universidade de lisboarepositorio.ul.pt/bitstream/10451/6211/1/ulsd062640_td_pedro_fale.pdf ·...

UNIVERSIDADE DE LISBOA

FACULDADE DE CIÊNCIAS

DEPARTAMENTO DE QUÍMICA E BIOQUÍMICA

BIOLOGICAL ACTIVITIES OF Plectranthus barbatus

AQUEOUS EXTRACTS.

IN VITRO AND IN VIVO STUDIES OF ACTIVITY,

BIOAVAILABILITY AND METABOLISM.

Pedro Luis Vieira Falé

DOUTORAMENTO EM BIOQUÍMICA

(Bioquímica Farmacêutica e Toxicológica)

2011

UNIVERSIDADE DE LISBOA

FACULDADE DE CIÊNCIAS

DEPARTAMENTO DE QUÍMICA E BIOQUÍMICA

BIOLOGICAL ACTIVITIES OF Plectranthus barbatus

AQUEOUS EXTRACTS.

IN VITRO AND IN VIVO STUDIES OF ACTIVITY,

BIOAVAILABILITY AND METABOLISM.

Pedro Luis Vieira Falé

DOUTORAMENTO EM BIOQUÍMICA

(Bioquímica Farmacêutica e Toxicológica)

Tese orientada por

Prof. Doutora Maria Luísa Mourato Oliveira Marques Serralheiro

Prof. Doutora Lia Maria Pereira de Ascensão Santos e Sousa

2011

The work presented in this thesis was performed in the

Centro de Química e Bioquímica da Faculdade de

Ciências da Universidade de Lisboa, with the financial

support of Fundação para a Ciência e a Tecnologia

(SFRH/BD/37547/2007)

v

De acordo com o disposto no Artigo nº 41 do Regulamento de Estudos Pós‐Graduados da

Universidade de Lisboa, Deliberação nº 1506/2006, publicada no Diário da República, 2ª

série—Nº 209—30 de Outubro de 2006, foram incluidos nesta dissertação resultados dos

seguintes artigos:

Falé PL, Borges C, Madeira PJA, Ascensão L, Araújo MEM, Florêncio MH, Serralheiro MLM. 2009.

Rosmarinic acid, scutellarein 4’‐methyl ether 7‐O‐glucuronide and (16S)‐coleon E are the main

compounds responsible for the antiacetylcholinesterase and antioxidant activity in herbal tea of

Plectranthus barbatus (‘‘falso boldo”). Food Chem. 114: 798–805.

Porfirio S, Falé PL, Madeira PJA, Florêncio H, Ascensão L, Serralheiro MLM. 2010.

Antiacetylcholinesterase and antioxidant activities of Plectranthus barbatus tea, after in vitro

gastrointestinal metabolism, Food Chem. 122: 798–805.

Falé PL, Madeira PJ, Florêncio MH, Ascensão L, Serralheiro MLM. 2011. Function of Plectranthus

barbatus herbal tea as neuronal acetylcholinesterase inhibitor. Food Funct. 2: 130‐136.

Falé PL, Ascensão L, Serralheiro MLM, Haris PI. 2011. Interaction between Plectranthus barbatus herbal

tea components and human serum albumin and lysozyme: binding and activity studies. Spectroscopy.

26: 79–92.

Falé PL, Ascensão L, Serralheiro MLM, Haris PI. Interaction between Plectranthus barbatus herbal tea

components and acetylcholinesterase: binding and activity studies. Submitted to Food Funct.

Falé PL, Ascensão L, Serralheiro MLM. Bioavailability of rosmarinic acid, luteolin, apigenin‐modelling

plant herbal teas through Caco‐2 cell monolayers. To be submitted to Food Chem.

Falé PL, Filipe MA, Ascensão L, Serralheiro MLM, Mira L. Activity of Plectranthus barbatus extract against

inflammatory response in human neutrophils. To be submitted to Plant Food Hum. Nutr.

No cumprimento do disposto na referida deliberação, esclarece‐se serem da minha

responsabilidade a execução das experiências que estiveram na base dos resultados

apresentados, assim como a interpretação e discussão dos mesmos.

vii

Aknowledgements

Many people have contributed, directly or indirectly, to the successful completion of

this work. First, I would like to express my deepest gratitude to Professor Maria Luísa

Serralheiro for her guidance and counseling, all the challenges, the knowledge and, most of all,

the practical (engineer-like) point of view. The elaboration of this work under her guidance

was not only a scientifically rewarding experience, but also a pleasant one.

I would like also to thank to Professor Lia Ascensão for the guidance and attention to

details the elaboration of this work. I would also like to thank her for the support, friendship,

and continuous encouragement all these years.

To Professor Parvez Haris, his group, and staff at the De Montfort University, in

Leicester, I would like to thank for the hospitality and for all the support in the studies

developed there.

To Professor Maria Helena Florêncio I would like to thank the expertise in the

compound identification by mass spectrometry. To Paulo Madeira I would also like to thank,

not only the help in mass spectrometry anlysis, but also for the lighter moments and all his

patience for listening expressions such as “Are you sure about this structure? I don’t like it!”.

I would also like to thank to Sara Porfírio, Ines Sousa Lima, Rita Carilho, Susana Santos,

Catarina Costa, Leticia Silva, Pedro Cleto, Neusa Figueiredo, Ana Margarida Rodrigues,

Francesca Amaral and Catarina Ferreira, the Master students that helped, even if only by their

enthusiasm.

Last, but not least, I am grateful to my friends from the “family of C4 library”, for

providing the encouragement and a stable (geek) environment. To Charlie, for lightening (and

enlightening) my life even in the toughest moments. And to my family, especially my Mother,

for the support through this period, but also in my dedication to such an odd area as Science,

instead of Arts.

ix

Summary

The Plectranthus barbatus herbal tea is traditionally used to treat a wide range of

health conditions, including psychiatric problems, gastrointestinal disturbances and

inflammation‐related conditions. The aim of this work was to determine if P. barbatus herbal

tea may be useful in the treatment of acetylcholinesterase or inflammation related diseases,

such as Alzheimer’s disease.

In vitro activities of the P. barbatus aqueous extract were determined, namely anti‐

acetylcholinesterase activity, antioxidant activity as radical scavenger and preventing lipid

peroxidation, and anti‐inflammatory activity by decreasing the amount of hypochlorous acid

produced by activated neutrophils. The values obtained for the in vitro activities were very

promising and related with its content in rosmarinic acid, flavonoid glucuronides and abietane

diterpenoids. After in vitro digestion with gastric and pancreatic juices, bacterial

β‐glucuronidase and metabolization by Caco‐2 cells, the anti‐acetylcholinesterase activity of

the plant extract suffered a slight decrease due to the loss of one active diterpenoid.

The bioavailability of the plant extract was determined by administering it to rats and

analyzing the rat plasma and brain. The extract components suffered glucuronidation, sulfation

and methylation by the intestine and by the liver, but the plant compounds were found in rat

brains and brain acetylcholinesterase activity showed an inhibition reaching 30%. As the

bioavailability of rosmarinic acid was not the same when in the extract and when alone, the

interference of plant phenolics on the permeability of each other was tested in Caco‐2

monolayers using rosmarinic acid and two flavonoids, apigenin and luteolin. This study showed

that the compounds have higher intestinal permeability when in a mixture due to the

inhibition of the efflux transporters that limit their bioavailability.

The compounds from the P. barbatus extract can bind to the protein structure of

acetylcholinesterase, human serum albumin and lysozyme by weak interactions such as

hydrophobic interactions and hydrogen bonds. These interactions are the cause of the

reversible inhibition of the enzymatic activity of acetylcholinesterase and lysozyme; therefore

the compounds are less susceptible to cause side effects when used therapeutically.

The P. barbatus herbal tea may be used to treat cholinesterase‐related problems such

as gastrointestinal conditions and Alzheimer’s disease as its active components may reach the

target organs, and brain acetylcholinesterase inhibition was detected. The compounds may

circulate in the plasma bound to albumin and lysozyme, and may decrease inflammation by

their radical scavenger activity, by decreasing neutrophil‐produced hypochlorous acid and by

inhibiting lysozyme activity.

x

Keywords: Acetylcholinesterase; Antioxidant; Bioavailability; Plectranthus barbatus;

Rosmarinic acid.

xi

Resumo

Infusões e decocções de Plectranthus barbatus são usadas tradicionalmente para uma

grande diversidade de fins terapeuticos, incluido o tratamento de problemas psiquiátricos,

distúrbios gastro‐intestinais e doenças relacionadas com processos inflamatórios. O objectivo

deste estudo é determinar se o extracto aquoso de P. barbatus, preparado como decocção,

poderá ser útil no tratamento de problemas relacionados com a actividade do enzima

acetilcolinesterase ou com processos inflamatórios.

Um extracto aquoso de P. barbatus foi preparado como decocção e foram

determinadas, in vitro, as actividades anti‐acetilcolinesterase, antioxidante e anti‐inflamatória.

Nestas actividades in vitro foram obtidos resultados muito promissores, com valores de IC50

baixos para a inibição da actividade do enzima acetilcolinesterase, actividade antioxidante no

sequestro de radicais livres e protegendo lipidos de peroxidação, e na actividade anti‐

inflamatória pela diminuição da quantidade de ácido hipocloroso produzido por neutrófilos

activados. Estas actividades estavam relacionadas com a composição do extracto de P.

barbatus, sendo ácido rosmarinico o composto maioritário, mas encontrando‐se também

presentes flavonóides glucuronados (apigenina 7‐O‐glucurónido, luteolina 7‐O‐glucurónido e

acacetina 7‐O‐glucurónido) e diterpenoides. Após a digestão in vitro do extracto vegetal com

sucos gástrico e pancreático artificiais, da acção de β‐glucuronidase de bactérias da microflora

intestinal e da metabolização por células Caco‐2, como modelo de células intestinais humanas,

a actividade do extracto sofreu uma diminuição devido à perda de um diterpenoide activo, no

entanto os outros compostos activos permaneceram intactos.

A biodisponibilidade da decocção de P. barbatus foi determinada por administração

intragástica e intraperitoneal a ratos, recolhendo‐se e analisando‐se o plasma e o cérebro dos

ratos por HPLC. Os componentes da decocção foram metabolisados no intestino e no fígado,

sofrendo glucuronidação, sulfatação e metilação. Encontrou‐se ácido rosmarínico no cérebro

dos ratos após a administração do extracto, e o cérebro apresentava uma inibição da

actividade da acetilcolinesterase atingindo aproximadamente 30%, sugerindo que outros

compostos ou metabolitos dos componentes da decocção pudessem estar presentes em

quantidades inferiores ao limite de detecção, mas que mesmo assim iriam influenciar a

actividade enzimática. Como se encontraram algumas alterações entre a biodisponibilidade do

ácido rosmarínico administrado no extracto ou administrado isolado, procedeu‐se ao estudo

da interferência que compostos fenólicos possam ter na permeabilidade uns dos outros

quando administrados em misturas, em membranas de células Caco‐2. Nesse estudo

recorreu‐se ao método de “central composite design” CCD para determinar se existiriam

xii

diferenças entre a permeabilidade e metabolização de ácido rosmarínico, apigenina e luteolina

em membranas de células Caco‐2, induzidas pela presença uns dos outros. Foram também co‐

administrados substractos de dois sistemas transportadores conhecidos, o transportador de

ácidos monocarbóxilicos (MCT) e a glicoproteina‐P (Pgp), e concluiu‐se que os compostos da

decocção de P. barbatus tinham maior permeabilidade por estarem em conjunto, uma vez que

uns compostos inibiam os transportadores de efluxo dos outros. A biodisponibilidade dos

substratos dos transportadores de efluxo presentes na membrana apical aumenta com a

inibição dos transportadores, pois estes transportam activamente os seus substratos para o

lúmen do intestino, limitando assim a passagem para a circulação sanguínea.

A interacção entre os compostos do extracto de P. barbatus e proteinas foi avaliado

através de técnicas de espectrometria de fluorescência e FTIR. Os componentes da decocção

de P. barbatus têm a capacidade de se ligar à estrutura proteica da acetilcolinesterase, da

albumina de soro humano e da lisozima através de interacções fracas, nomeadamente

interacções hidrofóbicas e pontes de hidrogénio. Por se tratarem de ligações fracas e não se

terem encontrado alterações da estrutura secundária das proteinas, estas interacções serão

responsáveis pela inibição reversivel da actividade enzimática da acetilcolinesterase e da

lisozima. Este tipo de interacções são as mais recomendadas para compostos a ser utilizados

para fins terapeuticos pois evitam efeitos secundários resultantes da inibição destes enzimas

por inibidores que formam complexos através de ligações mais fortes, como ligações

covalentes, e induzem alterações profundas na estrutura proteica levando à desnaturação.

Decocções de P. barbatus poderão ser utilizadas para tratar problemas relacionados

com a actividade da acetilcolinesterase, como a doença de Alzheimer, pois após a

administração os seus componentes poderão ser encontardos a a nivel do intestino, plasma e

cérebro, inibindo a actividade da acetilcolinesterase no cérebro. Os componentes da decocção

poderão circular na corrente sanguínea associados à albumina e à lisozima, e poderão reduzir

processos inflamatórios devido à sua actividade sequestradora de radicais livres, à sua

capacidade de diminuir a quantidade de àcido hipocloroso produzido por neutrófilos

activados, e por inibir a atividade enzimática da lisozima.

Palavras‐chave: Acetilcolinesterase; Ácido Rosmarínico; Antioxidante; Biodisponibilidade;

Plectranthus barbatus.

xiii

Table of Contents

Acknowledgments vii

Summary ix

Resumo xi

Table of Contents xiii

Figure List xvii

Table List xx

Abbreviations xxii

CHAPTER 1 – GENERAL INTRODUCTION 1

1. Literature Review 3

1.1. Plectranthus species and their ethnobotanical uses 3

1.2. Herbal tea components and their bioavailability 4

1.2.1. Phenolic acids 4

1.2.2. Flavonoids 6

1.3. Biological activities of herbal teas and their components 9

1.3.1. Acetylcholinesterase inhibitors 9

1.3.1.1. Acetylcholinesterase inhibitors and Alzheimer’s disease 10

1.3.1.2. Acetylcholinesterase inhibitors to treat gastrointestinal disorders 13

1.3.1.3. Finding acetylcholinesterase inhibitors ‐ from in vitro to in vivo studies 14

1.3.2. Inflammation and Antioxidants 16

2. Thesis overview 19

CHAPTER 2 – MATERIALS AND METHODS 21

1. Plant material 23

2. Animals 23

3. Chemicals 23

4. Extract preparation 24

5. Acetylcholinesterase inhibition 24

6. Determination of antioxidant activity 25

7. HPLC analysis 26

8. NMR spectroscopy 26

9. Mass spectrometry experiments 26

10. In vitro intragastric metabolism assays 27

10.1. In vitro metabolism by the gastric juice 27

10.2. In vitro metabolism by pancreatic juice 27

10.3. Glucuronidase activity 28

10.4. Metabolism by the Caco‐2 cells 28

10.5. Antiacetylcholinesterase and antioxidant activities of the digested extracts 28

11. In vitro conjugation studies for metabolites identification 29

11.1. Preparation of cell‐free extracts 29

11.2. Glucuronidation assay 29

xiv

11.3. Synthesis and identification of methyl rosmarinic acid 29

12. In vivo studies protocol 30

12.1. Intragastric and intraperitoneal administration 30

12.2. Plasma and brain sample preparation 30

12.3. Determination of glucuronidated and sulfated metabolites 30

12.4. Preparation of samples for HPLC analysis 31

12.5. Determination of acetylcholinesterase activity in brain samples 31

13. Protein fluorescence measurements 31

14. FTIR measurements 32

15. Lysozyme activity measurements 33

16. Measurement of hypochlorous acid in activated human neutrophils 33

16.1. Isolation of human neutrophils 33

16.2.Measurement of hypochlorous acid formation by human neutrophils 33

17. Caco‐2 bioavailability experiments 34

17.1. Study of the permeability and the metabolism of rosmarinic acid, luteolin and

apigenin 34

17.2. Study of the effect of transport systems (MCT and Pgp) on the permeation of

apigenin, luteolin and rosmarinic acid 35

17.3. HPLC analysis and bioavailability quantification 35

18. Statistical analysis 36

CHAPTER III ‐ SCREENING FOR ANTIACETYLCHOLINESTERASE AND

ANTIOXIDANT ACTIVITIES IN PLECTRANTHUS SPECIES. IN VITRO METABOLISM

AND ANTI‐INFLAMMATORY STUDIES 37

1. Screening for antiacetylcholinesterase and antioxidant activities in Plectranthus

species 39

1.1. Introduction 39

1.2. Materials and Methods 40

1.3. Results 40

1.3.1. General 40

1.3.2. Acetylcholinesterase inhibition 41

1.3.3. Antioxidant Activity 42

1.3.4. Identification of the main component of the Plectranthus extracts,

responsible for the enzyme inhibition activity 42

1.4. Discussion 45

1.5. Conclusion 46

2. In vitro Digestion Activities of Plectranthus barbatus Aqueous Extract and Activities

of the Digested Product 47



2.3. Results and Discussion 48

2.3.1. Main composition of P. barbatus herbal tea 48

2.3.2. In vitro metabolism of the extract by the gastric and pancreatic juices.

Biological activity of the resulting products 50

xv

2.3.3. Metabolism of the plant extract by Caco-2 cells and biological activity of

the final products 53

2.3.4. Metabolism of the plant extract by the β-glucuronidase from E. coli,

biological activities and Caco-2 cells permeation of the final products 55

2.4. Conclusion 57

3. Activity of Plectranthus barbatus extract against inflammatory response in human

neutrophils 59



3.3. Results and Discussion 60

3.4. Conclusions 63

4. Conclusions 64

CHAPTER IV – BIOAVAILABILITY STUDIES IN RATS AND CACO-2 CELL MONOLAYERS 65 1. Plectranthus barbatus aqueous extract bioavailability and resulting neuronal acetylcholinesterase inhibition in rats 67 1.1. Introduction 67 1.2. Materials and Methods 68 1.3. Results and Discussion 68

1.3.1. Intragastric administration of P. barbatus extract 68 1.3.2. Intraperitoneal administration of P. barbatus extract 71

1.4. Conclusions 76 2. Bioavailability of mixtures of rosmarinic acid, luteolin and apigenin through Caco-2 cell monolayers, modeling the bioavailability of plant herbal teas. 77 2.1. Introduction 77 2.2. Material and Methods 78 2.3. Results 78

2.3.1. Bioavailability of Plectranthus barbatus herbal tea 79 2.3.2. Bioavailability of a mixture of the standards rosmarinic acid, apigenin and luteolin 80 2.3.4. Effect of MCT and Pgp transporter systems on the bioavailability of the polyphenol mixture 82

2.4. Discussion 86 2.5 Conclusions 89 3. Conclusions 91

CHAPTER V – INTERACTIONS BETWEEN THE Plectranthus barbatus HERBAL TEA AND THE PROTEINS ACETYLCHOLINESTERASE, HUMAN SERUM ALBUMIN AND LYSOZYME 93 1. Interaction between the Plectranthus barbatus extract and acetylcholinesterase. Binding of herbal tea components to the protein structure and inhibition of enzymatic activity 95 1.1. Introduction 95 1.2. Materials and Methods 96 1.3. Results 97

1.3.1. Fluorescence studies on the binding of P. barbatus water extract to acetylcholinesterase 97

xvi

1.3.2. Analysis of binding equilibria of P. barbatus water extract to acetylcholinesterase

99

1.3.3. Determination of interaction forces between P.barbatus extract and AChE 101 1.3.4. Determination of protein structure changes caused by P. barbatus extract and its plasma metabolites by FTIR spectroscopy 102

1.4. Discussion 105 1.5. Conclusions 107 2. Interaction between Plectranthus barbatus herbal tea components and human serum albumin and lysozyme: binding and activity studies 109 2.1. Introduction 109 2.2. Material and Methods 110 2.3. Results 110

2.3.1 Binding of P. barbatus to albumin and lysozyme 110 2.3.2. Analysis of binding equilibria 112 2.3.3. Determination of interaction forces between P. barbatus extract metabolites and HSA and lysozyme 115 2.3.4. Determination of protein structure changes caused by P. barbatus extract and its plasma metabolites by FTIR 116 2.3.5. Effect of P. barbatus extract on lysozyme activity 118

2.4. Discussion 118 2.5. Conclusions 122 3. Conclusions 123

CHAPTER VI – GENERAL DISCUSSION AND CONCLUSIONS 125

References 133

xvii

Figure List

Figure 1.1. Plectranthus barbatus 4

Figure 1.2. Chemical structures of some common phenolic acids. 5

Figure 1.3. Basic chemical structures of the main classes of flavonoids. 7

Figure 1.4. Structures of (a) quercetin, (b) apigenin and (c) luteolin. 7

Figure 1.5. Active gorge of acetylcholinesterase (Abu‐Donia, 2003). 10

Figure 1.6. Tacrine binding to the active gorge of acetylcholinesterase, and details showing the

amino acid residues involved in the interaction between the two molecules (PDB 1ACJ, Harel

et al., 1993). 11

Figure 1.7. Biochemical pathways associated with the formation of amyloid plaques and

neurofibrillary tangles in Alzheimers’s Disease patients. (www.calbiochem.com/alzheimers). 12

Figure 1.8. Reaction associated with the Ellman assay to quantify acetylcholinesterase activity.

The final product TNB can be spectrofotometrically detected due to its absorption at a

wavelength of 405 nm. (adapted from Frasco et al., 2005) 15

Figure 1.9. Antioxidant strategies in Alzheimer’s disease. Solid arrows represent the

mechanisms of the disease and dashed arrows represent the mechanisms of antioxidant

therapy (Adapted from Dumond and Bael (2011)). 18

Figure 3.1. HPLC chromatogram of decoctions: (a) Plectranthus barbatus and (b) Plectranthus

verticillatus.

43

Figure 3.2. UV spectra obtained by HPLC‐diode array of (a) compounds with retention time

19.2 min and (b) caffeic acid. 43

Figure 3.3. NMR spectra of (a) Plectranthus barbatus extract, (b) Plectranthus verticillatus

extract and (c) rosmarinic acid standard. 44

Figure 3.4. Structure of the compounds with retention time 19.2 min, rosmarinic acid. 44

Figure 3.5. Overlay of UV spectra obtained by HPLC‐diode array of compounds with a retention

time 19.2 min (——) and the standard rosmarinic acid (___). 45

Fig. 3.6. HPLC chromatogram of Plectranthus barbatus herbal tea: 1, luteolin 7‐O‐glucuronide

(retention time 8.68 min); 2, rosmarinic acid (RT: 9.38); 3, apigenin 7‐O‐glucuronide (RT: 10.11);

4, hydrolysed abietane (RT: 12.58); 5, acacetin 7‐O‐glucuronide (RT: 13.62); 6, abietane

diterpenoid (RT: 14,15); 7, (16S)‐coleon E (RT: 18.54). 49

Figure 3.7. Chemical structure of compounds present in P. barbatus herbal tea:

1, luteolin 7‐O‐ glucuronide; 2, rosmarinic acid; 3, apigenin 7‐O‐glucuronide; 4, hydrolysed

abietane; 5, acacetin 7‐O‐glucuronide; 6, abietane diterpenoid; 7, (16S)‐coleon E. 50

xviii

Figure 3.8. HPLC chromatograms before and after the incubation of Plectranthus barbatus

extract with: (a) gastric juice, (b) pancreatic juice. *Indicates the residue of pancreatin. For the

identification of the peak numbers, refer to Figure 3.4.

51

Figure 3.9. Variations in peak areas of compounds present in herbal tea after 4 h digestion with

artificial pancreatic juice. () Luteolin 7‐O‐glucuronide; () rosmarinic acid;

() apigenin 7‐O‐glucuronide; () hydrolysed abietane; (○) acacetin 7‐O‐glucuronide; (●)

abietane diterpenoid. 53

Figure 3.10. Chromatograms of compounds inside the Caco‐2 cells (homogenates) after 6 h

contact with (a) P. barbatus extract, (b) rosmarinic acid standard. 2, rosmarinic acid; *, residual

peak from the Caco‐2 cells. 55

Figure 3.11. HPLC chromatogram after the action of β‐glucuronidase from E. coli on the herbal

tea. Compounds identified with numbers 1–7, refer to Figure 3.6. Peaks signaled with arrows: 8,

luteolin; 9, apigenin; 10, acacetin. 56

Figure 3.12. Variations in peak areas of: (a) apigenin and (b) luteolin. Each figure shows the

metabolites after 6 h in the presence of Caco‐2 cells. Peak areas are presented as percentage of

the initial area. () aglycones and () glucuronides. 57

Figure 3.13. Decrease of taurine chloration in the presence of several concentrations of

P. barbatus extract (a and b) or standard rosmarinic acid (b). The concentration of P. barbatus is

expressed in µg.mL‐1 (a) or by its content in rosmarinic acid in µM (b). 61

Figure 4.1. HPLC analysis, 30 and 60 min after intraperitoneal administration of P. barbatus

extract, of (a) plasma and (b) brain. 1: luteolin 7‐O‐glucuronide (retention time 9.6 min); 2:

rosmarinic acid (RT: 10.4); 3: apigenin 7‐O‐glucuronide (RT: 11.2); 4: abietane diterpenoid (RT:

13.8); 5: acacetin 7‐O‐glucuronide (RT: 15.1); 2m: monomethylated rosmarinic acid; 1g: luteolin

glucuronide derivative; 3’: apigenin. 72

Figure 4.2. Permeation surfaces for (a) rosmarinic acid, (b) luteolin and (c) apigenin with

different concentrations of the other two components, built with the CCD experimental plan.

The relative errors are 0.930 (a), 0.718 (b), and 0.817 (c). 81

Figure 4.3. Glucuronidation surfaces for (a) luteolin and (b) apigenin with different

concentrations of rosmarinic acid and of the other flavonoid, built with the CCD experimental

plan. The relative errors are 0.981 (a), and 0.766 (b). 82

Figure 4.4. Distribution of (a) rosmarinic acid, (b) luteolin and (c) apigenin six hours after being

placed in the apical side of a Caco‐2 cell monolayer. The effects of co‐administration of digoxin

and benzoic acid were analysed. Statistical analysis: c ‐ different from the control (P<0.05); c* ‐

different from the control (P<0.1); s – different from the other substrate (P<0.05).

83

xix

Figure 4.5. Distribution of (a) benzoic acid and (b) digoxin six hours after being placed in the

apical side of a Caco‐2 cell monolayer. The effects of co‐administration of a standard mixture

(SM) with rosmarinic acid, luteolin and apigenin, 50µM each, was analysed. Statistical analysis:

* ‐ different from the control (P<0.05).

84

Figure 4.6. Glucuronidation of (a) luteolin and (b) apigenin six hours after being placed in the

apical side of a Caco‐2 cell monolayer. The effects of co‐administration of digoxin and benzoic

acid were analysed. Statistical analysis: c ‐ different from the control (P<0.05); c* ‐ different

from the control (P<0.1); s – different from the other substrate (P<0.05). 85

Figure 5.1. Molecular structure of (a) rosmarinic acid, (b) quercetin, (c) luteolin, and (d)

apigenin. 96

Figure 5.2. Fluorescence emission spectra of acetylcholinesterase with the addition of

P. barbatus aqueous extract. Arrow points to increasing concentrations of P. barbatus plant

extract, ranging 0; 0.5; 1; 5; 10; 33; 50; 100 µg.ml‐1. 97

Figure 5.3. Stern‐Volmer plot (a) and plot of log(F0‐F)/F vs. log[Q] (b) of acetylcholinesterase

with P. barbatus aqueous extract. [Q] is the concentration of P. barbatus extract in mg.ml‐1. 99

Figure 5.4. FTIR spectra of the acetylcholinesterase alone, with P. barbatus extract, rosmarinic

acid (RA), luteolin (Lut), apigenin (Api) and quercetin (Quer). The spectra obtained are shown (a)

in the form of absorbance spectra and (b) the second derivative spectra (negative peaks). 103

Figure 5.5. Percentage of decrease of hydrogen‐deuterium exchange rate of AChE in the

presence of P. barbatus extract, rosmarinic acid, luteolin, apigenin and quercetin. 104

Figure 5.6. Fluorescence emission spectra of (a) HSA and (b) lysozyme with the addition of

P. barbatus aqueous extract. Arrow points to increasing concentrations of P. barbatus plant

extract, ranging 0; 0.5; 0.75; 1; 2.5; 5; 7.5; 100 μg.ml−1. 111

Figure 5.7. Stern–Volmer plots of HSA and lysozyme with P. barbatus aqueous extract. [Q] is the

concentration of P. barbatus in μg.ml−1. 112

Figure 5.8. Plots of log([F0 − F]/F) vs. log[Q] for HSA and lysozyme with P. barbatus aqueous

extract. [Q] is the concentration of P. barbatus in g.l−1. 114

Figure 5.9. FTIR spectra of the proteins alone, with P. barbatus extract, rosmarinic acid (RA),

luteolin (Lut) and apigenin (Api). The absorbance spectra obtained for HSA are shown in (a) and

the second derivatives are in (c). The absorbance spectra obtained for lysozyme are shown in (b)

and the second derivatives are in (d). 117

Figure 5.10. Percentage of change in protein (HSA and lysozyme) hydrogen–deuterium

exchange rate, determined from the analysis of the amide II band, in the presence of

P. barbatus extract, rosmarinic acid, luteolin or apigenin, in comparison with the

hydrogen–deuterium exchange rate of the protein alone. 118

xx

Table List

Table 2.1. Concentrations of rosmarinic acid, luteolin and apigenin used in the phenolic

mixtures to determine the bioavailability by Caco‐2 cells, according to the CCD experimental

plan. 35

Table 3.1. Amount of dry plant extract obtained. 40

Table 3.2. Inhibition of AChE activity (%), antioxidant activity and rosmarinic acid content of

water extracts of the leaves of several Plectranthus species. 41

Table 3.3. Detected ions and attribution errors (ppm) for the collected fractions corresponding

to peaks 1, 3, 4, 5, 6 and 7. 49

Table 3.4. Antiacetylcholinesterase and antioxidant activity of P. barbatus herbal tea after in

vitro gastrointestinal digestion. The action of the pancreatic juice on the inhibition activity of

rosmarinic acid (RA) is also shown. 52

Table 4.1. Concentration of rosmarinic acid, its metabolites and flavonoid glucuronide

derivatives in the plasma and in the brain, 30 and 60 min after the intragastric and

intraperitoneal administration of P. barbatus extract. 69

Table 4.2. Brain acetylcholinesterase inhibition (%) 30 and 60 min after administration

(intragastric and intraperitoneal) of rosmarinic acid and P. barbatus extract. Results significantly

different from the control are marked with * (P < 0.05) and ** (P < 0.1). Values that are not

significantly different (P < 0.05) are marked from a to d. 70

Table 4.3. Retention time of compounds from the P. barbatus extract found in plasma and their

decrease from 30 to 60 min after the extract intraperitoneal administration. 74

Table 4.4. Permeation of the P. barbatus aqueous extract constituents through the Caco‐2 cell

monolayer. 79

Table 5.1. Stern‐Volmer binding parameters (KSV, Kq), binding equilibria parameters (Kb, n) and

thermodynamic parameters (ΔHo, ΔSo, ΔGo) for the binding of P. barbatus extract, rosmarinic

acid (RA), luteolin (Lut), apigenin (Api) and quercetin (Quer) to acetylcholinesterase (R2>0.99 to

all linear regressions). Rates of Amide II/Amide I variation, reflecting the rate of hydrogen

deuterium exchange in the presence of similar amount of RA, Lut, Api and Quer or 10mg/mL of

P. barbatus extract (for AChE without ligand, ‐2.345mAU.min‐1). IC50 values for the inhibition of

acetylcholinesterase activity by P. barbatus, RA, Lut and Api. (*) and Quer for P. barbatus

extract the values are expressed in l.mg‐1, l.mg‐1s‐1, and mg.ml‐1 but the molarity was estimated

based on the content of rosmarinic acid, luteolin and apigenin (1.1122 mmol.g‐1). 100

xxi

Table 5.2. Binding parameters (KSV, Kq, Kb, n) and thermodynamic parameters (ΔHo, ΔSo, ΔGo) for

the binding of P. barbatus extract, rosmarinic acid (RA), luteolin (Lut) and apigenin (Api) to HSA

and to lysozyme (R2>0.99 to all linear regressions). Rates of Amide II/Amide I variation in the

presence of similar amount of RA, Api and Lut, or 10mg/mL of P. barbatus extract (without

ligand, ‐0.716 for HSA and ‐1.815 for lysozyme). IC50 values for the inhibition of lysozyme

activity. For P. barbatus extract the values are expressed in l.mg‐1(a), l.mg‐1s‐1(b) and mg.l‐1(c).

113

xxii

Abbreviations

ABC transporters ATP‐binding cassette transporters

AChE acetylcholinesterase

AChI acetylthiocholine

AD Alzheimer’s disease

ANOVA Analysis of Variance

Api apigenin

APP amyloid precursor protein

Asp aspartic acid

ATP adenosine triphosphate

AU absorbance unit

Aβ β‐amyloid

BCRP breast cancer resistance protein

BHT butylated hydroxytoluene

CCD central composite design

COMT catechol o‐methyl transferase

DMEM Dulbecco's Modified Eagle Medium

DMSO dimethyl sulfoxide

DMSO‐d6 deuterated dimethyl sulfoxide

DPPH di(phenyl)‐(2,4,6‐trinitrophenyl)iminoazanium

DTNB 5,5'‐dithiobis‐(2‐nitrobenzoic acid)

DTT dithiothreitol

EDTA ethylenediamine tetraacetic acid

ESI‐MS electrospray ionization mass spectrometry

FBS fetal bovine serum

FDA Food and Drug Administration

FTIR Fourier transform infrared spectroscopy

Glu glutamic acid

h hour

HBSS Hank’s balanced salt solution

HEPES 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid

His histidine

xxiii

HPLC high precision liquid chromatography

HPLC‐DAD high precision liquid chromatography coupled with a diode array

detector

HSA human serum albumine

I(%) inhibition in percentage

IC50 concentration of inhibitor causing 50% inhibition

Kb binding constant

Kq scatter collision quenching constant

KRC Krebs‐Ringer solution with calcium chloride

KSV Stern‐Volmer quenching constant

Lut luteolin

MCT monocarboxylic acid transporter

min minute

MRP multidrug resistance protein

MS mass spectrometry

MTT 3‐(4,5‐Dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide

NADPH nicotinamide adenine dinucleotide phosphate

NMR nuclear magnetic resonance spectroscopy

PBS phosphate buffer saline

PDB Protein Data Bank

PET positron emission tomography

Pgp P‐glycoprotein

PMA phorbol myristate acetate

Quer quercetin

RA rosmarinic acid

ROS radical oxygen species

RT retention time

SAM S‐adenosyl methionine

Ser serine

TEER trans‐epithelial electric resistance

TFA trifluoroacetic acid

TMB 3,3’,5,5’‐Tetramethylbenzidine

TNB 2‐nitro‐5‐thiobenzoate

Tris tris(hydroxymethyl)aminomethane

xxiv

Trp tryptophan

Tyr tyrosine

UDPGA 5'‐diphospho‐glucuronic acid

UV ultraviolet

UV‐Vis ultraviolet‐visible

ΔGo free energy change

ΔHo enthalpy change

ΔSo entropy change

Chapter I

General Introduction

Chapter I

3

1. Literature review

1.1. Plectranthus species and their ethnobotanical uses

The genus Plectranthus L’Hér., belonging to the Mint family (Lameaceae), comprises

about 300 species widely distributed in the savannahs and forest regions of Africa, Asia and

Australia. The majority of the Plectranthus species can be described as tender shrubs or

groundcovers growing in the shade of large forests or in the partial shade of the forest edge.

A few species occur in drier regions with rocky soils and have succulent or semi-succulent leaves

and stems, which help the plants to survive in these habitats (Codd, 1975; van Jaarsveld, 2006).

The geographical distribution of Plectranthus is highly variable. Some species, as

P. laxiflorus, occur in smaller areas in the south-east of the African continent, while others, like

P. barbatus, may naturally occur in vast areas, in Africa, India and South America (Codd, 1975;

Lukhoba et al., 2006). The later species, endemic form India, was taken to Brazil probably during

the colonial period (Codd, 1975; Lukhoba et al., 2006; Lorenzi & Matos, 2002).

The value of Plectranthus species as garden plants has been long recognized. They have

ornamental leaves and flowers, are easy and fast growing plants, resistant to most pests and

diseases, and make colourful displays in autumn when other flowers are often scarce. Several

Plectranthus species were introduced in Europe during the expansion of the British Empire and

the founding of the Royal Botanic Gardens of Kew. In the nineteenth century some species were

very popular in Scandinavia and commonly planted in window boxes and hanging baskets, as

P. oertendahii, nowadays known as “Swedish ivy”, and other Plectranthus groundcover species

(P. madagascariensis, P. verticillatus, P. strigosus, etc.) Some other species are tall shrubs that

can be pruned and form colourful hedges (P. ecklonii, P. fruticosus, P. barbatus, etc.).

A great variety of ethnobotanical uses are reported for Plectranthus species. Some, due

to their aromatic nature, are used as culinary plants to flavour food, or as insect repellents

(P. ornatus). Others have tubers with high starch content and are eaten as vegetables

(P. esculentus, P. rotundifolius). However, the most common uses are to treat a wide range of

diseases (Lukhoba et al., 2006). Infusions of several Plectranthus species are used to treat

coughs and colds (P. hadiensis, P. laxiflorus), liver complaints (P. hereroensis), and ailments in

respiratory system and skin (P. madagascariensis).

Among the Plectranthus species used in traditional medicines P. barbatus is one of the

most important (Luckhoba et al., 2006). P. barbatus (Figure 1.1) is used in Hindu and Ayurveda

treataments, and in traditional medicine from Brazil, tropical Africa and China (Alabashi &

Melzig, 2010a). Extensive reviews have been written about the ethnobotanical importance of

P. barbatus (Alabashi & Melzig 2010a, 2010b); Lukhoba et al., 2006). The leaves from P. barbatus


4

are usually prepared as a decoction or infusion and used in the treatment of pain from different

aetiologies, inflammation, infections, colds and coughs (Lukhoba et al., 2006), which suggest

anti‐inflammatory and antioxidant properties. P. barbatus is also used to treat psychiatric

disorders in Tanzania (Lukhoba et al., 2006).

Figure 1.1. Plectranthus barbatus

Other significant uses for P. barbatus reported by Lukhoba et al. (2006) and Alabashi &

Melzig (2010a) are as food, the leaves being cooked as vegetable in Kenya and Yemen, and as

ornamental plant or garden herb, in hedges, fences or boundary markers.

1.2. Herbal tea components and their bioavailability

The main components of most medicinal teas, used traditionally by people, are phenolic

acids and flavonoids (Cai et al., 2004). The flavonoids in herbal teas are usually present in

glycosylated forms, which have higher solubility in water, although some flavonoid aglycones

may also occur in aqueous extracts (Cai et al., 2004). As herbal teas are usually taken orally,

many studies have focused on the metabolism and bioavailability of their active components.

1.2.1. Phenolic acids

Phenolic acids are a class of polyphenols commonly found in herbal medicines and plant‐

derived food (Lafay and Gil‐Izquierdo, 2008). The most commonly found phenolic acids, shown

in Figure 1.2, are gallic acid, a component of hydrolysable tannins, and hydroxycinnamic acids, as

cinnamic acid, caffeic acid (3,4‐dihydroxycinnamic acid), coumaric acid (4‐hydroxycinnamic acid),

ferulic acid (3‐methoxy‐4‐hydroxycinnamic acid) and the ester derivatives of caffeic acid, the

chlorogenic acid (3‐caffeoylquinic acid) and the rosmarinic acid.

Chapter I

5

Figure 1.2. Chemical structures of some common phenolic acids.

When phenolic acids are ingested their bioavailability is dependent on the permeability

through the intestinal barrier, and on the conjugation (glucuronidation, sulfation or methylation)

by intestinal and liver cells. Bioavailability studies using rats as model animals show that

hydroxycinnamic acids such as coumaric acid and caffeic acid can be found in plasma after

intragastric administration, in unconjugated, sulfated and/or glucuronidated forms (Konishi et

al., 2004; Konishi et al., 2005). Both coumaric and caffeic acids presented higher bioavailability

than gallic acid, which is a simpler phenolic acid (Konishi et al., 2004 & 2005). Ferulic acid, like

coumaric acid showed a higher bioavailability than caffeic acid (Konishi et al., 2006). The caffeic

acid esters rosmarinic acid and chlorogenic acid showed much lower bioavailability than simpler

hydroxycinnamic acids and similar to gallic acid (Konishi et al., 2005 & 2006). Some authors

could not detect chlorogenic acid circulating in rat plasma after its administration, but found

traces of caffeic acid and ferulic acid conjugates 6 hours after administration, suggesting that

hydrolysis of chlorogenic acid might have occurred and the resulting metabolites might be more

easily available (Azuma et al., 2000).

