological actions of the flavonoids are similar to those reported for A.
afra, which may give a lead as to the possible active compounds within the plant.
2.7 Quality considerations for herbal materials
Quality may be defined as the sum of variable characteristics that may significantly
impact upon a product. For herbal medicines, such variable characteristics include the
origins of the herb, botanical identity, purity, potency, stability and content of specified
marker compounds (McCutcheon, 2002). Apart from these, issues of Good Agricultural
Practices and Good Manufacturing Practices are also important and have a direct bearing
on the final quality of the product (EMEA, 2001).
2.7.1 Standardization of herbal materials
Standardization refers to measures taken to ensure that there is a consistent quantity of a
defined marker compound within a herbal material, as herbal materials are known to be
highly variable in their make up (Gilani & Atta-ur-Rahman, 2005). Intrinsic factors (e.g.
genetics) and extrinsic factors (e.g. growing, harvesting, storage and drying processes)
may lead to variations in the chemical profiles of the herb (McCutcheon, 2002; Yang et
al., 2005; Zidorn et al., 2006). In order to achieve reproducible biological data in terms of
safety and efficacy, it is recommended that the herbal material be standardized to the
active ingredients when they are known, or to specific markers when the actives are not
yet known (Bombardelli, 2001).
Standardization is commonly achieved through blending different batches of the plant
material (McCutcheon, 2002; Bombardelli, 2001). The assumption is that the content of
the other constituents will also vary in proportion to the marker compound, and that if
each batch contains the same amount of marker compound, other constituents will also be
relatively consistent (Bombardelli, 2001). Alternatively, normalization may be employed.
This is the process of adjusting the extraction ratio and/or adding fillers to achieve the
targeted marker content. However, this is usually acceptable within narrow limits and
20
large adjustments are only permissible in cases where it has been established that the
marker is responsible for the pharmacological activity (McCutcheon, 2002). Certain
authorities consider this method similar to adulteration, particularly if this is not declared
on the product label and consequently this method of standardization is not encouraged
(McCutcheon, 2002).
2.7.2 Chemical fingerprints of herbal materials
Chemical fingerprints are commonly used to confirm the identity, authenticity and lot-lot
consistency of a plant (Fan et al., 2006). Currently chromatographic and/or
electrophoretic techniques are used to generate chromatographic patterns of plant
materials (Springfield et al., 2005; Fan et al., 2006). Chemical fingerprints allow most of
the phytochemical constituents of the herb to be determined and in this way, the full
herbal plant can be considered as the active as opposed to the use of a single constituent
(which may or may not be responsible for clinical efficacy). In this way, a form of
chemical standardization can thus be applied as a quantitative ratio among the various
constituents can be established, i.e. fixing a quantitative relationship between classes of
compounds, the active principles or some characteristic compounds present in the plant
(Bombardelli, 2001). It is generally accepted that materials with similar chromatographic
fingerprints are likely to possess similar properties and as a result, the fingerprints allow
the concept of phytoequivalence to be applied. This is the comparison of the chemical
fingerprints for a material or product under question with that of a clinically proven
reference product and is meant to guarantee patient safety as well as protection from
adulterated products (Liang et al., 2004).
2.8 Quality assessment of herbal materials
Monographs exist for some of the commonly used and popular herbal medicines and the
European Pharmacopoeia contains over 130 monographs to which the respective herbs
must comply. However where monographs are not available, quality assessment is
usually based on the results of test procedures performed on the plant material itself
21
(Springfield et al., 2005). Such tests include analysis of the starting material, tests on
microbial quality or contaminants such as pesticides and fumigation agents. Quantitative
determination of marker compounds with known activity is also assessed as a quality
criterion (EMEA, 2001).
2.8.1 Quality assessment of herbal starting materials
An assessment of the quality of the starting material and excipients is required. Firstly,
information on the site of collection, time of harvesting, stage of growth, drying and
storage conditions should be documented (WHO, 2004) and in the case of herbal drugs
with constituents with known activity, assays of their content using validated methods are
required. This content must be stated as a range in order to ensure reproducibility
(EMEA, 2001). Where the constituents are not known, suitable marker compounds may
be selected and used (WHO, 2004; EMEA, 2001).
Generally, herbal materials must be tested for microbial contamination, pesticides and
fumigation agents, toxic metals and other likely contaminants and adulterants.
Acceptance criteria and limits exist but are diverse and there appears to be lack of
consensus on these (WHO, 1998). For instance, the limits for some pesticides published
in the Pharmeuropa 1993 are more restrictive than the WHO limits. In addition, the
limits specified for microbial contamination in the European Pharmacopoeia 2002 are
more restrictive than the WHO 1998 limits.
2.9 Quality assessment of herbal preparations
Guidelines require that the particulars of the characteristics, identification tests and purity
tests for the product be established (EMEA, 2001). These may include details of tests on
the performance of the dosage form such as dissolution or infusion. Chemical fingerprints
can be used to trace the stability of the herbal preparation. Since the whole herbal drug or
preparation may be considered to be the active, a determination of the stability of a single
marker compound may not suffice; an analysis of the whole herbal material may be more
22
appropriate (EMEA, 2001). Heigl et al in a study to examine possible changes in the
flavonoid pattern of common herbal drugs during long term and stress testing storage
conditions, used HPLC fingerprint comparisons to demonstrate differences in stability of
individual flavonoid components (Heigl et al., 2003). Such comparisons may allow
determinations on substances present in the herbal preparations with respect to their
stability and proportions, for quality purposes (Bombardelli, 2001).
2.9.1 Dissolution / infusion testing
In vitro drug release from dosage forms may be characterized as a quality control
procedure as well as to establish in- vitro in- vivo correlations. Compendial methods such
as the BP or USP apparatus I (basket), II (paddle) and III (reciprocating cylinder) may be
used to evaluate the in vitro release characteristics of the dosage form.
Several factors are known to influence drug dissolution and include the effective surface
area of the drug, dissolution coefficient of the drug, thickness of the diffusion layer, the
saturation solubility of the drug, volume of the dissolution media and the amount of drug
in the solution (Missaghi & Fassihi, 2005). Generally, the compendial methods of
dissolution testing have been designed to test for dissolution from conventional
pharmaceutical dosage forms and although a test for dissolution from medicated patches
exists in the British Pharmacopoeia 2000, utilizing an extraction cell to hold the patch in
place, by and large the methods are intended for tablets and capsule dosage forms.
In characterizing the infusion of tea bags, which are both floating and swelling matrices
(a non conventional dosage form), the selection of an appropriate dissolution method
requires special considerations in order to attain sensitive and reproducible dissolution
data. Missaghi and Fassihi (2005) evaluated the effect of various hydrodynamic
conditions on drug release from a floating and eroding matrix. The USP apparatus I, II,
III and a modified apparatus II (paddle over mesh) were evaluated for similarity over
various agitation speeds and the results from the determinations of these investigations
showed that the paddle apparatus at 50 revolutions per minute (rpm) has similar
23
hydrodynamics to the paddle over a mesh at 50 rpm and to the basket apparatus at 100
rpm. These results may aid in the selection of an appropriate dissolution method and
respective agitation rate, for similar non-conventional dosage forms such as tea bags.