Many studies have focused on the bioavailability of rosmarinic acid, in the rats and

humans, as it has a number of interesting biological activities such as antioxidant, antiviral,

antibacterial and anti‐inflammatory activities that can be potentially important for public health

(Peterssen and Simmonds, 2003). When rosmarinic acid was administered to rats, intact

rosmarinic acid, monomethyl‐rosmarinic acid, and coumaric acid were found in plasma, mostly

in conjugated forms (Baba et al., 2004). In humans the same compounds plus caffeic acid and

ferulic acid were found circulating in plama, also in conjugated forms, after the intake of a

rosmarinic acid‐rich Perilla extract (Baba et al., 2005). In both models the maximum amount of


6

free rosmarinic acid in plasma was reached around 0.5h after administration, while conjugated

forms of rosmarinic acid had maximum peaks later, from 40 mins to 2h after administration

(Baba et al., 2004 & 2005; Konishi et al., 2005). Eighty three per cent of the ingested rosmarinic

acid was excreted between 8 and 18 hours after intake in rats (Baba et al., 2004), and 75%

within the first 6h in humans (Baba et al., 2005), suggesting that rosmarinic acid metabolites

may circulate for a relatively long time before being excreted and repeated intakes may be

favourable to the accumulation of these active compounds in the bloodstream.

The bioavailability results obtained with Caco‐2 cell monolayers are in agreement with

the in vivo results obtained in the rat model for the order of bioavailability of the phenolic acids,

which is gallic acid = rosmarinic acid = chlorogenic acid < caffeic acid < p‐coumaric acid (Konishi

et al., 2005), which suggests that the same kind of transport mechanisms for phenolic acids may

be present in both models. The high permeation of caffeic acid, p‐coumaric acid and ferulic acid

in Caco‐2 cells, and presumably in rat intestine, seems to be due to the fact that these

compounds are substrates of the monocarboxylic acid transporter (MCT) (Konishi and

Kobayashi, 2004a & 2004b). Rosmarinic acid and chlorogenic acid, which showed a very low

transepithelial permeation, do not seem to be substrates of MCT, being transported by

paracellular diffusion (Konishi and Kobayashi 2004a; Konishi and Kobayashi, 2005).

Compounds that were reported as metabolites of caffeic acid by gut microflora such as

m‐coumaric acid and m‐hydroxyphenylpropionic acid are also substrates of the MCT and showed

a higher influx than caffeic acid (Konishi et al., 2004b), suggesting that the action of gut

microflora may increase the bioavailability of these active compounds.

1.2.2. Flavonoids

Flavonoids are a group of polyphenolic compounds that share the same structural

features, namely a C6‐C3‐C6 carbon framework (Figure 1.3) or, more specifically, a

phenylbenzopyran functionality. Depending on the position of linkage of the aromatic ring to the

benzopyrano moiety, this group can be divided in three classes: flavonoids (2‐

phenylbenzopyrans) as shown in Figure 1.2; isoflavonoids (3‐phenylbenzopyrans) as the

isoflavone in Figure 1.3; and neoflavonoids (4‐phenylbenzopyrans) the rarest group of flavonoids

(not shown). In some flavonoids the C3 moiety between the two aromatic rings may not form a

heterocyclic ring (ring C) in their structure, such as the chalcones (Figure 1.3) (Marais et al.,

2006). Depending on the degree of saturation and oxidation of the C‐ring, some sub‐groups may

be considered, the most common structures can be seen in Figure 1.3 (Marais et al., 2006). The

pattern of hydroxylation of flavonoids is determinant for their identification and properties, as in

Chapter I

7

the case of quercetin, apigenin and luteolin (Figure 1.4), which are among the most commonly

occurring flavonoids.

Figure 1.3. Basic chemical structures of the main classes of flavonoids.

Figure 1.4. Structures of (a) quercetin, (b) apigenin and (c) luteolin.

Flavonoids often occur in the form of flavonoid glycosides, in which the sugar residue is

bound to the flavonoid structure by one of the hydroxyl groups. The most common

monosaccharide residues found in flavonoid glycosides are glucose, rhamnose and glucuronic

acid.

Ingested flavonoids may undergo several changes before reaching target organs where

they may have beneficial effects. Studies with quercetin, which is often used as a model of

flavonoid, showed the presence of glucuronidated, sulfated and methylated forms of quercetin

OH

O

O

OH

OH

OH

OH

OH

O

O

OH

OH

OH

OH

O

O

OH

OH

a

b

c


8

in rat plasma after its intragastrical administration to rats (Morand et al., 1998; da Silva et al.,

1998; Justino et al., 2004). Although the bioavailability of quercetin is low, the antioxidant

activity in rat plasma after its administration increased significantly (da Silva et al., 1998; Justin

et al., 2004). Liver sulfotransferases and glucuronyltranferases seem to be highly responsible for

the metabolization of flavonoids, as in vitro studies showed that liver enzymatic extracts are able

to conjugate quercetin into sulfated and glucuronidated metabolites (da Silva et al., 1998;

Justino et al., 2004). Studies with liver cell lines, such as HepG2, also showed that liver cells may

conjugate quercetin through sulfation, glucuronidation, and methylation reactions (O’Leary et

al., 2003), and the multidrug resistant protein 2 (MRP2) may promote the efflux of quercetin

conjugates from the cells (O’Leary et al., 2003). Although quercetin is mainly found circulating in

plasma in the conjugated form, one of the characteristic features of quercetin conjugates is a

slow elimination, with reported half‐times from 11 to 28 hours, favouring its accumulation in

plasma after repeated intakes (Manech et al., 2005).

Several studies on the bioavailability and metabolism of flavonoids by the intestine have

been performed using Caco‐2 cell monolayers as models for the intestinal barrier. Flavonoid

aglycones have shown a high permeability through these membranes, which is due to active

transport (Walgren et al., 1998; Walle et al., 1999). As flavonoids in food are present mainly in

glycosylated form, some studies suggest that the gut microflora may increase their

bioavailability by hydrolysing into flavonoid aglycones (Liu & Hu, 2002; Kobayashi et al., 2008).

Studies of bioavailability of hesperetin (aglycone) and hesperidin (hesperetin glycoside) showed

that the transepithelial transport of the aglycone is mainly active transport, while the glycoside

passes through Caco‐2 cell monolayers mainly by passive diffusion (Kobaiashi et al., 2008), which

is in good agreement with results obtained by Walgren et al. (2008) for quercetin and its

glycosides. Transporters from the MCT family seem to be related to the transepithelial transport,

across the Caco‐2 monolayers, of flavonoids such as hesperetin, naringenin, erydictiol and

epicatechins (Kobayashi & Konishi, 2008; Vaidyanathan and Walle, 2003). Several reports show

that flavonoids may be conjugated – preferentially glucuronidated, but also sulfated – by Caco‐2

cells, and transported back to the culture medium by transporters of the type multidrug

resistance pump (MRP) after conjugation (Ng et al., 2005; Walle et al., 1999).

The relationship between flavonoids and intestinal membrane transporters is highly

complex. Flavonoid aglycons may be transported to the bloodstream by some active

transporters (Walle et al., 1999). Some flavonoids may be substrate to efflux transporters such

as P‐glycoprotein (Pgp) and MRP and may return to gut lumen (Wang et al., 2009), while others

may act as inhibitors of Pgp and MRP, without being their substrates (Brand et al., 2006).

Chapter I

9

Although these complex relationships are still under study, it seems that the flavonoid

bioavailability depends on the balance of active influx transport, conjugation, efflux transport

and inhibition of the active transporters by other compounds.

1.3. Biological activities of herbal teas and their components

Many therapeutic properties have been reported for flavonoids and phenolic acids,

especially as enzyme inhibitors and antioxidants. The use of herbal teas for medicinal purposes

usually gives an indication of these activities.

The present thesis focuses on the anti‐acetylcholinesterase, anti‐inflammatory and

antioxidant activities, which can be useful in the treatment of several health disorders, such as

gastrointestinal disturbances, or the symptomatic treatment of Alzheimer’s disease.

1.3.1. Acetylcholinesterase inhibitors

Acetylcholine is usually found in the synaptic clefts in peripheral and central nervous

systems and has a neurotransmitter function. Acetylcholine is synthesized by neurons, released

to the synaptic cleft upon stimulation, and binds to acetylcholine receptors on the other end of

the synaptic cleft.

Acetylcholinesterase is an abundant enzyme in the synaptic cleft as it hydrolyses

acetylcholine into the inactive metabolites choline and acetate, clearing acetylcholine from the

synapse and ceasing the stimulation of the post‐synaptic receptors (Randall et al., 2000). The

active site of acetylcholinesterase is located in an active gorge (Figure 1.5) where some groups of

amino acid residues play an important role by positioning the acetylcholine molecule, the

peripheral binding site (Trp286, Tyr72, Tyr124 and Asp74), the choline binding site (Trp86,

Glu202, Tyr337) and the acyl pocket (Phe295, Phe297). The hydrolysis of acetylcholine is then

directly catalysed by the catalytic triad (Glu334, His447, Ser203), located in the bottom of the

active gorge (Abu‐Donia, 2003).

Acetylcholinesterase inhibitors can be used for the treatment of some dysfunctions, such

as Alzheimer’s disease and gastrointestinal disturbances, as will be discussed in the next

sections. These drugs, for instance tacrine, often bind to the active gorge with non‐covalent

bonds usually to the acetylcholine binding and positioning sites and to the catalytic site, as it is

shown in Figure 1.6 (Harel et al., 1993). Acetylcholinesterase inhibitors found by chemical

synthesis, as carbamates and organophosphates, are irreversible inhibitors binding to

acetylcholinesterase by ionic or covalent bonds, which makes them highly toxic and suitable to

be used as pesticides (Abu‐Donia, 2003; Eyer et al. 2007).


10

Figure 1.5. Active gorge of acetylcholinesterase (Abu‐Donia, 2003).

Galanthamine, an acetylcholinesterase inhibitor approved by the Food and Drug

Administration (FDA) to be used in the treatment of Alzheimer’s disease, is an alkaloid first

discovered in Galanthus species, and nowadays obtained for commercialization from Narcissus

spp. or synthetically (Heinrich & Teoh, 2004). The successful application of galanthamine led to

the search of new acetylcholinesterase inhibitors, more effective and with less side effects, in

plants used in traditional medicine (Houghton et al., 2004, Adsersen et al., 2006, Vinutha et al.,

2007, Ferreira et al., 2006).

1.3.1.1. Acetylcholinesterase inhibitors and Alzheimer’s disease

Alzheimer’s disease (AD) is the most common form of dementia among older people,

and its occurrence is of great concern in occidental populations where the life expectancy

increased in the 20th century.

AD is a degenerative terminal disease characterized by a progressive memory loss and

bodily functions, and symptoms generally include confusion, irritability, aggression, mood

swings, language breakdown, long term memory loss, and decline of the senses, which leads to

the withdrawal of the patients (Alzheimer’s Association, 2006).

On a biochemical point of view, AD is a protein misfolding disease, as it results from the

accumulation of abnormally folded amyloid beta peptides (Aβ). The Aβ may accumulate in the

cell within the neuronal endoplasmatic reticulum, and outside the cell forming senile plaques in

the brain of AD patients (Hashimoto et al., 2003).

Chapter I

11

Figure 1.6. Tacrine binding to the active gorge of acetylcholinesterase, and details showing the amino acid

residues involved in the interaction between the two molecules (PDB 1ACJ, Harel et al., 1993).

Aβ is a short peptide that results from the proteolytic cleavage by β‐secretase and

ϒ‐secretase of the amyloid precursor protein (APP) (Figure 1.7), a transmembrane protein whose

function is still unknown but seems to be involved in the early development of the nervous

system (Kerr & Small, 2005). Although Aβ are usually water soluble innocuous peptides with

short regions of β‐sheet and predominately α‐helix secondary structures in solution, they

undergo dramatic conformational changes at high concentrations to a form rich in β‐sheet


12

secondary structure that aggregates to form amyloid fibrils (Onishi & Takano, 2004). The

extracellular deposition of these water insoluble fibrils into senile plaques is characteristic of AD.

As the senile plaques are located extracellularly, they may not interfere directly with neuronal

metabolism, but may hinder cellular exchanges with the extracellular environment (Tanasalli et

al., 2006) and interact with glia cells, activating the release of pro‐inflammatory signals (Figure

1.7) (Stuchbury & Munch, 2005). A resulting inflammatory response and neurodegeneration is

the cause of the dementia and other symptoms felt by AD patients (Stuchbury & Munch, 2005).

Figure 1.7. Biochemical pathways associated with the formation of amyloid plaques and neurofibrillary

tangles in Alzheimers’s Disease patients. (www.calbiochem.com/alzheimers).

Due to an abnormal aggregation of tau protein, AD is also known as a tauopaty. Tau

protein is a microtubule associated protein that stabilizes neuronal cytoskeleton, being

Chapter I

13

regulated by phosphorylation. In AD patients tau protein is hyperphosphorylated and

accumulates as paired helical filaments, which aggregates into masses known as neurofibrillary

tangles inside the nerve cell bodies, also associated with the amyloid plaques (Figure 1.7)

(Stuchbury & Munch, 2005). Although some studies approach the treatment of AD by

modulating the activity of ϒ‐secretase, the enzyme mainly responsible for Aβ formation, the

most studied approach is by the administration of inhibitors of acetylcholinesterase activity

(Salawu et al., 2011).

Patients of AD show low concentration of acetylcholine in the brain as a consequence of

the cellular dysfunctions caused by the abnormal protein aggregations. The administration of

acetylcholinesterase inhibitors has proved to be effective in the symptomatic treatment of AD

patients (Rauf et al., 2002). This strategy to increase the acetylcholine levels in the brain is

presently the most commonly used, and the drugs that have been approved by the Food and

Drug Administration (FDA) to treat AD in the US are the acetylcholinesterase inhibitors tacrine,

rivastigmine, donepezil and galanthamine, which have all been successful in slowing down the

neurodegenerative process in AD patients (McGleenon et al., 1999; Heinrich & Teoh, 2004).

The major problem related with acetylcholinesterase inhibition treatments of AD is the

bioavailability of the inhibitors, as they must reach the brain passing through the blood‐brain

barrier to inhibit the brain acetylcholinesterase. If the inhibitors are too potent, or very high

concentrations are needed for the therapeutic effect, unwanted side effects may arise, such as

gastrointestinal and hepatic disturbances (McGleenon et al., 1999, Heinrich & Teoh, 2004). The

research for new acetylcholinesterase inhibitors for the treatment of AD continues, with the aim

of finding reversible inhibitors with higher specificity to brain acetylcholinesterase that may

cause less side effects than the currently used acetylcholinesterase inhibitors.

1.3.1.2. Acetylcholinesterase inhibitors to treat gastrointestinal disorders

The intestinal wall consists mainly of layers of muscle, which contract and relax in a

coordinated fashion, propelling food through the intestine to the anus. This complex pattern of

motility is coordinated by excitatory and inhibitory pathways of the enteric nervous system, to

which acetylcholine is the major excitatory neurotransmitter responsible for the peristaltic

contractions (Holzer and Maggi, 1994). Acetylcholine in gut epithelial cells is also responsible for

controlling ion transport, and therefore, water secretion for gut hydration. This process, very

important in establishing a proper aqueous environment for the enzymatic digestion and

absorption of nutrients, also provides surface lubrication to propel intestinal contents by

peristaltic movements (Hirota and McKay, 2006).


14

Acetylcholinesterase inhibition within the enteric nervous system prevents the

degradation of acetylcholine, increasing gastrointestinal motility (Jarvie et al., 2008). The

acetylcholinesterase inhibitor neostigmine has been used to treat conditions related to

impairment of gastrointestinal motility, such as colonic pseudo‐obstruction (Ponec et al., 1999)

and post‐operative impairment after colorectal surgery (Kreis et al., 2001). Other inhibitors have

also been used to treat gastric motility dysfunctions, such as metochlopramide and vinitidine

(Sasha et al., 1995). Conditions that may be associated with disturbances of gastrointestinal

motility and treated with acetylcholinesterase inhibitors include dysphagia, gastric stasis,

achalasia, abdominal pain, paralytic ileus, vomiting and constipation (Sasha et al., 1995).

However the side effects of acetylcholinesterase inhibitors as discussed in section 3.2. may be an

inconvenient of these therapies, being nausea and diarrhoea the most commonly associated

with neostigmine (Jarvie et al., 2008).

Some plant extracts are empirically used to treat gastrointestinal disorders, such as

extracts from Plectranthus species, and specifically, P. barbatus (Lukhoba et al., 2006). This

therapeutic activity may be related to the inhibition of acetylcholinesterase activity by the plant

extract components, which may be a potential alternative for the currently used medicines.

1.3.1.3. Finding acetylcholinesterase inhibitors ‐ from in vitro to in vivo studies

The research of acetylcholinesterase inhibitors led to the development of methods to

quantify acetylcholinesterase activity. The direct detection of acetylcholine (substrate) or choline

(product) is not easy. Therefore several indirect methods were developed, as the titration of the

acetic acid formed by the hydrolysis of acetylcholine (Jacobson et al., 1957), the use of

acetylcholine analogues as substrates, such as acetylthiocholine (Ellman et al., 1961) or

radiolabelled acetylcholine (Johnson and Rumel, 1975), or by hydrolysing the choline

enzymatically with choline oxidase, the H2O2 formed in the hydrolysis, with the addition of

luminol and peroxidase, can be detected by chemoluminescence (Birman et al., 1985).

The most commonly used in vitro assay was the one described by Ellman et al. (1961), in

which acetylthiocholine is used as substrate for acetylcholinesterase, forming acetate and

thiocholine. The thiocholine reacts with 5,5’‐dithio‐bis‐(2‐nitrobenzoic acid) (DTNB), in a very

fast chemical reaction, originating 2‐nitrobenzoate‐5‐mercaptothiocholine and

5‐thio‐2‐nitrobenzoate (TNB), which has a peak of light absorbance at 405nm (Figure 1.8)

(Ellman et al., 1961). This method is very used due to its sensitivity, fastness and accuracy, which

is mostly due to the fast reaction between thiocholine and DTNB, however it has some

limitations related with inhibitors to be tested (Sinko et al., 2007). Some compounds may react

Chapter I

15

directly with DTNB, such as thiols (Ellman, 1958), or may hydrolyse chemically acetylthiocholine,

such as oximes (Sinko et al., 2007), and in both cases there is an increase in the rate of TNB

formation that is not related with acetylcholinesterase activity.

Figure 1.8. Reaction associated with the Ellman assay to quantify acetylcholinesterase activity. The final

product TNB can be spectrofotometrically detected due to its absorption at a wavelength of 405 nm.

(adapted from Frasco et al., 2005)

In spite of its limitations, the Ellman method is very versatile and can be adapted in

order to determine the effect of acetylcholinesterase inhibitors by measuring cortical

acetylcholinesterase activity after administration (Chattipakorn et al., 2007). However, this is an

ex vivo technique that involves the sacrifice of animals, and therefore must be justified by

preliminary studies.

Acetylcholinesterase activity can also be measured in vivo by positron emission

tomography (PET), as described by Iyo and co‐workers (1997), who compared

acetylcholinesterase activity in brains of healthy controls and in brains of Alzheimer’s disease

patients. This method was also used to assess the efficiency of acetylcholinesterase inhibitors in

the brains of Alzheimer disease patients, such as donepezil (Bohnen et al., 2005) and

galanthamine (Kadir et al., 2008). However, this technique involves the administration of

Acetylthiocholine

Acetylcholinesterase

H2O

ThiocholineAcetate

5,5’‐Dithio‐bis(2‐nitrobenzoic acid)

(DTNB)

5‐thio‐2‐nitrobenzoic acid (TNB)

(detection at 405nm)


16

radiolabelled acetylcholine analogues to humans, and so it is considered just for potentially

active compounds that have been selected by in vitro and animal testing.

1.3.2. Inflammation and Antioxidants

Inflammation is a complex biological response to harmful stimuli involving the vascular

system, the immune system and the injured cells, in a highly coordinated process involving

multiple factors acting in a complex network as stimulators or inhibitors. Upon a stimulus

(infection) endogenous mediators are released, such as cytokines and chemokines, that

contribute to the recruitment of circulating leukocytes to the inflammation site (Gouwy et al.,

2005). These cells, highly specialized in phagocytosis, have developed mechanisms for

intracellular digestion of particles, such as pathogens and cell debris, involving the production of

radical oxygen species (ROS) and a range of hydrolytic and proteolytic enzymes.

Phagocytosis in leucocytes is followed by a sharp but transient increase in oxygen

uptake, which is used to produce O2‐ (superoxide ion) by the one‐electron reduction of oxygen, a

reaction catalysed by NADPH oxidase at the expense of NADPH. Most of the superoxide reacts

with itself forming H2O2 (hydrogen peroxide), and from these agents a large number of highly

reactive oxidants are formed, including HOCl (hypochlorous acid), which is produced by the

myeloperoxidase‐ catalyzed oxidation of Cl‐ by H2O2; OH. (hydroxyl radical), produced by the

reduction of H2O2 by Fe2+ or Cu+; ONOO‐ (peroxynitrite), formed by the reaction between O2

‐ and

NO‐; and many others. (Babior, 2000; Klebanoff, 2005) This battery of ROS not only kill the

invading particles but it also may damage nearby tissues. Although inflammation has primarily a

protective function, the destructive effects can largely surpass the gravity of the stimulus. Severe

inflammatory responses are often associated with a large number of diseases such as

emphysema, acute respiratory distress syndrome, atherosclerosis, reperfusion injury,

malignancy and rheumatoid arthritis (Babior, 2000).

Lysozyme, one of the lytic enzymes produced by the neutrophils, is a glycoside hydrolase

whose main function is to hydrolyse peptidoglycans in bacterial cell walls, especially in Gram

positive bacteria (Laible & Germaine, 1985). During inflammation lysozyme is discharged from

lysosomes of neutrophils to destroy the phagosomes, however it also may damage the animal

tissue itself, increasing the inflammation process. Therefore, excessive lysozyme activity is

known to be related to allergic conditions and violent inflammatory responses of the immune

system against pathogens (Makino et al., 2003; Ronca et al., 1998; Wu et al., 2006).

Inflammation is also involved in Alzheimer’s disease, with ROS acting as secondary

messengers (Stuchbury and Munch, 2005). Microglia are a type of glial cells that are responsible

Chapter I

17

for the immune defence in the central nervous system, acting as macrophages in the brain and

spinal chord, while astroglia, another kind of glial cells, provide biochemical support and have an

important role in repair and scarring processes of brain and spinal chord after traumatic injuries.

The β‐amyloid plaques in Alzheimer disease patients lead to the rapid activation of glial cells and

an over‐expression of glial mediators, namely free radicals (ROS) and citokines (Dickson et al.,

1993; Meda et al., 1995, 2001). The propagation of the consequent oxidative stress and

inflammation by the cytotoxic activation of glial cells plays a key‐role in the pathogenesis and

progressive degeneration characteristic of Alzheimer’s disease (Figure 1.7) (Meda et al., 2001).

The use of antioxidants to treat Alzheimer’s disease patients is still a controversial issue.

Many studies focused on the treatment of Alzheimer’s disease with antioxidant compounds that

may act through different mechanism, from free radical scavengers (citoplasmatic and

mitochondrial) and metal chelators to anti‐inflammatory agents and transcriptional activators

(Figure 1.9). Although the results obtained from animal models were usually promising, when

the compounds were tested in human clinical trials the results often did not present a significant

difference from the control groups (Dumond and Beal, 2011; Mecocci and Polidori, 2011).

Several factors were proposed to explain this disparity, being the bioavailability of the

compounds one of the most important, namely in the absorption, transport, distribution and

retention in the target area of the human body (Dumont and Bael, 2011). Other important

factors include the reaction kinetics, as the free radicals must be neutralized faster than the

damage they cause in the target tissue, and the mechanism of action itself (Mecocci and

Polidori, 2011; Viña et al., 2004; Dumont and Bael, 2011). Patients treated with vitamin E, for

instance, showed an improvement in cognitive function when a variation of antioxidant function

was detected, suggesting that the effectiveness of vitamin E over Alzheimer disease patients was

dependent on its function as antioxidant, which was highly variable among the patients (Viña et

al., 2004). Therefore it is suggested that a higher efficacy can be achieved by administering

several molecules acting with different antioxidant mechanisms, and monitoring the antioxidant

status of the patients (Mecocci and Polidori, 2011). However, many clinical trials of antioxidant

compounds are still undergoing, and most of the compounds known to be effective in in vitro

and in vivo systems are still untested in clinical trials (Dumont and Bael, 2011; Mecocci and

Polidori, 2011).


18

Figure 1.9. Antioxidant strategies in Alzheimer’s disease. Solid arrows represent the mechanisms of the

disease and dashed arrows represent the mechanisms of antioxidant therapy (Adapted from Dumond and

Bael (2011)).

Plant‐derived natural products have been and will continue to be extremely important to

mankind as sources of medicinal drugs. The continued interest of pharmaceutical industry in

plant‐derived drugs led to the screening of species used in traditional medicines to treat illnesses

and/or promote health. Therefore it is not surprising Cai et al. (2004) studied the antioxidant

activity of extracts of 112 plant species used in Chinese traditional medicine.

Chapter I

19

2. Thesis overview

Herbal teas are considered functional drinks, since almost all of the chemical

components of these aqueous extracts have biological functions. The ethnobotanical uses of

these drinks can often be explained through the biochemical activities found for the complete

extracts or for their isolated components. Despite of the great popularity of P. barbatus tea in

Brazil, Africa and India to treat a wide range of diseases, the present knowledge on its biological

activities is still very limited. The experimental work developed and presented in this thesis is a

contribution to the validation of the ethnobotanical medicinal uses.

The main objective of this thesis is to evaluate the potential of P. barbatus herbal tea in

therapies related with acetylcholinesterase inhibition, antioxidant and anti‐inflammatory

activities. Briefly, to approach the main objective, this thesis aimed to:

Provide information about the acetylcholinesterase inhibition, antioxidant and

anti‐inflammatory activities of P. barbatus herbal tea.

Study the bioavailability of the P. barbatus extract, and whether the active

components, or metabolites, can reach the target organs.

Analyse the interaction between the plant extract components and proteins to

provide information on some of the mechanisms involved in the extract’s

bioavailability and activities.

Contribute to the evaluation of the potential of P. barbatus herbal tea for the

symptomatic treatment of Alzheimer’s disease and gastrointestinal disorders.

This thesis is organized in six chapters.

Chapter I is a general introduction that compiles the existing information about the

ethnobotanic uses of extracts of Plectranthus species, especially the most used for medicinal

purposes, P. barbatus. The most common active components of aqueous extracts (herbal teas)

are reviewed especially regarding the current knowledge of their metabolism and bioavailability

when ingested. As the aim of this thesis is the potential of using the P. barbatus herbal tea as

acetylcholinesterase inhibitor, antioxidant and anti‐inflammatory, a brief review of the current

knowledge about these activities and of their most common therapeutic uses is presented.

Chapter II describes the general methodologies used throughout the practical work.

Chapter III reports in vitro studies of the activities of P. barbatus herbal tea. A screening

of antioxidant and acetylcholinesterase inhibition activities for several Plectranthus species is

presented. The aim of this screening was to prove that P. barbatus, the most used species for


20

medicinal treatments, was the most interesting in terms of its composition and activity. This

chapter also describes the in vitro digestion P. barbatus herbal tea with artificial gastric and

pancreatic juices, β‐glucuronidase from gut bacteria, and with Caco‐2 cells, modelling the

metabolism by intestinal cells, with the objective to know whether the extract components were

metabolized, and if the remaining acetylcholinesterase activity has a value that justifies carrying

out in vivo studies in rats. In this chapter the anti‐inflammatory activity of the P. barbatus extract

is also evaluated by measuring its ability to decrease the amount of hypochlorous acid produced

by activated neutrophils.

Chapter IV explores the bioavailability of the P. barbatus herbal tea in order to

investigate if, when administered to rats, the active compounds present in this aqueous extract

or their metabolites occur in the bloodstream and in the brain, and if the neuronal

acetylcholinesterase activity is affected by the administration of the P. barbatus extract. The

bioavailability of the herbal tea is also studied in Caco‐2 cell monolayers, to determine if the

extract components interfere with the permeability and metabolisation of each other.

Substrates of the transport systems MCT and Pgp are co‐administered in order to know the

involvement of these transporters in the bioavailability of the plant extract compounds.

Chapter V describes studies on the interactions between the P. barbatus herbal tea

components and proteins, using fluorescence and FTIR spectroscopic methods. The relationship

between the interactions and the inhibition of the enzymatic activity is also discussed. The

interactions between the plant extract and acetylcholinesterase are studied in order to elucidate

the mechanism by which the enzyme is inhibited by the herbal tea components. In addition, the

binding of the P. barbatus extract to the human plasma proteins albumin and lysozyme is also

analysed, to know if the extract components may be transported in the bloodstream bound to

these transport proteins, increasing their bioavailability in target organs. The influence of the

P. barbatus herbal tea on lysozyme activity is also evaluated with the objective of knowing if the

plant extract could be useful in alleviating inflammation and allergic conditions by inhibiting

lysozyme hydrolytic activity.

Chapter VI presents a global discussion and summarizes the main conclusions of this

thesis.

Chapter II

Materials and Methods

Chapter II

23

1. Plant material

The leaves of five Plectranthus species (P. barbatus, P. ecklonii, P. fruticosus,

P. lanuginosus and P. verticillatus) cultivated in the Botanic Garden of the University of Lisbon

were collected during spring (March–June) 2006 to prepare the extracts for the studies in the

section 1 of Chapter III and in September 2008 for all the other studies.

Vouchers specimens from each species have been deposited in the Herbarium of this

Botanic Garden. P. barbatus Andr. (LISU: 214625), P. ecklonii Benth. (LISU: 146895),

P. fruticosus L’Herit (LISU: 214627), P. lanuginosus (Benth.) Agnew. (LISU: 177258),

P. verticillatus (L.f.) Druce (LISU: 171088).

2. Animals

All experiments were carried out in accordance with the guidelines of the European

Communities Council Directive of 24th November 1986 (86/609/ECC). Adult male

Sprague–Dawley rats (3–4 months old) were obtained from Instituto de Investigação Cientifica

Bento da Rocha Cabral (Lisbon, Portugal). Two rats per cage were maintained in a room at 22 oC

under 12 h dark/ light cycling and ad libitum access to water and regular chow.

3. Chemicals

All chemicals were of analytical grade. Acetylcholinesterase (AChE) type VI-S, from

electric eel 349 U/mg solid, 411 U/mg protein, 5,50-dithiobis[2-nitrobenzoic acid] (DTNB),

acetylthiocholine iodide (AChI), tris[hydroxymethyl]aminomethane (Tris buffer),

dimethylsulphoxide (DMSO), 2,2-diphenyl-1-picrylhydrazyl (DPPH), linoleic acid, β-carotene,

2,6-di-tert-butyl-4-hydroxytoluene (BHT), Tween 40, β-glucuronidase type IX-A from Escherichia

coli, HEPES buffer, pancreatin, pepsin, thiazolyl tetrazolium bromide (MTT), polyethylenoglycol,

sulfatase from Helix pomatia, cathecol o-methyltransferase (COMT) from porcine liver,

S-adenosylmethionine (SAM), L-cystein non-animal source, human serum albumin (HSA),

lysozyme from human milk (EC 3.2.1.17), lyophilized Micrococcus lysodeiktcus quercetin,

luteolin, apigenin, caffeic acid and rosmarinic acid were obtained from Sigma. Trifluoroacetic

acid was acquired from Merck. DMEM (Dulbecco’s Modified Eagle Medium), HBSS (Hank’s

Balanced Salt Solution), glutamine, Pen–Strep (penicillin–streptomycin), PBS (phosphate

buffered saline), FBS (foetal bovine serum) and trypsin were bought from Lonza.


24

4. Extract preparation

Aqueous plant extracts were prepared as decoctions and infusions. Decoctions were

prepared using 10 g of grinded fresh plant material boiled for 10 min in 100 ml of distilled water.

Infusions were prepared using 10 g of grinded fresh plant material in 100 ml of freshly boiled

distilled water. Both infusions and decoctions were filtered through Whatman paper and

lyophilized.

The Plectranthus barbatus extract used in the all the work, except section 1 of Chapter III, was

the same, prepared as a decoction of 10 g of ground fresh leaves boiled for 10 min at 100oC, in

100 ml of distilled water and filtered through grade 1 Whatman paper. The extract was

lyophilised and the yield of extraction was approximately 140 mg of extract/g of plant.

5. Acetylcholinesterase inhibition

Acetylcholinesterase enzymatic activity was measured using adaptations of the method

described by Ingkaninan et al. (2003).

In microplates (in Chapter III, section 1): 90µl 50 mM Tris–HCl buffer pH 8, 30 µl sample

and 7.5 µl acetylcholinesterase solution containing 0.26 U/ml were mixed in a microplate and

left to incubate for 15 min. Subsequently, 22.5 µl of a solution of AChI (0.023 mg/ml) and 142 µl

of 3 mM DTNB were added. The absorbance at 405 nm was read when the reaction reached

equilibrium. A control reaction was carried out using water instead of extract and it was

considered 100% activity.

I(%) = 100 – (Asample/Acontrol) x 100

where I(%) is the percent inhibition of acetylcholinesterase, Asample is the absorbance of the

extract containing reaction and Acontrol the absorbance of the reaction control. Tests were carried

out in triplicate and a blank with buffer instead of enzyme solution was used.

In spectrophotometer cuvettes: 325 µl of 50mM Tris buffer (or 50 mM HEPES, when the

samples contained methanol) pH 8, 100 µl of sample and 25 µl acetylcholinesterase solution

containing 0.26 U/ml were mixed in a spectrophotometer cuvette and left to incubate for 15 min

at 25 oC. Subsequently, 75 µl of a solution of AChI (0.023 mg/ml) and 475 µl of 3 mM DTNB were

added. The absorbance at 405 nm was read during the first 5 min of the reaction and the initial

velocity was calculated. A control reaction was carried out using water, which was considered to

have 100% activity.

Chapter II

25

I(%) = 100 – (Vsample/Vcontrol) x 100

where I is the percent inhibition of acetylcholinesterase, Vsample is the initial velocity of the

extract containing reaction and Vcontrol is the initial velocity of the control reaction. Tests were

carried out in triplicate and a blank with buffer instead of enzyme solution was used. When

different volumes sample were used (maximum 200 µl), the final volume was corrected to 1ml

by adjusting the amount of buffer added in the beginning, and control reactions were carried

out in the same conditions.

6. Determination of antioxidant activity

Antioxidant activity was measured by DPPH and by a β-carotene/linoleic acid method,

both as described by Tepe et al. (2005), the latter with a slight modification.

DPPH assay: To a 2.5 ml solution of DPPH, (0.002% in methanol), was added 25 µl of

plant extract. The mixture was incubated for 30 min at room temperature. The absorbance was

measured at 517 nm against a corresponding blank. The antioxidant activity was calculated as

AA(%) = 100 x (ADPPH – Asample)/ADPPH

where AA is the antioxidant activity, ADPPH is the absorption of the DPPH solution against the

blank, Asample is the absorption of the sample against the blank. The tests were carried out in

triplicate and the extract concentration providing 50% of antioxidant activity (IC50) was obtained

by plotting the antioxidant activity against the plant extract concentration.

β-carotene–linoleic acid assay (just used in Chapter III, section 1): β-carotene (0.5 mg),

25 µl of linoleic acid and 200 mg of Tween 40 were dissolved in 1 ml of chloroform. The organic

solvent was evaporated. The residue was redissolved in 100 ml of aerated water. To an aliquot

of 2.5 ml of this solution, 300 µl of each sample were added. The test tubes were incubated in

hot water (50 oC) for 2 h, together with two blanks, one containing the standard antioxidant BHT

(as a positive control) and the other one with the same volume of distilled water instead of the

extracts. In the tube containing BHT the characteristic yellow colour of the β-carotene was

maintained. The absorbance was measured at 470 nm. The antioxidant activity (AA) of the

extracts was calculated using the following equation:

AA(%)= 100 x β-carotene content after 2 h assay/initial β-carotene content


26

The tests were carried out in triplicate and the extract concentration that gave 50% of

antioxidant activity (IC50) was obtained by plotting the percentage of antioxidant activity against

the extract concentration.

7. HPLC analysis

The HPLC analysis was carried out in a Liquid Chromatograph Finnigan™ Surveyor® Plus

Modular LC System equipped with a Purospher® STAR RP-18 column, from Merck and Xcalibur

software.

Generally, the composition of the extracts was analysed by injecting 25 µl and using a

gradient composed of solution A (0.05% trifluoroacetic acid acid), solution B (acetonitrile) and

solution C (methanol) as follows: 0 min, 90% A, 10% B and 50 min 5% A, 80% B, 15% C. Standard

compounds were run under the same conditions and the detection was carried out between 200

and 600 nm with a diode array detector. To purify the three main compounds from the

P. barbatus water extract, the HPLC system referred to above was also used. The peaks were

collected separately and this process was repeated several times.

8. NMR spectroscopy

1H-NMR spectra were recorded using a 400 MHz Bruker Advance 400. Samples were

dried, dissolved in DMSO-d6 and referenced to the residual solvent resonance at δH 2.50 ppm.

9. Mass spectrometry experiments

The mass spectrometric identification of rosmarinic acid was carried out on a

Thermoquest LCQ Duo quadrupole ion trap mass spectrometer equipped with an electrospray

interface. The flow rate of the electrospray solution to the ion source was 5 µl/min and the

instrumental parameters (sheath gas flow rate, ion spray voltage, capillary temperature,

capillary voltage, lens and octapole voltages) were optimised for maximum abundance of the

ions of interest. All mass spectrometric data were acquired in the negative ionisation mode. The

full scan mass spectra were measured using 50 ms for collection of the ions in the trap and the

three micro scans were summed. The data for mass spectra were based on 10–100 scans.

MS/MS experiments were performed in order to ascertain characteristic patterns. Helium was

used as collision gas and the collision energy was gradually increased until both the precursor

and product ions could be observed. MS/MS spectra were measured using 200 ms for collection

Chapter II

27

of the ions in the trap and the three micro scans were averaged. The data for mass spectra were

also based on 10–100 scans.

Further experiments were performed in an ApexQe FTICR Mass Spectrometer from

Bruker Daltonics (Billerica, MA, USA) equipped with an electrospray ion source and a 7 T actively

shielded superconducting magnet. The samples were introduced, by means of an infusion pump

from KD Scientific (Holliston, MA, USA), with a flow rate of 120 µl/h. The mass spectrometer was

calibrated using a 2.8 x 10-6 mol/l solution of polyethylenglycol acquired from Sigma-Aldrich

(St. Louis, MO, USA) in methanol HPLC grade acquired from Panreac (Barcelona, Spain) and

acidified with 0.1% (V/V) of acetic acid acquired from Fluka (Seelze, Germany).

The mass spectra were acquired in the positive ion broad band mode, with an

acquisition size of 512 k, in the mass range of 50-500. The nebuliser gas flow rate was set to

2.5 l/min, the dry gas flow rate was set to 4.0 l/min at a temperature of 220 oC. The capillary

voltage was set to 5000 V and the spray shield voltage was set to 4500 V. All mass spectra

presented are the average of 32 mass spectra.

10. In vitro intragastric metabolism assays

10.1. In vitro metabolism by the gastric juice

The assay was adapted from Yamamoto et al. (1999). Two and a half millilitres of gastric

juice were added to 2.5 ml of extract solution (5 mg/ml) or rosmarinic acid solution (2 mg/ml).

The mixture was left to incubate at 37 oC for 4 h. Samples (100 µl) were taken hourly, added to

900 µl of ice-cold methanol and analysed by HPLC. The gastric juice (100 ml) consisted of 320 mg

of pepsin, 200 mg NaCl, pH 1.2 (with HCl). Assays were done in triplicate.

10.2. In vitro metabolism by pancreatic juice

The assay was adapted from Yamamoto et al. (1999). Two and a half millilitres of

pancreatic juice were added to 2.5 ml of extract solution (10 mg/ml) or rosmarinic acid solution

(2 mg/ml). The mixture was left to incubate at 37 oC for 4 h. Samples (100 µl) were taken hourly,

added to 900 µl of ice-cold methanol, and centrifuged for 5 min at 5000g. The supernatant was

analysed by HPLC.

Two hundred microlitre samples were taken at the same time, centrifuged 5 min at

5000g and the supernatant was analysed for acetylcholinesterase activity, against a blank with

water instead of plant extract. The gastric juice consisted of 250 mg of pancreatin in 10 ml of

potassium-phosphate buffer 50 mM, pH 8. Assays were done in triplicate.


28

10.3. Glucuronidase activity

An adaptation of the method described by Justino et al. (2004) was used. The mixture

contained 567.3 U/ml of β-glucuronidase and 5 mg/ml aqueous extract of P. barbatus, in

potassium-phosphate buffer (50 mM, pH 7.4). The reaction was incubated for 2 h at 37 oC. After

this time the mixture was centrifuged at 5000g for 10 min and the supernatant was analysed by

HPLC. Two control reactions were analysed, one consisting of the reaction mixture without plant

extract and the other consisting of the reaction mixture without β-glucuronidase.

10.4. Metabolism by the Caco-2 cells

The assay for the metabolism of plant extracts by Caco-2 cells was adapted from Kern et

al. (2003). Caco-2 cells (ATCC#HTB-37), a colorectal adenocarcinoma epithelial cell line, were

cultured in DMEM medium supplemented with 10% FBS, 100 U/ml penicillin, 100 U/ml

streptomycin, and 2 mM L-glutamine, at 37 oC in an atmosphere with 5% CO2. The culture

medium was changed every 48–72 h.

For the assays, 2 x 104 cells were seeded in 4 cm diameter Petri dishes and grown for 10

days (to approximately 90% confluence). The medium was replaced with 2.5 ml HBSS plus

1–7 mg/ml of plant extract. For the assays with standards, the concentrations used were

1 mg/ml for rosmarinic acid and 50 µM for apigenin or luteolin. The cells were left in the

incubator and 100 µl samples of the medium were collected at 0, 1, 2, 4, and 6 h, added to 900

µl of ice-cold water, centrifuged 10 min at 5000g, and analysed by HPLC. Samples (100 µl) were

also collected at the same incubation times, centrifuged and analysed for acetylcholinesterase

inhibition, against a blank of HBSS incubated with Caco-2 cells under the same conditions. All

assays were done in triplicate.

After the 6-h assay, the cells in each Petri dish were washed with 500 µl HBSS and

incubated with 1 ml solution of 1:10 TFA (0.05% in methanol) for 30 min. The extract was

centrifuged for 10 min at 5000g, and the supernatant was analysed by HPLC.

10.5. Antiacetylcholinesterase and antioxidant activities of the digested extracts

Aliquots were withdrawn at the beginning of the experiment and after each hour. The

sample was centrifuged at 5000g to discard the bottom phase containing the enzyme, pepsin,

pancreatin or glucuronidase. The upper phase was used for the determination of

acetylcholinesterase inhibition activity as well as antioxidant activity by the DPPH method

according to the procedures described in the sections 4 and 5 of this Chapter. The aliquots from

Chapter II

29

the incubations were diluted to concentrations in the assays corresponding to the IC50 values of

the undigested extract.

11. In vitro conjugation studies for metabolites identification

11.1. Preparation of cell-free extracts

A cell-free liver extract was prepared, essentially as described by Justino and co-workers

(2004), to obtain the hepatic enzymes that metabolize flavonoids. In brief, rat livers were

collected and homogenized in ice-cold 10 mM phosphate buffer, pH 7.4, containing

10 mM 2-mercaptoethanol (1 g wet tissue/2.5 mL buffer). The homogenate was then

centrifuged for 90 min at 12 000 g at 4 oC. The protein content of the supernatant was

immediately determined by the Lowry method (Lowry et al., 1951) and the remainder

supernatant was lyophilized and stored at -20 oC until required.

11.2. Glucuronidation assay

The standard assay mixture contained 0.5 mg.mL-1, P. barbatus extract, or 40 mM

rosmarinic acid, or 50 mM flavonoid standards, 2 mM UDPGA, and cell-free extract (5 mg.mL-1)

in 10 mM potassium phosphate buffer, pH 7.4, final volume of 1 mL (Justino et al., 2004). The

reaction was initiated by addition of the cytosolic fraction extracted and the mixture was

incubated for 30 min at 37 oC without shaking. The reaction was stopped and the metabolites

were extracted as described in the HPLC analysis section below.

11.3. Synthesis and identification of methyl rosmarinic acid

The methylation of rosmarinic acid was done by the porcine liver enzyme catechol-o-

methyl-transferase (COMT), using the cofactor S-adenosylmethionine (SAM), with an adaptation

of the method described by Baba and co-workers (2004). Briefly, 20 U of COMT were dissolved

in 200 ml of deoxygenated 10 mM K-phosphate buffer, 20 mM L-cystein, 2 mM MgCl2, pH 7.4

and pre-incubated at 37 oC for 10 min under inert atmosphere. Five hundred microlitres of

7.1 mM SAM and 100 ml of rosmarinic acid, both dissolved in the reaction buffer, were added.

Two hundred microlitres of 7.1 mM were added every two hours, and the reaction was stopped

at 6 h by the process described in the section for HPLC analysis. The compounds were isolated by

HPLC, collected and analyzed by electrospray ionization mass spectrometry (ESI-MS). All

experiments were performed using a LCQ Duo ion trap mass spectrometer from Thermo

Scientific (San Jose, CA, USA) equipped with an ESI source. Samples were introduced, via a


30

syringe pump (flow rate of 5 mL.min-1), into the stainless steel capillary of the ESI source. The

applied spray voltage in the source was 4.5 kV, the capillary voltage was 10 V and the capillary

temperature was 220 oC. All the mass spectrometer parameters were adjusted in order to

optimize the signal-to-noise ratios for the ions of interest. Nitrogen was used as nebulising and

auxiliary gas in the source. All mass spectrometry data were acquired in the negative ion mode,

the full scan spectra were recorded in the range m/z 100–1000 and three micro-scans were

averaged. MS/MS experiments were performed with helium as collision gas.

12. In vivo studies protocol

12.1. Intragastric and intraperitoneal administration

P. barbatus herbal tea was administered to rats through intragastric procedure

(600 mg.kg-1, equivalent to 150 mmol RA kg-1) or intraperitoneal injection (1000 mg.kg-1,

equivalent to 250 mmol rosmarinic acid kg-1). For each experiment, twelve adult male Sprague–

Dawley rats weighing approximately 400 g and 16 h fasted were randomly divided into two

groups. Rosmarinic acid standard was administered dissolved in ethanol saline solution (20%

ethanol, 0.9% NaCl solution) in a concentration of 550 mmol kg-1. The high dose administered is

justified by the need of obtaining high levels of the different metabolites for analytical purposes.

To control animals only the ethanol saline solution was administered. Blood was withdrawn

from three rats, thirty minutes and 1 h after administration, by cardiac punction into K3EDTA

tubes and kept on ice. The brains were than collected in ice cold K-phosphate buffer 10 mM pH

7.4 10 mM mercaptoethanol and also kept on ice.

12.2. Plasma and brain sample preparation

Plasma was separated from red blood cells by centrifugation at 5000 g for 5 min at 4 oC

and samples were stored at -80 oC until further studies. Each rat brain was weighed and

homogenized in 5 ml 10 mM K-phosphate buffer pH 7.4 10 mM mercaptoethanol. The

homogenate was then centrifuged for 90 min at 12 000 g, at 4 oC and the supernatant (brain

extract) stored at -80 oC until further studies.

12.3. Determination of glucuronidated and sulfated metabolites

An adaptation of the method described by Justino and co-workers (2004) was used. To

determine the amounts of glucuronidated, sulfated and glucuronidated-sulfated compounds

sets comprised one assay with b-glucuronidase, one assay with sulfatase, one assay with both

Chapter II

31

enzymes and a control with no addition of enzymes were prepared. To 500 ml of sample were

added 1000 U of b-glucuronidase, and/or 25 U sulfatase and incubated at 37 oC for 1 h.

The samples were then processed as described in the section 12.4 for HPLC analysis.

12.4. Preparation of samples for HPLC analysis

Methanol was added to 500 ml of plasma or brain extract to 1 ml solution and the

mixture was vortexed, left to precipitate for one hour at 4 oC, and centrifuged for 10 min at

10 000 g, at 4 oC. The supernatant was recovered, left to precipitate another hour and

centrifuged in the same conditions prior to HPLC analysis. Controls were done using rosmarinic

acid and caffeic acid as internal standards added to plasma and brain extracts to determine the

losses related to this purification process. This protocol showed negligible loss of the standard

compounds (less than 10%) and the yield was calculated taking into account the concentrations.

12.5. Determination of acetylcholinesterase activity in brain samples

Acetylcholinesterase enzymatic activity was measured using an adaptation of the

methods described by Chattipakorn and coworkers (2007) and in section 5. Four hundred

microlitres of 50 mM HEPES buffer pH 8 and 50 ml brain extract were mixed in

spectrophotometer cuvette and left to incubate for 15 min at 25 oC. Subsequently, 75 ml of a

solution of AChI (0.023 mg ml-1) and 475 ml of 3 mM DTNB in HEPES 50mM pH 8 were added.

The absorbance at 405 nm was read during the first five minutes of the reaction and the initial

velocity was calculated as mAU min-1 mg-1 and converted into nmole min-1 mg-1 of tissue mass. A

control reaction was carried out using the brain extract from the control rats, and it was

considered 100% activity for calculations of enzymatic inhibition.

13. Protein fluorescence measurements

The fluorescence quenching study on HSA was performed in 3 ml solutions containing

1 × 10−8 M HSA in Tris-HCl buffer (0.20 M, pH 7.4) containing 0.10 M NaCl and the appropriate

quantities of quencher – P. barbatus extract, rosmarinic acid, luteolin or apigenin (Xiao et al.,

2008).

Lysozyme intrinsic fluorescence quenching was studied in 3 ml solutions containing

0.02 mg.ml−1 and the appropriate amount of quencher in K-phosphate buffer (0.01 M, pH 7.4).

All samples were pre-incubated for 1 h at 37◦C (Xiao et al., 2008).

The fluorescence quenching study on of acetylcholinesterase was performed according

Xiao and coworkers (2008), in 3ml solutions containing 5U of AChE in Tris-HCl buffer (50mM,


32

pH 8) and the appropriate quantities of quencher – P. barbatus extract, rosmarinic acid,

quercetin, luteolin or apigenin.

All fluorescence measurements were performed on a LS55 spectrofluorometer (Perkin-

Elmer, UK) in a 1 cm quartz cell. Excitation and emission bandwidths were set to 10 nm. Emission

spectra were recorded from 300 to 500 nm, under excitation wavelength of 280 nm. The

fluorescence intensity was measured at 335 nm under the same excitation wavelength, at

different temperatures (293, 298 and 303 K).

14. FTIR measurements

FTIR measurements were carried out at 21◦C using a Bruker Vector 22 Spectrometer.

Each spectrum was acquired by averaging between 32 and 300 scans in the spectral range of

400–4000 cm−1 at 4 cm−1 resolution.

Fifty microliters of reaction mixture were transferred into a CaF2 infrared cell, fitted with

a 50 μm path Teflon spacer and placed in the spectrometer. Spectra were acquired every 2 min

for 1 h. For each sample a spectrum was acquired after 2 h, averaging 300 scans.

The hydrogen–deuterium exchange was initiated when HSA was dissolved in Tris buffer,

50 mM pH 7.4, containing 25 mM NaCl, prepared in deuterium oxide, and the compounds in

study in a final concentration of 500 μM HSA, 500 μM of standard compounds or 10 mg/ml of P.

barbatus extract.

For the measurements with lysozyme, the hydrogen–deuterium exchange was started

when lysozyme was dissolved in K-phosphate buffer 10 mM pH 7.4, prepared in deuterium

oxide, containing lysozyme in the final concentration of 2 mM, standard compounds in the same

concentration or P. barbatus extract in the final concentration of 10 mg · ml−1.

The AChE hydrogen-deuterium exchange started when AChE was dissolved in Tris buffer,

50mM pH8, prepared in deuterium oxide. and the compounds in this study in the final

concentrations of 7080 U.mL-1 AChE, 1 mM of standard compounds or 10mg/mL of P. barbatus

extract.

The FTIR spectra recorded were analysed using the Bruker OPUS software. FTIR spectra

of the buffer were recorded under identical conditions and the OPUS software was used to

subtract the spectrum of the buffer from the spectrum of the protein in buffer using previously

described procedures (Haris et al., 1990). Subsequently, second-derivative analysis was carried

on the absorbance spectra to reveal the overlapping amide I components (Haris et al., 1990).

Chapter II

33

15. Lysozyme activity measurements

Lysozyme activity was measured with an adaptation of a method previously described

(Laible and Germaine, 1985). Five hundred microliters of M. lysodeikticus suspension in

K-phosphate buffer (0.10 M, pH 7.4) were added to 500 μl of a solution containing lysozyme

(0.4 mg) and an appropriate amount of inhibitor – P. barbatus extract, rosmarinic acid, luteolin

or apigenin – in the same buffer. The decrease in absorbance was followed at 570 nm for the

first five minutes of reaction and compared with the appropriate controls (without inhibitor and

without enzyme). All assays were done in triplicate and IC50 values were calculated.

16. Measurement of hypochlorous acid in activated human neutrophils

16.1. Isolation of human neutrophils

The isolation of human neutrophils was carried out as described by Coelho (2004). A

dextran solution (6% in phosphate buffer 0.1M, pH7.4, NaCl 0.9%) was added to the blood

collected form human donors in a proportion 5:1 (V:V), gently mixed by inversion, and left for

90 min in the dark to separate the red blood cells. A NH4Cl 0.87% (m:V) solution was added to

the supernatant in the proportion 1:2 (V:V) and left to incubate for 5 min at room temperature

for the lysis of the remaining red blood cells, and the leucocytes were collected by centrifuging

5 min at 250g. The cells were washed 3 times with phosphate buffer 16mM, pH 7.4, containing

NaCl 122 mM, KCl 4.89 mM, MgSO4 1.22 mM, D-glucose 1 mg/mL and EDTA 0.03% (m:V) and

collected by centrifuging 5 min at 250g. The cells were ressuspended in 1.5 mL ice cold KRC

solution (phosphate buffer 16mM, pH 7.4, containing NaCl 122 mM, KCl 4.89 mM, MgSO4

1.22 mM, D-glucose 1 mg/mL and CaCl2 0.6 mM), and kept on ice for no longer than 3h.

16.2.Measurement of hypochlorous acid formation by human neutrophils

The hypochlorous acid produced by human neutrophils was measured after activation

with phorbol 12-myristate 13-acetate (PMA), with an adaptation of the method described by

Dypbukt et al. (2005). The neutrophils were incubated for 30 min at 37oC in a 250 µL solution

containing 2x106 cells/mL, taurine 5 mM, PMA 100 ng/mL, and several concentrations of

rosmarinic acid or P. barbatus extract, in KRC buffer. The reaction was stopped by adding

20 µg /mL catalase and incubating on ice for 10 min. The cells were removed by centrifuging

5 min at 11000g, and the hypochlorous acid in the supernatant was measured by the taurin

chlorination method. Briefly, 40 µL of a revealing solution (3,3’,5,5’-tetrametilbenzidina (TMB)

2mM, KI 100µM, in acetate buffer 400mM pH 5.4, 30% dimethylformamide) were added to


34

160 µL of the supernatant in a 96-well plate. The absorbance was read at 655 nm after 5 min

incubation at room temperature.

17. Caco-2 bioavailability experiments

Caco-2 cells (ATCC#HTB37), a human colorectal adenocarcinoma epithelial cell line, were

cultured as described in section 10.4.

For transport and metabolism experiments, the cells were seeded at a density of

2-4x104 cells /cm2 in 12-well Transwell plate inserts with 10.5mm diameter, 0.4 µm pore size

(BD Falcon™). The monolayers were formed after 21-26 days. The integrity of the monolayers

was evaluated by measuring the permeability of phenol red or the transepithelial electrical

resistance (TEER). The membranes were considered fit when the permeability of phenol red

from apical to basolateral sides was less than 1% in one hour, or the TEER was higher than

250 Ω.cm2.

In general, to start the assays, the cells were washed with HBSS and 0.5 mL of the

solutions to be tested, in HBSS, were applied into the Transwell inserts (apical side of the cells).

1.5mL of HBSS were added to the plate well (basolateral side of the cells). After 6 hours of

incubation at 37 ºC, 5% CO2, the solutions in both sides of the cells were collected and analysed

by HPLC. The cells were washed with HBSS, and ressuspended by scraping with HBSS. The cells

were sonicated 5x10 s, centrifuged 10 min ant 5000 g, and the supernatant was analysed by

HPLC.

Before carrying out the assays the concentrations of the solutions were proved not to

affect Caco-2 cells growth by performing the MTT viability test (Mosmann, 1983). All

concentrations used showed 0% toxicity and did not affect the Caco-2 cell membrane integrity.

17.1. Study of the permeability and the metabolism of rosmarinic acid, luteolin and

apigenin

The effect of rosmarinic acid, luteolin and apigenin on the permeation and

metabolization of each other on Caco-2 cells was studied by applying several concentrations of

these compounds in the apical side of the cells according to a factorial design known as CCD

(Barker, 1985). The solutions to be used in this study were prepared in HBSS, and the

concentrations of rosmarinic acid, luteolin and apigenin were listed in Table 2.1.

Chapter II

35

Table 2.1. Concentrations of rosmarinic acid, luteolin and apigenin used in the phenolic mixtures

to determine the bioavailability by Caco-2 cells, according to the CCD experimental plan.

Solution

Concentration (µM)

Rosmarinic

Acid

Luteolin Apigenin

1 30 30 30

2 70 30 30

3 30 70 30

4 70 70 30

5 30 30 70

6 70 30 70

7 30 70 70

8 70 70 70

9 10 50 50

10 90 50 50

11 50 10 50

12 50 90 50

13 50 50 10

14 50 50 90

15 50 50 50

17.2. Study of the effect of transport systems (MCT and Pgp) on the permeation of

apigenin, luteolin and rosmarinic acid

Benzoic acid (0.1 mM) and digoxin (0.1 mM) were used as substrates of the known

transport systems in Caco-2 cells, MCT and Pgp, respectively. The assays were performed as

previously described (section…) with the substrates. Competition between rosmarinic acid,

luteolin and apigenin with the substrates of the transport system was also studied with solutions

containing a mixture of these 3 model phenolic compounds (50 µM each) and each of the

substrates, in the absence of inhibitors.

17.3. HPLC analysis and bioavailability quantification

The HPLC analysis was carried out in slightly different conditions from the ones decribed

in section 7, in Liquid Chromatograph Finnigan® Surveyor® Plus Modular LC System, Thermo-

Finningan, Germany equipped with a LiChroCART® 250-4 LiChrospher® 100 RP-8 (5 µm) column,

from Merck, Darmstadt, Germany, and Xcalibur software. The extracts were analysed by HPLC

injecting 25 µl with an auto injector, and using a linear gradient composed of solution A (0.05%

trifluoroacetic acid), and solution B (methanol) as following: 0 min, 70% A, 30% B; 20 min 20% A,

80% B; 25 min, 20% A, 80% B. The detection was carried out between 200 and 500 nm with a

diode array detector.


36

The amount of rosmarinic acid, apigenin and luteolin in the samples were determined by

calibration curves of concentration/peak area, and the bioavailability was quantified by the

distribution of each compound in each compartment of the Caco-2 monolayer (apical,

basolateral, and intracellular).

18. Statistical analysis

All results are presented as mean ± standard deviation of three replicates (or five replicates, in

the in vitro digestions with gastric and pancreatic juices) and the software used was Microsoft

Excel®. Additionally analysis of variance (ANOVA) was performed with p = 0.05.

Chapter III

Screening for antiacetylcholinesterase and antioxidant

activities in Plectranthus species. In vitro metabolism and anti‐

inflammatory studies

Falé PL, Borges C, Madeira PJM, Ascensão L, Araújo MEM, Florêncio MH, Serralheiro MLM. 2009.

Rosmarinic acid, scutellarein 4’‐methyl ether 7‐O‐glucuronide and (16S)‐coleon E are the main

compounds responsible for the antiacetylcholinesterase and antioxidant activity in herbal tea of

Plectranthus barbatus (‘‘falso boldo”). Food Chem. 114: 798–805.

Porfirio S, Falé PLV, Madeira PJA, Florêncio H, Ascensão L, Serralheiro MLM. 2010.


gastrointestinal metabolism, Food Chem. 122: 798–805.

Falé PLV, Filipe MA, Ascensão L, Serralheiro MLM, Mira L. Activity of Plectranthus barbatus extract

against inflammatory response in human neutrophils. To be submitted to Plant Food Hum. Nutr.

Chapter III

39

1. Screening for antiacetylcholinesterase and antioxidant

activities in Plectranthus species

1.1. Introduction

The genus Plectranthus L’Hér., belonging to the Mint family (Lameaceae), comprises

about 300 species widely distributed in Africa, Asia and Australia. A great variety of

ethnobotanical uses are reported for Plectranthus species. However, the most common uses are

to treat a wide range of diseases, including digestive problems, inflammation‐related conditions

and nervous system disorders (Lukhoba et al., 2006). Infusions of several Plectranthus species

are used to treat coughs and colds, liver complaints, and ailments in respiratory system and skin.

Plectranthus barbatus (Lamiaceae) is drunk as an herbal tea to treat a wide range of

diseases, and is also used in food recipes, particularly in South America, Africa and Eastern

regions (Lukhoba et al., 2006). In order to know which were the main compounds that could be

responsible for the healing properties attributed to P. barbatus, infusions and decoctions of this

species were analysed. The inhibition capacity of the extracts on the enzyme

acetylcholinesterase (AChE) was studied. Acetylcholinesterase is the enzyme that catalyses the

hydrolysis of acetylcholine, a neurotransmitter found in the synaptic gap, and the inhibition of

this enzyme could give some explanation for the action of the herbal tea on the nervous system.

Nowadays the AChE inhibition is used in the treatment of the Alzheimer’s disease (AD) (Heinrich

& Teoh, 2004). Therefore the most successful therapy for AD, at present, consists of increasing

the levels of acetylcholine through the inhibition of acetylcholinesterase activity (Heinrich &

Teoh, 2004).

Most of the compounds isolated from the plant polar extract fraction are polyphenols

(Cai et al., 2004; Trouillas et al., 2003). These compounds also have a high antioxidant activity

(Cai et al., 2004; Djeridane et al., 2006). The antioxidant activity found in some compounds has

been connected to the capacity to scavenge the free radicals that are formed during the

inflammation processes (Gomes et al., 2008). It is becoming the objective of research to develop

or find new drugs that exhibit many simultaneous activities, as most of the diseases are not the

result of a unique mechanism, but the consequence of several biochemical processes

(Bembenek et al., 2008). This is one of the reasons why, nowadays, the herbal teas are gaining

popular acceptance. They are a mixture of several compounds, so they can act on different

diseases simultaneously. A bioguided study has been carried out where the inhibition of AChE

was looked for and the P. barbatus water extracts were analysed in order to explain some of its

In vitro Studies of Activity and Metabolism

40

traditional uses. In order to confirm the results obtained with P. barbatus in this work, four

different Plectranthus species were also analysed. P. ecklonii and P. fructicosus also have

medicinal applications but are used mostly to treat skin conditions such as infections and burns

(Lukhoba et al., 2006). P. lanuginosus herbal tea is used for digestive conditions and P.

verticillatus is only used as ornamental.

1.2. Materials and Methods

The materials and methods are described in detail in the Chapter II.

The plant aqueous extracts were prepared as infusions and decoctions. Aliquots of 1 ml

of each extract were lyophilized and used to determine the dry weight. Data present in Table 3.1

are the average of three independent measures.

Table 3.1. Amount of dry plant extract obtained.

Plant species Plant extract (mg dry extract/g fresh plant)

Infusion Decoction

P. barbatus 16 ± 1 36 ± 2

P. ecklonii 36 ± 5 48 ± 5

P. fruticosus 25 ± 3 22 ± 4

P. lanuginosus 12 ± 4 13 ± 3

P. verticillatus 3 ± 1 3 ± 1

Acetylcholinesterase inhibition was determined in microplates, as described in Chapter II

section 5, and the antioxidant activity was quantified by the DPPH and β‐carotene‐linoleic acid

methods, as described in Chapter II, section 6. The composition of the plant extracts was

analysed by HPLC as described in Chapter II section 7, and the main components, isolated by

HPLC, were identified by NMR spectroscopy (Chapter II, section 8) and mass spectrometry

(Chapter II, section 9.).

1.3. Results

1.3.1. General

In this work the aerial part of P. barbatus, P. ecklonii, P. fructicosus, P. lanuginosus and P.

verticillatus was separated into a vegetative part (leaves) and floral part (flowers), which were

analysed separately, not only for their action as AChE inhibitors but also for their antioxidant

activity. The antioxidant activity was evaluated using two tests, the DPPH and the β‐

carotene/linoleic acid bleaching test. The first gives information about the ability of the tested

compounds to scavenge free radicals. The second gives information on the ability of the tested

Chapter III

41

extracts to delay the effect of lipid peroxidation on biological compounds by reacting with chain‐

propagating peroxyl radicals faster than these radicals can react with proteins or fatty acid side‐

chains. In order to establish the results as a function of the dry weight of the extracts, 1 ml of

each extract was lyophilized and its dry weight was determined. As it was shown in Table 3.1,

P. ecklonii and P. barbatus gave the highest yield of plant extract. All the calculations carried out

in the next sections are expressed in dry weight of the extracts.

1.3.2. Acetylcholinesterase inhibition

The inhibition activity of leaves and flowers of P. barbatus demonstrated that the

extracts obtained with the leaves were more active than those obtained with the flowers (<10%

when using 0.5 mg of extract/ml). All the subsequent results refer to the leaves of the

Plecthrantus species.

All the decoctions showed higher activities than the infusions. The decoction of P.

barbatus inhibited the activity of AChE by approximately 31.5%, when using 0.5 mg of extract/ml

of test solution (Table 3.2). This result is within the range of the values found in the literature for

the inhibition of this enzyme with plant extracts (Mata et al., 2007; Orhan et al., 2004) and

corresponds to approximately one small bag of tea made from the leaves (14 mg). The values

obtained with other species of Plectranthus showed that other species also inhibit AChE (Table

3.2). P. verticillatus and P. ecklonii were the species studied that gave higher values, 59.6% and

62.8%, respectively. With the decoctions of P. lanuginosus and P. fruticosus, the inhibition values

obtained when using 0.5 mg/ml, were 10.0% and 31.3%, respectively.

Table 3.2. Inhibition of AChE activity (%), antioxidant activity and rosmarinic acid content of water

extracts of the leaves of several Plectranthus species.

Plectranthus

species

AChE inhibition (%)

with 0.5 mg/ml

dry extract

DPPH

(IC50 µg dry extract/ml)

β-Carotene

(IC50 µg dry extract/ml)

Rosmarinic

acid

(mg/mg

extract)

Infusion Decoction Infusion Decoction Infusion Decoction Decoction

P. barbatus 17.0 ± 5.4 31.5 ± 3.3 10.4 ± 0.3 45.8 ± 0.5 226.4 ± 11.5 69.8 ± 3.1 0.14

P. ecklonii 46.7 ± 0.4 62.8 ± 6.5 2.4 ± 0.1 3.8 ± 0.1 142.1 ± 8.0 109.0 ± 3.1 0.48

P. fructicosus 11.3 ± 1.8 31.3 ± 2.6 6.2 ± 0.9 4.4 ± 0.2 367.1 ± 15.6 234.6 ± 5.4 0.28

P. lanuginosus 13.7 ± 7.2 10.0 ± 1.1 5.2 ± 0.4 3.3 ± 0.2 362.9 ± 5.9 196.7 ± 2.3 0.18

P. verticillatus 47.4 ± 5.4 59.6 ± 7.6 1.5 ± 0.7 1.2 ± 0.4 37.7 ± 1.5 28.9 ± 1.5 0.44

BHTa - 15.7 ± 0.2 12.0 ± 0.7 - aMata et al. (2007)


42

1.3.3. Antioxidant Activity

Antioxidant activity of infusions and decoctions of the Plectranthus species was

evaluated using the DPPH method. Results are presented in Table 3.2 as IC50 (µg dry weight

extract/ml of the test solution). The antioxidant activity of these extracts, on what concerns their

ability to scavenge free radicals, determined by the DPPH, is very high, and similar to the

antioxidant activity of a known synthetic standard, BHT, previously determined, IC50 = 15.7 ± 0.2

µg/ml (Mata et al., 2007). Both extracts of P. verticillatus show the highest antioxidant activity,

IC50 = 1.5 µg/ml (infusion) and IC50 = 1.2 µg/ml (decoction). Also the extracts from P. ecklonii, P.

fructicosus and P. lanuginosus have a comparatively high antioxidant activity. These values are

lower (higher activity) than those recently reported for other Lamiaceae species from genera

Mentha and Salvia (Kanat et al., 2007; Mata et al., 2007; Parejo et al., 2004; Tepe et al., 2005)

and have the same magnitude as those reported for green tea extract, a reference extract highly

commercialised for its antioxidant activity (Atouin et al., 2005).

The capacity to prevent the lipid peroxidation, measured through the β‐carotene–

linoleic acid test, gave values higher, which means less activity, than the one obtained with the

standard BHT. An exception is P. verticillatus that gave a value of the same magnitude as BHT

(Table 3.2). P. barbatus (decoction) and P. verticillatus (infusion and decoction) were the species

showing higher antioxidant activity when measured by this test, 69.8 µg/ml, 37.7 µg/ml and 28.9

µg/ml, respectively.

1.3.4. Identification of the main component of the Plectranthus extracts, responsible for

the enzyme inhibition activity

In order to assess whether the extracts contained a large number of different

compounds, and in an attempt to identify these constituents, HPLC–DAD analysis was carried

out with the decoctions of the Plectranthus species. The decoctions were selected to identify the

main compounds because, in general, they gave higher enzyme inhibition activity than the

infusions. The chromatograms for P. barbatus and P. verticillatus are shown in Figure 3.1a and b,

respectively. The chromatograms of P. ecklonii, P. fruticosus and P. lanuginosus are similar to the

one obtained with P. verticillatus. In all the extracts there is a major peak with the retention time

at ~19.2 min. The decoctions of P. barbatus showed two additional peaks on the total absorption

of the wavelength, Figure 3.1a. In P. barbatus chromatogram the main peak (retention time of

19.2 min) represents 51% of the total area, whilst in all the other chromatograms this peak

represents ~80%. The compounds with retention times of 23.4 min and 33.0 min found in P.

Chapter III

43

barbatus chromatogram represent 6% and 20%, respectively. The amount of the compound,

with a retention time of 19.2 min in the extracts, depends on the plant analysed.

Figure 3.1. HPLC chromatogram of decoctions: (a) Plectranthus barbatus and (b) Plectranthus verticillatus.

Analysing the UV spectra of the peak with a retention time of 19 min, Figure 3.2a, a

similar spectrum to the one obtained with caffeic acid was observed, Figure 3.2b. These

compounds have characteristic absorptions at 330 nm with a shoulder between 300 and 295

nm, Figure 3.2.

Figure 3.2. UV spectra obtained by HPLC‐diode array of (a) compounds with retention time 19.2 min and

(b) caffeic acid.

The extracts of P. barbatus and P. verticillatus were subjected to NMR spectroscopy,

Figure 3.3a and b, respectively. As the P. verticillatus extract was very pure, it was used to

identify the peak with a retention time 19.2 min. 1H‐NMR spectra showed the presence of two

doublets (7.3 ppm and 6.1 ppm, J = 15.88 Hz) corresponding to the double bond of a cinnamic

derivative. The presence in the aromatic region of the spectra, 7.1–6.4 ppm, of two singlets and

four doublets, each one integrating for one proton, suggested the presence of two aromatic

rings with the same pattern of substitution.

The existence of two double doublets at 3.0 ppm and 2.7 ppm coupled with another

double doublet at 4.8 ppm indicated that these two aromatic rings should have different carbon

chain substitutions: a α,β unsaturated and a saturated one, results not shown. 13C‐NMR spectra


44

showed 18 signals corresponding to: two different carbon carbonyls, eight unsaturated

methines, four aromatic quaternary carbons, one deshielded methylene and one methane group

bonded to an oxygen. Compound was identified as rosmarinic acid. 1H‐NMR spectra of the

standard rosmarinic acid was identical with the one obtained with the pure compound (Figure

3.3c). The identification of rosmarinic acid was further confirmed by MS and MS/MS. For this

compound and for the standard rosmarinic acid, an ion at m/z 359, corresponding to [M–H]‐,

was found using ESI–MS (not shown). MS/MS analysis of this ion (not shown) generated a

product ion at m/z 161, which is in agreement with the observation of m/z 163 in the positive

ion mode, considered to be typical of caffeic acid esters (Grayer et al., 2003). The MS data

obtained confirmed the identification of this major compound as rosmarinic acid, which is

consistent with a previous literature report by Parejo et al. (2004). The structure of rosmarinic

acid is shown in Figure 3.4a.

Figure 3.3. NMR spectra of (a) Plectranthus barbatus extract, (b) Plectranthus verticillatus extract and (c)

rosmarinic acid standard.

Figure 3.4. Structure of the compounds with retention time 19.2 min, rosmarinic acid.

Chapter III

45

Standard rosmarinic acid was analysed by HPLC and the retention time could be

confirmed, as well as the UV spectra (Figure 3.5). In this figure a complete overlay between the

spectra of the compound with retention time 19.2 min and rosmarinic acid can be seen.

Figure 3.5. Overlay of UV spectra obtained by HPLC‐diode array of compounds with a retention time 19.2

min (——) and the standard rosmarinic acid (___).

Rosmarinic acid was quantified in plant extracts, by HPLC, using a calibration curve with

a range of 1–50 µg/ml rosmarinic acid, R2 = 0.99. The results are indicated in Table 3.2. The

rosmarinic acid content correlates with the activity found for P. ecklonii, P. fructicosus, P.

lanuginosus and P. verticillatus, R2 = 0.998. The values found with P. barbatus extract cannot

correlate with rosmarinic acid content because it contains other compounds that also have AChE

inhibition activity.

The inhibition of AChE by rosmarinic acid was determined and an IC50 of 0.44 ± 0.03

mg/ml was obtained. Orhan et al. (2008) found a value of 85.8% inhibition when using 1 mg/ml

of rosmarinic acid, a value similar to the one obtained in this study, showing a linear relationship

between the rosmarinic acid content and the IC50 values. The activity found with P. barbatus

extract could not be completely explained by the presence of rosmarinic acid, suggesting that

the other extract components may also be active.

1.4. Discussion

Rosmarinic acid was the main compound found in all the aqueous extracts of the

Plectranthus species analysed and it is known to be a rather common compound in the

Lamiaceae family (Abdel‐Mogib et al., 2002). We could only find a report of the presence of

rosmarinic acid in P. fruticosus (Pederson, 2000). Rosmarinic acid was also found to be the major

constituent in Coleus aromaticus (Kumaran & Karunakaran, 2007), a species with several

synonyms that is also referred to in the literature as P. aromaticus and P. amboinicus (Lukhoba

et al., 2006). The AChE inhibition activity of P. ecklonii, P. fructicosus, P. lanuginosus and


46

P. verticillatus was due only to the presence of rosmarinic acid. The activity found in the

analysed species correlated rather well with the quantity of rosmarinic acid present (R2 = 0.998).

Rosmarinic acid is the main component of polar extracts from many plants of the

Lamiaceae family, commonly used in human diet. Besides being an enzyme inhibitor, it also has

antioxidant activity, with a value of IC50 = 1.68 ± 0.21 µg/ml in the present study. The antioxidant

activity of rosmarinic acid is well studied and the presence of this compound is considered one

of the reasons for the several biological activities found in species of the Lamiaceae family.

Besides its antioxidant activity, rosmarinic acid also inhibits several enzymes, for instance, the

angiotensin‐converting enzyme (Li et al., 2008) and phospholipase A2 from snake venom (Ticli et

al., 2005). It is also known to interfere with gene expression (Lee et al., 2007) and signalling

pathways related to cancer prevention (Lee et al., 2007; Lin et al., 2007). All these biological

activities may contribute to the large uses that the people in the South Latin America and in

other parts of the world do with these Plectranthus species.

It may be speculated that the rosmarinic acid may fit into the active site gorge of AChE.

This active site is composed of tryptophan at the peripheral site, phenylalanine and tyrosine

residues down the entrance of this gorge (Silman & Sussman, 2008). These residues may interact

with the aromatic rings of rosmarinic acid, and occupy the enzyme active site due to the affinity

to the amino acid residues localized at the entrance of the gorge. This peripheral binding site

defines whether the compounds will act on the enzyme or not as described in the literature

(Eastman et al., 1995).

Rosmarinic acid is the main compound responsible for the results obtained with all the

Plectranthus species P. barbatus, P. ecklonii, P. fructicosus, P. lanuginosus and P. verticillatus

studied in this work. Although the last two species are not traditionally applied in folk medicine,

they could be a good source of rosmarinic acid for future research.

1.5. Conclusion

The inhibition of acetylcholinesterase by the water extracts of Plectranthus species

seems to correlate mainly with the concentration of rosmarinic acid present. The most active

extracts were from P. verticillatus, P. ecklonii and P. barbatus; however the activities found in

the extracts of the first two species were exclusively due to the content in rosmarinic acid. The

beneficial health activity described with the use of P. barbatus may be attributed to the

presence not only of rosmarinic acid, but to other active components. The rosmarinic acid in the

water extracts and decoctions, consumed by the population, may explain some of the health

benefits described through its traditional applications.

Chapter III

47

2. In vitro Digestion Activities of Plectranthus barbatus Aqueous

Extract and Activities of the Digested Product.

2.1. Introduction

Plectranthus barbatus, is one of the most used Plectranthus species, especially for

therapeutic applications. In Brazil, where it is known as ‘‘falso boldo”, infusions or decoctions of

the leaves are drunk to treat a wide range of diseases such as digestive and nervous system

disorders (Lukhoba et al., 2006). The same species is also used to prevent or alleviate

inflammation conditions. It is also sold as food supplement due to its healing properties. P.

barbatus is also reported to be edible in Africa where the leaves are cooked as a vegetable

(Lukhoba et al., 2006).Previous work demonstrated that this tea has antiacetylcholinesterase as

well as antioxidant activity, and rosmarinic acid was the main compound present in this herbal

tea. (Falé et al., 2009). Rosmarinic acid is widespread in Nepetoideae species of Lamiaceae

family, such as rosemary (Rosmarinus officinalis), mint (Mentha arvense), basil (Ocimum

basilicum) (Petersen & Simmonds, 2003). In opposition to the other Plectranthus species in the

previous section, rosmarinic extract was not the only constituent responsible for the activities

shown by P. barbatus extract. Also for this reason, but mainly because P. barbatus is one of the

most used Plectranthus species, the present studied continued with a decoction of P. barbatus.

Several compounds can be found in different herbal teas and it is important to

determine their metabolism after being consumed as a beverage in order to evaluate the final

biological activity. The first step in the digestion process is the contact between the beverage

and the gastrointestinal tract, since saliva does not seem to have a strong effect on the

metabolism of these compounds (Spencer 2003). Rosmarinic acid may be subject to hydrolysis

either at acidic or basic pH, as it is an ester of caffeic acid, which means that it may be degraded

under the physiological conditions of the gastrointestinal tract before reaching the intestinal

barrier. The acidic pH of the gastric juice has been reported to hydrolyze flavanol oligomers to its

monomers, (Spencer et al., 2000) and also saponins from ginseng (Kong et al., 2009), which may

facilitate their further degradation. In spite of studies focusing on the metabolism in the

gastrointestinal tract being scarce, research on this topic has demonstrated that when subjecting

polyphenols, from fruit beverages, to acidic stomach conditions followed by pancreatic medium,

a decrease in their content can be observed. The exact amount of the decrease depends on the

compound under evaluation (Cilla et al., 2009; Yoshino et al., 1999).


48

To simulate the metabolism occurring during the absorption process, Caco‐2 cells are

commonly used, either in studies with intact cells or in the homogenate form (Galijatovic et al.,

2000). Studies carried out with intact cells indicated that polyphenols were not metabolised by

Caco‐2 cells (Cilla et al., 2009); on the other hand, studies of the action of cell homogenates on

polyphenolic compounds, namely flavonoids, indicated that metabolites might be obtained

during the permeation process (Galijatovic et al., 2000). In addition, when rosmarinic acid was

administered orally to rats, only approximately 5% appeared in the blood stream, and most of it

in derivatized forms, such as methyl‐rosmarinic acid, caffeic acid, ferulic acid and coumaric acid

(Baba et al., 2004). Very few reports tackled the problem of the remaining biological activity

after the biodegradation process and it is important to know if metabolism increases or

diminishes the initial biological activities of the herbal teas. Therefore, the aims of this study

were to investigate if the active compounds found in the herbal tea of P. barbatus were stable

under gastrointestinal tract conditions, what metabolites could be found in contact with the

intestinal wall and whether the metabolised extract had similar antiacetylcholinesterase and

antioxidant activities as the ones previously determined (Falé et al., 2009). This study will allow

an insight into the metabolism of herbal teas containing a mixture of polyphenols and terpenoid

compounds.