2.9.2 Dissolution / Infusion profile comparison methods
Dissolution/infusion profile comparison methods have been developed in order to answer
questions regarding the similarity in performance of pharmaceutical dosage forms. These
comparisons also aid the development of in- vitro in- vivo correlations, which can help
reduce costs, speed up product development and reduce the need to perform costly
bioavailability/ bioequivalence studies. They also have applications in establishing final
dissolution specifications for pharmaceutical dosage forms (O’Hara et al., 1998; Freitag,
2001).
A dissolution profile may be defined as the measured fraction (or percentage) of the
labelled amount of drug that is released from a dosage unit at a number of predetermined
time points when tested in a dissolution apparatus (O’Hara et al., 1998). The methods to
compare dissolution profile data can be categorized as either, exploratory data analysis
which includes graphical and numerical summaries of the data; mathematical methods
that typically use a single number to describe the difference between the profiles and
statistical and modelling methods which take both the variability and underlying
correlation structure in the data into account in the comparison.
Exploratory data analysis methods as stated above, involve graphical and numerical
analysis. The data may be illustrated graphically by plotting the mean dissolution profile
data for each formulation with error bars extending to two standard errors at each
dissolution time point. The dissolution profiles may be considered to differ significantly
from each other if the error bars at each time point do not overlap (O’Hara et al., 1998).
As a complement to the graphical summary, the data may be summarized numerically by
presenting the mean and standard deviation at each time point. In addition, the difference
between the mean profiles and a 95% confidence interval for the difference in the mean
24
profiles may be presented. If the 95% confidence interval does not contain zero, then the
difference at that time point are considered significantly different at the 5% significance
level (O’Hara et al., 1998). The exploratory method is not currently endorsed by the
United States Food and Drug Administration (FDA) and another deficiency is that it is
difficult to definitively conclude that profiles are different if the error bars overlap only at
some time points and not at others, or if the 95% confidence interval for the difference in
the profiles contains zero only at some time points. In addition, if several formulations
are being compared, graphical and numerical illustrations may become too cluttered, with
all the error bars and columns in the tables, making the data evaluation difficult (O’Hara
et al., 1998).
Two mathematical comparison methods are described in the literature. The first described
by Moore and Flanner and the second by Roscigno. Only the Moore and Flanner method
shall be discussed here, as it is the most popular and accepted of the two. Moore and
Flanner described two equations; a ‘difference factor’ f1 (equation 2.1) and a ‘similarity
factor’ f2 (equation 2.2).
Where n is the number of dissolution time point, Rt and Tt are the reference and test
dissolution values at time tt , respectively and wt is an optional weighting factor (Freitag,
2001).
The f1 equation is the sum of the absolute values of the vertical distances between the test
and reference values. The f1 equation is zero when the profiles are similar and increases
proportionally as the difference between the profiles increases and values between 0 and
15 mean similarity between profiles (FDA, 1997). The f2 equation is a logarithmic
transformation of the average of the squared vertical distances between the test and
Eqtn: 2.1
Eqtn: 2.2
25
reference dissolution values at each time point. The f 2
value approaches 100 when the
profiles are identical and values between 50 and 100 ensure similarity between the
profiles (FDA, 1997).
This mathematical method is the most popular of the comparison methods as it is easy to
compute and is accepted and recommended by the FDA. The main disadvantage is that it
does not take into consideration the variability or correlation structure in the data.
However, the FDA implies that the f2 equation should only be used when the within-
batch variation, in terms of coefficient of variation, is less than 15% (FDA, 1997). The f1
and f2 are also sensitive to the number of time points used and it has been criticized that
the basis of the criteria for deciding the difference or similarity between the dissolution
profiles is unclear. In other words, the difference between the dissolution profiles at
which the difference is considered to be of practical importance or likely to affect in vivo
performance is not clear (O’Hara et al., 1998; Saranadasa & Krishnamoorthy, 2005).
Statistical methods include the one-and two-way analysis of variance (ANOVA)
methods, mixed effects model (multivariate methods), modelling-based methods and
Chow and Ki’s methods (O’Hara, 1998). Statistical methods take into consideration the
variability and correlation structure of the data. However, they possess several
disadvantages which limit their use. For example, ANOVA comparisons may only be
statistically significant at some dissolution time points and not at others, thus making it
difficult to conclude whether there is any difference. As a result, ANOVA methods are
only recommended in the case of immediate release data at a single dissolution time point
(O’Hara, 1998; Freitag, 2001). The other statistical methods also have similar pitfalls
which limit their use, i.e. ambiguous in interpretation, tedious to perform, difficulty in
implementation using standard statistical software, sometimes inefficient and a lack of
regulatory endorsement. However, the multivariate method is recommended by the FDA
when the within-batch variability has a coefficient of variation greater than 15%, but this
method also suffers from some of the pitfalls mentioned above (FDA, 1997; O’Hara et
al., 1998).
26
Chapter 3
Work Plan
3.1 Study objectives
The objectives of the study were:
To prepare standardized dried leaves and freeze-dried aqueous extract powder of
A. afra plant leaves and to characterise the physiochemical properties of the
materials,
To develop and validate an analytical method for the detection and quantitation
of luteolin in the A. afra plant materials,
To quantitate and determine the inter- and intra batch variation in the luteolin
content of the plant materials,
To prepare the standardized dried leaves and freeze-dried aqueous extract powder
in tea bag dosage form and determine and compare the infusion profiles (using
luteolin as a marker compound), of the tea bag dosage form and the loose leaves
(traditional dosage form),
To design and evaluate credible placebos for the dried leaves and the freeze-dried
aqueous extract powder of A. afra.
3.2 Hypotheses
It was hypothesised that:
The intra batch variation in the luteolin levels decreases in the order dried leaves
> standardized dried leaves > freeze-dried aqueous extract,
The tea bag dosage form is a suitable dosage form for the traditional plant leaves
and will meet pharmacopoeial quality requirements,
The infusion profile of luteolin from the tea bag dosage form will be similar to
that of the loose leaves (traditional form) and therefore be suitable for use in
clinical trials and,
27
It is possible to design credible placebos for the Artemisia afra plant materials
devoid of pharmacological activity.
3.3 Study approach
This study intended to enable clinical trials on the plant to be conducted by, firstly,
standardizing the A. afra plant materials and preparing a freeze-dried aqueous extract
powder of the plant leaves and secondly, by preparing the plant materials in a tea bag
dosage form and determining the similarity of the infusion profiles of the tea bags and the
loose leaves (in order to ensure interchangeability) and finally, by designing credible
placebos for the plant materials devoid of pharmacological activity.
3.3.1 Why Artemisia afra?
Artemisia afra (Wilde als) has several uses attributed to it in literature. Several authors
describe it as being one of the most widely used and popular plant medicines in South
Africa. It is therefore important that such a popular medicinal plant be validated in terms
of its safety and efficacy using clinical trials. In addition, its many uses, widespread
distribution throughout Africa and several different methods of preparation and
administration made the plant an ideal candidate to model as a plant standardized and
developed into a suitable dosage form for evaluation in clinical trials following which,
regulatory approval for marketing can be granted.