The P. barbatus extract was prepared as a decoction, as described in Chapter II section 4,

and subjected to in vitro digestion by gastric juice (Chapter II, section 10.1), pancreatic juice

(Chapter II, section 10.2), β‐glucuronidase from E. coli (Chapter II, section 10.3) and Caco‐2 cells

(Chapter II, section 10.4). During the digestions the composition of the plant extract was

followed by HPLC (Chapter II, section 7), and the biological activities – antioxidant activity (DPPH

in Chapter II, section 6) and acetylcholinesterase inhibition (Chapter II, section 5). The

identification of compounds was carried out by mass spectrometry, as described in Chapter II,

section 9..

2.3. Results and Discussion

2.3.1. Main composition of P. barbatus herbal tea

The water extract of P. barbatus leaves was analysed by HPLC‐DAD (Figure 3.6). The

compounds after being separated by HPLC were identified by mass spectrometry comparatively

to standard libraries. The compounds from the that P. barbatus that had not been identified

Chapter III

49

before are shown in Table 3.3. All the compounds have already been identified in plants of

Plectranthus species (Abdel‐Mogib et al., 2002; Batista et al., 1996). The structures are shown in

Figure 3.7.

Fig. 3.6. HPLC chromatogram of Plectranthus barbatus herbal tea: 1, luteolin 7‐O‐glucuronide (retention

time 8.68 min); 2, rosmarinic acid (RT: 9.38); 3, apigenin 7‐O‐glucuronide (RT: 10.11); 4, hydrolysed

abietane (RT: 12.58); 5, acacetin 7‐O‐glucuronide (RT: 13.62); 6, abietane diterpenoid (RT: 14,15); 7, (16S)‐

coleon E (RT: 18.54).

Table 3.3. Detected ions and attribution errors (ppm) for the collected fractions corresponding to peaks 1,

3, 4, 5, 6 and 7.

Chromatogram

peak Attribution

m/z Error

(ppm) Experimental Theoretical

1 [C21H18O12 + H]+ 463.08571 463.08710 3.0

3 [C21H18O11 + H]+ 447.09064 447.09219 3.5

4 [C22H26O8 + H]+ 419.16973 419.17004 0.8

5 [C22H20O5 + H]+ 461.10754 461.10784 0.6

6 [C22H26O7 + H]+ 403.17361 403.17513 3.8

7 [C20H22O5 + H]+ 343.15341 343.15657 1.7


50

Figure 3.7. Chemical structure of compounds present in P. barbatus herbal tea: 1, luteolin 7‐O‐

glucuronide; 2, rosmarinic acid; 3, apigenin 7‐O‐glucuronide; 4, hydrolysed abietane; 5, acacetin 7‐O‐

glucuronide; 6, abietane diterpenoid; 7, (16S)‐coleon E.

2.3.2. In vitro metabolism of the extract by the gastric and pancreatic juices. Biological

activity of the resulting products

The digestion of the P. barbatus water extract was carried out for 4 h under stomach

acidic conditions (pepsin, pH 1). Although the digestion in the stomach may take 60–110 min

(Kong et al., 2009), it was decided to continue the process to see if the bioconversion increased

with the time. Aliquots were withdrawn during the 4 h and the HPLC of each aliquot was

analysed. None of the extract constituents was hydrolyzed under the stomach acidic conditions,

even after 4 h digestion (Figure 3.8.a). Pepsin, an enzyme that hydrolysis proteins, does not have

the capacity to hydrolyze esters, unlike other proteases, because the ester, rosmarinic acid, was

not hydrolyzed at pH 1.

The antiacetylcholinesterase and antioxidant activities of P. barbatus water extract were

analysed during the 4 h of the in vitro gastric process; the results are shown in Table 3.4. It can

be seen that after the gastric digestion the inhibition capacity of the extract showed a small

decline that is not statistically significant at the 95% level. The same was noticed for the

antioxidant activity. The standard rosmarinic acid was subject to the same in vitro conditions and

the results were similar to those obtained in the herbal tea mixture.

Chapter III

51

Figure 3.8. HPLC chromatograms before and after the incubation of Plectranthus barbatus extract with: (a)

gastric juice, (b) pancreatic juice. *Indicates the residue of pancreatin. For the identification of the peak

numbers, refer to Figure 3.4.

There is very scarce information about the gastric metabolism of phenolic acids and even

less about terpenes. Polyphenols from procyanidin oligomers were hydrolyzed by stomach acid

conditions (Spencer et al., 2000). When analysing the effect of the stomach pH on ginsenosides,

deglycosylation was observed (Kong et al., 2009). In the present study deglucuronidation was

not observed at acidic pH and the hydrolysis of rosmarinic acid, forming caffeic acid, could not

be detected.

Pancreatic juice contains a mixture of hydrolytic enzymes, pancreatin (amylase, lipase

and protease), at basic pH, which was used to simulate the behaviour of the P. barbatus extract

under small intestine conditions (Yamamoto et al., 1999). The effect of the buffer solution at pH

8, without pancreatin, on compound degradation was also analysed. The study was carried out

for 4 h with five replicates. The extracts were analysed by HPLC and the chromatograms (Figure

3.8b) allowed the establishment of the graph shown in Figure 3.9. The abietane diterpenoid

decreased, whilst its hydroxylated form increased. Rosmarinic acid showed a 25% decrease in its

concentration (Figure 3.9). Luteolin 7‐O‐glucuronide, apigenin 7‐O‐glucoronide and acacetin 7‐O‐


52

glucoronide were kept constant. It seems that the glucuronide moiety was not hydrolysed from

the flavonoid by pancreatin.

When the activity after pancreatic digestion was analysed a decrease in the extract

inhibition capacity was noticed (Table 3.4). After 4 h, the inhibition activity of the extract

towards AChE decreased by 50%. The statistical analysis indicated that after the first hour of

digestion there was a significant decrease in the inhibition capacity of the extract, as well as

after 3 h digestion, meaning that the pancreatic juice produces a decrease in the biological

activity of the extract. The antioxidant activity was kept constant throughout the digestion.

Rosmarinic acid alone has a reduction of AChE inhibition activity of 25%, when subject to

pancreatic digestion (Table 3.4). It seems that the transformation of the abietane diterpenoid

must account for the other 25% reduction in the AChE inhibition activity. The rosmarinic acid

showed the same behaviour either pure or in the mixture, in both situations it was transformed

by 25%.

Table 3.4. Antiacetylcholinesterase and antioxidant activity of P. barbatus herbal tea after in vitro

gastrointestinal digestion. The action of the pancreatic juice on the inhibition activity of

rosmarinic acid (RA) is also shown.

Time (h) 0 1 2 3 4 6

AChE (%)

Gastric 100.0 ± 0.4 94.6 ± 0.6 92.9 ± 4.1 89.3 ± 2.5 89.6 ± 3.9 -

Pancreatic 100.0 ± 10.5 76.5 ± 8.3 61.2 ± 5.5 49.6 ± 6.9 44.2 ± 5.3 -

Pancreatic

(RA) 100.0 ± 1.1 86.0 ± 2.1 81.1 ± 3.0 79.8 ± 2.0 75.6 ± 2.4 -

Glucuronidase 100.0 ± 1.6 - 81 ± 8.2 - - -

Caco-2 100.0 ± 1.4 120.7 ± 1.7 119.2 ± 1.3 116.3 ± 3.6 - 105.6 ± 1.7

DPPH (%)

Gastric 100.0 ± 2.4 90.4 ± 1.0 100.5 ± 1.7 89.7 ± 0.6 88.5 ± 0.6 -

Pancreatic 100 ± 4.3 97.2 ± 6.2 94.2 ± 11.0 93.7 ± 6.1 103.3 ± 6.9 -

Glucuronidase 100.0 ± 0.6 - 81.7 ± 0.4 - - -

Caco-2 100.0 ± 5.0 86.1± 4.4 87.2 ± 2.6 - 96.1 ± 5.9 86.7 ± 3.9

The transformation of the abietane diterpenoid into its hydroxylated form was obtained with

the buffer used in the artificial pancreatic juice, even in the absence of the pancreatin, so this is

a chemical reaction that occurs at the pancreatic pH (pH 8.0). Epigallocatechin gallate, a

polyphenol, was also unstable under the alkaline conditions found in the intestinal juice (Yoshino

et al., 1999). On the other hand the alkaline conditions of the pancreatic juice brought about

hydrolysis of the abietane diterpenoid compound, but apigenin, luteolin and acacetin

glucuronide remained intact. Rosmarinic acid is 25% hydrolyzed. Cilla et al. (2009) also found a

Chapter III

53

reduction in all the phenolic compounds of a fruit beverage after pancreatic digestion, although

the chemical composition of the juice was not specified.

Figure 3.9. Variations in peak areas of compounds present in herbal tea after 4 h digestion with artificial

pancreatic juice. () Luteolin 7‐O‐glucuronide; () rosmarinic acid; () apigenin 7‐O‐glucuronide; ()

hydrolysed abietane; (○) acacetin 7‐O‐glucuronide; (●) abietane diterpenoid.

The results found here may give an explanation the traditional use as purgative

described for P. barbatus herbal tea (El‐Kamali, 2009). It is known that inhibitors of AChE

stimulate gastrointestinal motility (Cellek et al., 2008; Jarvie et al., 2008). The results shown here

point out that the activity of the herbal tea in the stomach is not reduced by the gastric juice,

and although by passing through the pancreatic juice it reduces its inhibitory capacity to 50% of

its initial activity, it may go on with its purgative activity through the gut. The fact that P.

barbatus herbal tea can act as an inhibitor of AChE may facilitate stomach and gut motility,

explaining the purgative effect described for this herbal tea.

2.3.3. Metabolism of the plant extract by Caco‐2 cells and biological activity of the final

products

The aim of this study was to simultaneously investigate if the extract components might

disappear from the culture medium, being incorporated into the Caco‐2 cells, or if any

extracellular protein could produce a change in the chemical composition of the extract.

The permeation of phenolic acids or flavonoids is not completely established.

Polyphenols seem to be absorbed through the intestinal barrier, since after their ingestion there

is a rising in the antioxidant capacity of blood plasma (Scalbert & Williamson, 2000). Flavonoids

like glycosylated quercetin seem to permeate into the bloodstream more easily than the parent


54

flavonoid (Hollman & Katan, 1997). On the other hand there are very few studies pointing out

that absorption of flavonoids occurs preferentially when they are deglycosylated by the

intestinal bacteria (Liu & Hu, 2002) and that the absorption of phenolic acids depends largely on

their structure. Rosmarinic acid is very slowly absorbed, whilst coumaric or ferulic acids, for

instance, are more easily absorbed (Konishi & Kobayashi, 2005). Therefore, further studies are

necessary in order to elucidate these procedures. As the extract used in the present

investigation is a mixture of glycosylated flavonoids and phenolic acids (like rosmarinic acid), this

study was carried out in order to further elucidate the absorption process.

After having analysed that the culture medium did not produce any change to the

extract chemical composition, the herbal tea was added to the culture medium in the presence

of Caco‐2 cells. The extract was used at concentrations from 1 to 10 mg/ml in the culture

medium. After evaluating the cells’ viability by the MTT method (results not shown) 1 mg/ml of

extract was chosen in order to accomplish this study. At this concentration the P. barbatus water

extract was not toxic to the Caco‐2 cells.

The extract remained in contact with the cells for 6 h and the aliquots withdrawn and

analysed by HPLC; this analysis revealed that there were no changes in the extract composition.

The results indicated that the Caco‐2 cells did not produce extracellular enzymes capable of

metabolizing the constituents of the herbal tea, and these compounds were not able to

penetrate the cells in an appreciable quantity, since the peak areas were kept constant. At the

end of the 6 h, a homogenate of the cells was prepared and the content inside the cells was

analysed. This test had the objective of knowing if any compound could be detected inside the

cellular system. In fact, a very small quantity of rosmarinic acid could be detected inside the cells

(retention time 9.3 min; Figure 3.10a).

The permeation of rosmarinic acid standard through Caco‐2 cells was studied. It could be

seen that only 10% of its concentration disappeared from the medium. Nevertheless when the

content inside the cells was analysed, a very small quantity of rosmarinic acid was found inside

the cell homogenate (Figure 3.10b). The UV–Vis spectrum of the compound present with

retention time around 9.38 min confirms that it is rosmarinic acid. Rosmarinic acid was

previously studied regarding its permeation through Caco‐2 cell layers, and it was verified that

permeation occurred mainly by paracellular diffusion. In addition, rosmarinic acid was not

hydrolyzed by Caco‐2 esterases (Konishi & Kobayashi, 2005). When a tablet containing

rosmarinic acid was given to healthy humans, 6.3% of the ingested rosmarinic acid appeared in

the blood in several metabolized forms (Baba et al., 2004). The behaviour of rosmarinic acid was

identical in the different extracts in spite of being inside a mixture of several flavonoids.

Chapter III

55

Figure 3.10. Chromatograms of compounds inside the Caco‐2 cells (homogenates) after 6 h contact with

(a) P. barbatus extract, (b) rosmarinic acid standard. 2, rosmarinic acid; *, residual peak from the Caco‐2

cells.

The final activity of the extract, that is, the activity after being in contact with the Caco‐2

cells, was not significantly modified. The antioxidant activity showed a slight decrease but it was

not statistically significant either (Table 3.4).

Polyphenols were neither metabolized nor absorbed by Caco‐2 cells (Cilla et al., 2009);

the glycosyl moiety may hamper the absorption. On the other hand, flavanols like catechin and

epicatechin may undergo significant metabolism and conjugation during absorption in the small

intestine and in the colon (Spencer, 2003). The work of Liu and Hu (2002) indicated that the

absorption through Caco‐2 cells was very low for glycosidic flavonoid derivatives.

2.3.4. Metabolism of the plant extract by the β‐glucuronidase from E. coli, biological

activities and Caco‐2 cells permeation of the final products

The enzyme β‐glucuronidase from E. coli was used to analyses the possibility of

transforming the herbal tea extract in order to obtain luteolin and apigenin, simulating some of

the enzymes of the intestinal bacteria. The HPLC chromatograms at the beginning of the

experiment and after 2 h of digestion with glucuronidase can be seen in Figure 3.11. It is noticed

that the peaks of the compounds containing glucuronide moiety diminish or even disappear

from the chromatogram, confirming the hypothesis of deglucuronidation of the compounds 1, 3

and 5. The aglycones appear in the chromatogram, peaks 8, 9 and 10. When analyzing the


56

remaining biological activity, it could be seen that there were no significant changes after this

enzymatic process (Table 3.4).

Figure 3.11. HPLC chromatogram after the action of β‐glucuronidase from E. coli on the herbal tea.

Compounds identified with numbers 1–7, refer to Figure 3.6. Peaks signaled with arrows: 8, luteolin; 9,

apigenin; 10, acacetin.

Probably the extract did not lose activity due to the fact that the aglycones (flavonoids)

also have antiacetylcholinesterase activity. Apigenin, luteolin and acacetin are the compounds

formed by the digestion with glucuronidase. The IC50 values for the inhibition of AChE were

determined in order to know if these compounds also might inhibit the enzyme after the extract

has passed through the intestine. Values of 100.6 ± 7.0 and 92.1 ± 2.4 µM were obtained for

apigenin and luteolin, respectively. The transformation of glycosylated flavonoids into their

aglycones does not cause any decreases in the antiacetylcholinesterase activity, because these

aglycones also show low IC50 values for the enzyme inhibition.

As the extract tends to form apigenin and luteolin aglycones by bacterial glucuronidase

activity, these compounds were used in Caco‐2 metabolic assay, as standards. The metabolism

was followed for 6 h by HPLC analysis of the cell medium. For each aglycone assay, a peak with

the retention time and UV spectrum of the respective glucuronide showed up. The evolution of

these peaks in the culture medium is shown in Figure 3.12a and b. After 6 h of assay, the

aglycone could still be found in the medium, but the majority of the flavonoid was present as

glucuronide. Inside the cells, the quantity of flavonoid found in the cell homogenates,

corresponded to 2.6 ± 0.1% glucuronide and 6.2 ± 1.0% aglycone for apigenin and 6.6 ± 4.1%

glucuronide and 8.4 ± 3.7% aglycone for luteolin. It seemed that the flavonoid could get inside

the cells, be glucuronidated and transported again to the medium. Ng and coworkers (2005) also

Chapter III

57

reported that glucuronidation of flavones occurs inside Caco‐2 cells, and the glucuronides were

transported back to the culture medium.

Figure 3.12. Variations in peak areas of: (a) apigenin and (b) luteolin. Each figure shows the metabolites

after 6 h in the presence of Caco‐2 cells. Peak areas are presented as percentage of the initial area. ()

aglycones and () glucuronides.

Gut microflora plays an important role in the digestion of phenolic compounds. Bacterial

glucuronidase activity affected the composition of P. barbatus extract by hydrolyzing the

flavonoid glucuronides into their respective aglycones. Although flavonoid glycosides showed

low absorption in Caco‐2 cells, their aglycones can get inside the cells, as evidenced by the study

with luteolin and apigenin. These aglycones were metabolized by these cells into their

glucuronides, in similar way to the one described by Ng et al. (2005). Both aglycones and

glucuronides were found within the cells; they do not seem to accumulate there but are

excreted to the medium.

2.4. Conclusion

The herbal tea of P. barbatus leaves, composed mainly of rosmarinic acid and lesser

quantities of acacetin 7‐O‐glucuronide, (16S)‐coleon E, apigenin 7‐O‐glucuronide, and luteolin 7‐

O‐glucuronide, can pass through the stomach without any modification, neither in the chemical

structure of the compounds nor in its biological activities. The rosmarinic acid and the abietane

diterpenoids present in the herbal tea were 25% and 100% hydrolyzed, respectively, through

action of the pancreatic juice. The antiacetylcholinesterase activity drops to 50% although the

antioxidant activity is kept constant. The permeation into Caco‐2 cells was only verified for

rosmarinic acid and only in a minute amount. The aglycones from apigenin 7‐O‐glucuronide and

luteolin 7‐O‐glucuronide could also permeate the cells and be transformed into the glucuronides

again. There was no change in the biological activity after 6 h in contact with the cells. These


58

results point out that P. barbatus herbal tea will reach the intestinal barrier with a reduction in

50% of its AChE inhibitory activity and without modification in its antioxidant activity. The

permeation through the intestinal barrier will be very low for the extract compounds, but may

increase due to their glucuronidase digestion by the gut microflora. The aglycones thus can be

formed and then metabolized by intestinal cells back into their glucuronides, as was shown here

for the Caco‐2 model. The P. barbatus extract showed the same activity with its flavonoids

glucuronidated or as aglycones. The AChE inhibition activity found during the gastrointestinal

digestion may explain, at least, the purgative effect found for this herbal tea.

Chapter III

59

3. Activity of Plectranthus barbatus extract against inflammatory

response in human neutrophils

3.1. Introduction

Inflammation is a complex biological response to harmful stimuli involving the vascular

system, the immune system and the injured cells. Although it has primarily a protective function,

the destructive effects can largely surpass the gravity of the stimulus. Therefore, serious

inflammatory responses are often associated with diseases, as atherosclerosis, allergies and

myopathies.

Polymorphonuclear neutrophils are white blood cells in mammals and form have an

essential part in the innate immune system. These cells, highly specialized in phagocytosis, have

developed mechanisms for intracellular digestion of particles, such as pathogens and cell debris,

involving the production of radical oxygen species (ROS) and an range of hydrolytic and

proteolytic enzymes.

Microorganisms areusually coated with opsonins (generally complement and/or

antibody) that bind to specific receptors on the surface of the phagocyte, inducing the

invagination of the cell membrane and the incorporation of microorganisms into an intracellular

phagosome. There follows a sharp but transient increase in oxygen uptake, which is used to

produce O2‐ (superoxide ion) by the one‐electron reduction of oxygen, a reaction catalysed by

NADPH oxidase at the expense of NADPH. Most of the superoxide reacts with itself forming H2O2

(hydrogen peroxide), and from these agents a large number of highly reactive oxidants are

formed, including HOCl (hypochlorous acid), which is produced by the myeloperoxidase‐

catalyzed oxidation of Cl‐ by H2O2; OH. (hydroxyl radical), produced by the reduction of H2O2 by

Fe2+ or Cu+; ONOO‐ (peroxynitrite), formed by the reaction between O2‐ and NO‐; and many

others. (Babior, 2000; Klebanoff, 2005) This battery of reactive oxidizing agents not only kills the

invading particles but also inflicts harm on nearby tissues, and is thought to be of pathogenic

significance in a large number of diseases such as emphysema, acute respiratory distress

syndrome, atherosclerosis, reperfusion injury, malignancy and rheumatoid arthritis (Babior,

2000). The myeloperoxidase‐catalysed reaction is directly related with neutrophil stimulation,

the resulting inflammatory process, tissue injury and related pathologies, and therefore serves

as a good marker for neutrophil production of reactive oxygen species (Klebanoff, 2005; Deby‐

Dupont et al., 1999; Zeraik et al., 2011).


60

Plectranthus barbatus Andrews (Lamiaceae), known as “falso boldo”, is used in South

Africa and South America for a wide range of therapeutic purposes. Some of the traditional uses

of P. barbatus suggest that its extract may have anti‐inflammatory activity, such as the

treatment of aches from diverse etiologies, burns, sores, insect bites, allergies, and to reduce

swelling on bruises (Lukhoba et al., 2006). Previous in vitro studies showed that a P. barbatus

extract, prepared as decoction, has promising antioxidant and antiacetylcholinesterase

properties, and the extract composition is almost unaltered during an in vitro simulated

digestion, especially in the content of its main component, rosmarinic acid (Porfirio et al., 2010).



The P. barbatus extract was prepared as a decoction, as described in Chapter II section 4.

Neutrophils were isolated from blood of human donors as described in Chapter II section 17.1.

and were activated by phorbol 12‐myristate 13‐acetate, to stimulate the production of reactive

oxygen species (Chapter II section 17.2.). The hypochlorous acid produced was measured by the

taurine chlorination method as described in section Chapter II section 17.2., in the presence and

in the absence of P. barbatus extract and rosmarinic acid.


The taurine chlorination by hypochlorous acid is a rapid chemical reaction that allows to

quantify hypochlorous acid in a sample, as the amount chlorotaurine formed (measured

spectrophotometrically) is directly proportional to the amount of hypochlorous acid (Weiss et

al., 1982). Therefore the decrease of taurin chlorination represented in Figure 3.13 shows a

decrease in hypochlorous acid production by human neutrophils relatively to a untreated control

(no antioxidants added), where 0% means that the neutrophils produced the same amount of

hypochlorous acid as the control, and 100% means that no hypochlorous acid was produced. To

evaluate the effect of the P. barbatus extract on the modulation of hipochlorous acid production

in neutrophils, we tested several concentrations of the extract. As it is shown in Figure 3.13a, the

effect is dose‐dependent, decreasing the production of hypochlorous acid with higher

concentrations of the plant extract. The value that inhibits 50% of the hypochlorous acid

production (IC50) was estimated as 10.7±3.1 µg.mL‐1. For isolated rosmarinic acid, the main

component of the P. barbatus extract, the same dose‐dependent effect was also observed

(Figure 3.13b), and the IC50 value was estimated as 3.62 ± 0.42 µM.

Chapter III

61

The amount of rosmarinic acid present in the plant extract was estimated by the peak

area in the HPLC chromatogram, comparing with a calibration curve with standard rosmarinic

acid. By plotting the amount of rosmarinic acid in neutrophil assay with P. barbatus extract

versus the decrease in taurine chlorination, together with the data with standard rosmarinic acid

(Figure 3.13b), it can be observed that the decrease in taurine chlorination is very similar. This

observation suggests that the decrease in the hypochlorous acid production by the P. barbatus

extract is mainly due to its content in rosmarinic acid.

Figure 3.13. Decrease of taurine chloration in the presence of several concentrations of P. barbatus

extract (a and b) or standard rosmarinic acid (b). The concentration of P. barbatus is expressed in µg.mL‐1

(a) or by its content in rosmarinic acid in µM (b).

Several studies have demonstrated the decrease of ROS released by activated

neutrophils in the presence of several plant extracts. The P. barbatus extract induced a decrease

in the release of hypochlorous acid in the same magnitude of Ribus nigrum water‐acetone

extract (Tabart et al., 2011), but was less potent than the ethanolic extract of Baccharis trimera

(Padua et al., 2010) and pomegranate peel extract (Bachoual et al., 1998), which showed IC50

values of approximately 0.5 µg.mL‐1 and 0.1µg.mL‐1, respectively. These values, however,


62

correspond to the decrease in total amount of ROS, not only hypochlorous acid as in the present

study.

To our knowledge no studies were performed on the effect of rosmarinic acid in

activated neutrophils. However, it has been shown that, when administered to rats, rosmarinic

acid reduces the inflammatory response and the myeloperoxidase activity after induction of

inflammation by 12‐tetracanoylphorbol 13‐acetate (TPA) (Osakabe et al., 2004) or by septic

shock (Jiang et al., 2009). The effect of rosmarinic acid in TPA‐induced inflammation was also

observed by the administration of Perilla frutescens extract, another Lamiaceae species, whose

main component is rosmarinic acid, but this effect was not observed with the flavonoid luteolin

(Osakabe et al., 2004).

In the present study the activity of the plant extract seems to be mainly due to the

rosmarinic acid content, however other studies showed that flavonoids may also be active.

Myricetin induced 72.7% decrease of hypochlorous acid in neutrophils with a concentration of

100 µM (Meotti et al., 2008), and isoorientin – a luteolin glucoside derivative – decreased 66% of

the release of ROS in with a concentration of 8.9 µM (Zeraik et al., 2011). These effects seemed

to be due to the inhibition of myeloperoxidase, as the same authors observed that the enzyme

activity was inhibited in the presence of those flavonoids (Meotti et al., 2008; Zeraik et al.,

2011). The hydroxylation pattern seems also to play an important role in the inhibition of

myeloperoxidase (Meotti et al., 2008), and quercetin isoquercetin and rutin showed IC50 values

in nanomolar concentrations for in vitro assays (Regasini et al., 2008). In in vivo studies a

decrease superior to 50% in myeloperoxidase activity in the lungs of rats treated with

70µmol.Kg‐1 of luteolin after acute lung injury (Kuo et al., 2011).

The decrease of ROS released by activated neutrophils in the presence of plant phenolic

compounds seems to be mainly due to the inhibition of myeloperoxidase. Bachoual and co‐

workers (1998) showed that pomegranate peel extract does not inhibit NADPH oxidase or the

formation of superoxide anion, and does not scavenge hydrogen peroxide. The same authors

observed that the inhibition of myeloperoxidase was the only biochemical effect related to ROS

production in the presence of pomegranate extract, and therefore concluded that it may be the

primary cause of the ROS decrease.

This preliminary study suggests that P. barbatus extract and rosmarinic acid may be

potentially useful in the treatment of inflammation related health conditions, especially if

P. barbatus extract components are to be found circulating in the bloodstream after the oral

administration of the water extract. The decrease of the amount of hypochlorous acid produced

by neutrophils may explain the traditional use of P. barbatus in the treatment of swellings and

Chapter III

63

pain related to inflammatory processes (Lukhoba et al., 2006). However further studies would

be useful to enlighten some aspects related to the mechanism of action of the plant extract

components, especially if the action is intracellular or extracellular. Intracellular ROS are

responsible for the destruction of phagocytised particles, while extracellular myeloperoxidase

and ROS, although playing a role in the defence mechanisms, are the main cause of damage by

inflammation (Deby‐Dupont, 1999; Klebanoff, 2005).

3.4. Conclusions

P. barbatus extract was able to scavenge hypochlorous acid produced by the activated

neutrophils in a dose‐dependent process.

The content of rosmarinic acid in the plant extract seems to be the main responsible for

the hypochlorous scavenging activity presented by the extract.

The low IC50 values shown by the plant extract and its main component for the

scavenging of hypochlorous acid, produced by human neutrophils, allow to consider that

P. barbatus herbal tea and rosmarinic acid may be useful in the treatment of inflammation

related health problems.


64

4. Conclusions

In summary, the in vitro tests allowed to conclude that all the Plectranthus species

tested showed antioxidant and anticholinesterase activity due to their content in rosmarinic

acid. In contrast, the activities shown by Plectranthus barbatus were not only due to rosmarinic

acid, but also to other compounds, which may explain some health benefits that are attributed

to this aqueous extracts in folk medicine.

The major component of P. barbatus aqueous extract is rosmarinic acid, but it also

contains flavonoid glucuronides (apigenin 7‐O‐glucuronide, luteolin 7‐O‐glucuronide and

acacetin 7‐O‐glucuronide), and diterpenoids. During in vitro digestion of the extract with gastric

juice and pancreatic juice all compounds are stable except one of the abietane diterpenoids,

which suffers hydroxylation in pancreatic conditions, and a consequent reduction of the

acetylcholinesterase inhibitory activity was observed. The flavonoid glucuronides were

deglucuronidated by β‐glucuronidases from gut microbiota, but the aglycones can be

glucuronidated by intestinal cells, as it was shown with the Caco‐2 cell model.

The P. barbatus aqueous extract also showed a high in vitro anti‐inflammatory activity,

by reducing the amount of hypochlorous acid formed by activated neutrophils. The content of

rosmarinic acid in the plant extract seems to be the main responsible for this activity.

Chapter IV

Bioavailability studies in rats and Caco‐2 cell monolayers

Falé PL, Madeira PJ, Florêncio MH, Ascensão L, Serralheiro ML. 2011. Function of Plectranthus barbatus

herbal tea as neuronal acetylcholinesterase inhibitor. Food Funct. 2: 130‐136.

Falé PL, Ascensão L, Serralheiro MLM. Bioavailability of rosmarinic acid, luteolin, apigenin‐modelling

plant herbal teas through Caco‐2 cell monolayers. To be submitted to Food Chem.

Chapter IV

67

1. Plectranthus barbatus aqueous extract bioavailability and

resulting neuronal acetylcholinesterase inhibition in rats

1.1. Introduction

Herbal teas may be considered as functional drinks (Percival et al., 2007), as indeed

almost all of the chemical components of these water extracts possess some biological function

in the human body. Some of the ethnobotanical uses of these herbs can be explained through

the biochemical activities found and described in the literature either for the complete extracts

or for the isolated compounds. Leaves of Plectranthus barbatus (Lamiaceae) were studied

previously concerning the antiacetylcholinesterase as well as the antioxidant activity. The

activities found in the P. barbatus herbal tea could be attributed to its main constituent,

rosmarinic acid, together with other compounds, although present in much lesser quantity,

abietane diterpenoids and flavonoid glucuronides (Falé et al., 2009; Porfirio et al., 2010). The

function of the herbal teas depends on the metabolism that the compounds present in the

extract may be subject to during the gastrointestinal digestion process. The compounds may be

transformed into metabolites with different biological activity compared to the one initially

determined. In the case of P. barbatus, some of the active compounds found were transformed

when the extract was submitted to in vitro conditions simulating the gastrointestinal tract, what

caused a small decrease in the biological activity (Porfirio et al., 2010). The fact that the herbal

tea could pass the digestive tract and keep some of its function lead to an in vivo experiment in

order to see if the compounds present in the water extract could reach the brain and still be

active there. Recently, for instance, an ethanol extract of Tabernaemontana divaricata proved to

be effective in inhibiting neuronal acetylcholinesterase when administered to rats (Chattipakorn

et al., 2007).

The quantification of the metabolites of the herbal tea components in the blood stream,

and the study of the remaining biological activity in the target organ, especially on what

concerns the acetylcholinesterase activity, are topics that are seldom referred to in scientific

papers. Therefore, the aim of this study was to investigate if the active compounds present in

the herbal tea of P. barbatus, or their derivatives, when administered to rats were found in the

blood stream and in the brain, and if the neuronal acetylcholinesterase activity was affected by

the herbal tea administration.

Bioavailability Studies

68




To know what metabolites to expect after the extract administration to rats, standards of the

metabolites were produced in vitro (Chapter II section 11). Rat liver enzymes (Chapter II section

11.1) were used to produce glucuronide derivatives of the plant extract components as

described in Chapter II section 11.2. Catechol o‐methyltransferase (COMT) from porcine liver

was used to make methylated derivatives of rosmarinic acid, as described in section Chapter II

section 11.3.

The P. barbatus extract and its main component, rosmarinic acid, were administered to

rats as described in Chapter II section 12.1. Plasma and brain were collected from the rats 30

and 60 minutes after administration, and treated as described in section II.12.2. The samples

were analyzed by HPLC‐DAD (Chapter II sections 12.4. and II.7.) and the HPLC samples were pre‐

treated as mentioned in Chapter II section 12.3 to determine the glucuronidated and sulfated

metabolites. The acetylcholinesterase activity of the rat brains was determined as described in

Chapter II section 12.5.


P. barbatus herbal tea proved to have antiacetylcholinesterase activity in previous

studies, with an IC50 of 1.02 ± 0.02 mg of dry leaves ml‐1 and this activity was kept constant after

in vitro gastric digestion and lost approximately 50% after the in vitro pancreatic studies (Falé et

al., 2009; Porfirio et al., 2010). Rosmarinic acid, luteoline 7‐O‐glucuronide, apigenine 7‐O‐

glucuronide, two abietane diterpenoid, acacetin 7‐O‐glucuronide and (16S)‐coleon E, were the

compounds identified in the water extract (Porfirio et al., 2010), being rosmarinic acid the main

component. Although all the compounds demonstrated inhibition activity relatively to AChE, an

IC50 of 0.44 mg.ml‐1 for the main component was determined (Falé et al., 2009). Inhibition

studies demonstrated that the process was reversible (unpublished studies). Due to these

previous results, the study was continued by analyzing the action of P. barbatus herbal tea in

vivo. In the present work the metabolism of the herbal tea after intragastric and intraperitoneal

administration to laboratory animals was analyzed.

1.3.1. Intragastric administration of P. barbatus extract

Plasma. After intragastric administration of P. barbatus extract, the plasma was analyzed

by HPLC and rosmarinic acid was the only compound detected. In order to confirm if this was

Chapter IV

69

indeed the only compound present in the plasma or if some metabolites could be present but in

a low amount that was beneath the detection limit of the system, β‐glucuronidase and sulfatase

were added and allowed to react under the conditions described in Chapter II, section 12.3. The

plasma was analyzed once again by HPLC, and this time the rosmarinic acid showed an increase

in its area and the aglycones from the flavonoid derivatives were also detected. The results from

this study confirmed that not only the presence of rosmarinic acid glucuronide and sulfo‐

derivatives, but also the presence of the flavonoid glucuronide derivatives in the plasma. The

concentrations of these rosmarinic acid metabolites found in the plasma are shown in Table 4.1.

It can be seen that the derivatives of rosmarinic acid in circulation in the blood stream decrease

after 30 min. This study indicated that, in fact, the flavonoid derivatives found initially in the

herbal tea could pass through the gastrointestinal barrier and appear in the plasma (Table 4.1). A

vestigial peak with retention time corresponding to acacetin was also found. The quantity of

total rosmarinic acid found in the plasma relatively to the amount of rosmarinic acid

administered in the extract can be calculated, assuming that a male Sprague‐Dawley rat contains

4.12 ml plasma per 100 g body weight (Probst et al., 2006), as 0.009% and 0.005% for 30 and 60

min, respectively.

Table 4.1. Concentration of rosmarinic acid, its metabolites and flavonoid glucuronide derivatives in the plasma and in

the brain, 30 and 60 min after the intragastric and intraperitoneal administration of P. barbatus extract.

Compound

Concentration in plasma Concentration in the brain

Intragastric administration (nM)

Intraperitoneal administration (µM)

Intraperitoneal administration (µM)

Time (min) 30 60 30 60 30 60

P. barbatus extract

Rosmarinic Acid (RA)

<0.5 10.1 1440±144 745282 24.11.1 20.40.4

RA Glucuronides

112.1 87.8 - - - -

RA Sulfates 210.4 77.3 - - - -

RA methyl - - 17.82.3 11.13.6 - -

Luteolin glucuronide

1.9 - 40.55.9 29.98.0 - -

Apigenin glucuronide

6.5 - 20.03.4 3.70.3 - -

Rosmarinic acid standard

RA 4.9 x 103 1.9 x 103 1042.3192.1 - 25.40.9 -

RA methyl - - 16.75.6 - - -


70

Rosmarinic acid was also intragastrically administered, in a higher quantity than that

found in the herbal extract, and analyzed under the same conditions. Rosmarinic acid was

present in the plasma 30 min and 60 min after the intragastric administration. The quantities

found (Table 4.1) corresponded to 0.036% and 0.015% of the amount of rosmarinic acid

administered to rats, 30 and 60 min after administration, respectively. The results one hour after

administration showed a decrease in the rosmarinic acid concentration in circulation.

The test with β‐glucuronidase was also carried out, but an increase in the rosmarinic acid

area in the HPLC chromatogram was not detected. This means that there were no

glucuronidated metabolites of this phenolic acid formed in the plasma. It seems that rosmarinic

acid, as pure compound or inside the herbal tea, behaves differently.

Brain AChE. When rat brains were analyzed by HPLC no compounds or metabolites of P.

barbatus extract or rosmarinic acid were found. However, when the brain acetylcholinesterase

activity was measured, a decrease of 10.0 ± 1.8% and 5.5 ± 1.7% of the enzymatic activity was

observed, 30 and 60 min after the extract administration, respectively (Table 2), from the control

value of activity of 0.149 ± 0.002 nmol min‐1 mg‐1. In the case of rosmarinic acid injection a 13.5%

decrease in AChE activity was detected after 30 min.

Table 4.2. Brain acetylcholinesterase inhibition (%) 30 and 60 min after administration (intragastric and

intraperitoneal) of rosmarinic acid and P. barbatus extract. Results significantly different from the control

are marked with * (P < 0.05) and ** (P < 0.1). Values that are not significantly different (P < 0.05) are

marked from a to d.

Brain acetylcholinesterase inhibition (%)

Intragastric

administration

Intraperitonial

administration

P. barbatus 30 minutes 10.0±1.8*a 29.0±2.3*c

60 minutes 5.5±1.7**b 24.9±3.7*d

Rosmarinic acid 30 minutes 13.5±1.7*a 10.7±5.0**ab

60 minutes 12.8±2.4**a -

When standard rosmarinic acid was administered intragastrically the concentration of

rosmarinic acid found in blood showed values in the same magnitude of those found in other

studies (Baba et al., 2004; Konishi et al., 2005). However, in the present work, when rosmarinic

acid was present within the extract, a lower quantity of this acid was found in plasma, suggesting

that the other extract components may interfere in the gut permeation of rosmarinic acid.

Previous pharmacokinetic studies also reported that the highest concentration of rosmarinic

acid was reached in less than 30 min after intragastric administration. Methylated and sulfated

Chapter IV

71

forms of rosmarinic acid were found, as well as simpler hydroxycinnamic acids generated by the

hydrolysis of rosmarinic acid (Baba et al., 2004; Konishi et al., 2005; Nakazawa & Ohsawa, 1998).

The present study also observed a decrease in the rosmarinic acid concentration after 30

min, but the methylated form of rosmarinic acid was not detected. This suggests that when

rosmarinic acid is administered alone, it can pass the intestinal barrier without undergoing

metabolisation by the intestinal cells, but it can be glucuronidated and sulfated as confirmed by

the action of β‐glucuronidase and sulfatase.

Taking into account that the amount of rosmarinic acid in the extract was approximately

one third of the rosmarinic acid standard, it is seen that the decrease in the enzyme activity is

not linearly correlated with the rosmarinic acid concentration. This is due to the fact that the

extract contains other compounds besides rosmarinic acid. These compounds also inhibit

acetylcholinesterase (Porfirio et al., 2010) and the present study indicates that they can also

reach the brain and act together with rosmarinic acid as enzyme inhibitors. This fact is

corroborated by the analysis of the rosmarinic acid concentration in the plasma (less than

0.5 nM when the herbal tea is administered and 4.9 mM when the standard is given to the

laboratory animals) and the enzyme inhibitory activity, which is similar in both situations. These

results can only be explained with the inhibitory activity of all the compounds, including the

rosmarinic acid derivatives, towards AChE.

1.3.2. Intraperitoneal administration of P. barbatus extract

Plasma. After intraperitoneal administration of P. barbatus aqueous extract, rat plasma

was analysed by HPLC (Figure 4.1a). With the exception of the diterpenoid 6, all the extract

compounds were present in rat plasma in higher amount at 30 min than 60 min after injection

(Figure 4.1a). The abietane diterpenoids 4 has a structure similar to the abietane diterpenoid 6,

and is present in the extract as a minor compound. Previous studies reported that it may be

formed from abietane diterpenoid 6 in the small intestine conditions (Porfirio et al., 2010), this

suggests that abietane diterpenoid 6 might have been completely transformed into abietane

diterpenoid 4 under the biological blood‐circulation conditions.

Rosmarinic acid, compound 2, is the major compound found in the chromatogram

(Figure 4.1a). The amount of rosmarinic acid present in the plasma was calculated to be

1439.8 ± 144.0 mM and 744.5 ± 282.3 mM for 30 and 60 min after the extract administration,

respectively (Table 4.1). The amount of rosmarinic acid found in the plasma corresponded to

23.7 ± 2.4% and 12.3 ± 4.3% of the administered amount. In the chromatogram, a peak of

methylated rosmarinic acid, 2m, could be detected. According to a standard previously


72

developed by using a commercial COMT from porcine liver and SAM as cofactor in a

methodology described in the Experimental section, it was possible to prove the formation of a

methylated derivative from rosmarinic acid. The amount of methyl rosmarinic acid was 17.8 ±

2.3 mM and 11.1 ± 3.6 mM for the same administration time‐points (Table 4.1). The search for

glucuronidated/sulfated metabolites was carried out by reacting β‐glucuronidase and sulfatase

with the plasma. The only increase in area was observed for flavonoid aglycons, corresponding

to the decrease in the corresponding area of the respective glucuronide peaks. The fact that no

significant amounts of glucuronidated/sulfated rosmarinic acid metabolites were found after

intraperitoneal administration could be explained by the predominance of methylation occurring

in the liver over other metabolisation reactions, such as sulfation and glucuronidation. A similar

situation was also reported by O’Leary and co‐workers (2003).

Figure 4.1. HPLC analysis, 30 and 60 min after intraperitoneal administration of P. barbatus extract, of (a)

plasma and (b) brain. 1: luteolin 7‐O‐glucuronide (retention time 9.6 min); 2: rosmarinic acid (RT: 10.4); 3:

apigenin 7‐O‐glucuronide (RT: 11.2); 4: abietane diterpenoid (RT: 13.8); 5: acacetin 7‐O‐glucuronide (RT:

15.1); 2m: monomethylated rosmarinic acid; 1g: luteolin glucuronide derivative; 3’: apigenin.

In the chromatogram shown in Figure 4.1a, a different glucuronide of luteolin than the

one initially present in the water extract (compound numbered 1g) could be detected.

Compound 1g was identified after preparing, in vitro, the glucuronidation derivatives according

to Materials and Methods. In this chromatogram the aglycone apigenin (compound numbered

3’) could also be seen. No luteolin or acacetin aglycones were detected in the plasma samples.

These results suggest that the deglucuronidation of plasma apigenin glucuronide may be the

Chapter IV

73

main reaction occurring to this compound. Studies concerning the oral administration of luteolin

suggested that the presence of the unconjugated luteolin in plasma might be due to the

permeation of the aglycone without undergoing conjugation by the intestinal cells (Shimoi et al.,

1998). However, in the present study apigenin was already administrated in its glucuronidated

form, and so the high amount of aglycon found in plasma must be due to deglucuronidation of

the apigenin glucuronide. The deglucuronidation of flavonoid glucuronides by the liver was

previously reported by O’Leary and co‐workers (2003). These authors remarked that the

conversion of the quercetin‐7‐ and quercetin‐3‐glucuronide to the mono‐sulfate conjugate

showed an intermediate step of deglucuronidation by β‐glucuronidase activity, allowing

transient contact of the free aglycone with the cellular environment.

The detection only of the apigenin aglycone in plasma may be due to a higher

deglucuronidation of the apigenin glucuronide, comparatively to the other flavonoids and/or to

a lower re‐glucuronidation of apigenin by the liver. The in vitro glucuronidation assay showed a

lower rate of glucuronidation for apigenin than for luteolin, which may explain the existence of

apigenin aglycone relatively to luteolin. This hypothesis is also supported by the appearance of a

luteolin glucuronide, which could only be produced by deglucuronidation of the extract’s

luteolin 7‐O‐glucuronide, followed by glucuronidation in a different position (Figure 4.1a). Little

is known about these de‐conjugation and re‐conjugation reactions in flavonoids, especially when

they are administered in their glucuronidated forms. The present study differs from the majority

of the in vivo flavonoid administration reports in the fact that the several flavonoids were

simultaneously administered as glucuronide derivatives, as they are commonly found in herbal

teas, instead of their aglycones.

All extract compounds showed a decrease from 30 to 60 min after extract

administration, with rates of elimination ranging from 25% to 85% (Table 4.3), the lowest value

for the luteolin glucuronide with retention time 13.8 min, numbered 1g (24.4 ± 6.9%), and the

highest value for the apigenin glucuronide, numbered 3 (84.5% ± 4.6%). Only the aglycone from

apigenin (retention time of 17.3 min, numbered 3’) (Figure 4.1a) was seen. No luteolin or

acacetin aglycones were detected in the plasma.

Brain AChE. After intraperitoneal administration of P. barbatus extract, brains from rats

were also analyzed after 30 and 60 min by HPLC (Figure 4.1b). The chromatograms showed only

the presence of unconjugated rosmarinic acid, the other extract compounds or their metabolites

were not detected (Figure 4.1b). No increase of the rosmarinic acid peak area or the aglycones

from the flavonoid derivatives were observed after the reactions with β‐glucuronidase or

sulfatase. The amount of rosmarinic acid in rat brains was quantified and a relationship with the


74

amount of rosmarinic acid in the plasma was calculated (Table 4.1). The amount of rosmarinic

acid found in the brain was 0.40 ± 0.02% and 0.34 ± 0.01% of the administered amount, for 30

and 60 min after administration, respectively. Although the plasma rosmarinic acid

concentration showed a decrease from 30 to 60 min (around 52%, Table 4.3) the brain

rosmarinic acid concentration decreased around 15% (14.90 ± 7.11%, Table 4.1). This difference

is reflected in the ratio of brain/plasma rosmarinic acid concentration, which is higher 60 min

after the extract administration than at 30 min (0.027 at 60 min and 0.016 at 30 min). The

brain/plasma concentration ratio reflects the permeability of the compounds from the plasma to

the brain, which may be restricted primarily by the blood brain barrier, and ranges from 0 (no

permeability) to 1 (total permeability) (Zheng et al., 2006). These ratios indicate that rosmarinic

acid has the ability to pass to the brain, but the permeability is low.

Table 4.3. Retention time of compounds from the P. barbatus extract found in plasma and their decrease

from 30 to 60 min after the extract intraperitoneal administration.

compound Retention Time (min) Decrease from 30 to 60 minutes

(%)

Luteolin 7-O-glucuronide (1) 9.7 44.91±13.48

Rosmarinic Acid (2) 10.5 51.66±15.60

Apigenin 7-O-glucuronide (3) 11.3 84.57±4.59

Monomethylated Rosmarinic Acid (2m) 12.0 41.7±14.5

Abietane Diterpenoid (4) 13.5 36.33±15.61

Luteolin glucuronide (1g) 13.8 24.36±6.88

Acacetin 7-O-glucuronide (5) 15.2 56.75±0.10

Apigenin (3’) 17.3 33.4±12.8

Rosmarinic acid diffuses slowly to the brain, but it is retained there longer than in the

plasma, where a higher decrease was observed from 30 to 60 min. Brains of rats to which

standard rosmarinic acid was administered, (results not shown) showed the same concentration

of rosmarinic acid as when rosmarinic acid was given within the extract, leading to the

conclusion that the other extract compounds do not interfere with the permeability of

rosmarinic acid in the blood‐brain barrier, in opposition to the gut barrier where a interference

of the other extract compounds was observed in the permeability of rosmarinic acid.

Thirty minutes after the extract administration the rat brain showed an inhibition in

acetylcholinesterase activity of 29.0 ± 2.3%, and after 60min an inhibition of 24.9 ± 3.7% (Table

4.2), in comparison with the control rats, where the activity was 0.145 ± 0.012 nmol.min‐1.mg‐1.

Brains of rats where standard rosmarinic acid was administered showed the same concentration

Chapter IV

75

of rosmarinic acid as that detected when the extract was administered, but demonstrated to

have less acetylcholinesterase inhibition (10.7 ± 5.0% after 30 min). This confirms the inhibition

of AChE by the other extract components, although they were not detected by HPLC.

The percentages of acetylcholinesterase inhibition achieved were of the same

magnitude of the ones reported to galanthamine in the rat model, however the amounts of

interaperitoneal injection of galanthamine were lower: 3 mg.kg‐1 for 10% inhibition (Geerts et

al., 2005) and 10 mg.kg‐1 for 28% inhibition (Chattipakorn et al., 2007). Higher inhibitions can be

obtained with lower amounts of donepezil, 39% with 3 mg.kg‐1 (Geerts et al., 2005), however,

this inhibitor may present a higher incidence of adverse effects in high amounts. The

dosage/activity in humans for galanthamine is 16–24 mg.day‐1 to 30–40% brain

acetylcholinesterase inhibition (Kadir et al., 2008), and for donepezil is 10 mg.day‐1 to 19–27%

inhibition (Bohnen et al., 2005). As there is an interest in decreasing the adverse peripheral

effects of acetylcholinesterase inhibition, mostly related to gastrointestinal and hepatic

disturbances, herbal teas may offer an alternative for mild treatments of Alzheimer’s disease. P.

barbatus water extracts are traditionally prepared as decoctions (as in the present report) and

drunk by people and those secondary effects were not reported (Lukhoba et al., 2006). In fact,

the use of this plant extract to treat gastrointestinal and hepatic conditions is widely reported

(Lukhoba et al., 2006). Rosmarinic acid when administered in low doses seems to act as an

anxiolytic‐like compound, only when administered in high doses it seem to act on the peripheral

nervous system (Pereira et al., 2005).

Several recent reports showed that flavonoids, such as quercetin and its metabolites,

have the ability to reach the brain (Huebbe et al., 2010). Coleta and co‐workers (2008)

demonstrated that intragastrically‐administered luteolin, or its metabolites, should reach the

brain, causing anxiolytic‐like effects. Gujinski and co‐workers (2009) reported that ethanolic

extracts from Melissa officinalis, a Lamiaceae species, and its main component rosmarinic acid,

when administered to rats produced dose‐related antinociception in several models of chemical

pain, through mechanisms that involved cholinergic systems. Other studies also show that

rosmarinic acid, when administered through intraperitoneal injection (2–8 mg.kg‐1), induced

neurobehavioral changes in rats (Pereira et al., 2005). Although rosmarinic acid in the brain was

not quantified in those studies, its presence in the brain was suggested by the activities it

exerted in vivo.

The bioavailability of the currently used acetylcholinesterase inhibitors is highly variable,

depending mostly to the structure of the compounds, their metabolisation and elimination, and

their ability to permeate the intestinal and blood‐brain barriers (McGleenon et al., 1999). Their


76

time of action also depends on the mode of acetylcholinesterase inhibition. Donepezil, for

instance, has a peak plasma level 4 h after administration and a half‐life of over 70 h, while

tacrine shows low bioavailability when taken orally and has a half‐life in plasma of 1.4 h

(McGleenon et al., 1999). However, tacrine reaches a brain concentration that is 10‐fold that of

plasma, its dosage and brain bioavailability are not proportional, and so higher doses and

multiple dosing are used to increase bioavailability and the half‐life (McGleenon et al., 1999).

In the present study rosmarinic acid and the other components of the extract also do not

seem to follow a direct relation between dose and response. First, because the intestinal

absorption and metabolisation of rosmarinic acid is influenced by the presence of the other

extract compounds. Also, although the concentration of rosmarinic acid was always much lower

in brain than in the plasma, the relationship between plasma and brain rosmarinic acid

concentrations did not seem to be linear, as a decrease of more than 50% in plasma rosmarinic

acid concentration corresponded just to a slight decrease in the brain (Table 4.1). However, the

amount of detected rosmarinic acid and the activity seem to be related, as the decrease in the

brain acethylcholinesterase inhibition was similar to the decrease in the content in rosmarinic

acid. As the phenolic compounds from the plant extract seem to act in the brain in very low

concentrations, and their amount in brain has less fluctuations than the amount in the plasma,

taking a large volume of herbal tea during a long period of time, and several times per day, as it

is usually taken, may be the correct way of increasing its effects.

Apart from the beneficial effect of increasing the acetylcholine mediated neuronal

transmission, the treatment of Alzheimer’s disease with acetylcholinesterase inhibitors also

seems to protect free radical toxicity, β‐amyloid injury and to attenuate cytokine release from

microglial cells (Tabet, 2006). Apart from this ‘‘cholinergic anti‐inflammatory pathway’’,

rosmarinic acid is also a well‐known antioxidant and free radical scavenger (Falé et al., 2009). As

far as we know, this is the first report confirming the neurologic activity of rosmarinic acid by the

quantification its amount in the brain.

1.4. Conclusions

In this study it was demonstrated that the rosmarinic acid, present in herbal teas, may

cross the intestinal barrier, as well as the blood brain barrier, and be detected in the brain,

where it inhibits the enzyme AChE. The metabolisation and bioavailability of the herbal tea

components is different from the administration of pure compounds.

Chapter IV

77

2. Bioavailability of mixtures of rosmarinic acid, luteolin and

apigenin through Caco‐2 cell monolayers, modeling the

bioavailability of plant herbal teas.

2.1. Introduction

Herbal teas are among the oldest and most common drinks in human culture and are

drunk for their pleasant flavour and/or for their health improving properties. These plant

aqueous extracts are often composed of phenolic compounds such as phenolic acids and

flavonoids, which have proved to be therapeutically active (see Komes et al., 2010 for the

composition of several taxonomically unrelated medicinal plants).

Previous studies demonstrated that the Plectranthus barbatus herbal tea has relevant in

vitro antioxidant and anti‐acetylcholinesterase activities, which are due to its main components:

rosmarinic acid, flavonoid glucuronides (apigenin, luteolin and acacetin glucuronides), and

abietane diterpenoids (Falé et al., 2009; Porfirio et al., 2010). Rosmarinic acid and the flavonoid

components remained after in vitro digestion of the P. barbatus extract with gastric and

pancreatic juices, but flavonoid glucuronides were hydrolysed into their aglycones by the

β‐glucuronidase of the microorganisms in the gut flora (Porfirio et al., 2010). When P. barbatus

extract was intragastrically administered to rats its components were found in the plasma and a

decrease in brain acetylcholinesterase activity was observed (Falé et al., 2011).

As herbal teas are complex mixtures of compounds, we aimed to determine the effect of

the presence of rosmarinic acid, luteolin and apigenin in the permeation and metabolisation of

each other in Caco‐2 cell monolayers, an in vitro model for the intestinal absorption and

metabolism. Rosmarinic acid, apigenin and luteolin are the main constituents of P. barbatus

herbal tea, but they are very commonly found together in many other herbal teas and

foodstuffs. However, little is known about the way that these compounds affect the

bioavailability of each other. Nevertheless, it is known that phenolic acids and flavonoids

interact with transport systems in intestinal cells, such as the monocarboxylic acid transporters

(MCT) (Konishi et al., 2003; Konishi et al., 2004), the ABC transporters P‐glycoprotein (Pgp) and

multidrug resistance proteins (MRP), being actively transported or inhibiting them (for review

see Brand et al., 2006). The MCT4, MCT5, MRP1, MRP3 and MRP5 transporters, located in the

basolateral membrane of the intestinal cells, promote the transport of their substrates to the

bloodstream, while Pgp and MRP2, located in the apical membrane, promote the transport of

their substrates from the intracellular compartment to the intestinal lumen (Gill et al., 2005,


78

Konishi et al., 2003, 2004; Brand et al., 2006). The MCT1 transporter is located in the apical

membrane, but transports its substrates from the intestinal lumen to the intracellular

compartment (Gill et al., 2005).

The aim of this study is to evaluate the permeation through Caco‐2 cell monolayers of

the aqueous extract of P. barbatus, and of mixtures of its main phenolic compounds (rosmarinic

acid, luteolin, apigenin), in different concentrations, but in a relative proportion similar to the

usually found in herbal teas. The effect of two intestinal epithelium transporters, Pgp and MCT,

on the bioavailability of the polyphenolic mixture was also investigated.

2.2. Material and Methods



The bioavailability studies were performed in Caco‐2 cell monolayers as referred in Chapter II

section 17. This methodology was used to evaluate the bioavailability of P. barbatus extract and

of mixtures of phenolic compounds (rosmarinic acid, luteolin and apigenin) according to a

factorial design known as central composite design (CCD) (Barker, 1985), as described in Chapter

II section 17.1. The concentrations used in the mixtures are listed in Table 2.1.

The involvement of the transport systems MCT and Pgp in the bioavailability of a mixture

of rosmarinic acid, luteolin and apigenin was evaluated using the substrates for these systems,

benzoic acid and digoxin, as described in Chapter II section 17.2.

The HPLC analysis was carried out as described in Chapter II section 17.3.

2.3. Results

Previous studies with Plectranthus barbatus herbal tea revealed the presence of

rosmarinic acid and flavonoid glucuronide derivatives together with two abietane diterpenoids

(Porfirio et al., 2010). Rosmarinic acid, the poplyphenol present in higher amount in the water

extract, is a characteristic secondary metabolite of species from the Lamiaceae family, where

P. barbatus is included. In the present study we evaluated the intestinal bioavailability of the

aqueous extract of P. barbatus as a model extract, and of mixtures of three of its main

components, rosmarinic acid, luteolin and apigenin. We aimed to evaluate the differences in

bioavailability of these compounds, which are very commonly found in plant infusions and

decoctions.

Chapter IV

79

2.3.1. Bioavailability of Plectranthus barbatus herbal tea

In order to study the bioavailability of the major polyphenol compounds in the herbal

tea of P. barbatus, the water extract was applied to the apical side of a Caco‐2 cell monolayer

and the polyphenols were quantified in the apical and basolateral sides after 6h of incubation.

Forty three per cent of the rosmarinic acid in the plant extract permeated the Caco‐2 membrane

to the basolateral side. The flavonoid glucuronides initially present permeated the cell

membranes 34‐35% and the abietane diterpenoids permeated the same system in a slightly

higher amount, 43‐49% (Table 4.4).

Table 4.4. Permeation of the P. barbatus aqueous extract constituents through the Caco‐2 cell monolayer.

Compound Permeation (%)

Rosmarinic acid 43.3±5.3

Luteolin 7-O-glucuronide 35.3±3.4

Apigenin 7-O-glucuronide 34.1±4.6

Acacetin 7-O-glucuronide 33.8±4.1

Abietane diterpenoid 1 43.5±4.1

Abietane diterpenois 2 49.9±1.8

Although P. barbatus abietane diterpenoids were the compounds that showed highest

bioavailability in Caco‐2 cells, previous studies showed that they may be hydroxylated in the

small intestine by the pancreatic juice into non‐active metabolites (Porfirio et al., 2010).

Furthermore, when the plant extract was administered intragastrically and intraperitonially to

rats these compounds were found in the plasma just in the non‐active form (Falé et al., 2011),

suggesting that their metabolisation is very fast processes. For this reason the abietane

diterpenoids shall not be considered in ulterior experiments.

In the digestive tract, rosmarinic acid does not undergo extensive degradation or

metabolisation and the flavonoid glucuronides may be deglucuronidated by the gut microflora,

as it was demonstrated by the in vitro digestion (Porfirio et al., 2010). Pursuing the investigation

of P. barbatus aqueous extract, the effect of rosmarinic acid, apigenin and luteolin on the

intestinal permeation and metabolisation of each other by Caco‐2 cells were analysed, as well as

the interaction of these compounds with two known transport systems to further enlighten the

mechanism by which these active compounds are made bioavailable.

In summary, when the bioavailability of the P. barbatus aqueous extract was tested with

the Caco‐2 cell system all the compounds permeated the cell membrane after 6h, including the

glucuronide derivatives.


80

2.3.2. Bioavailability of a mixture of the standards rosmarinic acid, apigenin and luteolin

An experimental plan involving the effect of several concentrations of rosmarinic acid,

luteolin and apigenin was set in order to evaluate the effect of the concentration of these

compounds on each other’s permeability through a Caco‐2 cell monolayer. The response surface

methodology allowed drawing the permeability curves (response surface) of each compound in

function of the concentration of the other two compounds (Figure 4.2). Using this methodology

it was possible to study interactions in the permeability by the presence of different compounds,

the real situation found in the herbal teas.

The curves show that rosmarinic acid has higher permeation in the Caco‐2 cells in the

presence of very high amounts of both flavonoids (Figure 4.2a). This means that the flavonoids

interfere with the phenolic acid permeation, probably by inhibiting the efflux transport systems.

The permeation of luteolin was facilitated by rosmarinic acid and by apigenin (Figure 4.2b), as

they may have facilitated the transport or inhibited the efflux mechanisms. The permeation of

luteolin increased with increasing concentrations of rosmarinic acid, being this effect more

pronounced than the one found with increasing concentrations of apigenin. The behaviour of

apigenin permeation is more complex (Figure 4.2c). A maximum value for apigenin permeation

can be obtained with a luteolin concentration, and this maximum can be obtained with a higher

concentration of luteolin at low concentration of rosmarinic acid, or with lower concentration of

luteolin at high concentration of rosmarinic acid. On the other hand, minimum apigenin

permeability can be seen by varying the concentration of rosmarinic acid, which can be obtained

with lower concentrations of rosmarinic acid at low concentration of luteolin, or with higher

concentration of rosmarinic acid at high concentration of luteolin. This complex permeation

behaviour suggests that these compounds may inhibit the efflux systems. When they are both

present they may also compete with apigenin for an uptake transport system, or inhibit it.

Apigenin permeation shows the lowest permeability in the presence of high concentrations of

rosmarinic acid and luteolin together.

Although only the aglycones were applied in the cell culture medium of the apical side of

the Caco‐2 system, the glucuronides derivatives could be detected on the basolateral side of the

cells. Caco‐2 cells have the ability to glucuronidate flavonoids, as was previously reported for

apigenin and luteolin (Porfirio et al., 2010). The effect of the concentration of rosmarinic acid,

luteolin and apigenin on the glucuronidation and bioavailability of these flavonoids was assessed

using the same CCD factorial design. The response – glucuronidation – surfaces curves obtained

are shown in Figure 4.3. For the glucuronidation of each flavonoid, the presence of the other

flavonoid and rosmarinic acid seemed to have a similar effect whether in presence of high or low

Chapter IV

81

amount of each other, increasing with the increase of the concentration of rosmarinic acid and

decreasing with the increase of the concentration of the other flavonoid, which suggests that

the both flavonoids are substrates of the same enzyme and may compete for the same

glucuronosyltransferases.

Figure 4.2. Permeation surfaces for (a) rosmarinic acid, (b) luteolin and (c) apigenin with different

concentrations of the other two components, built with the CCD experimental plan. The relative errors are

0.930 (a), 0.718 (b), and 0.817 (c).


82

Figure 4.3. Glucuronidation surfaces for (a) luteolin and (b) apigenin with different concentrations of

rosmarinic acid and of the other flavonoid, built with the CCD experimental plan. The relative errors are

0.981 (a), and 0.766 (b).

2.3.4. Effect of MCT and Pgp transporter systems on the bioavailability of the polyphenol

mixture

The monocarboxylic acid transporter system (MCT), located in the basolateral

membrane of intestinal cells, is known to act on the permeability of monocarboxylic acids and it

is also recognized that the flavonoids may interfere with these systems (Konishi et al., 2003). As

rosmarinic acid, a monocarboxylic acid, is present in a mixture containing flavonoids, the

involvement of the MCT transporter system on the mixture bioavailability was analysed. On the

other hand Pgp efflux system is known to be inhibited by flavonoids (Brand et al., 2006), so the

participation of these systems on the bioavailability of the two flavonoids together with the

monocarboxylic acid (rosmarinic acid) were also analysed. In order to evaluate the participation

of MCT and Pgp on the bioavailability of a mixture of rosmarinic acid with the two flavonoids,

the studies were conducted in the absence and in the presence of the transporter system

substrates, benzoic acid and digoxin, respectively (Tsuji et al., 1994).

Chapter IV

83

In the present study, rosmarinic acid was found in lower amount in the basolateral

compartment in the presence of the digoxin or benzoic acid (Figure 4.4a), while apigenin and

luteolin showed similar distribution in the presence of the two substrates (Figure 4.4b and 4.4c).

This suggests that both benzoic acid and digoxin interact with the transporters responsible for

rosmarinic acid permeability.

Figure 4.4. Distribution of (a) rosmarinic acid, (b) luteolin and (c) apigenin six hours after being placed in

the apical side of a Caco‐2 cell monolayer. The effects of co‐administration of digoxin and benzoic acid

were analysed. Statistical analysis: c ‐ different from the control (P<0.05); c* ‐ different from the control

(P<0.1); s – different from the other substrate (P<0.05).


84

In the case of benzoic acid, transported by the system MCT (Figure 4.5a) the standard

mixture of polyphenols affected the permeation of the substrate, because it was found in higher

amounts in the basolateral side of the cells, comparatively to the amounts found in the absence

of the mixture. In the case of the Pgp system, digoxin decreased on the apical side and increased

slightly in the basolateral side when in the presence of polyphenols (Figure 4.5b). This result may

indicate that the Pgp transporter is inhibited by this system and the polyphenol mixture

(rosmarinic acid, apigenin and luteolin) may indeed alter the distribution of Pgp substrates

around the cell membranes and in the intracellular compartment (Figure 4.5b).

Figure 4.5. Distribution of (a) benzoic acid and (b) digoxin six hours after being placed in the apical side of

a Caco‐2 cell monolayer. The effects of co‐administration of a standard mixture (SM) with rosmarinic acid,

luteolin and apigenin, 50µM each, was analysed. Statistical analysis: * ‐ different from the control

(P<0.05).

On what concerns the bioavailability of the flavonoid glucuronide derivatives, studies

were also conducted in the absence and in the presence of the MCT and Pgp substrates. In this

study it is not possible to distinguish the effect of the transporter substrates on the

glucuronosyltransferase enzymatic activity and on the bioavailability of the flavonoid

Chapter IV

85

glucuronides derivatives. The results indicated the joint effect of both actions, as the glucuronyl

derivatives were quantified in each side of the Caco‐2 monolayer, after 6h incubation, where

both processes may be occurring simultaneously. The results showed that the presence of

flavonoid derivatives was affected by the substrates of the transport systems. The basolateral

concentration of luteolin derivative increased in the presence of digoxin and benzoic acid (Figure

4.6a), and the basolateral concentration of apigenin increased in the presence of digoxin and

decreased in the presence of benzoic acid (Figure 4.6b). The amount of luteolin glucuronides

found in the basolateral compartment increased from 24% (control) to 46% in the presence of

digoxin and benzoic acid, and from 1% (control) to 10% inside the cells (Figure 4.6a). By contrast,

the amount of apigenin glucuronides slightly increased in the basolateral compartment and

inside the cells in the presence of digoxin, while it decreased in the basolateral compartment in

the presence of benzoic acid (Figure 4.6b).

Figure 4.6. Glucuronidation of (a) luteolin and (b) apigenin six hours after being placed in the apical side of

a Caco‐2 cell monolayer. The effects of co‐administration of digoxin and benzoic acid were analysed.

Statistical analysis: c ‐ different from the control (P<0.05); c* ‐ different from the control (P<0.1); s –

different from the other substrate (P<0.05).


86

2.4. Discussion

In order to evaluate the involvement of two membrane transporter systems on the

permeability pattern found for rosmarinic acid, luteolin and apigenin in the herbal decoction

through Caco‐2 cell monolayers, the permeability of these phenolic compounds was analysed in

the presence of substrates of known transport systems, namely benzoic acid (substrate of MCT

(Tsuji et al., 1994)) and digoxin (substrate to Pgp).

The permeability of rosmarinic acid through the Caco‐2 cell monolayer decreased in the

presence of benzoic acid (Figure 4.4a), suggesting a competition for the transport across this

membrane system indicating that it may be via one transporter of the MCT family. Previous

studies pointed out that the permeability of rosmarinic acid through MCT may be very low

because it showed lower affinity to the transporter than ferulic acid, for instance (Konishi and

Kobayashi 2005). The low affinity is also suggested by the increase of benzoic acid in the

basolateral compartment in the presence of the standard mixture (Figure 4.5a), suggesting that

the affinity of benzoic acid to MCT is much higher than rosmarinic acid.

The distribution of the rosmarinic acid was also affected by digoxin, decreasing in the

same way as benzoic acid did, from the apical to the basolateral side. Digoxin permeates the

Caco‐2 monolayer predominantly by transcellular passive diffusion, and Pgp, responsible for

digoxin efflux, is the only known transporter for this molecule (d’Souza et al., 2003). Pgp does

not seem involved in the efflux of rosmarinic acid, as the competition of digoxin with rosmarinic

acid would lead to an increase in the amount of rosmarinic acid in the basolateral compartment,

which was not observed. Therefore these results suggest that digoxin may inhibit the

transporters involved in the permeation of rosmarinic acid from the apical to the basolateral

side. Although these transporters were not studied in the present work, digoxin is known to

interact with active transport in muscle cells, as an inhibitor of Na+‐K+‐ATPase pump (Rochetti et

al., 2003).

The flavonoids luteolin and apigenin are known inhibitors of the ABC transporters

(Schutte et al., 2008). Our results suggest that luteolin and apigenin are not substrates of MCT

or Pgp transporters, since benzoic acid and digoxin did not affect the transport of these

flavonoids across the Caco‐2 cell monolayers (Figures 4.4b and 4.4c). However, the

bioavailability of apigenin is strongly affected by the presence of luteolin, this means that other

transport system beyond Pgp or MCT must be involved. Flavonoids such as quercetin,

kaempferol and isorhamnetin have been shown to be substrates of Pgp, which limits their

bioavailability (Wang et al., 2005).

Chapter IV

87

The effect of benzoic acid and digoxin on the distribution of flavonoid glucuronides

(Figure 4.6) suggests that both transporter systems seem to be involved in the permeability of

these compound because significant differences were found on the bioavailability of the

glucuronides in the presence of the substrates.

The behaviour of the flavonoids and rosmarinic acid towards the permeation of one

another seems to be a balance of two factors: competition for the transepithelial transporters

and inhibition of the efflux via ABC transporters. Previous studies showed that flavonoid

aglycones may permeate the Caco‐2 cell monolayers by active transport (Kobayashi et al., 2008)

and the co‐administration of flavonoids inhibited the efflux of hesperitin, a flavonoid aglycone,

by ABC transporters (Brand et al., 2010). The result of this balance may be seen in the

permeation surfaces of Figure 4.2. For rosmarinic acid the flavonoids caused a decrease of

permeability when a low concentration of apigenin and a high concentration of luteolin is used.

These changes may be due to the inhibition of the membrane transporters by the flavonoids. In

lower concentrations of flavonoids, the inhibition of transporters, such as MCT, may be

dominant effect, decreasing the bioavailability of rosmarinic acid. By contrast, in higher

concentrations of flavonoids may be dominant the inhibition of efflux transporters, such as the

ABC transporters. Luteolin and apigenin were previously reported to inhibit MCT transporters

(Wang and Morris, 2007) and ABC transporters (Brand et al., 2006). The permeation of

rosmarinic acid increases when the concentration of apigenin also increases, this may be due to

a higher inhibition of the efflux transporters by this flavonoid relatively to luteolin (Figure 4.2a).

The efflux transporter of rosmarinic acid that seems to be inhibited by the flavonoids may

belong to the ABC family, but it is not Pgp, since the co‐administration of digoxin did not

increase the bioavailability of rosmarinic acid (Figure 4.4a)

For luteolin the effect seems to be the inhibition of the efflux systems by rosmarinic acid

and apigenin (Figure 4.2b), while for apigenin there is a joint effect competition for the same

transporter of luteolin and inhibition of the efflux transporters by rosmarinic acid and luteolin

(Figure 4.2c). The clearly competitive effect shown by luteolin in the permeation surface of

apigenin, suggests that the transporter of both molecules has more affinity to luteolin than

apigenin. The permeation of luteolin across the Caco‐2 monolayer increases with higher

concentrations of rosmarinic acid, and the same happens with the permeation of apigenin when

the concentration of luteolin is low, which suggests that rosmarinic acid may inhibit the

transporters responsible for the efflux of both flavonoids. Although rosmarinic acid has been

reported not to have effect on Pgp or MRP1 (Kobayashi et al., 2003), it is known that it inhibits


88

the efflux of some molecules, like ovalbumin, involving other transporters (Nabekura et al.,

2010).

The competition of luteolin and apigenin for the same transporters is also clearly evident

in the glucuronidation surfaces of both flavonoids (Figures 4.3a and 4.3b). As rosmarinic acid

may decrease the efflux of the flavonoids in the apical membrane, their concentration in the

intracellular compartment, increasing the glucuronidation. Apigenin and luteolin may also be

glucuronidated by the same enzymes, and for this reason compete for the

glucuronosyltransferases, resulting in the decrease of glucuronidation of one of these flavonoids

in the presence of the other one, as shown in Figures 2a and 2b. Again, luteolin seems to have

higher affinity to the glucuronosyltransferases, as the glucuronidation of apigenin was more

affected by luteolin than the glucuroidnation of luteolin was affected by apigenin, suggesting

that the presence of an extra hydroxyl group may increase the affinity for the enzyme active site.

Other studies also showed this difference in affinity for the enzymes, mainly those involved in

phase II of the metabolism. For instance, a decrease of phase II metabolisation of the flavanone

hesperetin by the co‐administration of other flavonoid, especially flavonols and flavones as

luteolin and apigenin, may be also due to the higher affinity of the phase II enzymes to these

compounds (Brand et al., 2010).

To confirm the involvement of PgP and MCT in the polyphenol bioavailability, the effect

of the herbal tea on the permeation of bezoic acid and digoxin was also evaluated. The increase

of bioavailability of benzoic acid in the presence of the standard mixture (Figure 4.5a) may be

due to the inhibition of a benzoic acid efflux mechanism by the standard polyphenolic mixture.

Although the efflux mechanism of benzoic acid is not clearly elucidated, flavonoids are inhibitors

of a wide variety of efflux processes related with ABC transporters (as reviewed by Brand et al.,

2006). Therefore it can be concluded that the consumption of herbal teas containing rosmarinic

acid together with these two flavonoids may interfere with drugs based on carboxylic acid

structures.

An increase of the amount of digoxin in the basolateral compartment was observed also

in the presence of the standard mixture (Figure 4.5b), which may be assumed as due to the

inhibition of the efflux system Pgp, probably due to apigenin and luteolin that are known

inhibitors of this transporter (Brand et al., 2006). However, previous studies suggest that

rosmarinic acid is not a substrate and has no inhibitory effect on Pgp (Konishi and Kobayashi

2005; Nabekura et al., 2010). Nevertheless, the efflux of digoxin may also be partially inhibited

by the competition with the luteolin and apigenin glucuronides, which appear to be substrates

of Pgp, as discussed previously.

Chapter IV

89

The permeation of rosmarinic acid in the P. barbatus aqueous extract and in the

standard mixture, in the presence of luteolin and apigenin, was similar. However, the

permeation of rosmarinic acid after 6h was higher than that expected according to the literature

(Konishi & Kobayashi 2005), which may be due to the inhibition of the efflux mechanisms by the

flavonoids. The inhibition of the intake and efflux mechanisms points out that rosmarinic acid

may be substrate of intestinal transporters. As far as we know this is the first report suggesting

that the flux of rosmarinic acid may be mediated by transport systems, although with low

affinity. To this conclusion may contribute the larger times of incubation performed, six hours in

contrast to 40 minutes in similar studies (Konishi et al., 2005). The flavonoid glucuronides also

showed a higher bioavailability than expected, which may be due to the inhibition of the efflux

mechanisms by the co‐administration of the two flavonoids together with rosmarinic acid. A

similar increase of flavonoid glucuronides in the basolateral compartment by the co‐

administration of flavonoids has been reported (Brand et al., 2010). These authors have

demonstrated that the increase of flavonoid glucuronides was induced by the inhibition of the

efflux mechanisms exclusively in the apical membrane.

The results obtained in this study with the P. barbatus aqueous extract may be

extrapolated to other teas, since rosmarinic acid seems to be one key component in aqueous

extracts of Lamiaceae species (Janicsak et al., 1999), to which belong some of the most common

species used in herbal teas (mints, rosemary, thyme, sage, lavender and basil). The occurrence of

flavones in these species, especially luteolin and apigenin glucuronides, together with rosmarinic

acid is very common and widely described in the literature (Fecka & Turek 2007, for peppermint,

melissa and sage).

2.5 Conclusions

The compounds of P. barbatus aqueous extract have shown higher bioavailability

together than the values expected for the isolated standards, since the bioavailability of

mixtures of phenolic compounds depends on the balance between competition for the same

transporters (reducing bioavailability), and the inhibition of efflux transport systems (leading to

an increase of bioavailability).

Rosmarinic acid seems to be substrate of MCT, but not of Pgp. The flavonoids apigenin

and luteolin do not seem to be substrates of MCT or Pgp, but they may inhibit Pgp. The

glucuronide derivatives of apigenin and luteolin, however, seem to be substrates of Pgp.


90

The high diversity of compounds often present in medicinal plant extracts is favourable

to a high bioavailability of the compounds, as it decreases the competition for the same intake

transporters while promoting the inhibition of efflux transporters.

These results may also predict an interaction between herbal teas containing these

mixtures of polyphenols with pharmaceuticals presenting carboxylic acid groups in their

structure.

Chapter IV

91

3. Conclusions

The bioavailability studies allowed to conclude that rosmarinic acid, present in most

herbal teas, may cross the intestinal barrier (in rats and Caco‐2 monolayers), as well as the blood

brain barrier (in rats), and be detected in the brain.

The administration (intragastric and intraperitoneal) of the P. barbatus aqueous extract

and the rosmarinic acid caused the inhibition of the brain acetylcholinesterase activity.

The metabolisation and bioavailability of the herbal tea components is different from

the administration of pure compounds, as observed in the bioavailability studies with rats and

with Caco‐2 monolayers.

The compounds of P. barbatus aqueous extract have shown, in the Caco‐2 cell

monolayer model, a higher bioavailability together than the values expected for the isolated

standards. This fact is due to a predominant effect of inhibition of efflux transport systems,

leading to an increase of bioavailability.

The high diversity of compounds often present in medicinal plant extracts is favourable

to a high bioavailability of the compounds, as it decreases the competition for the same intake

transporters while promoting the inhibition of efflux transporters.

Chapter V

Interactions between the Plectranthus barbatus herbal tea

and the proteins acetylcholinesterase, human serum albumin

and lysozyme

Falé PL, Ascensão L, Serralheiro MLM, Haris PI. Interaction between Plectranthus barbatus herbal tea

components and acetylcholinesterase: binding and activity studies. Submitted to Food Funct.

Falé PL, Ascensão L, Serralheiro MLM, Haris PI. 2011. Interaction between Plectranthus barbatus herbal

tea components and human serum albumin and lysozyme: binding and activity studies. Spectroscopy. 26:

79–92.

Chapter V

95

1. Interaction between the Plectranthus barbatus extract and

acetylcholinesterase. Binding of herbal tea components to the

protein structure and inhibition of enzymatic activity

1.1. Introduction

Acetylcholinesterase (AChE) is the enzyme that catalyses the hydrolysis of acetylcholine,

a neurotransmitter found in the synaptic gap, thus terminating the synaptic transmission.

Alzheimer’s disease is a very complex ailment, and the most successful therapy to present

consists of increasing the levels of acetylcholine in the brain through the inhibition of AChE

activity by reversible inhibitors (Heinrich & Teoh, 2004). However, irreversible inhibitors cannot

be used for therapeutic purposes, as complete inactivation of AChE leads to a toxic accumulation

of acetylcholine and failure of synaptic transmission, with consequent deterioration of

neuromuscular junctions, flaccid muscle paralysis and seizures in the central nervous system

(Harel et al., 2008).

Plants have proven to be an important source of useful AChE inhibitors for the

symptomatic treatment of Alzheimer’s disease (Mukjerjee et al., 2007), as for instance,

galantamine, a commercially available compound that was first isolated from Galanthus species

(Heinrich & Teoh, 2004). The herbal tea of Plectranthus barbatus leaves, composed mainly of

rosmarinic acid and flavonoid glucuronides, has been shown to be a promising in vitro AChE

inhibitor (Falé et al., 2009), even after being subjected to an in vitro simulation of the

gastrointestinal digestion (Porfirio et al., 2010). Furthermore, when the tea was intraperitoneal

administered to rats, its components and their metabolites were found in rat plasma and brain,

and a significant decrease in brain AChE activity was observed. The flavonoid glucuronides from

the plant extract were metabolized when the extract was administered to rats, and were found

in the plasma as glucuronides and as aglycones (Falé et al., 2011).

The aim of the present work was to determine the mode of interaction of AChE with

rosmarinic acid, the main constituent of P. barbatus extracts, as well as with luteolin and

apigenin (Figure 5.1), the flavonoid aglycons found in the plasma after the extract

administration. Furthermore the interaction of AChE with quercetin was also studied, as this

flavonoid has an extra hydroxyl group in position 3 of the C ring (Figure 5.1) that makes it

interesting for a comparative study. Although quecetin is not present in P. barbatus extract, it is

very common in plants, and plant‐derived food (Erlund, 2004). These findings will allow the

determination of structural features that may be more favourable to the binding of plant

Interactions with Proteins

96

phenolic compounds to the enzyme, and which forms of interaction may be more useful to

increase the reversible inhibition of AChE. This knowledge may lead to the prediction of the

potential of a plant extract in the treatment of Alzheimer’s disease by AChE inhibition, based on

its chemical constitution.