3.3.2 Why a tea bag dosage form?
Traditionally A. afra is administered in the form of a tea. This tea is commonly prepared
by infusing the loose leaves in hot water (Roberts, 1990). Therefore, in order to mimic
the traditional method of preparation, which would be necessary for the evaluation of the
safety and efficacy of the plant as used traditionally, a dosage form which closely
resembles the loose leaves, yet overcomes the disadvantages associated with the loose
leaves was needed. Previous studies have attempted to manufacture tablet dosage forms
of the plant leaves (Komperlla, 2005). These have by and large proved unsuitable due to
28
the hygroscopic nature of the extracts used to manufacture the tablets. In addition, these
extracts may not contain all the constituents found within the leaves. Therefore, a tea bag
containing the A. afra leaves would allow the complete plant leaves to be administered to
the patient and closely resemble the way the plant is used traditionally.
3.3.3 Why luteolin?
It is generally recommended that marker compounds be used to identify and characterize
the quality and stability of plant materials especially where the active ingredient is
unknown (EMEA, 2005). In this study, the flavonoid luteolin was chosen as a marker
compound for Artemisia afra leaves. This flavonoid has been found consistently in
various batches of A. afra plant leaves, and was found to be stable in assay procedures
applied (Waithaka, 2004). In addition, several literature sources describe the
pharmacological effects of luteolin, e.g. anti-oxidant, anti-inflammatory, anti-allergic,
anti-cancer and blood glucose modifying effects (Shimoi et al, 2004; Mino et al, 2004)
and these effects, in relation to the traditional uses of A. afra, may suggest that this
compound may play a role in the observed therapeutic effects of the herb. Also, much
was known on the extraction, detection and quantitation of flavonoids in plant matrices,
making the flavonoids and luteolin in particular, suitable marker compounds to
characterise plant materials.
29
Chapter 4
Preparation and evaluation of Artemisia afra plant material and design
of placebos
4.1 Introduction
In this chapter, the methods used in the preparation of the standardized dried leaves and
freeze-dried aqueous extract powder (herewith referred to as SDL and FDAE,
respectively), of Artemisia afra leaves are presented. The development and validation of
the HPLC assay used for evaluation of the plant materials is also described in detail.
Finally, methods used in the design of the placebos as well as the results obtained are
presented and discussed.
4.2 Equipment and materials
The following equipment were used;
-85˚C freezer (Lozone CFC Freezer, Model U855360, New Brunswick Scientific,
USA), balance (Scaltec SPB42, Model SPB71, Scaltec Instruments, Heiligenstadt,
Germany), centrifuge (Labofuge 200, Germany), furnace (Naber Model L47T.
Industrieofenbau 2804, Lilienthal/Bremen, West Germany), freeze-drier (Virtis
Freeze Mobile 72SL, The Virtis Company Gardner, New York, USA), light
microscope (Nikon Monocular Model Sc, Japan), HPLC filter unit (Millipore
Cameo 25 AS, DDA 02025So MSI: Micro separation Inc., USA), solvent filter
paper (47mm filter membrane 0.45µm PVDF, Millipore, USA), filtration system
(SUPELCO) connected to vacuum pump (Medi-Pump Model 1132-2, Thomas
Industries, Inc., USA), filter paper (Whatman No. 41 & Whatman No.1,
Whatman, England), laboratory blender (Waring Commercial Laboratory
Blender 8010, Model 32BL79, New Hartford, Connecticut, USA), oven (Model
Memmet 854 Schwabach, West Germany), pH meter (Basic 20 Crison
Instruments, S.A, Italy), spectrophotometer (DU 640 spectrophotometer,
Beckman, USA), test sieve shaker (Endecott sieve shaker, E.F.L. 1MK11,
Endecotts (test sieve) LTD, London, England), vortex machine (Vortex-2, G-
30
560E, Scientific Industries, Inc. Bohemia, N.Y. 11716 USA), water bath (Labcon,
CDH 110, Maraisburg, South Africa).
The HPLC system used consisted of an auto sampler (Beckman Gold Module 507), a
programmable binary gradient pump (Beckman Gold Module 126 series), a diode array
detector (DAD) (Beckman Gold Module 168 series) with a 32-KaratTM-software package
and a Synergy® Hydro- reverse phase column (Phenomenex, USA) having 4μm particle
size and a column length of 250 x 4.60 mm.
The following materials were used;
Artemisia afra plant, quinine hydrochloride dihydrate Ph Eur (Fluka Chemie
GmbH, Germany), acetonitrile (Burdick & Jackson, M.I. USA), ethyl acetate CP
(Saarchem, South Africa), luteolin, morin hydrate, hesperetin (Sigma Aldrich,
Germany), sodium hydroxide (UnivAR, Saarchem, South Africa), methanol AR,
hydrochloric acid, potassium hydrogen phosphate (Riedel-de Haen AG,
Germany), lactose (UniLab, Saarchem, South Africa), D.C brown (AR Agenicies
cc, South Africa), sodium starch glycollate (UniLab, Saarchem, South Africa),
potato starch (Saarchem, South Africa), calcium chloride, microcrystalline
cellulose, di-sodium hydrogen phosphate heptahydrate (Merck, South Africa),
sodium saccharin (Supelco. Bellefonte, USA), linalool (Sigma-Aldrich Chemie
GmbH, Germany), silica gel (BDH, South Africa), dimethyl sulphoxide (Sigma-
Aldrich, Germany).
4.3 Methods
4.3.1 Collection and preparation of the plant material
Whole Artemisia afra plants were collected from the Montagu district of the Western
Cape Province of South Africa in February 2005. A sample of the collected plants was
filed as a voucher specimen (Voucher No: 6735) in the herbarium of the Botany
Department of the University of the Western Cape. This collected plant material was
considered as batch 001. Additional plant material batches 002 and 003 were collected
31
from the same district in the months of March and April 2005, respectively. The freshly
collected plant material was separated from earthy and other foreign material and the
leaves (for use in the study) manually separated from the stems and flowers of the plant.
The collected leaves were then weighed before being dried in an oven at a temperature of
30oC for 3 days (Komperlla, 2005). After drying, the leaves were weighed again to obtain
the final dry mass. Finally, the dried leaves were cut up into smaller fragment sizes using
a laboratory blender and the resultant leaf fragments stored sealed in plastic bags in a cool
dark place away from direct light.
4.3.2 Standardization of the Artemisia afra dried leaves
There are various ways of achieving standardization of plant materials. The most
common and accepted method is blending different batches of plant material together
(McCutcheon, 2002), and in this study, the 3 different batches of A. afra plant leaves
were blended together. Equal portions of each plant batch were separately weighed and
placed together in a large plastic bag before being mixed by shaking vigorously for 5
minutes. The resultant leaves blend was then stored in sealed plastic bags in a cool dry
place, protected from direct light.
4.3.3 Preparation of the freeze-dried aqueous extract powder.
The extraction procedure used to prepare the extract, as far as possible, mimicked the
methods described in the literature for the preparation of the traditional dosage.
Commonly, a quarter cup quantity of A. afra leaves to a cup of boiling water is allowed to
stand and seep for 10 minutes (Roberts, 1990). Therefore, the aqueous extract was
prepared by adding boiling distilled water to the appropriate mass of standardized dried
leaves (SDL) in 1:35 ratio of leaves to solvent (Komperlla, 2005) and the extraction
mixture left to seep and draw for 10 min. After this period, the mixture was filtered using
Whatman no. 1 filter paper and the filtrate frozen at -85oC in a freezer. The frozen extract
was then dried at -44oC under vacuum over 3 days using a freeze drier. The resultant
extract powders were then combined and weighed and the percentage yield calculated.
32
The FDAE powder was then placed in amber glass containers and these stored in
desiccators until used (Komperlla, 2005).
4.3.4 Irradiation of the plant materials
There is a requirement for microbiological decontamination of materials of natural origin
due to the high levels of microbes inherent in these materials. However, methods of
decontamination are few and are often restricted. For example the use of ethylene oxide is
forbidden, and the efficacy of some other methods has not been well established (WHO,
1998). Ionizing irradiation is a well established decontamination method which has been
proved effective and safe, having been used for decontamination of spices and seasonings
for several years (Razem & Katusin-Razem, 2002).
In this study, the SDL as well as the FDAE powder of A. afra were decontaminated using
ionizing radiation from a Cobalt (Co) source at a level of 18 kGy. Pre- and post-
irradiation microbial tests were conducted to check and confirm the effectiveness of the
procedure in the decontamination of the plant materials.
4.3.5 Determination of the plant material organoleptic characteristics
Organoleptic characteristics refer to the appearance, colour, odour, and taste of a
substance. An examination to determine these characteristics is normally the first step
towards establishing the identity and degree of purity of materials (WHO, 1998; EMEA,
2005). Characterization of these properties is also a primary step in setting of
specifications, as well as setting a standard to which a placebo should resemble if it is to
be credible. In addition, changes in the stability of a material may be recognized by
changes in the organoleptic properties. Sweeteners, flavorants and aroma chemicals may
be added to mask objectionable odours and tastes present in a material (Zheng & Keeney,
2006).
33
For this study, the appearance, colour, odour and taste of the SDL and FDAE powder of
A. afra were characterized using the human sensory evaluation (i.e. by eye, nose and
tongue).
4.3.5.1 Determination of the plant material particle size and shape.
Particle size and shape have an effect on the physico-chemical properties of a dosage
form. They play a great role in producing homogeneity of dosage forms and, in the case
of tea bags, influence the rate of infusion of the soluble constituents. Smaller fragment
sizes are known to infuse faster than larger sized ones (Wichtl, 1994; Jaganyi &
Mdletshe, 2000) and this difference in the rate of infusion may affect the dose the patient
receives for a given tea brewing time. The size and shape of the particles also has a direct
impact on mixing and significant deviations in the particle size can result in the
segregation of the mixture, with smaller fragments settling at the bottom and larger
fragments at the top of the mixture (Fayed & Otten, 1984). Sieve and microscopic
methods are the commonly used methods to determine particle size and shape and were
the methods used in this study.
4.3.5.1.1 The sieve method
The degree of coarseness or fineness of a powder is classified according to the nominal
aperture size expressed in micrometers of the mesh of the sieve through which the
powder will pass (BP, 2000a). Using this criterion, powders are divided into ‘coarse’,
‘moderately fine’, ‘fine’, and ‘very fine’. The degree of fineness of the powder is
expressed as a weight-to-weight percentage of the powder passing through different size
sieves. If a single sieve number is given, not less than 95% of the powder passes through
the sieve of that number, unless otherwise indicated.
For the evaluation of the A. afra plant materials, sieves having numbers of 1400, 355,
180, 125 and 90 µm were assembled in a descending order, i.e. 1400 µm size sieve on top
and 90 µm at the bottom. The assembled set of sieves was placed on a test sieve shaker
34
before 10 g of the SDL or FDAE powder was placed onto the top sieve, and the assembly
shook for 30 minutes. Thereafter the powder collected on each of the sieves was weighed
and the percentage (w/w) of each fraction determined.
4.3.5.1.2 Determination of the plant material particle shape
The microscopic method allows the determination of the shape of the particles and a
Nikon light microscope was used to determine the particle shape of the plant materials
used. A few milligrams of the SDL or FDAE powder were sprinkled onto a slide and
viewed under the microscope. The particle shapes were observed and described as either
irregular, round, cylindrical or rectangular.
4.3.5.2 Determination of the plant material colour
Colour can be used as a means of identifying a particular substance. Several
Pharmacopoeias include the colour of the substance as part of the substances monograph.
In this study, the colour of the A. afra SDL and FDAE powder as well as the colour of the
hues produced when these materials are infused in hot water to produce teas was
determined. This information was crucial for the design of the placebos which should be
similar in appearance to the plant materials.
In determining the colour of the materials, 1 g of the material being examined was placed
against a white background and its colour described. In determining the colour of the
hues produced, 4 g of the SDL and 1g of the FDAE powder were infused in 200 ml of
distilled water at a temperature of 80˚C, stirred for 2 minutes, and allowed to stand for 8
minutes. The resultant hue was then visually examined and described accordingly.
4.3.5.3 Determination of the plant material odour
Odour can be used as a quick method for identification of a substance. The presence or
absence of odour in A. afra plant material was determined using the method prescribed in
35
the British Pharmacopoeia 2000. One gram of the SDL or the FDAE powder of A. afra
was placed on a watch glass about 5 cm in diameter. The material was then allowed to
stand for 15 minutes, smelt and the presence or absence of odour from the material noted
(BP, 2000b).
4.3.5.4 Determination of the plant material bitterness value
The bitterness value is defined as the reciprocal of the dilution of a compound, liquid or
extract that still has a bitter taste (EP, 2002a). The bitter properties of plant materials are
determined by comparing the threshold bitter concentration of a preparation of the
material with that of a dilute solution of quinine hydrochloride. The threshold bitter
concentration is defined as the lowest concentration at which the material continues to
provoke a bitter sensation 30 seconds after tasting it. The bitterness value is expressed in
units equivalent to the bitterness of a solution containing 1g of quinine hydrochloride, the
bitterness value of which is set at 200 000 (WHO, 1998). In the literature A. afra is
described as having an extremely bitter taste (Van Wyk et al., 2000; Roberts, 1990). In
order to aid effective design of the herbal placebos it was necessary to characterize the
bitterness value of the plant materials.
For the bitterness value determinations, the SDL and FDAE of A. afra were dispersed in
boiling distilled water at a concentration of 0.01 g/ml, stirred and allowed to stand for 10
minutes before being filtered using Whatman No. 1 filter paper. The solutions were
allowed to cool before tasting. Serial dilutions of quinine hydrochloride and the plant
materials, i.e. 0.00001 g/ml and 0.001, 0.0001, 0.00002, 0.00001 g/ml respectively, were
then prepared as described in the European Pharmacopoeia 2002 (EP, 2002a).
A panel comprising of 6 tasters was set up. In order to correct for individual differences
in tasting bitterness, a correction factor for each taster was established using quinine
hydrochloride. The panel members then tasted serially diluted concentrations of the SDL
and the FDAE solutions, starting with the most dilute concentration. The bitterness value
36
was determined as prescribed in the European Pharmacopoeia 2002 using the following
equation:
Bitterness Value = Y x k Eqtn: 4.1
X x 0.1
Where; Y = Dilution factor of threshold bitter concentration
k = Correction factor for each panel member
X = Number of ml of threshold bitter concentration which when diluted to
10 ml still has a bitter taste.
4.3.6 Determination of the plant material ash values
Ash values can be considered as quality standards to indicate identity, purity or possible
adulteration of a herbal material. The ash remaining following ignition of medicinal plant
materials is determined by different methods which measure total ash, acid-insoluble ash
and water-soluble ash. The total ash method is designed to measure the total amount of
material remaining after ignition. This includes both “physiological ash”, which is
derived from the plant tissue itself, and “non-physiological” ash, which is the residue of
the extraneous matter (e.g. sand and soil) adhering to the plant surface (WHO, 1998).