Figure 5.1. Molecular structure of (a) rosmarinic acid, (b) quercetin, (c) luteolin, and (d) apigenin.




The interactions between the plant extract, or the standards rosmarinic acid, luteolin, apigenin

and quercetin, with acetylcholinesterase were evaluated by fluorescence spectroscopy, as

described in Chapter II section 13, and by FTIR spectroscopy, as referred in Chapter II

section 14. The acetylcholinesterase inhibition was determined as described in Chapter II

section 5.

OHOOH

OH

O

O

OH

OH

OH

O

O

OH

OH

OH

OH

OH

O

O

OH

OH

OH

OH

O

O

OH

OH

a

b

c

d

Chapter V

97

1.3. Results

1.3.1. Fluorescence studies on the binding of P. barbatus water extract to

acetylcholinesterase

The binding of P. barbatus extract to AChE was studied by fluorescence spectroscopy.

AChE, alone, showed an excitation spectrum with a maximum wavelength at 280 nm and

emission spectrum with a maximum at 335 nm, under excitation at 280 nm. Figure 5.2 shows the

emission fluorescence spectra of AChE with the addition of increasing concentrations of P.

barbatus plant extract. The intensity of the intrinsic fluorescence of the protein decreased

significantly – was quenched – with the increase of concentration of the plant extract, which

acted as a quencher.

Figure 5.2. Fluorescence emission spectra of acetylcholinesterase with the addition of P. barbatus

aqueous extract. Arrow points to increasing concentrations of P. barbatus plant extract, ranging 0; 0.5; 1;

5; 10; 33; 50; 100 µg.ml‐1.

The dominant fluorophore in proteins is the indole group of the tryptophan residues and

the fluorescence quenching by the plant extract is caused by the interaction between the plant

compounds and the protein, in the vicinity of the tryptophan residues (Lakowicz, 2006). The

emission spectrum of the indole may be shifted towards lower wavelengths (blue shifted), if the

group is buried within the native protein, or its emission may be shifted towards larger

wavelengths (red shifted), when the protein is unfolded (Lakowicz, 2006). These effects were not

observed with the plant extract in the present study (Figure 5.2), suggesting that, although the

extract components may bind to the protein in close proximity of the tryptophan residues for

the fluorescence quenching to occur, they do not change the tryptophan residue exposition by

altering the secondary or tertiary structure of the protein. The quenching of AChE intrinsic


98

fluorescence by the standards rosmarinic acid, apigenin, luteolin and quercetin also occurred in

a similar way, without a red or blue shift of the peak maximum.

The fluorescence data were analysed by the Stern‐Volmer equation (Equation 1), which

allows to calculate the Stern‐Volmer quenching constant (KSV) and the quenching rate constant

(Kq) of the fluorescence quenching reaction.

1 1 (Equation 1)

Where F0 and F are the steady state fluorescence in the absence and presence of

quencher, respectively, [Q] is the concentration of quencher and τ0 the average lifetime of the

protein fluorescence in the absence of quencher.

The Stern‐Volmer plot for the fluorescence quenching of AChE by P.barbatus extract can

be found in Figure 5.3a, where it is shown that the fluorescence quenching by the plant extract

follows the behaviour of single‐compound binding to the protein. As the extract is composed of

many compounds, the concentration is expressed in µg.ml‐1 and the KSV values cannot be

calculated in M‐1, just estimated in l.mg‐1. The molarity of the P. babatus extract was estimated,

based on the concentration of rosmarinic acid, luteolin glucuronide and apigenin glucuronide as

1.11 mmol per gram of plant extract, calculated by the HPLC chromatogram peak areas (Porfirio

et al., 2010). The enzyme intrinsic fluorescence quenching was also analysed for the main

component of the plant extract, rosmarinic acid, and for luteolin and apigenin, which were also

found as aglycons in the plasma of rats after the intraperitoneal administration of P. barbatus

herbal tea (Falé et al., 2011). The values for the Stern‐Volmer quenching constant (KSV) and the

quenching rate constant (Kq) for AChE in the presence of rosmarinic acid, apigenin, luteolin or

quercetin are shown in Table 5.1. The estimated KSV and Kq for the P. barbatus extract at 298K

were respectively 93598 M‐1 and 9.36x1011 M‐1s‐1, which were approximate to the values found

for the extract’s main component, rosmarinic acid. The maximum scatter collision quenching

constant (Kq) value of various quenchers with a biopolymer is reported to be 2x1010 l.M‐1s‐1.9 As

the Kq values obtained in the present work were higher (ranging from 6x1010 to 8x1011 l.M‐1s‐1)

than the Kq obtained for a scatter mechanism, it is implied that the quenching was not initiated

by dynamic collision but rather originated by the formation of a complex (Lakowicz & Weber,

1973).

Chapter V

99

Figure 5.3. Stern‐Volmer plot (a) and plot of log(F0‐F)/F vs. log[Q] (b) of acetylcholinesterase with P.

barbatus aqueous extract. [Q] is the concentration of P. barbatus extract in mg.ml‐1.

1.3.2. Analysis of binding equilibria of P. barbatus water extract to acetylcholinesterase

Fluorescence quenching data also allows one to determine the equilibrium constants

between free and bound molecules, as well as the number of binding sites on a macromolecule,

by using Equation 2 (Jin & Zhang, 2008),

log log log (Equation 2)

Where F0, F and Q have the same meaning as in Equation 1, Kb is the binding constant

and n is the number of binding sites.

As it is shown in Figure 5.3b, the binding of P. barbatus extract follows the behaviour of

a single molecule binding to a macromolecule that is modelled by Equation 2. Similarly to the

Stern‐Volmer relationship, the Kb value could not be determined in M‐1 to the plant extract, just

be expressed in terms of its concentration in mg.L‐1, and estimated based on concentration of

rosmarinic acid, luteolin glucuronide and apigenin glucuronide as 91395 M‐1. The fluorescence

quenching effects by rosmarinic acid, luteolin, apigenin and quercetin standards were analysed

independently, and the values of Kb and n can be found in Table 5.1.


100

Table 5.1. Stern‐Volmer binding parameters (KSV, Kq), binding equilibria parameters (Kb, n) and

thermodynamic parameters (ΔHo, ΔSo, ΔGo) for the binding of P. barbatus extract, rosmarinic acid (RA),

luteolin (Lut), apigenin (Api) and quercetin (Quer) to acetylcholinesterase (R2>0.99 to all linear

regressions). Rates of Amide II/Amide I variation, reflecting the rate of hydrogen deuterium exchange in

the presence of similar amount of RA, Lut, Api and Quer or 10mg/mL of P. barbatus extract (for AChE

without ligand, ‐2.345mAU.min‐1). IC50 values for the inhibition of acetylcholinesterase activity by

P. barbatus, RA, Lut and Api. (*) and Quer for P. barbatus extract the values are expressed in

l.mg‐1, l.mg‐1s‐1, and mg.ml‐1 but the molarity was estimated based on the content of rosmarinic acid,

luteolin and apigenin (1.1122 mmol.g‐1).

T (K) P.barbatus RA Lut Api Quer

KSV (M-1)

293 - 84696.54 51112.81 10516.96 49757.31

298 104.10 l.mg-1 (90598

M-1*) 83788.32 42921.28 7178.62 59176.54

303 - 78474.03 57734.37 6457.27 62860.43

Kq (M-1s-1)

293 - 8.47E+11 5.11E+11 1.05E+11 4.98E+11

298 1.04E+09 L.mg-1s-1 (9.36E+11 M-1s-1*)

8.38E+11 4.29E+11 7.18E+10 5.92E+11

303 - 7.85E+11 5.77E+11 6.46E+10 6.29E+11

Kb (M-1)

293 - 62517.27 2854.96 8661.63 5845208

298 101.65 l.mg-1 (91395

M-1*) 144178.3 4390.36 7314.75

1.31E+07 303 - 287342.6 5057.08 6180.16 1.47E+08

n 293 - 0.9871 0.7365 0.9911 1.4692 298 1.0371 1.0595 0.7721 1.0005 1.4298 303 - 1.1281 0.7783 0.9987 1.7651

ΔH0 (J.mol-1)

- - 112636.2 42417.34 -24915.1 237336.6

ΔS0 (J.mol-1K-1)

- - 476.4026 210.9631 -9.64146 937.2904

ΔG0 (J.mol-1)

293 - -26949.81 -19494.85 -22090.20 -37289.49 298 - -29331.83 -20549.67 -22041.99 -41975.94 303 - -31713.84 -21604.48 -21993.78 -46662.4

Rates of AAmideII/AmideI

intensity ratios (mAU.min-1)

298 0.752 0.497 1.119 1.129 0.588

IC50 (µM) 298 1.02±0.02 mg.ml-1

(1134±22 µM*) 1221±73 92.12±2.36 100.57±6.93 50.88±7.25

The Kb and n values estimated for the interaction of P. barbatus extract with AChE were

within the range of those found for the isolated extract compounds/metabolites, suggesting that

all compounds bind to the protein in just one and the same site. Quercetin however showed

higher Kb values and seems to be able to bind to two sites in the enzyme.

The values for Kb and n for rosmarinic acid, luteolin and quercetin to AChE increased

with the temperature (Table 5.1), suggesting that temperature may increase the binding of the

molecules to the AChE structure. For apigenin these values decreased with temperature,

Chapter V

101

suggesting that higher temperature may destabilise the interaction, which is in agreement with

the decrease of Kq to values closer to the ones reported to scatter mechanisms (Lakowicz &

Weber, 1973).

Apigenin, in spite of decreasing Kb with temperature, showed a higher Kb values than

luteolin, suggesting that the lack of the catechol group may increase the hydrophobic interaction

between the flavonoids and the proteins. This interaction, although it may be stronger than the

weak polar interactions made by the catechol group, is more susceptible to breakage by

increasing the energy in the system.

1.3.3. Determination of interaction forces between P.barbatus extract and AChE

The thermodynamic parameters, enthalpy change (ΔHo) and entropy change (ΔSo) of the

interaction of the plant phenolic compounds and the proteins, allow elucidating the nature of

the bond between the plant extract components and AChE macromolecules. For this purpose,

the variation of the binding constant Kb with temperature was studied at three different

temperatures: 293, 298 and 303K.

The thermodynamic parameters were determined by the Van’t Hoff equation (Equation

3), considering that the enthalpy change does not vary significantly with temperature (Jin &

Zhang, 2009). The free energy change (ΔGo) was determined by equation 4.

ln (Equation 3)

(Equation 4)

Where R is the gas constant and T is the temperature. The values of the thermodynamic

parameters of the interaction of rosmarinic acid, apigenin, luteolin and quercetin with AChE can

be found in Table 5.1.

The negative values of ΔGo indicate that the binding processes occurred spontaneously

in all studied cases. From the water structure point of view, a positive ΔSo value is a typical

evidence of hydrophobic interaction (Yang et al., 2008). In the case of rosmarinic acid, luteolin

and quercetin, both ΔHo and ΔSo values were positive (Table 5.1), suggesting that hydrophobic

association is the dominant form of interaction of these plant phenolics with AChE (Ross &

Subramanian, 1973). The values of ΔHo and ΔSo in the interaction of apigenin with AChE were

negative, suggesting that van der Waals interactions between phenolic compounds and certain

domains of AChE may originate from the higher affinity of apigenin to the hydrophobic moiety


102

and, consequent higher proximity between the flavonoid and the enzyme (Ross & Subramanian,

1973). In this case, the major contribution to ΔGo arises from the ΔHo term rather than from ΔSo,

which implies that the binding process is enthalpy driven.

1.3.4. Determination of protein structure changes caused by P. barbatus extract and its

plasma metabolites by FTIR spectroscopy

FTIR spectroscopy may provide additional evidence on the interaction between protein

and a ligand in aqueous media. The protein secondary structure is reflected in the amide I band,

in the region 1600‐1700 cm‐1 (mainly C=O stretch), and in the amide II band, ≈1540 cm‐1 (C‐N

stretch coupled with NH bending mode). Amide I is more sensitive to secondary structure

changes than amide II (Haris & Severcan, 1999; Xiao et al., 2008; Tantipolphan et al., 2006),

hence its analysis is more useful for the study of structural changes induced by diverse factors.

The FTIR spectra of AChE, obtained by subtracting the absorption of the buffer solution,

are shown in Figure 5.4a, in the absence and in the presence of P. barbatus extract, rosmarinic

acid, luteolin, apigenin or quercetin. The second derivative spectra are shown in Figure 5.4b. As

can be seen from the absorbance and second derivative spectra, the position and the overall

shape of the amide I band in the presence and absence of the ligands are virtually identical. As

the amide I is composed of the sum of absorbance from different secondary structures (β‐turn,

β‐sheet and α‐helix) and irregular structures (random coil) (Haris & Severcan, 1999; He et al.,

2006), it can be concluded that the presence of P. barbatus extract or its metabolites did not

interfere with the secondary structure or irregular structure of AChE. This is in good agreement

with the fluorescence results, where no shifts were observed in the peak maximum of the

emission spectra.

The amide I band maximum is located at 1938 cm‐1. Several shoulders can be observed

and features of the protein secondary structure can be attributed to them, according to the

general attributions of secondary structure features to amide I peak components for proteins in

deuterium oxide (Haris et al., 1996; Jackson et al., 1991), or the specific case of

acetylcholinesterase (Gorne‐Tschelnokow et al., 1993): β‐sheet to 1630 cm‐1; α‐helix or random

coil to 1645 cm‐1; α‐helix to 1651 cm‐1 and 1658 cm‐1 (different populations); 3(10) helices to

1665 cm‐1; β‐turns to 1675 cm‐1; and antiparallel β‐sheet structure to 1682 cm‐1, which might

also have contributed to the shoulder at 1630 cm‐1. The maximum band at 1638 cm‐1 is usually

associated with predominantly β‐sheet structures (Haris & Severcan, 1999). X‐ray

crystallography data for this protein indicate that its secondary structure is predominantly

helical (35%, 25 helices, 194 residues) (Bourne et al., 1999). Previous studies have shown that

Chapter V

103

the peaks corresponding to the α‐helices in AChE may be found at lower wavenumbers

(Gorne‐Tschelnokow et al., 1993). The α‐helices with more than 11 amino acid residues present

in AChE structure may present peaks with wavenumber approximately 20 cm‐1 lower

(Gorne‐Tschelnokow et al., 1993), which may overlap the peaks corresponding to β‐sheet that

correspond to 16% of the protein sequence (27 strands, 91 amino acids) (Bourne et al., 1999).

Another reason for the decrease of wavenumber of α‐helices may be their high flexibility and

level of solvation of the α‐helices. X‐ray crystallography data showed that acetylcholinesterase is

a loosely planar tetramer with of two four‐helix bundles in antiparallel alignment and a large

space in the center, which gives a high conformational flexibility to the whole protein structure

(Bourne et al., 1999). The α‐helix flexibility may increase the access to the solvent and, when the

solvent is deuterium oxide, significantly reduce the wavenumber of the peaks attributable to the

highly flexible α‐helices, as it was shown for the predominantly α‐helical protein calmodulin

(Jackson et al., 1991).

Figure 5.4. FTIR spectra of the acetylcholinesterase alone, with P. barbatus extract, rosmarinic acid (RA),

luteolin (Lut), apigenin (Api) and quercetin (Quer). The spectra obtained are shown (a) in the form of

absorbance spectra and (b) the second derivative spectra (negative peaks).

FTIR spectroscopy also allows the determination of the rate of hydrogen‐deuterium

exchange in proteins (Haris & Severcan, 1999). Hydrogen‐deuterium exchange is the chemical

reaction in which covalently bonded hydrogen atoms are replaced by deuterium atoms.

Following the rate of hydrogen‐deuterium exchange of a protein molecule in deuterium oxide

provides information about solvent accessibility in various parts of the molecule, which reflects

the tertiary structure of the protein (Haris & Severcan, 1999; Hvidt & Wallevik, 1972). The

hydrogen‐deuterium exchange can be followed by a decrease in the intensity of the amide II

band, during the first moments of contact of the protein with deuterium oxide, till amide II band


104

stabilizes, when all exchangeable hydrogen’s atoms were replaced by deuterium atoms (Haris &

Severcan, 1999; Hartshorne & Stracher, 1965).

When P. barbatus extract (10 mg.mL‐1) was in contact with AChE, the rate of

hydrogen/deuterium exchange decreased by 67.93% compared to the protein in the absence of

the extract (Figure 5.5). The effects of rosmarinic acid, luteolin and apigenin were analysed

separately for each of the compounds in the same concentration of the protein. Rosmarinic acid

caused a 78.81% decrease in the rate of hydrogen‐deuterium exchange of AChE, while the

flavonoids luteolin and apigenin caused a decrease of 52.28 % and 51.86 %, respectively

(Figure 5.5). Quercetin caused a decrease in hydrogen‐deuterium exchange of 74.93%. As there

was no apparent difference in the shape of the amide I band of AChE in the presence of

P.barbatus extract, the differences found in the rate of hydrogen‐deuterium exchange are not

likely to be due to changes in the overall secondary structure of the protein (Haris et al., 1986;

Hartshorne & Stracher, 1965).

Figure 5.5. Percentage of decrease of hydrogen‐deuterium exchange rate of AChE in the presence of P.

barbatus extract, rosmarinic acid, luteolin, apigenin and quercetin.

The decrease in the rate of hydrogen‐deuterium exchange of the amide protons in AChE,

in the presence of P. barbatus extract, appears to be due to rosmarinic acid, as it is the major

component, while the flavonoids play a smaller role in the global hydrogen‐deuterium exchange.

The decrease in the protein hydrogen‐deuterium exchange in the presence of the plant extract

may be due to a blocking of the accessibility of the solvent to the peptide bonds connecting the

amino acid residues within the protein. The exchangeable hydrogens in the peptide groups may

be protected from the hydrogen‐deuterium exchange process due to the binding of the extract

components to the active gorge of the enzyme, preventing solvent accessibility. An alternative

explanation could be that the binding of the compounds results in an overall rigidification of the

Chapter V

105

structure such as through increased hydrogen‐bonding. However, as we did not detect any

major shifts in the amide I components, the former explanation is more likely.

1.4. Discussion

The in vitro acetylcholinesterase (AChE) inhibition for P. barbatus aqueous extract, used

in the present study, and its main component, rosmarinic acid, was previously determined

(Porfirio et al., 2010; Falé et al., 2011), showing IC50 values of 1.02±0.02 mg.ml‐1 (P. barbatus)

and 1221±73 µM (rosmarinic acid), as shown in Table 5.1. The standards apigenin and luteolin

also inhibit AChE activity, with IC50 values of 100.57±6.93 µM and 92.12±2.36 µM, respectively

(Porfirio et al., 2010). Quercetin is a flavonoid that, although it is not present in P. barbatus

extract, is very common in plants, and plant‐derived food, as its glycoside and aglycon forms

(Erlund, 2004). The extra hydroxyl group in position 3 of the C ring (Figure 5.1) makes this

flavonoid interesting to add to this study. Quercetin showed an IC50 for AChE inhibition of

50.88±7.25 µM, the lowest value found for the flavonoids under evaluation.

Spectroscopic analysis show that the interaction between the plant extract, or the

isolated compounds, with the AChE do not change the protein secondary structure, as no shift

was observed in either the fluorescence emission peak or on the amide I peak in the FTIR

spectrum. The fact that drastic structural changes do not occur in the protein is in agreement

with previous reports showing that the enzyme recovers its activity when the concentration of

the extract components decreases (Falé et al., 2011), which confirms that enzyme inhibition is

reversible.

The inhibition of AChE by quercetin was shown to be competitive (Khan et al., 2009) and,

several docking studies pointed out that the inhibitory activity of flavonoids to this enzyme is

related to the binding to amino acid residues in the active gorge of the enzyme (Khan et al.,

2009; Sheng et al., 2009). The active gorge of AChE is a narrow and deep gorge in the enzyme

structure, with amino acid residues that trap and position the substrate, acetylcholine, to the

active site, which is located at the base of the gorge (Harel et al., 2008). The decrease of

hydrogen‐deuterium exchange rate observed in AChE with the plant extract and its isolated

constituents indicated a decrease in the accessibility of the exchangeable amide hydrogens in

the peptide groups. This situation is probably due to the binding of the plant phenolic

compounds to the active gorge of the enzyme that may cause small changes in the protein

structure as it was already seen by X‐ray crystallography studies of AChE‐inhibitor complexes

from mouse (Bourne et al., 2010) and Torpedo (Sanson et al., 2011). These changes are usually

small orientation changes of amino acid residues, especially related to aromatic amino acids,


106

which may affect considerably the enzymatic activity, and occur near the four‐helix bundles of

AChE (Bourne et al., 2010; Sanson et al., 2011; Harel et al., 1993). As the peak at 1638 cm‐1 by

FTIR is not shifted to higher or lower wavenumbers, the final amount of hydrogen‐deuterium

exchange of the α‐helices may not be altered by the binding of the inhibitors. However, the

decrease of the rate of hydrogen‐deuterium exchange in AChE complexes may be due to small

changes, maybe a decrease in α‐helix‐related flexibility (Bourne et al., 1999), which may hinder

the accessibility of deuterium oxide to certain regions of AChE.

The thermodynamic data (Table 5.1) suggest that the dominant interactions in the

binding of rosmarinic acid, luteolin and quercetin to AChE are hydrophobic interactions (Ross &

Subramanian, 1981). Docking studies suggest that the hydrophobic interactions in the binding

and positioning of quercetin to the active gorge of electric eel AChE are related to the residues

Trp86, Glu202, His447 and Ser125 (Khan et al., 2009). Other authors presented a similar docking

of flavonoids to rat brain acetylcholinesterase, relating the interactions to homologous amino

acid residues (Sheng et al., 2009). Docking studies also refer the occurrence of hydrogen bonds

between the hydroxyl groups of quercetin and amino acid residues of the active site of electric

eel AChE (Khan et al., 2009). The higher Kb and lower ΔG values shown by quercetin, when

compared to luteolin and apigenin (Table 5.1), may be due to a hydrogen bond between the

Trp86 of the enzyme and the hydroxyl group in position 3 of the B ring of quercetin (Khan et al.,

2009), which is absent in the other two flavonoids. This interaction caused by this particular

hydroxyl group in quercetin seems to play an important role in stabilizing the complex flavonoid‐

protein, causing a significantly higher competitive inhibition of AChE by quercetin and, a

decrease in the accessibility of the solvent molecules to the active gorge, which is suggested by

the low hydrogen‐deuterium exchange rate presented by AChE with quercetin (Table 5.1), and

may explain the low IC50.

Although all flavonoids presented negative ΔG values for the binding to AChE, apigenin

showed the lowest for ΔHo and ΔSo, and quercetin the highest. If a plot is drawn with the

number of hydroxyl groups in the flavonoid and the values of ΔHo and ΔSo, the linear regressions

have R2 of 0.93 for ΔHo and 0.91 for ΔSo, showing that there is a direct relationship between the

number of hydroxyl groups and the thermodynamic parameters. Docking studies showed that

the hydroxyl groups in flavonoids may establish hydrogen bridges with the amino acid residues

in the active gorge, being thus positioned in a specific manner (Khan et al., 2009). The lowest the

number of hydroxyl groups, the looser is the positioning of the flavonoid in the active site, which

may explain the decrease in ΔSo with a lower the number of hydroxyl groups (Table 5.1). The

changes in ΔSo are compensated with changes in ΔHo to give negative ΔGo values in an enthalpy

Chapter V

107

driven binding. The increase in number of hydroxyl groups is also related to a decrease in the

ΔGo values (Table 5.1), suggesting that the binding is more favourable with the higher number

hydrogen bridges that may be established between the flavonoid and the amino acid residues in

the active gorge.

Rosmarinic acid presented ΔSo and ΔHo values that suggested that the dominant forces

in the rosmarinic acid‐protein complex were hydrophobic interactions (Ross & Subramanian,

1981). The enzymatic inhibition by rosmarinic acid was lower than the one presented by the

flavonoids (Table 5.1). Conversely to rosmarinic acid, the structure of flavonoids present a

hydrophobic back opposite to a hydrophilic side, with a higher number of hydroxyl groups,

allowing them to hold a clear position inside the active gorge, by establishing hydrophobic

interactions in one side and hydrogen bonds in the other with the amino acid residues (Khan et

al., 2009; Sheng et al., 2009). Also the B ring in the flavonoids molecule can rotate around the

C2‐C1’ bond, letting the hydroxyl groups to adjust the conformation, binding to the active sites of

diverse target proteins (Ji & Zhang, 2006). For these reasons, rosmarinic acid may be able to bind

to AChE through hydrophobic interactions but, may not establish such a complex interactions

with the protein as the flavonoids do. Consequently, rosmarinic acid may be a weaker

competitive inhibitor of the enzyme activity, as the 10‐fold higher IC50 seems to confirm

(Table 5.1).

To our knowledge, this is the first spectroscopic study on the interaction of plant

phenolic compounds to AChE, and relating these interactions with the enzymatic inhibition.

Although docking studies are available, spectroscopic data may present an important

complement to this knowledge, confirming docking studies and clarifying some details related to

the dominance of the binding forces, as for instance, the dominance of the hydrophobic

interactions in the binding of plant phenolics to AChE.

1.5. Conclusions

P. barbatus extract components bind to AChE forming a complex. The analysis of the

binding of the isolated standard compounds suggested that the interaction of the plant extract

components with the amino acid residues, in the active gorge of acetylcholinesterase, may be

responsible for the AChE inhibition by P. barbatus extract.

The interactions between the plant phenolics and AChE do not cause secondary

structure changes, which might be too drastic and lead to the denaturation of the enzyme. Thus

the change appears to be subtle and the drastic reduction in hydrogen‐deuterium exchange can


108

be explained by the binding of the phenolic compounds in the region of the active gorge

blocking solvent accessibility to a significant portion of the protein molecule.

Flavonoids seem to be better AChE inhibitors than rosmarinic acid, since this acid has a

structure less able to establish interactions with the amino acid residues in the active gorge,

while the flavonoid structure seems to fit to the active gorge of the enzyme, interacting

dominantly by hydrophobic interactions but also by hydrogen bonds established between the

hydroxyl groups of the flavonoids and the amino acid residues, which allow the flavonoid to hold

a specific position inside the gorge.

The position of the hydroxyl groups in the flavonoids seems to play an important role in

the binding to the enzyme and in the inhibitory activity, particularly in some positions, such as

the position 3 of the C ring, that is responsible for the binding of quercetin to Trp86, increasing

its inhibitory activity to the double, when compared to luteolin. Therefore, spectroscopic studies

proved to be useful in confirming and clarifying docking studies related to AChE inhibitors.

Chapter V

109

2. Interaction between Plectranthus barbatus herbal tea

components and human serum albumin and lysozyme: binding

and activity studies

2.1. Introduction

Plasma proteins play an important role in transportation and deposition of substances in

the circulatory system, such as fatty acids, hormones and medicinal drugs. Therefore, it is

important to reveal the interaction between drugs and proteins in the bloodstream, as it may

affect the bioavailability, distribution and elimination of pharmaceutical or nutraceutical active

compounds. Albumin is the major protein of plasma, and its main function is the regulation of

colloidal osmotic pressure and the binding and transport of substances in the bloodstream (Li et

al., 2007). The interaction of human serum albumin with chemically synthesized drugs used in

medicine may influence their bioavailability and effectiveness, and so many recent studies have

focused on these interactions (Varshney et al., 2010). The interaction between HSA and plant

secondary metabolites traditionally taken by people as “natural medicines” has also been

subject to many reports (Rawel et al., 2006). Lysozyme is also known to play a role in the

transportation of drugs (Jin & Zhang, 2010), although its main function is to hydrolyse

peptidoglycans, which are found in bacterial cell walls (especially in Gram positive bacteria), as

part of the innate immune system (Laible & Germaine, 1985). The most dramatic

lysozyme‐related conditions are caused by the decrease or lack of its activity (Laible & Germaine,

1985). However, its excessive activity is known to be related to allergic conditions and worsening

of inflammation in the immune response to pathogens (Makino et al., 2003; Ronca et al., 1998;

Wu et al., 2006).

Plectranthus barbatus (Lamiaceae) herbal tea has antiacetylcholinesterase as well as

antioxidant activity (Falé et al., 2009), which is related to its main components: rosmarinic acid,

flavonoid glucuronides (glucuronidated apigenin, luteolin and acacetin) and abietane

diterpenoids. After the oral administration of the plant extract to rats, its main component,

rosmarinic acid, was found in the plasma, as well as the flavonoids, in both the glucuronidated

and as aglycone forms (Falé et al., 2011).

The aim of the present study was to investigate if P. barbatus water extract compounds

and metabolites, that were previously found circulating in the plasma can interact with human

albumin and lysozyme, by binding and being transported by these proteins, or affecting

lysozyme activity.


110

2.2. Material and Methods



The interaction between the plant extract, or the standards rosmarinic acid, luteolin and

apigenin, with the human proteins albumin and lysozyme were evaluated by fluorescence

spectroscopy, as described in Chapter II section 13, and by FTIR spectroscopy, as referred in

Chapter II section 14. The lysozyme inhibition was determined as described in Chapter II

section 15.

2.3. Results

2.3.1 Binding of P. barbatus to albumin and lysozyme

Both lysozyme and human serum albumin (HSA) showed excitation spectra with a

maximum wavelength at 280 nm and emission spectra with a maximum at 335 nm, under

excitation at 280 nm. Figure 5.6 shows the emission fluorescence spectra of HSA and lysozyme

with the addition of increasing concentrations of P. barbatus aqueous extract. The intensity of

the intrinsic fluorescence of both proteins decreased significantly – was quenched – with the

increase of concentration of the plant extract, which acted as a quencher.

The dominant fluorophore in these proteins is the indole group of the tryptophan

residues, and the fluorescence quenching by the plant extract is caused by the interaction

between the plant compounds and the protein, in the vicinity of the tryptophan residues

(Lakowicz, 2006). The emission spectrum of the indole may be shifted towards lower

wavelengths (blue shifted) if the group is buried within the native protein, or its emission may be

shifted towards larger wavelengths (red shifted) when the protein is unfolded (Lakowicz, 2006).

These effects were not observed in the present study (Figure 5.6a and b), suggesting that,

although the flavonoids may bind to the proteins in close proximity of the tryptophan residues

for the fluorescence quenching to occur, they do not change the tryptophan residue exposition

by altering the secondary or tertiary structure of the protein.

The fluorescence data were analysed by the Stern–Volmer equation (Equation 1), which

allows to calculate the Stern–Volmer quenching constant (KSV) and the quenching rate constant

(Kq) of the fluorescence quenching reaction, as previously shown in this Chapter, section 1.2.1.

Chapter V

111

Figure 5.6. Fluorescence emission spectra of (a) HSA and (b) lysozyme with the addition of P. barbatus

aqueous extract. Arrow points to increasing concentrations of P. barbatus plant extract, ranging 0; 0.5;

0.75; 1; 2.5; 5; 7.5; 100 μg.ml−1.

The Stern–Volmer plots for the fluorescence quenching of HSA and lysozyme by

P. barbatus extract can be found in Figure 5.7, where it is shown that the fluorescence

quenching by the plant extract follows the behaviour of single‐compound binding for both

proteins. As the extract is composed of many compounds, the concentration is expressed in

μg.ml−1 and the KSV values cannot be calculated in M−1, just estimated in l.mg−1. The molarity of

the P. barbatus extract was estimated, based on the concentration of rosmarinic acid, luteolin

glucuronide and apigenin glucuronide as 1.1122 mmol per gram of plant extract, calculated by

the HPLC chromatogram peak areas (Porfirio et al., 2010). The protein intrinsic fluorescence

quenching was also analysed for the main component of the plant extract, rosmarinic acid, and

for luteolin and apigenin, which could be found in plasma as aglycones after the administration

of P. barbatus herbal tea, due to the liver β‐glucuronidase activity (Falé et al., 2011).


112

Figure 5.7. Stern–Volmer plots of HSA and lysozyme with P. barbatus aqueous extract. [Q] is the

concentration of P. barbatus in μg.ml−1.

The values for the Stern–Volmer quenching constant (KSV) and the quenching rate

constant (Kq) for HSA and lysozyme in the presence of rosmarinic acid, apigenin or luteolin are

shown in Table 5.2.

The estimated KSV and Kq for the P. barbatus extract were, respectively, 461,518 M−1 and

4.61 × 1013 M−1s−1 for HSA, and 68,513M−1 and 6.85×1012 M−1s−1 for lysozyme, which were

approximate to the values found for the extract’s main component, rosmarinic acid. The

maximum scatter collision quenching constant (Kq) value of various quenchers with a biopolymer

is reported to be 2 × 1010 l.M−1s−1 (Lakowicz & Weber, 1973). As the Kq values obtained in the

present work were higher, ranging from 1011 to 1013, than the Kq obtained for a scatter

mechanism, it is implied that the quenching was not initiated by dynamic collision but rather

originated by the formation of a complex.

2.3.2. Analysis of binding equilibria

Fluorescence quenching data also allows to determine the equilibrium constants

between free and bound molecules, as well as the number of binding sites on a macromolecule,

by using Equation 2 (Jin & Zhang, 2010), as shown in section 1.2.2, in this Chapter.

In Figure 5.8, the binding of P. barbatus extract follows the behaviour of a single

molecule binding to a macromolecule that is modelled by Equation 2, in both cases of HSA and

lysozyme.

Similarly to the Stern–Volmer relationship, the Kb value could not be determined in M−1

to the plant extract, just be expressed in terms of its concentration in mg.l−1, and estimated

based on concentration of rosmarinic acid, luteolin glucuronide and apigenin glucuronide as

5.84×105 and 7.92×105 M−1 for HSA and lysozyme, respectively. The fluorescence quenching

Chapter V

113

effects by rosmarinic acid, luteolin and apigenin standards were analysed independently, and

the values of Kb and n can be found in Table 5.2.

Table 5.2. Binding parameters (KSV, Kq, Kb, n) and thermodynamic parameters (ΔHo, ΔSo, ΔGo) for the

binding of P. barbatus extract, rosmarinic acid (RA), luteolin (Lut) and apigenin (Api) to HSA and to

lysozyme (R2>0.99 to all linear regressions). Rates of Amide II/Amide I variation in the presence of similar

amount of RA, Api and Lut, or 10mg/mL of P. barbatus extract (without ligand, ‐0.716 for HSA and ‐1.815

for lysozyme). IC50 values for the inhibition of lysozyme activity. For P. barbatus extract the values are

expressed in l.mg‐1(a), l.mg‐1s‐1(b) and mg.l‐1(c).

T(K) Human Serum Albumin Lysozyme

P.barbatus RA Lut Api P.barbatus RA Lut Api

KSV(M-1)

293 - 245192 194500 28029 - 87179 4191.9 5344.4

298 0.5133(a) 355856 250748 8090.5 0.0762(a) 98081 3381.4 3805.1

303 - 482757 326653 5886.5 - 143377 3533.7 3250.8

Kq(M-1s-1)

293 - 2.45E+13 1.95E+13 2.8E+12 - 8.72E+12 4.19E+11 5.34E+11

298 5.13E+7(b) 3.56E+13 2.51E+13 8.09E+11 7.62E+6(b) 9.81E+12 3.38E+11 3.81E+11

303 - 4.83E+13 3.27E+13 5.89E+11 - 1.43E+13 3.53E+11 3.25E+11

Kb(M-1)

293 - 1.32E+08 3.03E+08 3.07E+04 - 3069729 2760.57 7084.35

298 0.6493(a) 3.96E+08 1.12E+09 1.85E+06 0.8806(a) 205541.7 3498.64 642.98

303 - 7.79E+08 6.92E+09 1.61E+09 - 84781.29 4926.06 363.74

n

293 - 1.4998 1.5849 1.0101 - 1.1132 0.9551 0.9935

298 1.4643 1.5796 1.5403 1.4283 1.4667 1.0673 0.9839 0.8253

303 - 1.6267 1.6267 2.0964 - 0.9617 0.9789 0.7814

ΔH0

(kJ.mol-1) - - 131.16 230.67 96.35 - -265.66 42.69 -219.90

ΔS0

(J.mol-1K-1) - - 603.68 948.93 338.72 - -784.91 211.45 -679.24

ΔG0

(kJ.mol-1)

293 - -45.551 -47.576 -25.172 - -36.389 -19.302 -21.598

298 - -49.048 -51.619 -35.757 - -36.389 -20.218 -16.021

303 - -51.577 -57.081 -53.413 - -28.588 -21.420 -14.855

Rates of

Amide

II/Amide I

intensity

ratios

(mAU/min)

298 -0.979 -1.071 -0.071 -0.715 -1.425 -1.820 -0.086 -0.449

IC50 (µM) 298 - - - - 9.02±0.40(c) 97.2±3.9 106.5±9.6 91.0±1.4


114

Figure 5.8. Plots of log([F0 − F]/F) vs. log[Q] for HSA and lysozyme with P. barbatus aqueous extract. [Q] is

the concentration of P. barbatus in g.l−1.

The n value estimated for the interaction of P. barbatus extract with HSA was within the

range of the one found for the isolated compounds, while the Kb value was lower, suggesting

that the extract components may interfere with the binding of each other to HSA. For lysozyme,

the plant extract seemed to bind to more than one binding site (n = 1.5, Table 5.2) while for the

isolated compounds n is approximately 1, which may be the cause of the higher Kb value

estimated for the plant extract than for the isolated compounds (Table 5.2).

The values for Kb and n for the three compounds to HSA increased with the temperature,

suggesting that temperature may increase the binding of the molecules to two sites in the HSA

structure. The binding of luteolin showed higher Kb values than apigenin, which agrees with the

results of Xiao and coworkers (2008) for the binding of flavonols to bovine serum albumin,

where it is reported that the binding constants increased with the number of hydroxyl groups in

the B‐ring of flavonoids.

A decrease in Kb and n was observed for lysozyme with rosmarinic acid and apigenin,

suggesting that temperature may be unfavourable to the binding in these cases. For the binding

of luteolin with lysozyme, the value of n remained similar through the changes of temperature,

while the Kb value increased. This suggests that the catechol group that is present in luteolin, but

not in apigenin, may play an important role in stabilizing the complex flavonoid–protein at

higher temperatures. Apigenin, in spite of decreasing Kb with temperature, showed a higher Kb

with a lower temperature than luteolin, suggesting that the lack of the catechol group may

increase the hydrophobic interaction between flavonoids and proteins.

This interaction, although it may be stronger than the weak polar interactions made by

the catechol group, is more susceptible to be broken by increasing the energy in the system. At

higher temperatures, closer to body temperature, the binding affinities of luteolin to HSA and

lysozyme are higher than the ones for the binding of apigenin. These facts, shown by Xiao and

Chapter V

115

coworkers (2011) for total plasma proteins extracted from rat blood, demonstrate that the

hydroxyl group in position 3’ of the B ring in the flavonoid structure has great influence in

flavonoid affinity to the proteins.

2.3.3. Determination of interaction forces between P. barbatus extract metabolites and

HSA and lysozyme

The thermodynamic parameters enthalpy change (ΔHo) and entropy change (ΔSo) of the

interaction of the plant phenolic compounds and the proteins allow to elucidate the nature of

the bond between the plant extract components and the HSA/lysozyme macromolecules. For

this purpose, the variation of the binding constant Kb with temperature was studied at three

temperatures: 293, 298 and 303 K.

The thermodynamic parameters were determined by the van’t Hoff equation (Equation

3) and the free energy change (ΔGo) was determined by Equation 4, as shown in this Chapter,

section 1.2.3. The values of the thermodynamic parameters of the interaction of rosmarinic acid,

apigenin and luteolin with HSA and Lysozyme can be found in Table 5.2.

The negative values of ΔGo indicate that the binding processes occurred spontaneously

in all studied cases. From the view‐point of water structure, a positive ΔSo value is a typical

evidence of hydrophobic interaction (Yang et al., 2008). In the case of HSA, both ΔHo and ΔSo

values were positive (Table 5.2), suggesting that hydrophobic association is the dominant form

of interaction of the tested plant phenolics with HSA (Ross & Subramanian, 1981).

Previous reports on the interaction of apigenin with HSA (Yuan et al., 2007), and luteolin

with bovine serum albumin (Yang et al., 2008), also showed positive values for ΔSo. However,

Yang and coworkers (2008) presented a negative value for ΔHo, which may be due to a

strengthening of the interaction by other forces, such as van der Waals or hydrogen bonds,

introduced by hydrophobic effect in higher temperature conditions than the interval used in the

present study (Ross & Subramanian, 1981).