Acid-insoluble ash is the residue obtained after boiling the total ash with dilute
hydrochloric acid, and igniting the remaining insoluble matter. This measures the amount
of silica present, especially as sand and siliceous earth. Water-soluble ash is the
difference in weight between the total ash and the residue remaining after treatment of the
total ash with water (EMEA, 2005).
Total ash and acid-insoluble ash determinations were performed on the standardized dried
leaf material and FDAE powder of A. afra according to the prescribed methods in the
British Pharmacopoeia 2000. Water-soluble ash determination is a non-pharmacopoeial
test and was performed by adding 25 ml of water to the crucible containing the total ash
and boiling for 5 minutes. Thereafter, the insoluble matter was collected in a sintered-
37
glass crucible, washed with hot water, ignited in a crucible for 15 minutes at a
temperature of 450°C and the remaining ash retrieved. The difference in weight (in mg)
of this residue from the weight of the total ash was the water-soluble ash (WHO, 1998).
4.3.7 Determination of the plant material moisture content
The moisture content of the plant materials was determined using the method prescribed
in the European Pharmacopoeia 2002 which involves the use of a drying oven. For this,
0.5g quantities of the SDL and the FDAE powder of A. afra were placed in a flat-
bottomed dish about 50mm in diameter and dried in an oven for 3 hours at a temperature
range of 100-105ºC (EP, 2002b). The moisture content was calculated as a mass
percentage using the following formula:
% Moisture = Initial weight (Wet mass) – Final weight (Dry mass) x 100 Eqtn: 4.2
Initial Weight
4.3.8 Determination of microbial contamination of the materials
Medicinal plant materials normally carry a great number of bacteria and moulds, often
originating from the soil. Bacterial and fungal contamination is of serious concern
especially in terms of health risks (EMEA, 2005). Unacceptable levels of microbial
contamination are often due to poor agricultural practices or improper drying and storage
of the harvested plant material and typically result in degradation of the plant materials.
In some cases microbial contamination may render the plant material toxic either by
transforming benign chemicals in the plant into harmful substances, or through the
microbes’ production of toxic compounds (McCutcheon, 2002).
The European Pharmacopoeia provides guidance on which microbes are to be tested for
as well as their minimum acceptable limits (EP, 2002c). Following this guidance, the
total viable aerobic count, total yeasts and moulds, absence of Escherichia coli,
Salmonella and enterobacteria were consequently tested for in the SDL and FDAE
38
powder of A. afra. In addition to this, tests not specified in the pharmacopoeia were also
conducted. These were tests for absence of Staphylococcus aureus and Pseudomonas
aeruginosa. Microbiological testing on the materials was conducted before and after
irradiation of the materials, using standard microbial test methods as described in
Appendix A7.
4.3.9 Development and validation of the HPLC assay
For the quantitation of luteolin in the plant materials, a reversed phase HPLC method
using an internal standard was developed. The compounds hesperetin, p-coumaric acid,
mefenamic acid, morin hydrate, cinnamic acid and salicylic acid were evaluated as
possible internal standards. The requirements were that the compound should not be
present in the A. afra plant materials, should absorb strongly at the luteolin absorption
wavelength maximum (349 nm), be extractable in ethyl acetate, have similar
physiochemical characteristics to luteolin, and have a chromatographic retention time that
did not interfere with other peaks from the plant.
Different chromatographic conditions were evaluated for the separation of the luteolin
peak from the internal standard peak and other possible interfering peaks. Three different
types of reverse phase chromatographic columns (i.e. silica C-18 (250 x 4.6 mm, 5 µm),
Luna ® C-18 (150 x 4.60 mm, 5 µm), Luna® C-18 (250 x 4.6 mm, 5 µm)), different
mobile phase compositions and flow rates were evaluated and optimized. The mobile
phases consisted of either methanol or acetonitrile, with a mixture of acetic acid or formic
acid in water or a phosphate buffer at different concentrations to control the pH and to
reduce peak tailing. To validate the assay, the linearity, recovery, limit of detection
(LOD), limit of quantitation (LOQ) and intra- and inter-day precision, for the assay were
determined.
In performing the validation, firstly, stock solutions of luteolin and morin hydrate
(1mg/ml) were prepared in dimethyl sulphoxide (DMSO), wrapped with aluminium foil
to protect from light and stored at -21˚C. Standard solutions were freshly prepared each
39
day by appropriate dilution of the stock solutions with the mobile phase (acetonitrile
30%, phosphate buffer 70% at pH 2). Concentrations of 2.5, 5, 10, 20, 50 and 100 µg/ml
of luteolin and 100 µg/ml of morin were prepared.
For the preparation of the calibration curve, the stock solutions were diluted with mobile
phase to produce the concentrations described above. To each concentration of luteolin,
50 µg of morin hydrate was added. 50 µl of each standard was injected onto the column
and the peak height ratios between luteolin and morin plotted against the corresponding
concentrations of the injected luteolin. The complete procedure was repeated over 3
consecutive days.
The limit of detection (LOD) was obtained by successively decreasing the concentration
of luteolin as long as a signal to noise ratio of 3: 1 appeared. The limit of quantitation
(LOQ) was defined as the lowest concentration where an accuracy better than 20 % was
achieved (Ruckert et al., 2004).
Finally, for the determination of intra-day precision and accuracy, 6 replicates of the
standards were analysed on the same day. The precision and the accuracy for the inter-
day analysis were determined on 3 different days.
4.3.10 Quantitation of luteolin in the plant materials
The flavonoid luteolin was used as a marker compound for the A. afra plant materials. As
part of the characterization of the plant materials, different batches of the plant leaves, the
SDL and the FDAE were evaluated for the average content of total, free and glycosidic or
conjugated luteolin using the validated HPLC assay. In addition, the intra- and inter-batch
variation in the content of the different forms of luteolin in the various plant materials,
were also determined.
A defined procedure was used for sampling from the various plant material batches. This
entailed randomly taking 20g of the plant material under evaluation from the batch bulk
40
and spreading it on a tile 10 x 10cm. From here, 6 sampling positions were defined (as
shown in Figure 4.1) and 250 mg samples were removed from each position.
The sampled plant material was then prepared following the common traditional method
for the preparation of the tea. The 250 mg sample was placed in 10 ml of hot (80˚C)
distilled water (25 mg/ml), vortexed for 2 minutes and then allowed to stand for 8
minutes before being filtered using Whatman no. 1 filter paper. The filtrate was separated
into two 1 ml portions for analysis of the total luteolin (hydrolysed) and free luteolin
(unhydrolysed).
For the total luteolin assay, the 1 ml sample was transferred to a tube containing 4.8 ml of
2N HCl and an appropriate amount of internal standard. The mixture was vortexed for 30
seconds, before being left to ‘hydrolyse’ in a water bath at 80˚C for 40 minutes.
Following this, the sample was allowed to cool, 5 ml of ethyl acetate added, vortexed for
2 minutes (to ensure the luteolin was extracted into the ethyl acetate layer) and then
centrifuged at 3000 rpm for 10 minutes to separate the ethyl acetate from the aqueous
1 2
3 4
5 6
10cm
10cm
Figure 4.1: Sampling positions used in the quantitation of
luteolin in the A. afra plant materials.