The values of ΔHo and ΔSo in the interaction of rosmarinic acid and apigenin with

lysozyme were negative, suggesting van der Waals interactions between phenolic compounds

and certain domains of lysozyme (Ross & Subramanian, 1981). In these cases, the major

contribution to ΔGo arises from the ΔHo term rather than from ΔSo, which implies that the

binding processes are enthalpy driven. In the interaction of luteolin and lysozyme, both ΔHo and

ΔSo values were positive, which points to hydrophobic association as the main form interaction

between the molecules. The difference in thermodynamic parameters found between the

apigenin and luteolin may have been caused by the higher polarity and solubility in water of


116

luteolin (Kim et al., 2008), that arises from the higher hydroxylation of the B‐ring. The higher

affinity to water of luteolin may cause a weaker hydrophobic association that did not induce the

electrostatic interactions in our interval of temperatures (Ross & Subramanian, 1981).

2.3.4. Determination of protein structure changes caused by P. barbatus extract and its

plasma metabolites by FTIR

FTIR spectroscopy may provide additional evidence on the interaction between a protein

and a ligand. The protein secondary structure is reflected in the amide I band, in the region

1600–1700 cm−1 (mainly C=O stretch), and in the amide II band, ≈1540 cm−1 (C–N stretch coupled

with NH bending mode). Amide I is more sensitive to secondary structure changes than amide II

(Haris & Severcan, 1999; Tantipolphan et al., 2007; Xiao et al., 2011), hence its analysis is more

useful for the study of structural changes induced by diverse factors.

The FTIR spectra of HSA and lysozyme, obtained by subtracting the absorption of the

buffer solution, are shown in Figure 5.9a and b, for HSA and Lysozyme, respectively, in the

absence and in the presence of P. barbatus extract, rosmarinic acid, luteolin or apigenin. The

second derivatives are shown in Figure 5.9c and d for both HSA and lysozyme, respectively. The

position of the amide I band in region 1651–1653 cm−1 indicates the presence of predominantly

α‐helical structure in these two proteins which is consistent with their known X‐ray structures.

As it can be seen in the absorbance spectra and second derivatives, the amide I band showed

the same peak and shape in the presence and absence of the ligands for both proteins. As the

amide I is composed of the sum of β‐turn, β‐sheet and α‐helix absorbance contributions (He et

al., 2006; Kim et al., 2008), it can be concluded that the presence of P. barbatus extract or its

metabolites did not interfere with the secondary structure of HSA or lysozyme.

FTIR spectroscopy also allows one to determine the rate of hydrogen–deuterium

exchange in proteins (for a review see Kim et al., 2008). Following the rate of hydrogen–

deuterium exchange of a protein molecule in deuterium oxide medium may give information

about solvent accessibility in various parts of the molecule, which reflects the tertiary structure

of the protein (Haris et al., 1990). The hydrogen–deuterium exchange can be followed by a

decrease in amide II band during the first moments of contact of the protein with deuterium

oxide (Haris & Severcan, 1999).

Chapter V

117

Figure 5.9. FTIR spectra of the proteins alone, with P. barbatus extract, rosmarinic acid (RA), luteolin (Lut)

and apigenin (Api). The absorbance spectra obtained for HSA are shown in (a) and the second derivatives

are in (c). The absorbance spectra obtained for lysozyme are shown in (b) and the second derivatives are

in (d).

When P. barbatus extract (10 mg.ml−1) was in contact with the proteins under study, the

rate of hydrogen‐deuterium exchange increased by 36.2% for HSA and decreased by 21.5% for

lysozyme (Figure 5.10).

The effects of rosmarinic acid, luteolin and apigenin were analysed separately for each of

the compounds in the same concentration of the protein. Rosmarinic acid caused a 49.6%

increase in the rate of hydrogen–deuterium exchange of HSA, while the increase in lysozyme

was negligible (0.3%). The flavonoids luteolin and apigenin caused a decrease in

hydrogen–deuterium exchange in both HSA and lysozyme. The highest decrease was observed

for luteolin (90.1% and 95.3% for HSA and lysozyme, respectively), while apigenin caused a

75.3% decrease in the hydrogen–deuterium exchange rate of lysozyme, and a negligible

decrease in HSA (0.1%) (Figure 5.10). There was no apparent difference in the overall shape and

position of amide I bands, as seen from both the absorbance and second‐derivative spectra

(Figure 5.9) of HSA and lysozyme, in the presence of P. barbatus extract. This suggests that the

interaction does not alter the secondary structure of these two proteins. Therefore, the


118

differences found in the rate of hydrogen–deuterium exchange may be due, predominantly to

changes in the tertiary structure of the proteins (Haris & Severcan, 1999).

Figure 5.10. Percentage of change in protein (HSA and lysozyme) hydrogen–deuterium exchange rate,

determined from the analysis of the amide II band, in the presence of P. barbatus extract, rosmarinic acid,

luteolin or apigenin, in comparison with the hydrogen–deuterium exchange rate of the protein alone.

The decrease of lysozyme hydrogen–deuterium exchange by P. barbatus extract may be

due to its flavonoid components, since its main component, rosmarinic acid, does not affect the

hydrogen–deuterium exchange of lysozyme. The increase in the rate of hydrogen–deuterium

exchange in HAS, when in contact with P. barbatus extract appears to be due to rosmarinic acid,

as it is the major component, while the flavonoids play a smaller role in altering the global

hydrogen–deuterium exchange.

2.3.5. Effect of P. barbatus extract on lysozyme activity

Lysozyme is a glycoside hydrolase able to hydrolyse the peptidoglycan that constitutes

the cell walls of bacteria such as Micrococcus sp. (Laible & Germaine, 1985).

The aqueous extract of P. barbatus inhibited lysozyme activity, with an IC50 value of 9.02

μg.ml−1 (Table 5.2). The IC50 values for the inhibition of lysozyme by rosmarinic acid, apigenin and

luteolin were similar and around 100 μM, as it is shown in Table 5.2.

2.4. Discussion

The structure of human serum albumin, alone and when bound to diverse molecules,

has been determined using X‐ray crystallography (for a review see Curry (2009)). In this study we

Chapter V

119

have demonstrated using fluorescence spectroscopy that flavonoids including rosmarinic acid

alter the tertiary structure of HSA. Furthermore, our FTIR spectroscopic analysis reveals

significant reduction in amide proton hydrogen– deuterium exchange rate for HSA complexed to

flavonoids. This is indicative of a decrease in the accessibility of the exchangeable hydrogens to

the solvent (deuterium oxide). However, we did not detect any changes in the secondary

structure of HSA, through the analysis of the amide I band, as a consequence of interaction with

the flavonoids. This finding is in good agreement with the results of X‐ray crystallographic

studies which have consistently shown that there are no gross changes in the secondary

structure of HSA associated with binding ligands although changes in tertiary structure such as

rigid‐body‐rotation of domains have been detected (for a review see Curry (2009)). The latter

changes in the tertiary structure may explain the differences observed in the fluorescence

spectra and hydrogen–deuterium exchange rates.

Among the different compounds tested, luteolin induced the highest decrease of

hydrogen–deuterium exchange and it also presented the highest values for the binding

constants and lowest ΔGo values, suggesting a higher stability for the HSA‐luteolin complex

compared to the HSA‐apigenin complex. The increased stability of the protein associated with

the bonding between the flavonoids and the HSA molecule makes it more difficult for the amide

protons to be substituted with deuterium. Furthermore, the interaction may result in the

shielding of certain segments of the protein from being accessible to solvent. All this explains the

significant reduction in hydrogen–deuterium exchange upon binding of the flavonoids to HSA.

Rosmarinic acid, like luteolin, had no effect on the secondary structure of HSA. However,

unlike luteolin, rosmarinic acid increased the hydrogen–deuterium exchange of HSA, suggesting

that the nature of its interaction with the transport protein is different. Our results suggest that

hydrophobic regions of the protein becomes accessible to solvent due to alterations in the

protein tertiary structure (Militello et al., 2004), without any changes in the secondary structure.

The nature of the change could be a large rigid‐body rotation of the subunits that have been

shown to occur on fatty acid binding to the main fatty acid binding sites in HSA, as was

extensively reviewed by Curry (2009). X‐ray crystallographic analysis has shown that fatty acid

binding results in a HSA molecule that is 10 Å wider than the defatted HSA. It is thus possible,

that whilst the binding may reduce hydrogen–deuterium exchange for molecules that are

interacting with rosmarinic acid, a much larger change, such as widening of the molecule, could

expose a greater number of amino acid residues to solvent resulting in an overall increase in

hydrogen–deuterium exchange. We propose that rosmarinic acid may induce a similar type of

tertiary structural change as induced by binding of fatty acids to HSA. In five of the seven known


120

fatty acid binding sites the lipid is anchored by the interaction of the carboxylic group with a

basic or polar group at the pocket entrance. Some non‐lipid acidic compounds are known to

interact with these binding sites, specifically by interacting with the carboxylic acid binding

residues (Curry, 2009). Rosmarinic acid and fatty acids both share a common functional group

(carboxylic acid group) which is not the case for luteolin and this may explain as to why they

behave differently in altering the structure of the HSA molecule.

Flavonoids have been reported to bind to the drug binding sites (Rawel et al., 2006; Yuan

et al., 2007), which are different from the fatty acid binding sites. The drug binding sites 1 and 2

comprise largely apolar cavities with defines polar features located in sub‐domains IIA and IIIA,

respectively. Most drugs bind to the drug binding site 1, which has preference for flat aromatic

compounds that fit between Leu238 and Ala291 in the centre of the pocket (Curry, 2009). The

planar structure of flavonoids may be responsible for their binding predominantly to the drug

site 1, as it was reported for apigenin (Yuan et al., 2007) and quercetin (Rawel et al., 2006).

Usually small adjustments in the side chains in the site 1 cavity can happen upon the binding of

the compounds. These changes are not so extensive that may lead to changes in the secondary

structure of HSA (Curry, 2009), but may be the cause of the decrease in hydrogen–deuterium

exchange rate in the binding of luteolin to HSA.

It is interesting that like with HSA, luteolin and apigenin interaction with lysozyme results

in a decrease in amide proton hydrogen–deuterium exchange rate. This can be due to one or

more of the following reasons:

(i) due to the amide protons participating in stronger H‐bonds within the secondary

structural elements as a consequence of interaction with these compounds. This will make it

more difficult to break the H‐bonds for the deuterium substitution to occur;

(ii) due to amide protons forming H‐bonds to other groups, including luteolin and

apigenin making it more difficult for the hydrogen–deuterium exchange reaction to occur;

(iii) or it could be due to movement of domains/regions of protein that results in a

reduction in solvent accessibility so that deuterium oxide is unable to penetrate into the core of

the protein to replace the amide protons with deuterium.

Similarly to HSA, luteolin induced a greater decrease in lysozyme hydrogen deuterium

exchange (Table 5.2), and presented a higher binding constant and lower ΔGo value at room

temperature, suggesting that the binding of luteolin to lysozyme is more stable at room

temperature than the binding of apigenin.

Chapter V

121

Once again, like with what we saw for HSA, the interaction of rosmarinic acid with

lysozyme is different to that observed for luteolin. Although, there is no significant increase in

hydrogen–deuterium exchange, it at least does not reduce the hydrogen–deuterium exchange.

This could be due to weaker interaction between rosmarinic acid and lysozyme or that the

nature of the binding does not alter the tertiary structure in a way that would make the amide

protons more vulnerable to hydrogen–deuterium exchange.

Although we have detected changes in hydrogen–deuterium exchange, no significant

alteration in the secondary structure was detected. However, other studies concerning the

interaction of flavonoids detected small secondary structure changes, as well as tertiary

structure changes, by the flavonoids alpinetin and cardamonin (He et al., 2006). Our findings

reveal that alterations in protein secondary structure were not found with either luteolin or

apigenin, by FTIR or fluorescence spectrometry, suggesting that the alterations in rate of

hydrogen deuterium exchange is mainly due to changes in the tertiary structure.

As the proteins and the plant phenolic compounds seemed to interact by weak forces by

fluorescence spectrometry, the bonds formed may be reversible and the compounds released

under certain conditions. Studies with total plasma protein extracted from rat blood report that

the affinity of flavonoids to the total protein may be even lower than to purified albumin, which

may be due to presence of metallic ions such as Zn2+, Cu2+, Ca2+ and Mg2+ (Xiao et al., 2011). The

weak and reversible interaction between these plant compounds and plasma proteins is in

agreement with in vivo studies that show a decrease in the concentration of plant phenolic

compounds in plasma, due to metabolism and excretion, as was observed in P. barbatus extract

components (Falé et al., 2011), rosmarinic acid (Baba et al., 2004; Falé et al., 2011) and the

flavonoids apigenin and luteolin (Shi et al., 2011). Therefore, it is important that the structural

changes in plasma proteins when bound to compounds are not so drastic that their function may

be compromised after the release of the compounds.

Lysozyme, apart from drug transport, has an important function related to the immune

response process (Laible & Germaine, 1985). When degranulation occurs, after neutrophils

reach the injured tissue by margination, adhesion and emigration, lysozyme is discharged from

lysosomes of neutrophils and destroys not only the phagosomes but also damages the animal

tissue itself, thus aggravating the response to inflammation (Ronca et al., 1998; Wu et al., 2006).

Previous reports showed that phenolic compounds from Leggera species extracts decreased the

release of lysozyme from the neutrophils to the serum (Wu et al., 2006). In the present study,

P. barbatus extract phenolic compounds inhibited directly lysozyme activity. It is long known

that some flavonoids may act as lysozyme inhibitors (Rodney et al., 1950). Other studies also


122

reported that plant extracts mainly composed of apigenin, luteolin and luteolin glycosides

decreased allergy symptoms caused by egg‐white lysozyme sensitization in rats (Iwaoka et al.,

2010). The same anti‐allergic effect was observed for Perilla frutescens water extract, and by its

major component, rosmarinic acid (Makino et al., 2003).

As P. barbatus components bind to lysozyme with weak interactions, our data suggests

that the enzyme does not undergo major conformational changes, the plant compounds are

eliminated from plasma in a short period of time (Falé et al., 2011), and the antibacterial activity

of lysozyme is not permanently compromised. Therefore, these compounds may be helpful in

decreasing the damage caused by a high lysozyme activity, in response to pathogens or in

allergic conditions.

P. barbatus extract has also proved to have antioxidant activity as a radical scavenger

more powerful than the commercial antioxidant BHT (Falé et al., 2009). This activity is mainly

due to its main component, rosmarinic acid. The present results suggest that P. barbatus water

extract may be useful to treat inflammatory conditions as its components and metabolites may

be transported in circulatory system binding to albumin, the most abundant transport protein,

and to lysozyme, and decrease the inflammation process by two mechanism: as free radical

scavengers, and as lysozyme inhibitors.

2.5. Conclusions

Protein intrinsic fluorescence quenching proved that the P. barbatus extract components

and its metabolites found in rat plasma are able to bind to the human transport proteins

albumin and lysozyme, allowing them to be carried in the bloodstream to organs where they can

have a beneficial activity. The spectroscopic data suggest that the interaction of the plant

phenolic compounds and the proteins in this study is made by weak interactions, causing some

changes in protein tertiary structure, but not in the secondary structure. This suggests that the

compounds may be released from the complexes they form with the proteins, without

compromising the albumin or lysozyme function in plasma.

P. barbatus extract components and metabolites also inhibit lysozyme activity, which

may be helpful in decreasing the aggravation of the inflammation caused by the immune system

in response to pathogens and in allergies. This mechanism can be added to the already well

known radical scavenger capacity of the extract components, making the anti‐inflammatory

activity of plant extract very promising.

Chapter V

123

3. Conclusions

P. barbatus extract components can bind to acetylcholinesterase, human serum albumin

and lysozyme with the formation of a complex protein‐phenolic compound.

The spectroscopic data suggest that the plant phenolic compounds bind to the proteins

through weak interactions, causing some changes in protein tertiary structure, but not in the

secondary structure. This suggests that the compounds may be released from the complexes

they form with the proteins, without compromising irreversibly the acetylcholinesterase,

albumin or lysozyme functions.

The extract components bind to the active gorge of acetylcholinesterase and the

position of the hydroxyl groups in the flavonoids seems to play an important role in the binding

to the enzyme and in the inhibitory activity.

The binding of the P. barbatus extract components to the human transport proteins

albumin and lysozyme allows them to be carried in the bloodstream to organs where they can

have a beneficial activity, and it may explain the bioavailability and brain acetylcholinesterase

inhibition in rats, shown in Chapter IV.

P. barbatus extract components and metabolites also inhibit lysozyme activity, which

may be helpful in decreasing the aggravation of the inflammation caused by the immune system

in response to pathogens and in allergies.

Chapter VI

General Discussion and Conclusions

Chapter VI

127

The use of plant aqueous extracts, prepared as infusions and decoctions, for medicinal

purposes by a wide range of cultures originates an interest in finding the scientific validation of

the activities attributed to these traditional therapies. The use of herbal teas in milder health

disorders may be an alternative therapy with fewer side effects than the use of regular

medications. Plant extracts proved to be important sources of bioactive compounds that may

modulate biochemical reactions, as galanthamine, an acetylcholinesterase inhibitor first

extracted from Galanthus species (Heinrich & Teoh, 2004).

Aqueous extracts of Plectranthus barbatus leaves are reported in the literature to be

used in the treatment of psychiatric and digestive conditions (Lukhoba et al., 2006), suggesting

that these aqueous extract may inhibit acetylcholinesterase activity. Although it’s extracts were

not the most active among the other extracts of Plectranthus species tested for

acetylcholinesterase inhibition, it was the only Plectranthus species whose activities were

associated with other compounds than rosmarinic acid, the main component of all plant extracts

in Chapter III. The IC50 values obtained for acetylcholinesterase inhibition by the components of

the P. barbatus extract ranged from 10‐3 to 10‐4 M, which are much higher (less potent) than the

IC50 value, 5.85 x 10‐7 M, obtained for galanthamine in our lab, in the same experimental

conditions. However, the efficacy of the acetylcholinesterase inhibitors in therapies depends on

their bioavailability, and on the mechanism of inhibition, as irreversible inhibitors are toxic. The

currently commercialized acetylcholinesterase inhibitors present a wide diversity of in vitro

acetylcholinesterase inhibition values, but their therapeutic value is controlled by the mode of

administration, which modulates their bioavailability (McGleenon et al., 1999).

The P. barbatus aqueous extract components also showed antioxidant activity as radical

scavengers, and anti‐inflammatory activity as lysozyme inhibitors and decreasing the amount of

hypochlorous acid in human neutrophils. These biological activities are potentially useful in the

treatment of several inflammation‐related pathologies.

The composition in phenolic compounds of the P. barbatus aqueous extracts is very

similar to extracts of other species from the Lamiaceae family, since it is composed mainly of

rosmarinic acid and flavonoid glucuronides (Abdel‐Mogib et al., 2002; Batista et al., 1996). Apart

from the phenolic compounds, the P. barbatus decoction is also composed of abietane

diterpenoids, compounds also very common in Plectranthus species. However the most active

diterpenoid was metabolized into a non‐active form during the in vitro digestion with pancreatic

juice. Although the diterpenoids showed a high permeability in Caco‐2 monolayers, when the

plant extract was intraperitonially administered to rats just the non‐active diterpenoid was

observed in plasma, suggesting that the active diterpenoid is quickly metabolized and its activity

128

lost. Diterpenes from Salvia milthiorrizha, also a species from the Lamiaceae family, have shown

a high acetylcolinesterase inhibitory activity, with IC50 values in micromolar concentrations

(Wong et al., 2010). These diterpenoids from S. milthiorrizha were also metabolized into

hydroxylated metabolites by rat liver (Li et al., 2006; Wong et al., 2010), in a similar way to the

observed to the diterpenoids of P. barbatus in Chapter III, in pancreatic juice, and in Chapter IV,

possibly by the liver after intraperitoneal administration of the aqueous extract. According to

Wong et al. (2010) the hydroxylation is followed by furan ring cleavage, oxidation, conjugation of

the resulting metabolites and excretion. Although in the present study the active diterpenoid

have a short life, and therefore a limited usefulness to the treatment of Alzheimer’s disease, a

diterpenoid isolated from P. barbatus was shown to inhibit the gastric H+K+ATPase, decreasing

this way the acid secretion in the stomach, which may be an approach in the prevention or

treatment of gastric ulcers (Schultz et al., 2007).

The flavonoids in the P. barbatus aqueous extract were found in the glucuronidated

form, which is more water soluble and therefore easier to extract from the plant material than

the aglycones. The flavonoid glucuronides were stable during the in vitro digestion with gastric

and pancreatic juices, which is in agreement with other reports in the literature for this class of

compounds (Bouayed et al., 2011; Gião et al, 2011; Laurent et al., 2007). The only flavonoid

whose degradation with gastric juice was reported in the literature was procyanidin B2 (Bouayed

et al., 2012), a dimmer of epicatechin that is hydrolysed into monomeric epicatechin in the

acidic conditions of the gastric juice. Anthocyanins were also reported to be hydrolysed to some

extent in the pancreatic juice (McDougall et al., 2005; McDougall et al., 2007).

A limitation in the bioavailability of ingested flavonoids, and phenolic compounds in

general, was reported to be due to the binding of these plant compounds to the protein in the

pancreatic juice (Laurent et al., 2007) and in protein‐rich food such as milk (Cilla et al., 2009), by

hydrophobic interactions and hydrogen bonds. These interactions between plant phenolics and

proteins were also observed in the present study for the components of the P. barbatus aqueous

extract and the proteins acetylcholinesterase, albumin and lysozyme. Although flavonoid

derivatives are generally stable in pancreatic conditions and in contact with intestinal cells

(Chapter III, section 2; Laurent et al., 2007), the glucuronide derivatives may be hydrolysed into

their aglycones by the gut flora β‐glucuronidase activity. Several other authors also reported the

same reaction for flavonoid glycosylated derivatives (Bouayed et al., 2012; Berthiller et al.,

2011), and showed that the bacterial glycosidases are the only responsible for flavonoid

deglycosylation, as flavonoid glycosides showed low affinity for human glycosidases (Berthiller et

al., 2011).

Chapter VI

129

In the present studies rosmarinic acid was stable in gastric but 25% of its amount in the

plant extract was hydrolysed in pancreatic conditions. Other authors also referred that some

degradation may occur in the passage from gastric to pancreatic conditions (Gião et al., 2011). A

similar degradation also occurs with chlorogenic acid, with formation of caffeic acid (Gião et al.,

2011; Fazzari et al., 2008; Bouayed et al., 2012). Both rosmarinic acid and chlorogenic acid are

esters of caffeic acid, and therefore may be relatively stable in acidic conditions but easily

hydrolysed in basic conditions.

When P. barbatus decoction was intragastrically administered to rats, and when it was

applied to the Caco‐2 cell monolayers for the permeability assay, low amounts of unconjugated

rosmarinic acid was found to permeate the intestinal barrier, which may be due to the low

affinity of rosmarinic acid to the MCT trasporters (Chapter IV section 2). The low affinity of

rosmarinic acid to the membrane transporters led other authors to suggest that no transporters

were involved in the permeability of rosmarinic acid. However, in the present study assays were

carried out during longer times, 6 hours instead of 40 minutes (Konishi & Kobayashi, 2005),

which allowed to detect the differences in permeability in the presence and absence of

competitors for the same transporters. Glucuronidated and sulfated metabolites of rosmarinic

acid were found in the plasma of rats after intragastric administration, while methylated

metabolites were present after intraperitonial administration. These facts suggest that

rosmarinic acid may be predominantly glucuronidated and sulphated in the intestine, and

undergo mainly methylation by the liver. In Caco‐2 cell monolayers rosmarinic acid remained

intact (Chapters III and IV), which is not in agreement with the results found in in vivo studies

with rats (Chapter IV), or with the literature (Konishi & Kobayashi, 2005). Nevertheless

rosmarinic acid in the Caco‐2 cell monolayer studies in the Chapters III and IV was co‐

administered with flavonoids that may have more affinity to phase II enzymes. Also the Caco‐2

monolayers and the rats are different bioavailability models that may present slightly different

results (Keldenish, 2009).

The flavonoid glucuronides in the plant extract permeated the intestinal barrier both in

the Caco‐2 and in the rat models. Most studies in the literature focus on flavonoid aglycones,

which can be metabolized by Caco‐2 cells in flavonoid glucuronides and sulfates (Walle et al

1999; Barrington et al., 2009). The conjugates are reported by the same authors to be effluxed

mainly to the apical side of the Caco‐2 monolayer, corresponding to the gut lumen in the

intestine, limiting their bioavailability. In chapter IV it is shown that the glucuronides can also be

transported from the apical to the basolateral side of the Caco‐2 monolayers, although the

apigenin and luteolin glucuronides may also be effluxed back to the apical side possibly with the

130

involvement of Pgp transporter. Studies regarding the bioavailability of flavonoid glycosylated

derivatives are scarce in the literature, however the quercetin derivative rutin was found to

permeate the Caco‐2 monolayers from apical to basolateral side (Gião et al 2011). The flavonoid

glycosylated derivatives may be hydrolysed into their aglycones as discussed above, or by the

liver β‐glucuronidases, as observed and discussed in Chapter IV. In the present study the

glucuronidation of flavonoid aglycones was observed in Caco‐2 cells, but sulfated metabolites

were not detected. Also, when the plant extract was intraperitonially administered to rats,

although liver β‐glucuronidase activity occurred, no sulfated metabolites of the flavonoids were

found. The sulfotransferase activity was shown to occur in the rat liver, as the sulfated

derivatives of the phenolic compounds were synthesised in vitro using rat liver protein extracts

(Chapter IV). Other authors have reported that the glucuronidation of flavonoids was

predominant, and occurred in more positions than the sulfation of the compounds (Barrington

et al., 2009). This suggests a lower affinity of the flavonoids to the sulfotransferases than to the

glucuronyltransferases.

Due to the differences found when administering rosmarinic acid in the extract or as an

isolated compound, a study was performed in Caco‐2 cell monolayers using central composite

design (CCD) to determine if the phenolic compounds interfere with the bioavailability of one

another, and if transporters were involved in the bioavailability of the compounds. It was found

that rosmarinic acid, apigenin and luteolin can permeate Caco‐2 cell monolayers through

transporters and, generally, have higher bioavailability together than when isolated. This

increase of bioavailability is mainly due to inhibition or competition for efflux transporters that

transport compounds back to the apical side of the Caco‐2 monolayers (Chapter IV section 2).

The P. barbatus extract components followed the same behaviour predicted by the model

mixture of rosmarinic acid, luteolin and apigenin.

Flavonoids are well known inhibitors of ABC transporters, such as Pgp, MRP and BCRP

(Brand et al., 2006). Some flavonoids are also substrates of these transporters, which limits their

bioavailability, such as quercetin, kaempferol, isorhametin (Wang et al., 2005), and chrysin

(Walle et al., 1999). As substrates and inhibitors of this type of efflux transporters, the

coadministration of many compounds increases their bioavailability by decreasing the transport

from basolateral to apical side, as it was described in Chapter IV. In the same way the inhibition

of effux systems may increase the bioavailability of some orally administered medicines (Brand

et al., 2006), as for instance, doxorubicin, an anticancer drug whose bioavailability is increased

by inhibiting Pgp with quercetin (Choi et al., 2011).

Chapter VI

131

As it was discussed in Chapter IV, the values for the inhibition of acetylcholinesterase

activity in rat brains after the administration of the plant extract were comparable to those

found in the literature for galanthamine, even though a lower amount of galanthamine was

administered (Geerts et al., 2005; Chattipakorn et al., 2007). Only a low amount of extract

compounds was found in the brain of rats. Although the amount of these plant compounds, or

their metabolites, was much higher, the amount found in the brain kept constant for a longer

period of time. This fact suggests that if the extract is administered several times a day, the

amount of extract components in the brain should be kept relatively constant, even though the

amounts of extract metabolites in the plasma may have high variations during the day.

Differences of plasma bioavailability, half‐life, and brain bioavailability lead to different

strategies in the administration of acetylcholinesterase inhibitors. Tacrine has high brain

bioavailability, but when taken orally, it has low plasma bioavailability and low half‐life.

Therefore the administration of tacrine follows a similar strategy to the one suggested for the

decoction of P. barbatus in this study, of several administrations during the day (McGleenon et

al., 1999).

Peripheral effects of acetylcholinesterase inhibition, such as gastrointestinal and hepatic

disturbances, are usually associated with the administration of large dosages of

acetylcholinesterase inhibitors. However, as mentioned in Chapters III (section 2) and IV (section

2), with 1mg/ml of the plant extract or 90 μM of any of the standards no toxicity was observed in

Caco‐2 cells, and 100μg/mL of plant extract or 20μM of pure compound would be the maximum

values accepted to consider the occurrence of toxicity (Cárdenas et al. 2006; Yu et al., 2007).

Furthermore, toxic effects are not reported for P. barbatus, on the contrary, the herbal tea of

this plant is usually drunk to treat gastrointestinal problems (Lukhoba et al., 2006). The effect of

the P. barbatus decoction on the gastrointestinal tract may be due to a mild inhibition of

acetylcholinesterase activity, increasing gastrointestinal motility and gut hydration (Jarvie et al.,

2008; Hirota and McKay 2006).

The binding of the plant extract components to the plasma proteins, albumin and

lysozyme, may also decrease the side‐effects of the acetylcholinesterase inhibition by reducing

the availability of the compounds to peripheral acetylcholinesterase while transporting them

through the bloodstream to the target organ, the brain. Although, as was discussed in chapter V,

the compounds seem to bind in the active gorge of acetylcholinesterase, the weak interactions

between the plant phenolics and acetylcholinesterase do not cause large conformational

changes in the enzyme and the enzymatic activity may be easily recovered after the excretion of

the compounds. Therefore, the weak interactions between the plant phenolics and

132

acetylcholinesterase may also be preferable, decreasing the side effects of excessive

acetylcholinesterase inhibition.

The P. barbatus extract showed radical scavenging activity with the DPPH and protected

fatty acids from peroxidation in the β‐carotene/linoleic acid assay, and therefore it may be

useful preventing oxidative stress‐related cell membrane disruption and apoptosis. This cell

death mechanism is similar to the one caused by the oxidative stress resultant from the

accumulation of amyloid plaques in Alzheimer’s disease patients. The plant extract and

rosmarinic acid also decreased the amount of hypochlorous acid produced by activated

neutrophils (Chapter III) and inhibited lysozyme activity (Chapter V), which can be associated

with a immune response, but also with a cytotoxic inflammatory reaction (Ronca et al., 1998;

Deby‐Dupont, 1999). The cognition status of Alzheimer’s disease patients has been correlated

with the systemic oxidative stress status (Viña et al., 2004), and therefore the plant extract may

be useful in this approach to the treatment of Alzheimer’s disease. Furthermore, the

components of the plant extract act as antioxidants through different mechanisms, which has

been proposed as the most efficient strategy in the antioxidant treatment of Alzheimer’s disease

(Macocci & Polidori, 2011).

In conclusion,

The P. barbatus decoction showed promising in vitro antioxidant, anti‐inflammatory and

anti‐acetylcholinesterase activities.

The P. barbatus herbal tea showed no toxicity in Caco‐2 cells. The low toxicity is also

supported by the weak interactions established with proteins without causing secondary

structure changes.

Some active components of the P. barbatus herbal tea undergo metabolization in the

gastrointestinal tract, but the phenolic compounds can cross the intestinal barrier and

reach the brain, where they can inhibit acetylcholinesterase activity, similarly to the

currently used acetylcholinesterase inhibitors.

The phenolic compounds can circulate in the bloodstream bound to the plasma proteins

and decrease oxidative stress‐related processes, such as inflammation, by decreasing

the amount of hypochlorous acid released by neutrophils and also by inhibiting lysozyme

activity.

The use of polyphenolic compound mixtures such as herbal teas may increase the

bioavailability of the active compounds and, as the herbal tea components show many

activities, may be useful in the treatments with multiple approaches, as it is thought to

be the most efficient for Alzheimer’s disease.

References

References

135

Abdel‐Mogib M, Albar HA, Batterjee SM. 2002. Chemistry of the genus Plectranthus.

Molecules, 7: 271–301.

Abou‐Donia MB. 2003. Organophosphorous ester‐induced chronic neurotoxicity. Arch

Environ Health. 58:484‐497.

Adsersen A, Gauguin B, Gudiksen L, Jäger AK. 2006. Screening of plants used in Danish folk

medicine to treat memory dysfunction for acetylcholinesterase inhibitory activity. J

Ethnopharmacol. 104: 418‐22.

Alasbahi RH, Melzig MF. 2010a. Plectranthus barbatus: a review of phytochemistry,

ethnobotanical uses and pharmacology ‐ Part 1. Planta Med. 76: 653‐661.

Alasbahi RH, Melzig MF. 2010b. Plectranthus barbatus: a review of phytochemistry,

ethnobotanical uses and pharmacology ‐ part 2. Planta Med. 76: 753‐765.

Albuquerque RL, Kentopff MR, Machado MIL, Silva MG, Matos FJA. 2007. Diterpenos tipo

abietano isolados de Plectranthus barbatus Andrews. Química Nova. 30: 1882–1886.

Alzheimer’s association. 2008. Alzheimer’s disease facts and figures. Alzheimer’s &

Dementia. 4: 110–133.

Atoui AK, Mansouri A, Bosku G, Kefalas P. 2005. Tea and herbal infusions: their antioxidant

activity and phenolic profile. Food Chem., 89: 27–36.

Azuma K, Ippoushi K, Nakayama M, Ito H, Higashio H, Terao J. 2000. Absorption of

Chlorogenic Acid and Caffeic Acid in Rats after Oral Administration. J. Agric. Food Chem. 48:

5496‐5500.

Baba S, Osakabe N, Natsume M, Terao J. 2004. Orally administered rosmarinic acid is present

as the conjugated and/or methylated forms in plasma, and is degraded and metabolized to

conjugated forms of caffeic acid, ferulic acid and m‐coumaric acid. Life Sci. 75: 165–178.

Baba S, Osakabe N, Natsume M, Yasuda A, Muto Y, Hiyoshi K, Takano H, Yoshikawa T, Terao

J. 2005. Absorption, metabolism, degradation and urinary excretion of rosmarinic acid after

intake of Perilla frutescens extract in humans. Eur J Nutr. 44: 1–9.

Babior BM. 2000. Phagocytes and oxidative stress. Am J Med. 109:33‐44.

Bachoual R, Talmoudi W, Boussetta T, Braut F, El‐Benna J. 2011. An aqueous pomegranate

peel extract inhibits neutrophil myeloperoxidase in vitro and attenuates lung inflammation

in mice. Food Chem Toxicol. 49:1224‐8.

Barker, T. B., in: Schilling, E. G. (Ed.), Quality by Experimental Design, Marcel Dekker, New

York 1985, pp. 310–325.

Barrington R, Williamson G, Bennett RN, Davis BD, Brodbelt JS, Kroon PA. 2009. Absorption,

conjugation and efflux of the flavonoids, kaempferol and galangin, using the intestinal Caco‐

2/TC7 Cell Model. J Funct Foods. 1:74‐87.

References

136

Batista O, Simões MF, Nascimento J, Riberio S, Duarte A, Rodríguez B, de la Torre MC. 1996.

A rearranged abietane diterpenoid from Plectranthus hereroensis. Phytochemistry. 41: 571–

573.

Bembenek SD, Keith JM, Letavic MA, Apodaca R, Barbier AJ, Dvorak L, Aluisio L, Miller KL,

Lovenberg TW, Carruthers NI. 2008. Lead identification of acetylcholinesterase inhibitors –

histamine H3 receptor antagonists from molecular modelling. Bioorg Med Chem. 16: 2968‐

2973.

Berthiller F, Krska R, Domig KJ, Kneifel W, Juge N, Schuhmacher R, Adam G. 2011. Hydrolytic

fate of deoxynivalenol‐3‐glucoside during digestion. Toxicol Lett. 206: 264‐267.

Birman S. 1985. Determination of acetylcholinesterase activity by a new chemiluminescence

assay with the natural substrate. Biochem. J. 225: 825‐828.

Bohnen NI, Kaufer DI, Hendrickson R, Ivanco LS, Lopresti BJ, Koeppe RA, Meltzer CC,

Constantine G, Davis JG, Mathis CA, Dekosky ST, Moore RY. 2005. Degree of inhibition of

cortical acetylcholinesterase activity and cognitive effects by donepezil treatment in

Alzheimer’s disease. J. Neurol., Neurosurg. Psychiatry. 76: 315–319.

Bouayed J, Deusser H, Hoffmann L, Bohn T. 2012. Bioaccessible and dialysable polyphenols

in selected apple varieties following in vitro digestion vs. their native patterns. Food Chem.

131: 1466‐1472.

Bouayed J, Hoffmann L, Bohn T. 2011. Total phenolics, flavonoids, anthocyanins and

antioxidant activity following simulated gastro‐intestinal digestion and dialysis of apple

varieties: Bioaccessibility and potential uptake. Food Chem. 128: 14‐21.

Bourne Y, Grassi J, Bougis PE, Marchot P. 1999. Conformational flexibility of the

acetylcholinesterase tetramer suggested by X‐ray crystallography. J. Biol. Chem. 274: 30370–

30376.

Bourne Y, Radić Z, Taylor P, Marchot P. 2010. Conformational remodeling of femtomolar

inhibitor − acetylcholinesterase complexes in the crystalline state. J Am Chem Soc. 132: 18292‐

18300.

Brand W, Padilla B, van Bladeren PJ, Williamson G, Rietjens IM. 2010. The effect of co‐

administered flavonoids on the metabolism of hesperetin and the disposition of its

metabolites in Caco‐2 cell monolayers. Mol Nutr Food Res. 54: 851‐60.

Brand W, Schutte ME, Williamson G, van Zanden JJ, Cnubben NH, Groten JP, van Bladeren PJ,

Rietjens IM. 2006. Flavonoid‐mediated inhibition of intestinal ABC transporters may affect

the oral bioavailability of drugs, food‐borne toxic compounds and bioactive ingredients.

Biomed Pharmacother. 60: 508‐519.

Brand W, van der Wel PA, Rein MJ, Barron D, Williamson G, van Bladeren PJ, Rietjens IM.

Metabolism and transport of the citrus flavonoid hesperetin in Caco‐2 cell monolayers. Drug

Metab Dispos. 36: 1794‐1802.

References

137

Cai Y, Luo Q, Sun M, Corke H. 2004. Antioxidant activity and phenolic compounds of 112

traditional Chinese medicinal plants associated with anticancer. Life Sci. 74: 2157‐2184.

Cárdenas M, Mardel M, Blank VC, Roguin LP. 2006. Antitumor activity of some natural

flavonoids and synthetic derivatives on various human and murine cancer cell lines. Bioorg

Med Chem. 14: 2966‐2971.

Cellek S, Thangiah R, Jarvie EM, Vivekanandan S, Lalude O, Sanger GJ. 2008. Synergy

between 5‐HT4 receptor activation and acetylcholinesterse inhibition in human colon and

rat forestomach. Neurogastroenterol Motil. 20: 539‐545.

Chattipakorn S, Pongpanparadorn A, Pratchayasakul W, Pongchaidacha A, Ingkaninan K,

Chattipakorn N. 2007. Tabernaemontana divaricata extract inhibits neuronal

acetylcholinesterase activity in rats. J. Ethnopharmacol. 110: 61–68.

Choi JS, Piao YJ, Kang KW. 2011. Effects of quercetin on the bioavailability of doxorubicin in

rats: role of CYP3A4 and P‐gp inhibition by quercetin. Arch Pharm Res. 34: 607‐13.

Cilla A, Gonzalez‐Sarrias A, Tomas‐Baeberan F, Espin JC, Barbera R. 2009. Availability of

polyphenols in fruit beverages subjected to in vitro gastrointestinal digestion and their

effects on proliferation, cell‐cycle and apoptosis in human colon cancer Caco‐2 cells. Food

Chem. 114: 813–820.

Codd LE. 1975. Plectranthus (Labiatae) and allied genera in southern Africa. Bothalia. 11:

371‐442.

Coelho M. 2004. Eixo Renina‐Angiotensina. Polimorfismos genéticos e interacções

fenotípicas. Lisboa: Universidade de Lisboa.

Coleta M, Camposa MG, Cotrim MD, Lima MCT, Cunha AP. 2008. Assessment of luteolin

(3’,4’,5,7‐tetrahydroxyflavone) neuropharmacological activity, Behav. Brain Res. 189: 75–82.

Curry S. 2009. Lessons from the crystallographic analysis of small molecule binding to human

serum albumin, Drug Metab Pharmacok. 24: 342–357.

da Silva EL, Piskula MK, Yamamoto N, Moon JH, Terao J. 1998. Quercetin metabolites inhibit

copper ion‐induced lipid peroxidation in rat plasma. FEBS Lett. 430: 405‐408.