41
layer. Finally, the top layer of ethyl acetate was pipetted off into a clean tube and
evaporated to dryness under a cool stream of nitrogen.
For unhydrolysed (free) luteolin assay, the 1 ml sample was transferred to a tube
containing 5ml of ethyl acetate and internal standard. This was vortexed for 2 minutes
before being centrifuged at 3000 rpm for 10 minutes. The top layer of ethyl acetate was
then removed and the sample evaporated to dryness as described above.
To quantify the amount of luteolin in the extracted samples, the dry residues were
reconstituted in 1ml of the mobile phase. 50 µl of the solution was then injected onto the
column and isocratically eluted at a flow rate of 1 ml/min at room temperature with a run
time of 30 minutes using a mobile phase of acetonitrile 30% and phosphate buffer 70% at
pH 2). Detection was conducted at a wavelength of 349 nm and the luteolin in the
samples quantitated from a standard curve based on the peak height ratio of luteolin and
internal standard.
4.3.11 Design of the plant material placebos
Designing credible herbal placebos is known to present special challenges especially with
regard to achieving and maintaining concealment. This is mainly due to the fact that
herbal medicines are more elaborate in their composition compared to the simple
chemical structures of drugs commonly encountered in pharmaceutical placebo design
(Kaptchuck, 2001). In this study, the objective was to design placebos which looked,
smelt and tasted similar to the A. afra SDL and FDAE.
4.3.11.1 Design of placebo for the standardized dried leaves
The SDL of A. afra were used to produce the placebo. Since the phytochemical
constituents, in particular the flavonoids, were thought to be responsible for the observed
therapeutic effects of the plant leaves, it was inferred that A. afra leaves without these
phytochemical constituents or with very low levels of them, would be devoid of
42
pharmacological activity. Therefore, to produce the placebo of the dried leaves, the
phytochemical constituents were to be extracted out of the A. afra leaves as much as
possible.
A method utilising exhaustive solvent extraction with hydrolysis was developed to extract
as much as possible of the flavonoids (as well as any other compounds soluble in the
extracting solvents) within the leaves. In selecting appropriate solvents, flavonoid
solubilities were taken into consideration. For example, for the extraction of water-
soluble flavonoids, a more polar solvent was selected. The safety or hazardous potential
of the solvents were also taken into consideration. Various solvents were evaluated for
their effectiveness in extracting the constituents using the minimal number of extractions.
Based on the above considerations, a total of 15 extractions using a combination of either
water, methanol/ water (50/50, v/v), methanol or ethyl acetate, (depending on the
particular extraction step) including a hydrolysis and neutralisation step (to free up the
sugar bound flavonoids for optimal extraction), were conducted. In all extractions a mass
to solvent ratio of 1:20 was used. In conducting the extractions, the leaf and solvent
mixture was stirred for 2 minutes then allowed to stand for 10 minutes before being
filtered using Whatman no. 1 filter paper. The leaf to solvent ratio was kept constant in
all extractions. From the supernatant, a 1 ml sample was retained for analysis using UV
spectrophotometry and to the remaining residue of leaves, additional solvent was added.
UV spectrophotometry was conducted at the luteolin absorbance maxima (349 nm) as
well as the absorbance maxima for the plant material extract. The process was repeated
using the next solvent in the procedure, while monitoring the UV absorbance of the
extract, until the absorbance was acceptably low. Samples of the initial and final extracts
were also analysed using the validated HPLC assay in order to obtain a chromatographic
fingerprint of the material as well as obtain a measure of the extraction process through a
comparison of the pre- and post extraction chromatograms.
43
4.3.11.2 Design of the placebo for the freeze-dried aqueous extract powder
There are few clinical trial studies in the literature, where placebos of herbal extracts have
been employed (Pan et al., 2000). In many of these trials, extracts in capsules or tablets
were used with the result that the patient could not see the extract itself. In making these
placebos, elaborate formulas of inorganic salts were employed to mimic the extract as
well as impart good flow properties to the powder to aid the manufacture of the tablets or
capsules. One example is that of a placebo of a Chinese herbal preparation containing
78.2% calcium hydrogen phosphate, 19.6% soy fibre, 0.3% cosmetic brown, 0.5%
cosmetic yellow, 0.01% edicol blue 0.09% identical liquorice dry flavour and 0.03%
bitter flavour (Bensoussan et al., 1998).
In the design of the placebo of the A. afra FDAE, various selected inorganic salts were
blended together. These salts were chosen based on their organoleptic properties relative
to the organoleptic properties of the FDAE. In addition, in order to match the hue
produced when the A. afra FDAE is dissolved in water, a colourant was included in the
placebo formulation. The appropriate proportions of inorganic salts to be added were
determined on a trial and error basis, where a formulation was evaluated against the
aqueous extract and modified as needed. The various formulations and proportions of
each salt tried are shown in Table 4.1 below.
44
Table 4.1: Various excipients and their proportions used in different
formulations in the design of the placebo for the A. afra freeze-dried
aqueous extract powder.
Amount of various excipients in the various formulations (g)
Excipient 1 2 3 4 5 6
Potassium dihydrogen
phosphate
0.275 --- --- 0.200 0.400 0.475
Microcrystalline
cellulose
0.225 0.300 0.300 --- --- ---
Sodium starch glycollate
--- 0.100 --- 0.100 --- ---
Potato starch
--- --- 0.130 0.130 0.050 0.03
Lactose
--- --- 0.050 --- 0.035 0.025
Calcium chloride
---- --- --- 0.050 --- ---
DC Brown 0.050 0.015 0.020 0.020 0.02 0.02
Di-sodium hydrogen
phosphate heptahydrate
--- 0.135 --- 0.050 --- ---
Total 0.55 0.55 0.55 0.55 0.55 0.55
4.3.11.3 Matching of odour between the placebos and the plant materials
A. afra is a highly aromatic plant. The dried leaves and infusions from the leaves
(aqueous extract) also possess a characteristic odour. In order to make the odours similar
between the placebos and the A. afra plant materials, an aroma chemical was employed to
provide an aroma to the placebos as well as to spike the A. afra plant materials so that the
plant material odours are matched.
An appropriate aroma chemical was selected from a range of commercially available
synthetic aroma chemicals. In selecting the appropriate chemical, the fragrance
characteristics, flavour properties, toxicity, fixative properties and substantivity (which
refers to the ability of the perfume to last on a specific surface) of the chemical was taken
into consideration. A chemical with a more natural or green fragrance with a high
tenacity and substantivity was preferred in order to resemble the herbal smell of A. afra.
45
The aroma chemical selected to mask the odour of the A. afra plant materials was linalool
(Figure 4.2). This chemical possesses a fruity, herbal woody, rosewood odour (BASF,
2006). Linalool is a substance widely used in functional and alcoholic perfumery. It is
also frequently used in fruit imitations (peach, apricot, pineapple, grape, berry flavours
etc), and for chocolate and spice complexes (BASF, 2006). It is accepted by the Council
of Europe for use in foods as an artificial flavouring and is considered Generally
Recognised as Safe (GRAS) by US Food and Drug Administration (OECD, 2004). Based
on this, the aroma chemical was deemed suitable for inclusion in the formulations.
In order to determine the appropriate concentration of the aroma chemical sufficient to
mask the smell of an A. afra tea solution, a modification of the European Pharmacopoeia
method for the determination of bitterness of materials was used. Serial dilutions of the
aroma chemical were prepared, i.e. 0.0042, 0.0048, 0.0052, 0.0056 and 0.0058 g/ml and
mixed with equal proportions of a 20 mg/ml hot water extract of A. afra. Using a panel of
four evaluators, the natural sense of smell was used to discriminate and evaluate the
odours. The concentration of the aroma chemical at which the odour of the A. afra in the
mixed solution could not be determined was taken to be the threshold masking
concentration. This concentration capable of masking the characteristic odour of A. afra
was therefore used in matching the smell properties of the materials.
In order to incorporate the aroma chemical into the materials, a method was devised
whereby a silica gel adsorbent would adsorb the aroma chemical and be able to release
the aroma when in solution. The aroma chemical was dissolved in 1ml of ethyl acetate
and the solution added to 0.125 g of silica gel powder. The ethyl acetate was then
evaporated off under a stream of nitrogen to leave a dry silica gel powder containing the
aroma chemical. Appropriate amounts of the aroma-impregnated silica gel were then
added to the A. afra SDL and FDAE, as well as the placebo materials.
46
Figure 4.2: Chemical structure of linalool.
4.3.11.4 Matching of taste between the placebos and the plant materials
A. afra is an extremely bitter tasting plant (Roberts, 1990). Therefore the inclusion of a
sweetening agent was considered in order to improve the patient acceptance of the plant
medicine as well as to match the taste properties of the plant with that of the placebo
(Zheng & Keeney, 2006). In the selection of an appropriate sweetener, the safety profile
and classification of the sugar compound was taken into consideration. Sodium saccharin,
a synthetic sweetener was preferred as opposed to a nutritive caloric natural sweetener
such as glucose. This was to avoid possible confounding of results if the preparation was
to be used in a trial such as a diabetes trial.
In order to determine the concentration of sodium saccharin required to mask the
bitterness of the plant materials, a method was devised based on a modification of the
European Pharmacopoeia bitterness value determination method. Serial dilutions of the
chemical were prepared and mixed with equal proportions of a 20 mg/ml hot water
extract of A. afra. Using a panel of 4 tasters, the natural sense of taste was used to
establish the solution in which the bitter taste could no longer be tasted and the solution
that was determined to be sweet. The amount of sodium saccharin in this solution was
then adsorbed onto the silica gel adsorbent as described in section 4.3.11.3 above and
incorporated into the A. afra and placebo materials.
47
4.4 Results and Discussion
4.4.1 Preparation of the plant materials
For this study, the leaves were removed from each of the 3 collected A. afra plant
batches. On average 43.5% of the total A. afra plant weight consisted of the leaves.
Drying of these collected leaves in the oven for 3 days at 30˚C in resulted in an average
57.36 ± 7.51% loss in weight. This result therefore indicates that approximately 50% of
the mass of harvested A. afra leaves is in the form of moisture. This information may be
useful to researchers in future studies on the plant, planning the amount of plant material
required in order to attain a certain dry mass of leaves. The complete data for the
collection and preparation are presented in Appendix A1.
To prepare the FDAE powder, a mass ratio of leaves to solvent of 1:35 was used. Several
extractions of the SDL were conducted and a summary of the yields obtained for each
extraction are presented in Appendix A3. The average yield for the FDAE powder
obtained was 21.96 ± 1.97%. This value was similar to that obtained previously by other
investigators, viz 27.88% and 19.9% by Komperlla (2005) and Mukinda (2006),
respectively. Figure 4.3 shows an image of the SDL and the resultant FDAE powder.
Figure 4.3: The standardized dried leaves of A. afra (left) and the freeze-dried aqueous extract
powder of the A. afra standardized dried leaves (right).
48
4.4.2 Organoleptic properties of the A. afra plant materials
The results of the organoleptic determinations on the A. afra SDL and FDAE powder are
shown in Table 4. 2 below. The leaves were observed to be grey-green in colour, while
the FDAE powder resulting from them was light brown in colour (see Figure 4.3). This
difference in colour may point to a possible breakdown or change in the compounds
present in the leaves occurring during the infusion process. When the FDAE was allowed
to stand for 15 minutes, the extract became darker in colour and the powder clumped
together. This observation was similar to that made by Komperlla (2005) and was
attributed to the hygroscopic nature of the extract. The taste, odour and colour of the teas
(hue) of the leaves and FDAE were however similar. The A. afra leaves and FDAE
powder were also both bitter in taste. The bitterness of the leaves is described in the
literature and in some traditional preparations sugar or honey is added to the preparations
to mask the bitterness and improve the patient acceptance of the medicine (Van Wyk &
Gerike, 2000; Roberts, 1990). The above information gathered during the organoleptic
evaluation was used to set specifications for the plant materials and with these
specifications, set standards for the placebo materials to meet.
Table 4.2: Organoleptic characteristics of the A. afra standardized dried leaves and
freeze-dried aqueous extract powder.
Organoleptic
characteristic
A. afra
Standardized dried leaves
A. afra
freeze-dried aqueous extract
powder
Appearance
Colour of material
Colour of hue
produced
Odour
Taste
Particulate leaves, finely divided,
hydrous, with a rough texture.
Grey-green.
Light brown to golden brown hue.
Characteristic odour, highly aromatic.
Odour still present after 15mins on
watch glass.
Extremely bitter.
Brittle, free flowing, small particulate
powder, which clumps together to
form a solid mass on prolonged
exposure to air.
Light brown powder changing to dark
brown on prolonged exposure to air.
Light brown to golden brown hue.
Characteristic odour, aromatic. Odour
still present after 15mins on watch
glass.
Extremely bitter.
49
4.4.2.1 Particle size and shape of the A. afra plant materials
Under the microscope, the particles of the SDL and FDAE powder were both observed to
be irregular in shape. The results of the particle size determinations on the materials are
presented in Appendix A4. Based on the BP 2000 classification system, both the SDL
and FDAE were classified as being coarse powders. The particles of the SDL passing
through the 1400 µm but retained by the 355 µm sieves i.e. those in the particle size
range 1400-355 µm were retained for use in this study, while the FDAE powder was used
un-sieved.
4.4.2.2 Bitterness value of the A. afra plant materials
The results of the determination and calculation of the bitterness values are presented in
Appendix A5. The bitterness value for both the A. afra SDL and FDAE was calculated to
be 600 000. From this, it meant that the A. afra leaves and FDAE were 3 times more
bitter than quinine hydrochloride which is known to be an extremely bitter tasting
compound. To our knowledge this is the first estimate of the bitterness of A. afra leaves
and FDAE powder, and this result can be added to the list of already known
specifications for A. afra plant materials. The fact that the bitterness values for the dried
leaves and the FDAE were similar, indicated that the compounds responsible for the
bitterness were extracted from the leaves during the infusion process and were probably
present in similar amounts in the two preparations. This therefore implied that in the
formulation of the placebo materials, the placebo of the FDAE powder and dried leaves
needed to be of similar taste. In addition, as a result of the bitterness, a sweetener
chemical was required in order to mask the bitter taste of the materials and therefore
improve the patient acceptance of the medicines.