Deby‐Dupont G, Deby C, Lamy M. 1999. Neutrophil myeloperoxidase revisited: it’s role in

health and disease. Intensivmed. 36: 500–513.

Dickson D, Lee S, Mattiace L, Yen S, Brosnan C. 1993. Microglia and cytokines in neurological

disease, with special reference to AIDS and Alzheimer’s disease. Glia. 7: 75‐83.

Djeridane A, Yousfi M, Nadjemi B, Boutassouna D, Stocker P, Vidal N. 2006. Antioxidant

activity of some Algerian medicinal plants extracts containing phenolic compounds. Food

Chem. 97: 654–660.

D'Souza VM, Shertzer HG, Menon AG, Pauletti GM. 2003. High glucose concentration in

isotonic media alters Caco‐2 cell permeability. AAPS PharmSci. 5:, E24.

References

138

Dumont M, Beal MF. 2011. Neuroprotective strategies involving ROS in Alzheimer disease.

Free Radic Biol Med. 51: 1014‐1026.

Dypbukt J, Bishop C, Brooks W, Thong B, Eriksson H,Kettle A. 2005. A sensitive and selective

assay for chloramine production by mieloperoxidase. Free Radic Biol Med. 39: 1468‐1477.

Eastman J, Wilson EJ, Cerveñansky C, Rosenberry TL. 1995. Fasciculin 2 binds to the

peripheral site on acetylcholinesterase and inhibits substrate hydrolysis by slowing a step

inhibiting proton transfer during enzyme action. J Biol Chem. 270: 19694‐19701.

Eliman GL, Courtney KD, Andres V.,Featherstone RM. 1961. A new and rapid colorimetric

determination of acetylcholinesterase activity. Biochem. Pharmacol. 7: 88‐95

El‐Kamali HH. 2009. Medicinal plants in East and Central Africa: Challenges and constraints.

Ethnobotanical Leaflets. 13: 364–369.

Ellman GL. 1958. A colorimetric method for determining low concentrations of mercaptans.

Arch Biochem Biophys. 74: 443‐450.

Erlung I. 2004. Review of the flavonoids quercetin, hesperetin, and naringenin. Dietary

sources, bioactivities, bioavailability, and epidemiology. Nutr. Res. 24: 851–874.

Eyer P, Szinicz L, Thiermann H, Worek F, Zilker T. 2007. Testing of antidotes of

organophosphorus compounds: Experimental procedures and clinical reality. Toxicol. 233:

108‐119.

Falé PL, Borges C, Madeira PJA, Ascensão L, Araújo MEM, Florêncio MH, Serralheiro MLM.

2009. Rosmarinic acid, scutellarein 4’‐methyl ether 7‐O‐glucuronide and (16S)‐coleon E are

the main compounds responsible for the antiacetylcholinesterase and antioxidant activity in

herbal tea of Plectranthus barbatus (‘‘falso boldo’’), Food Chem. 114: 798– 805.

Falé PLF, Madeira PJA, Florêncio MH, Ascensão L, Serralheiro MLM. Function of Plectranthus

barbatus herbal tea as neuronal acetylcholinesterase inhibitor. Food Funct. 2: 130‐136.

Fazzari M, Fukumoto L, Mazza G, Livrea MA, Tesoriere L, Marco LD. 2008. In vitro

bioavailability of phenolic compounds from five cultivars of frozen sweet cherries (Prunus

avium L.). J Agric Food Chem. 56: 3561‐3568.

Fecka I, Turek S. 2007. Determination of water‐soluble polyphenolic compounds in

commercial herbal teas from Lamiaceae: peppermint, melissa, and sage. J. Agric. Food

Chem. 55: 10908‐10917.

Ferreira A, Proença C, Serralheiro MLM, Araújo MEM. 2006. The in vitro screening for

acetylcholinesterase inhibition and antioxidant activity of medicinal plants from Portugal. J

Ethnopharm. 108: 31‐37.

Galijatovic A, Walle UK, Walle T. 2000. Induction of UDPglucuronosyltransferase by the

flavonoid chrysin and quercetin in Caco‐2. Pharm Res. 17: 21–26.

References

139

Geerts H, Guillaumat PO, Grantham C, Bode W, Anciaux K, Sachak S. 2005. Brain levels and

acetylcholinesterase inhibition with galantamine and donepezil in rats, mice, and rabbits,

Brain Res. 1033: 186–193.

Gião MS, Gomes S, Madureira AR, Faria A, Pestana D, Calhau C, Pintado ME, Azevedo I,

Malcata FX. 2011. Effect of in vitro digestion upon the antioxidant capacity of aqueous

extracts of Agrimonia eupatoria, Rubus idaeus, Salvia sp. and Satureja Montana. Food Chem.

doi:10.1016/j.foodchem.2011.09.030.

Gill RK, Saksena S, Alrefai WA, Sarwar Z, Goldstein JL, Carroll RE, Ramaswamy K, Dudeja PK.

2005. Expression and membrane localization of MCT isoforms along the length of the human

intestine. Am J Physiol Cell Physiol. 289: C846‐852.

Gomes A, Fernandes E, Lima JFC, Mira L, Corvo ML. 2008. Molecular mechanisms of anti‐

inflammatory activity mediated flavonoids. Curr. Med. Chem. 15: 1586–1605.

Gorne‐Tschelnokow U, Naumann D, Weise C, Hucho F. 1993. Secondary structure and

temperature behaviour of acetylcholinesterase Studies by Fourier‐transform infrared

spectroscopy. Eur J Biochem. 213: 1235‐1242.

Gouwy M, Struyf S,Proost P, Damme JV. 2005. Synergy in cytokine and chemokine networks

amplifies the inflammatory response. Cytokine Growth Factor Rev. 16: 561–580.

Grayer RJ, Eckert MR, Veitch NC, Kite GC, Marin PD, Kokubun T, Simmonds MS, Paton AJ.

2003. The chemotaxonomic significance of two bioactive caffeic acid esters, nepetoidins A

and B, in the Lamiaceae. Phytochemistry. 64: 519–528.

Grayer, R. J., Veitch, N. C., Kite, G. C., Paton, A. J., & Garnock‐Jones, P. J. (2002). Scutellarin

4’‐methyl ether glycosides as taxonomic markers in Teucridium and Tripora (Lamiaceae,

Ajugoideae). Phytochemistry, 60, 727–731.

Guginski G, Luiz AP, Silva MD, Massaro M, Martins DF, Chaves J, Mattos RW, Silveira D,

Ferreira VM, Calixto JB, Santos AR. 2009. Mechanisms involved in the antinociception caused

by ethanolic extract obtained from the leaves of Melissa officinalis (lemon balm) in mice,

Pharmacol., Biochem. Behav. 93: 10–16.

Harel M, Schalk I, Ehret‐Sabatier L, Bouet F, Goeldner M, Hirth C, Axelsen PH, Silman I,

Sussman JL. 1993. Quaternary ligand binding to aromatic residues in the active‐site gorge of

acetylcholinesterase. Proc Natl Acad Sci U S A. 90: 9031–9035.

Harel M, Sonoda LK, Silman I, Sussman JL, Rosenberry TL. 2008. Crystal structure of thioflavin

T bound to the peripheral site of Torpedo californica acetylcholinesterase reveals how

thioflavin T acts as a sensitive fluorescent reporter of ligand binding to the acylation site, J.

Am. Chem. Soc. 130: 7856–7861.

Haris PI, Chapman D, Harrison RA, Smith KF, Perkins SJ. 1990. Conformational transition

between native and reactive center cleaved forms of α1‐antitrypsin by Fourier transform

infrared spectroscopy and small‐angle neutron scattering. Biochemistry. 29: 1377–1380.

References

140

Haris PI, Lee DC, Chapman D. 1986. A Fourier transform infrared investigation of the structural

differences between ribonuclease A and ribonuclease S. Biochim Biophys Acta. 874: 255‐265.

Haris PI, Severcan F. 1999. FTIR spectroscopic characterization of protein structure in

aqueous and non‐aqueous media. J Mol Catal B Enzym. 7: 207‐221

Hartshorne DJ, Stracher A. 1965. Deuterium‐hydrogen exchange of muscle proteins,

Biochemistry. 4: 1917‐1923.

Hashimoto M, Rockenstein E, Crews L, Masliah E. 2003. Role of protein aggregation in

mitochondrial dysfunction and neurodegeneration in Alzheimer's and Parkinson's diseases.

Neuromolecular Med. 4: 21‐36.

He W, Li Y, Tang J, Luan F, Jin J, Hu Z. 2006. Comparison of the characterization on binding of

alpinetin and cardamonin to lysozyme by spectroscopic methods. Int. J. Biol. Macromol. 39:

165–173.

Heinrich M, Teoh HL. 2004. Galanthamine from snowdrop – the development of a modern

drug against Alzheimer’s disease from local Caucasian knowledge. J Ethnopharmacol. 92:

147–162.

Hirota CL, McKay DM. 2006. Cholinergic regulation of epithelial ion transport in the

mammalian intestine. Br J Pharmacol. 149: 463‐479.

Hollman PC, Katan MB. 1997. Absorption, metabolism and health effects of dietary

flavonoids in man. Biomed Pharmacother. 51: 305–310.

Holzer P, Maggi CA. 1994. Synergistic role of muscarinic acetylcholine and tachykinin NK‐2

receptors in intestinal peristalsis. Naunyn Schmiedebergs Arch Pharmacol. 349: 194‐201.

Hostettmann K, Borloz A, Urbain A, Marston A. 2006. Natural product inhibitors of

acetylcholinesterase. Curr Org Chem. 10: 825–847.

Houghton PJ, Agbedahunsi JM, Adegbulugbe A. 2004. Choline esterase inhibitory properties

of alkaloids from two Nigerian Crinum species. Phytochemistry, 65: 2893‐2896.

Huebbe P, Wagner AE, Boesch‐Saadatmandi C, Sellmer F, Wolffram S, Rimbach G. 2010.

Effect of dietary quercetin on brain quercetin levels and the expression of antioxidant and

Alzheimer’s disease relevant genes in mice. Pharmacol Res. 61: 242–246.

Hvidt A, Wallevik K. 1972. Conformational changes in human serum albumin as revealed by

hydrogen‐deuterium exchange studies. J. Biol. Chem. 247: 1530‐1535.

Ingkaninan K, Temkitthawon P, Chuenchon K, Yuyaem T, Thongnoi W. 2003. Screening for

acetylcholinesterase inhibitory activity in plants used in Thai traditional rejuvenating and

neurotonic remedies. J. Ethnopharmacol. 89: 261–264.

Iwaoka E, Oku H, Iinuma M, Ishiguro K. 2010. Allergy‐preventive effects of the flowers of

Impatiens textori. Biol Pharm Bull. 33: 714‐716.

References

141

Iyo M, Namba H, Fukushi K, Shinotoh H, Nagatsuka S, Suhara T, Sudo Y, Suzuki K, Irie T. 1997.

Measurement of acetylcholinesterase by positron emission tomography in the brains of

healthy controls and patients with Alzheimer’s disease. Lancet. 349: 1805–1809.

Jaarsveld E. 2006. The southern African Plectranthus and the art of turning shade to glade.

Fernwood Press. Simon’s Town, South Africa.

Jackson M, Haris PI, Chapman D. 1991. Fourier transform infrared spectroscopic studies of

calcium‐binding proteins. Biochemistry. 30: 9681‐9686.

Jacobsen CF, Leonis J, Linderstrom‐Lang K, Otteren M. 1957. The pH‐stat and its use in

biochemistry. Methods Biochem Anal. 4: 171‐210.

Janicsák G, Máthé I, Miklóssy‐Vári V, Blunden G. 1999. Comparative studies of the rosmarinic

and caffeic acid contents of Lamiaceae species. Biochem Syst Ecol. 27: 733‐738.

Jarvie E, Cellek S, Sanger G. 2008. Potentiation by cholinesterase inhibitors of cholinergic

activity in rat isolated stomach and colon. Pharmacol Res. 58: 297–301.

Ji H‐F, Zhang H‐Y. 2006., Theoretical evaluation of flavonoids as multipotent agents to

combat Alzheimer's disease. J Mol Struct THEOCHEM. 767: 3–9.

Jiang WL, Chen XG, Qu GW, Yue XD, Zhu HB, Tian JW, Fu FH. 2009. Rosmarinic acid protects

against experimental sepsis by inhibiting proinflammatory factor release and ameliorating

hemodynamics. Shock. 32: 608‐13.

Jin J, Zhang X. 2008. Spectrophotometric studies on the interaction between pazufloxacin

mesilate and human serum albumin or lysozyme. J Luminescence. 128: 81–86.

Johnson CD, Russell RL. 1975. A rapid, simple radiometric assay for cholinesterase, suitable

for multiple determinations. Anal Biochem. 64: 229‐238.

Justino GC, Santos MR, Canário S, Borges C, Florêncio MH, Mira L. 2004. Plasma quercetin

metabolites: structure–antioxidant activity relationships. Arch Biochem Biophys. 432: 109–

121.

Kadir A, Darreh‐Shori T, Almkvist O, Wall A, Grut M, Strandberg B, Ringheim A, B Eriksson,

Blomquist G, Langstrom B, Nordberg A. 2008. PET imaging of the in vivo brain

acetylcholinesterase activity and nicotine binding in galantaminetreated patients with AD.

Neurobiol Aging. 29: 1204–1217.

Kanatt SR, Chander R, Sharma A. 2007. Antioxidant potential of mint (Mentha spicata L.) in

radiation‐processed lamb meat. Food Chem. 100: 451–458.

Keldenich J. 2009. Measurement and prediction of oral absorption. Chem Biodivers. 6: 2000‐

2013.

Kern SM, Bennett RN, Needs PW, Mellon FA, Kroon PA, Garcia‐Conesa M‐T. 2003.

Characterization of metabolites of hydroxycinnamates in the in vitro model of human small

intestinal epithelium Caco‐2 cells. J Agric Food Chem. 51: 7884–7891.

References

142

Kerr ML, Small DH. 2005. Cytoplasmic domain of the beta‐amyloid protein precursor of

Alzheimer's disease: function, regulation of proteolysis, and implications for drug

development. J Neurosci Res. 80: 151‐159.

Khan MT, Orhan I, Senol FS, Kartal M, Sener B, Dvorská M, Smejkal K, Slapetová T. 2009.

Cholinesterase inhibitory activities of some flavonoid derivatives and chosen xanthone and

their molecular docking studies. Chem Biol Interact. 181: 383‐389.

Kim H, Kim H, Jung S. 2008. Aqueous solubility enhancement of some flavones by

complexation with cyclodextrins, Bull Korean Chem Soc. 29: 590–594.

Klebanoff SJ. 2005. Myeloperoxidase: friend and foe. J Leukoc Biol. 77: 598‐625.

Kobayashi S, Konishi Y. 2008. Transepithelial transport of flavanone in intestinal Caco‐2 cell

monolayers. Biochem. Biophys. Res. Commun. 368: 23‐29.

Kobayashi S, Tanabe S, Sugiyama M, Konishi Y. 2008. Transepithelial transport of hesperetin

and hesperidin in intestinal Caco‐2 cell monolayers. Biochim Biophys Acta. 1778: 33‐41.

Kobayashi S, Watanabe J, Fukushi E, Kawabata J, Nakajima M, Watanabe M. 2003.

Polyphenols from some foodstuffs as inhibitors of ovalbumin permeation through Caco‐2

cell monolayers. Biosci. Biotechnol. Biochem. 67: 1250‐1257.

Komes D, Belščak‐Cvitanović A, Horžić D, Rusak G, Likić S, Berendika M. 2011. Phenolic

composition and antioxidant properties of some traditionally used medicinal plants affected

by the extraction time and hydrolysis. Phytochem. Anal. 22: 172‐180.

Kong H, Wang M, Venema K, Maathuis A, van der Heijden R, van der Greef J, Xu G,

Hankemeier T. 2009. Bioconversion of red ginseng saponins in the gastro‐intestinal tract in

vitro model studied by high‐performance liquid chromatography–high resolution Fourier

transform ion cylcotron resonance mass spectrometry. J Chromatogr A. 1216: 2195–2203.

Konishi Y, Hitomi Y, Yoshioka E. 2004. Intestinal Absorption of p‐Coumaric and Gallic Acids in

Rats after Oral Administration. J. Agric. Food Chem. 52: 2527‐2532

Konishi Y, Kobayashi S, Shimizu M. 2003. Tea polyphenols inhibit the transport of dietary

phenolic acids mediated by the monocarboxylic acid transporter (MCT) in intestinal Caco‐2

cell monolayers. J Agric Food Chem. 51: 7296‐7302.

Konishi Y, Kobayashi S. 2004a. Microbial Metabolites of Ingested Caffeic Acid Are Absorbed

by the Monocarboxylic Acid Transporter (MCT) in Intestinal Caco‐2 Cell Monolayers. J. Agric.

Food Chem. 52: 6418‐6424

Konishi Y, Kobayashi S. 2004b. Transepithelial Transport of Chlorogenic Acid, Caffeic Acid,

and Their Colonic Metabolites in Intestinal Caco‐2 Cell Monolayers. J. Agric. Food Chem. 52:

2518‐2526.

Konishi Y, Kobayashi S. 2005. Transepithelial transport of rosmarinic acid in intestinal Caco‐2

monolayers. Biosci Biotechnol Biochem. 69: 583‐591.

References

143

Konishi Y, Zhao Z, Shimizu M. 2006. Phenolic Acids Are Absorbed from the Rat Stomach with

Different Absorption Rates. J. Agric. Food Chem. 54: 7539‐7543.

Konishi Y., Hitomi Y, Yoshida M, Yoshioka E. 2005. Pharmacokinetic Study of Caffeic and

Rosmarinic Acids in Rats after Oral Administration. J Agric Food Chem. 53: 4740–4746.

Kreis ME, Kasparek M, Zittel TT, Becker HD, Jehle EC. 2001. Neostigmine increases

postoperative colonic motility in patients undergoing colorectal surgery. Surgery. 130: 449‐

456.

Kumaran A, Karunakaran RJ. 2007. Activity‐guided isolation and identification of free radical‐

scavenging components from an aqueous extract of Coleus aromaticus. Food Chem., 100:

356–361.

Kuo MY, Liao MF, Chen FL, Li YC, Yang ML, Lin RH, Kuan YH. 2011. Luteolin attenuates the

pulmonary inflammatory response involves abilities of antioxidation and inhibition of MAPK

and NFκB pathways in mice with endotoxin‐induced acute lung injury. Food Chem Toxicol.

49: 2660‐2666.

Lafay S, Gil‐Izquierdo A. 2008. Bioavailability of phenolic acids. Phytochem Rev. 7: 301–311.

Laible NJ, Germaine GR. 1985. Bactericidal activity of human lysozyme, muramidase‐inactive

lysozyme, and cationic polypeptides against Streptococcus sanguis and Streptococcus

faecalis: inhibition by chitin oligosaccharides. Infect Immun. 48: 720–728.

Lakowicz JR, Weber G. 1973. Quenching of fluorescence by oxygen. Probe for structural

fluctuations in macromolecules. Biochemistry. 12: 4161‐4170.

Lakowicz JR. Principles of Fluorescence Spectroscopy, Springer, New York, 3rd edn., 2006, ch.

8, pp. 278‐292.

Laurent C, Besancon P, Caporiccio B. 2007. Flavonoids from a grape seed extract interact

with digestive secretions and intestinal cells as assessed in an in vitro digestion/Caco‐2 cell

culture model. Food Chem. 100: 1704–1712.

Lee HJ, Jeong YI, Lee TH, Jung ID, Lee JS, Lee CM, Kim JI, Joo H, Lee JD, Park YM. 2007.

Rosmarinic acid inhibits indoleamine 2,3‐dioxygenase expression in murine dendritic cells.

Biochem Pharmacol. 73: 1412‐21.

Lee J, Kim YS, Park D. 2007. Rosmarinic acid induces melanogenesis through protein kinase A

activation signaling. Biochem Pharmacol. 74: 960–968.

Li P, Wang GJ, Li J, Hao HP, Zheng CN. 2006. Identification of tanshinone IIA metabolites in

rat liver microsomes by liquid chromatography‐tandem mass spectrometry. J Chromatogr A.

1104:366‐369.

Li Q‐L, Li B‐G, Zhang Y, Gao X‐P, Li C‐Q, Zhang G‐L. 2008. The angiotensin‐converting enzyme

inhibitors from Rbdosia coetsa. Phytomedicine. 15: 386–388.

References

144

Li Y, He WY, Liu H, Yao X, Hu Z. 2007. Daidzein interaction with human serum albumin

studied using optical spectroscopy and molecular modeling methods, J Mol Struct. 831: 144–

150.

Lin C‐S, Kuo C‐L, Wang J‐P, Cheng J‐S, Huang Z‐W, Chen C‐F. 2007. Growth inhibitory and

apoptosis inducing effect of Perilla frutescens extract on human hepatoma HepG2 cells. J

Ethnopharmacol. 112: 557–567.

Liu G, Ruedi P. 1996. Phyllocladanes (13b‐kauranes) from Plectranthus ambiguus.

Phytochemistry. 41: 1563–1568.

Liu Y, Hu M. 2002. Absorption and metabolism of flavonoids in the caco‐2 cell culture model

and a perused rat intestinal model. Drug Metab Dispos. 30: 370‐377.

Lorenzi H, Matos FJA. 2002. Plantas medicinais no Brasil ‐ nativas e exóticas, 2nd Edition.

Instituto Plantarum de Estudos da Flora, Nova Odessa, Brasil.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin

phenol reagent, J. Biol. Chem. 193: 265–275.

Lukhoba CW, Simmonds MSJ, Paton AJ. 2006. Plectranthus: A review of ethnobotanical uses.

J Ethnopharmacol. 103: 1‐24.

Makino T, Furuta Y, Wakushima H, Fujii H, Saito K, Kano Y. 2003. Anti‐allergic effect of Perilla

frutescens and its active constituents. Phytother Res. 17: 240‐243.

Manach C, Williamson G, Morand C, Scalbert A, Rémésy C. 2005. Bioavailability and

bioefficacy of polyphenols in humans. I. Review of 97 bioavailability studies. Am J Clin Nutr.

81: 230S‐242S.

Marais JPJ, Deavours B, Dixon RA, Ferreira D. 2006. The stereochemistry of flavonoids in The

Science of Flavonoids, ed. Grotewold E, Springer, New York, pp 1‐46.

Mata AT, Proença C, Ferreira AR, Serralheiro MLM, Nogueira JMF, Araújo MEM. 2007.

Antioxidant and antiacetylcholinesterase activities of five plants used as Portuguese food

spices. Food Chem. 103: 778–786.

McDougall GJ, Fyffe S, Dobson P, Stewart D. 2005. Anthocyanins from red wine‐‐their

stability under simulated gastrointestinal digestion. Phytochemistry. 66: 2540‐2548.

McDougall GJ, Fyffe S, Dobson P, Stewart D. 2007. Anthocyanins from red cabbage‐‐stability

to simulated gastrointestinal digestion. Phytochemistry. 68: 1285‐94.

McGleenon BM, Dynan KB, Passmore AP. 1999. Acetylcholinesterase inhibitors in

Alzheimer’s disease. Br. J. Clin. Pharmacol. 48: 471–480.

Mecocci P, Polidori MC. 2011.Antioxidant clinical trials in mild cognitive impairment and

Alzheimer's disease. Biochim Biophys Acta. doi:10.1016/j.bbadis.2011.10.006.

References

145

Meda L, Baron P, Scarlato G. 2001. Glial Activation in Alzheimer's disease: the role of Abeta

and its associated proteins. Neurobilogy of Aging, 22: 885‐893.

Meda L, Cassatella M, Szendrei G, Otvos L, Baron P, Villalba M, Ferrari D, Rossi F. 1995.

Activation of microglial cells by β‐amyloid protein and interferron‐γ. Nature, 374: 647‐650.

Meotti FC, Senthilmohan R, Harwood DT, Missau FC, Pizzolatti MG, Kettle AJ. 2008.

Myricitrin as a substrate and inhibitor of myeloperoxidase: implications for the

pharmacological effects of flavonoids. Free Radic Biol Med. 44: 109‐20.

Militello V, Casarino C, Emanuele A, Giostra A, Pullara F, Leone M. 2004. Aggregation kinetics

of bovine serum albumin studied by FTIR spectroscopy and light scattering. Biophys Chem.

107: 175‐187

Morand C, Crespy V, Manach C, Besson C, Demigné C, Rémésy C. 1998. Plasma metabolites

of quercetin and their antioxidant properties. Am J Physiol Regulatory Integrative Comp

Physiol. 275: 212‐219.

Mosmann, T. 1983. Rapid colorimetric assay for cellular growth and survival: Application to

proliferation and cytotoxicity assays. J. Immunol. Methods. 65: 55‐63.

Mukjerjee P K, Kumar V, Mal M, Houghton PJ. 2007. Acetylcholinesterase inhibitors from

plants. Phytomedicine. 14: 289–300.

Nabekura T, Yamaki T, Hiroi T, Ueno K, Kitagawa S. 2010. Inhibition of anticancer drug efflux

transporter P‐glycoprotein by rosemary phytochemicals. Pharmacol. Res. 61: 259‐263.

Nakazawa T, Ohsawa K. 1998. Metabolism of rosmarinic acid in rats. J. Nat. Prod. 61: 993–

996.

Ng SP, Wong KY, Zhang L, Zuo Z. 2005. Evaluation of the first‐pass glucoronidation of

selected flavones in gut by Caco‐2 monolayer model. J Pharm Pharm Sci. 8: 1‐9.

Ohnishi S, Takano K. 2004. Amyloid fibrils from the viewpoint of protein folding. Cell Mol Life

Sci. 61: 511‐24.

O'Leary KA, Day AJ, Needs PW, Mellon FA, O'Brien NM, Williamson G. 2003. Metabolism of

quercetin‐7‐ and quercetin‐3‐glucuronides by an in vitro hepatic model: the role of human β‐

glucuronidase, sulfotransferase, catechol‐O‐methyltransferase and multi‐resistant protein 2

(MRP2) in flavonoid metabolism. Biochem. Pharmacol. 65: 479–491.

Orhan I, Aslant S, Kartal M, Sener B, Baser KHC. 2008. Inhibitory effect of Turkish Rosmarinus

officinalis L. on acetylcholinesterase and butyrylcholinesterase enzymes. Food Chem. 108:

663–668.

Orhan I, Sener B, Choudhary M I, Khalid A. 2004. Acetylcholinesterase and

butyrylcholinesterase inhibitory activity of some Turkish medicinal plants. J Ethnopharmacol.

91: 57–60.

References

146

Osakabe N, Yasuda A, Natsume M, Yoshikawa T. 2004. Rosmarinic acid inhibits epidermal

inflammatory responses: anticarcinogenic effect of Perilla frutescens extract in the murine

two‐stage skin model. Carcinogenesis. 25: 549‐557.

Pádua BC, Silva LD, Rossoni Júnior JV, Humberto JL, Chaves MM, Silva ME, Pedrosa ML, Costa

DC. 2010. Antioxidant properties of Baccharis trimera in the neutrophils of Fisher rats. J

Ethnopharmacol. 129: 381‐386.

Parejo I, Caprai E, Bastida J, Viladomat F, Jauregui O, Codina C. 2004. Investigation of

Lepechinia graveolens for its antioxidant activity and phenolic composition. J

Ethnopharmacol. 94: 175–184.

Pederson JA. 2000. Distribution and taxonomic implications of some phenolics in the family

Lamiaceae determined by ESR spectroscopy. Biochem. Syst Ecol. 28: 229–253.

Percival SS, Turner RE. 2007. Applications of Herbs to Functional Foods. In: Handbook of

Nutraceuticals and Functional Foods, 2nd Ed. (Edited by R. E. C. Wildman), CRC Press, pp 269–

284.

Pereira P, Tysca D, Oliveira P, Brum LFS, Picada JN, Ardenghi P. 2005. Neurobehavioral and

genotoxic aspects of rosmarinic acid, Pharmacol. Res. 52: 199–203.

Petersen M, Simmonds MSJ. 2003. Rosmarinic acid. Phytochemistry. 62: 121–125.

Ponec RJ, Saunders MD, Kimmey MB. 1999. Neostigmine for the treatment of acute colonic

pseudo‐obstruction. N Engl J Med. 341: 137‐141.

Porfírio S, Falé PL, Madeira PJA, Florêncio MH, Ascensão L, Serralheiro MLM. 2010.


gastrointestinal metabolism. Food Chem. 122: 179‐87.

Probst RJ, Lim JM, Bird DN, Pole GL, Sato AK, Claybaugh JR. 2006. Gender differences in the

blood volume of conscious Sprague‐Dawley rats. J. Am. Assoc. Lab. Anim. Sci. 45: 49–52.

Randall D, Burggren W, French K. 2000. Eckert Animal Physiology, Fourth Edition. W.H.

Freeman and Company. New York, USA.

Rawel HW, Frey SK, Meidtner K, Kroll J, Schweigert FJ. 2006. Determining the binding

affinities of phenolic compounds to proteins by quenching of the intrinsic tryptophan

fluorescence. Mol Nutr Food Res. 50:, 705–713.

Regasini LO, Vellosa JC, Silva DH, Furlan M, de Oliveira OM, Khalil NM, Brunetti IL, Young MC,

Barreiro EJ, Bolzani VS. 2008. Flavonols from Pterogyne nitens and their evaluation as

myeloperoxidase inhibitors. Phytochemistry. 69: 1739‐44.

Rocchetti M, Besana A, Mostacciuolo G, Ferrari P, Micheletti R, Zaza A. 2003. Diverse toxicity

associated with cardiac Na+/K+ pump inhibition: evaluation of electrophysiological

mechanisms. J. Pharmacol. Exp. Ther. 305: 765‐771.

References

147

Rodney G, Swanson AL, Wheeler LM, Smith GN, Worrel CS. 1950. The effect of a series of

flavonoids on hyaluronidase and some other related enzymes. J Biol Chem. 183: 739–747.

Ronca F, Palmieri L, Panicucci P, Ronca G. 1998. Anti‐inflammatory activity of chondroitin

sulphate. Osteoarthritis Cartilage. 6: 14‐21.

Ross PD, Subramanian S. 1981. Thermodynamics of protein association reactions: forces

contributing to stability, Biochemistry. 20: 3096‐3102.

Salawu FK, Umar JT, Olokoba AB. 2011. Alzheimer's disease: A review of recent

developments. Ann Afr Med. 10: 73‐79.

Sanson B, Colletier JP, Xu Y, Lang PT, Jiang H, Silman I, Sussman JL, Weik M. 2011. Backdoor

opening mechanism in acetylcholinesterase based on X‐ray crystallography and molecular

dynamics simulations. Protein Sci. 20: 1114‐1118.

Sasho S, Obase H, Ichikawa S, Kitazawa T, Nonaka H, Yoshizaki R, Ishii A, Shuto K. 1995.

Synthesis of 2‐imidazolidinylidene Propanedinitrile Derivatives as Stimulators of

Gastrointestinal Motility – III. Bioorg Med Chem. 3: 279‐287.

Scalbert A, Williamson G. 2000. Dietary intake and bioavailability of polyphenols. J Nutr. 130:

2073S–2085S.

Schultz C, Bossolani MP, Torres LM, Lima‐Landman MT, Lapa AJ, Souccar C. 2007. Inhibition

of the gastric H+,K+ ‐ATPase by plectrinone A, a diterpenoid isolated from Plectranthus

barbatus Andrews. J Ethnopharmacol. 111: 1‐7.

Schutte ME, Boersma MG, Verhallen DA, Groten JP, Rietjens IM. 2008. Effects of flavonoid

mixtures on the transport of 2‐amino‐1‐methyl‐6‐phenylimidazo[4,5‐b]pyridine (PhIP)

through Caco‐2 monolayers: an in vitro and kinetic modeling approach to predict the

combined effects on transporter inhibition. Food Chem Toxicol. 46: 557‐566.

Sheng R, Lin X, Zhang J, Chol KS, Huang W, Yang B, He Q, Hu Y. 2009. Design, synthesis and

evaluation of flavonoid derivatives as potent AChE inhibitors. Bioorg Med Chem. 17: 6692–

6698.

Shi R, Qiao S, Yu D, Shi X, Liu M, Jiang X, Wang Q, Zhang L. 2011. Simultaneous determination

of five flavonoids from Scutellaria barbata extract in rat plasma by LC‐MS/MS and its

application to pharmacokinetic study. J Chromatogr B Analyt Technol Biomed Life Sci. 879:

1625‐1632.

Shimoi K, Okada H, Furugori M, Goda T, Takase S, Suzuki M, Hara Y, Yamamoto H, Kinae N.

1998. Intestinal absorption of luteolin and luteolin 7‐O‐beta‐glucoside in rats and humans,

FEBS Lett. 438: 220–224.

Silman I, Sussman JL. 2008. Acetylcholinesterase: How is structure related to function? Chem

Biol Interact. 175: 3‐10.

References

148

Sinko G, Calić M, Bosak A, Kovarik Z. 2007. Limitations of the Ellman method: cholinesterase

activity measurement in the presence of oximes. Anal Biochem. 370: 223‐227.

Spencer JP, Chaudry F, Pannala AS, Srai SK, Debnam E, Rice‐Evans C. 2000. Decomposition of

cocoa procyanindins in the gastric milieu. Biochem Biophys Res Commun. 27: 236–241.

Spencer JPE. 2003. Metabolism of tea flavonoids in the gastrointestinal tract. J Nutr. 133:

3255S–3261S.

Stuchbury G, Münch G. 2005. Alzheimer’s associated inflammation, potential drug targets

and future therapies. J Neural Transm. 112: 429‐453.

Tabart J, Franck T, Kevers C, Pincemail J, Serteyn D, Defraigne J‐O, Dommes J. 2011.

Antioxidant and anti‐inflammatory activities of Ribes nigrum extracts. Food Chem. In Press.

doi:10.1016/j.foodchem.2011.09.076

Tabet N. 2006. Acetylcholinesterase inhibitors for Alzheimer’s disease: anti‐inflammatories

in acetylcholine clothing! Age Ageing. 35: 336–338.

Tantipolphan R, Rades T, McQuillan AJ, Medlicott NJ. 2007. Adsorption of bovine serum

albumin (BSA) onto lecithin studied by attenuated total reflectance Fourier transform

infrared (ATR‐FTIR) spectroscopy, Int. J. Pharm. 337: 40–47.

Temple MD, Perrone GG, Dawes IW. 2005. Complex cellular responses to reactive oxygen

species. Trends in Cell Biology, 15: 319‐326

Tepe B, Daferera D, Sokmen A, Sokmen M, Polissiou M. 2005. Antimicrobial and antioxidant

activities of essential oil and various extracts of Salvia tomentosa Miller (Lamiaceae). Food

Chem. 90: 333–340.

Ticli FK, Hage LI, Cambraia RS, Pereira PS, Magro AJ, Fontes MR, Stábeli RG, Giglio JR, França

SC, Soares AM, Sampaio SV. 2005. Rosmarinic acid, a new snake venom phospholipase A2

inhibitor from Cordia verbenacea (Boraginaceae): Antiserum action potentiation and

molecular interaction. Toxicon. 46: 318–327.

Trouillas P, Calliste C‐A, Allais D‐P, Simon A, Marfak A, Delage C, Duroux J‐L. 2003.

Antioxidant, anti‐inflammatory and antiproliferative properties of sixteen water plant

extracts used in Limousin countryside as herbal teas. Food Chem. 80: 399–407.

Tsuji A, Takanaga H, Tamai I, Terasaki T. 1994. Transcellular transport of benzoic acid across

Caco‐2 cells by a pH‐dependent and carrier‐mediated transport mechanism. Pharm Res. 11:

30‐37.

Vaidyanathan JB, Walle T. 2003. Cellular uptake and efflux of the tea flavonoid

(‐)epicatechin‐3‐gallate in the human intestinal cell line Caco‐2. J Pharmacol Exp Ther. 307:

745‐52.

Varshney A, Sen P, Ahmad E, Rehan M, Subbarao N, Khan RH. 2010. Ligand binding strategies

of human serum albumin: How can the cargo be utilized? Chirality 22: 77–87.

References

149

Viña J, Lloret A, Ortí R, Alonso D. 2004. Molecular bases of the treatment of Alzheimer’s

disease with antioxidants: prevention of oxidative stress. Mol Aspects Med. 25:117‐123.

Vinutha B, Prashanth D, Salma K, Sreeja SL, Pratiti D, Padmaja R, Radhika S, Amit A,

Venkateshwarlu K, Deepak M. 2006. Screening of selected Indian medicinal plants for

acetylcholinesterase inhibitory activity. J Ethnopharmacol. 109: 359‐363.

Walgren RA, Walle UK, Walle T. 1998. Transport of Quercetin and Its Glucosides across

Human Intestinal Epithelial Caco‐2 Cells. Biochem Pharmacol. 55: 1721‐1727.

Walle UK, Galijatovic A, Walle T. 1999. Transport of the flavonoid chrysin and its conjugated

metabolites by the human intestinal cell line Caco‐2. Biochem Pharmacol. 58: 431‐438.

Wang Q, Morris ME. 2007. Flavonoids modulate monocarboxylate transporter‐1‐mediated

transport of gamma‐hydroxybutyrate in vitro and in vivo. Drug Metab Dispos. 35: 201–208.

Wang Y, Cao J, Zeng SJ. 2005. Involvement of P‐glycoprotein in regulating cellular levels of

Ginkgo flavonols: quercetin, kaempferol, and isorhamnetin. J Pharm Pharmacol. 57: 751–

758.

Weiss S, Klein R, Slivka A, Wei M. 1982. Chlorination of Taurine by Human Neutrophils. J Clin

Invest. 70: 598‐607.

Wong KK, Ngo JC, Liu S, Lin HQ, Hu C, Shaw PC, Wan DC. 2010. Interaction study of two

diterpenes, cryptotanshinone and dihydrotanshinone, to human acetylcholinesterase and

butyrylcholinesterase by molecular docking and kinetic analysis. Chem Biol Interact. 187:

335‐339.

Wu Y, Zhou C, Song L, Li X, Shi S, Mo J, Chen H, Bai H, Wu X, Zhao J, Zhang R, Hao X, Sun H,

Zhao Y. 2006. Effect of total phenolics from Laggera alata on acute and chronic

inflammation models, J Ethnopharmacol 108: 243–250.

Xiao J, Cao H, Chen T, Yang F, Liu C, Xu X. 2011. Molecular property‐binding affinity

relationship of flavonoids for common rat plasma proteins in vitro. Biochimie 93: 134–140.

Xiao J, Suzuki M, Jiang X, Chen X, Yamamoto K, Ren F, Xu M. 2008. Influence of B‐ring

hydroxylation on interactions of flavonols with bovine serum albumin, J. Agric. Food Chem.

56: 2350–2356.

Yamamoto Y, Takahashi Y, Kawano M, Iizuka M, Matsumoto T, Saeki S, Yamaguchi H. 1999.

In vitro digestibility and fermentability of levan and its hypocholesterolemic effects in rats. J

Nutr Biochem. 10: 13–18.

Yang Y, Hua Q, Fan Y, Shen H. 2008. Study on the binding of luteolin to bovine serum

albumin. Spectrochim Acta A Mol Biomol Spectrosc. 69: 432–436.

Yoshino K, Suzuki M, Saski K, Miyase T, Sano M. 1999. Formation of antioxidants from (‐)‐

epigallocatechin gallate in mild alkaline fluids, such as authentic intestinal juice and mouse

plasma. J Nutr Biochem. 10: 223–229.

References

150

Yu JQ, Liao ZX, Cai XQ, Lei JC, Zou GL. 2007. Composition, antimicrobial activity and

cytotoxicity of essential oils from Aristolochia mollissima. Environ Toxicol Pharmacol. 23:

162‐167.

Yuan J, lv Z, Liu Z, Hu Z, Zou G. 2007. Study on interaction between apigenin and human

serum albumin by spectroscopy and molecular modelling, J Photochem Photobiol A Chem.

191: 104–113.

Zeraik ML, Serteyn D, Deby‐Dupont G, Wauters J‐N, Tits M, Yariwake JH, Angenot L, Franck T.

2011.Evaluation of the antioxidant activity of passion fruit (Passiflora edulis and Passiflora

alata) extracts on stimulated neutrophils and myeloperoxidase activity assays. Food Chem.

128: 259–265

Zheng GZ, Bhatia P, Kolasa T, Patel M, El Kouhen OF, Chang R, Uchic ME, Miller L, Baker S,

Lehto SG, Honore P, Wetter JM, Marsh KC, Moreland RB, Brioni JD, Stewart AO. 2006.

Correlation between brain/plasma ratios and efficacy in neuropathic pain models of

selective metabotropic glutamate receptor 1 antagonists, Bioorg. Med. Chem. Lett. 16:

4936–4940.

universidade de lisboarepositorio.ul.pt/bitstream/10451/6211/1/ulsd062640_td_pedro_fale.pdf ·...

Documents