4.4.3 Ash values of the A. afra plant materials
The complete data sets and results for the ash value determinations on the A. afra
materials are presented in Appendix A6. The average percentage of total ash, acid
50
insoluble ash and water soluble ash for the SDL were 9.9832 ± 0.0617%, 0.9308 ±
0.0012% and 4.4052 ± 0.0012%, respectively. These results are comparable to the
8.558% and 1.341% previously reported by Komperlla (2005) for the total ash and acid-
insoluble ash, respectively, of A. afra dried leaves.
The average percentage of total ash, acid insoluble ash and water-soluble ash for the
FDAE was found to be 21.820 ± 0.0886%, 0.0094 ± 0.0793% and 16.5952 ± 0.6760%,
respectively. These results were also comparable to those obtained by Komperlla (2005),
viz. 17.436% and 0.027% for total ash and acid insoluble ash, respectively. To the best of
our knowledge, values for water-soluble ash of A. afra have not previously been reported
and the values above are the first such for A. afra plant leaves and FDAE. These values
may be added to the current list of specifications for the plant materials.
Collectively, the ash values, in addition to the organoleptic characteristics results helped
to confirm the identity of the A. afra plant materials used. It was noted that the A. afra
materials used by the previous investigators mentioned above was collected from the
same source as that used in this study. It may therefore be inferred that A. afra plants
growing in similar conditions (similar climate and soil conditions) possess similar
organoleptic and physiochemical properties.
4.4.4 Moisture content of the A. afra plant materials
The results of the moisture content determinations on the A. afra SDL and FDAE powder
are shown in Tables 4.3 and 4.4, respectively.
Table 4.3: Residual moisture content of the A. afra standardized dried leaves
Sample
No
Initial mass of leaves (g) Final mass of leaves (g) Loss on drying
(g)
Percentage
moisture content
1
2
3
4
5
0.5253
0.5191
0.5206
0.5213
0.5149
0.4700
0.4650
0.4722
0.4613
0.4558
0.0553
0.0541
0.0484
0.0600
0.0591
10.53
10.42
9.38
11.51
11.48
Ave 0.5202 0.4649 0.0554 10.66
S. D 0.0038 0.0066 0.0046 0.88
51
Table 4.4: Residual moisture content of the A. afra freeze-dried aqueous extract
powder.
Sample
No
Initial mass of aqueous
extract (g)
Final mass of aqueous
extract (g)
Loss on drying
(g)
Percentage of
moisture
1
2
3
4
5
0.5039
0.5071
0.5145
0.5081
0.5020
0.4609
0.4645
0.4695
0.4642
0.4577
0.0430
0.0426
0.0450
0.0439
0.0443
8.53
8.40
8.75
8.64
8.82
Ave 0.5071 0.4634 0.0438 8.62
S. D 0.0048 0.0044 0.0010 0.17
The average moisture content for the SDL and FDAE powder of A. afra were 10.66 ±
0.88% and 8.62 ± 0.17%, respectively. Moisture content determinations are necessary,
particularly for materials known to be hygroscopic (EMEA, 2005). In the present study
the moisture content determinations were conducted immediately following the drying of
the plant materials and therefore were an indication the residual moisture present in the
materials at that stage. The values obtained in this part of the study were therefore used as
the moisture content specifications for the materials and were used in the later part of the
study to determine moisture absorption by the materials upon storage.
4.4.5 Microbial contamination of the plant materials
Microbial contamination tests were conducted before and after irradiation of the plant
materials. The tests conducted before irradiation included tests for total microbial
activity, and for absence/presence of Escherichia coli, Pseudomonas and Staphylococcus
aureus. All testing was conducted externally by a company contracted to do the testing.
The results of the microbial contamination tests are presented in Appendix A7. Before
irradiation, considerable contamination was detected within the dried leaves, i.e. a total
microbial activity of more than 106 and the presence of more than 104 and 300 colony
forming units (CFU) per gram, of Pseudomonas and E. coli bacteria respectively. Based
on the European Pharmacopoeia 2002 guidelines, which state that not more than 200
CFU/gram E. coli and not more than 107 bacteria should be present (EP, 2000c), the dried
leaves did not comply with the specification on microbial contamination of herbals. The
52
presence of E. coli in the plant leaves may have reflected the quality of the growing and
harvesting practices at the source of the material, as E. coli bacteria are not normally
found as part of the flora and fauna of the soil (WHO, 1998). The tests on the FDAE
powder before irradiation however indicated the presence of a reduced total microbial
activity of 984 000 and the absence of all the other organisms previously detected in the
dried leaves. This implies that hot water extractions, and presumably also the traditional
decoction/infusion process, reduce the microbial load present in the plant materials.
After irradiation of the plant materials, no contamination was detected for the tests
conducted (i.e. absence of total microbial activity, yeasts & moulds, E. coli, Salmonella
and Enterobacteriaceae) demonstrating the efficacy and usefulness of the irradiation
process. From these results, it was thus assured that the plant materials were suitable for
use in the manufacture of the medicinal dosage forms.
4.4.6 Development and validation of the HPLC assay
A reversed-phase HPLC method, using morin as internal standard, was developed for the
quantitation of luteolin in A. afra plant materials. The optimal conditions of separation
and detection were achieved on a Synergy® Hydro - reverse phase column, packed with a
4 μm particle size resin and a column length of 250 x 4.60 mm using a mobile phase of
acetonitrile (30%) and phosphate buffer 100mM (70%, pH 2) at a flow rate of 1 ml/min
and UV detection at 349 nm.
The standard curve of peak height ratio versus concentration of luteolin was linear in the
range of 2.5 - 100 µg/ml (i.e. 125 – 5000 ng luteolin on column) (Figures 4.4 and 4.5),
and was described by the equation Y = 0.03407X + 0.008233 (where Y = peak height
ratio, X = luteolin concentration in µg/ml) and had a regression correlation coefficient, r
value of 0.9996. The concentration of internal standard used was 50 µg/ml and the HPLC
injection volume was 50 µl.
53
Figure 4.4: Retention times and peak heights of morin and luteolin standards injected for
generation of the standard curve. The average retention times for morin and
luteolin were 11.90 ± 0.06 minutes and 15.86 ± 0.05 minutes, respectively.
025 50 75 100
0
1
2
3
4
Conc (μg/ml)
Peak Height Ratio
Figure 4.5: The standard curve and linear regression line values, used in the quantitation of
luteolin.
The results of the accuracy and recovery of the assay are shown in table 4.5 and the inter-
and intra- day precision results are presented in Table 4.6. The mean recovery of the
method was 94.70 ± 8.62%. The concentration of 2.5 µg/ml had accuracy greater than
20.0% and was therefore considered to be the limit of quantitation (LOQ) for the assay
(Ruckert et al., 2004). The LOD was found to be 50 ng on column, from an injected
volume of 20 µl corresponding to a concentration of 2.5 µg/ml. The average intra- and
inter-day precision were found to be 2.497 ± 0.6470% and 3.123 ± 1.068%, respectively.
Collectively, these values indicated the good validity and reproducibility of the assay.
Best-fit values
Slope 0.03407 ± 0.0004926
Y-intercept 0.008233 ± 0.02296
X-intercept -0.2416
1/slope 29.35
95% Confidence Intervals
Slope 0.03271 to 0.03544
Y-intercept when X=0; 0.05550 to 0.07196
X-intercept when Y=0; -2.171 to 1.587
Goodness of Fit
R 0.9996
Sy.x 0.04172
40
Minutes
0246810 12 14 16 18 20
mAU
0
50
100
150
200
250
mAU
0
50
100
150
200