Phytoestrogens as natural prodrugs in cancerprevention: a novel concept

Randolph R. J. Arroo Æ Vasilis Androutsopoulos ÆAsma Patel Æ Somchaiya Surichan Æ Nicola Wilsher ÆGerry A. Potter

Received: 11 October 2007 / Accepted: 28 February 2008 / Published online: 26 March 2008

� Springer Science+Business Media B.V. 2008

Abstract It has been generally accepted that regular

consumption of fresh fruits and vegetables is linked

with a relatively low incidence of cancers (e.g. breast,

cervix, and colon). A number of plant-derived com-

pounds have been identified that are considered to

play a role in cancer prevention. However, at present

there is no satisfactory explanation for the cancer

preventative properties of the above-mentioned com-

pound groups. The current review is an effort to

develop a consistent and unambiguous model that

explains how some plant-derived compounds can

prevent tumour development. The model is based on

recent discoveries in the fields of genomics and drug-

metabolism; notably, the discovery that CYP1 genes

are highly expressed in developing tumour cells but not

in the surrounding tissue, and that a variety of plant-

derived compounds are substrates for the CYP1

enzymes. Our hypothesis is that some dietary com-

pounds act as prodrugs, i.e. compounds with little or no

biological effect as such, but become pharmaceutically

effective when activated. More specifically, we

state that the abovementioned prodrugs are only

activated in CYP1-expressing cells—i.e. cells in the

early stages of tumour development—to be converted

into compounds which inhibit cell growth. Thus, the

prodrugs selectively kill precancerous cells early in

tumour development. The review focuses on the

identification of naturally-occurring prodrugs that are

activated by the tumour-specific CYP1 enzymes and

aims to assess their role in cancer prevention.

Keywords CYP1 � Drug metabolism �Flavones � Natural products � Prevention

Anticancer drug discovery

In the 1950s, the National Cancer Institute (NCI)

started a large-scale empirical anticancer drug screen-

ing program that continued in more or less the same

form until the mid-1970s. In this program, compounds

were tested against a panel of mouse tumour models,

mainly L1210 and P388 leukaemias (Suggit and

Bibby 2005). This screening method led to the

identification of a large number of compound classes

that have formed the basis for most of the anticancer

drugs that are currently in clinical use (Table 1). The

screening program was highly compound-oriented and

resulted in a golden era for phytochemistry, since

natural products formed a major source of lead

compounds. Indeed, a large proportion of the now

conventional anticancer drugs are derived from nat-

ural sources, e.g. Vincristine and Vinblastine from

Catharanthus roseus, Paclitaxel and Docetaxel from

Taxus species, Doxorubicin (Adriamycin) and

R. R. J. Arroo (&) � V. Androutsopoulos �A. Patel � S. Surichan � N. Wilsher � G. A. Potter

Leicester School of Pharmacy, De Montfort University,

The Gateway, Leicester LE1 9BH, UK

e-mail: rrjarroo@dmu.ac.uk

123

Phytochem Rev (2008) 7:431–443

DOI 10.1007/s11101-008-9093-5

Mitomycin from Streptomycin species, the podophyl-

lotoxin derivatives Etoposide and Teniposide from

Podophyllum species, Camptothecin and its semisyn-

thetic derivative Irinotecan from Camptotheca

acuminata. A recent estimate stated that approximately

60% of drugs in clinical trials for the multiplicity of

cancers are either natural products, compounds derived

from natural products, or drugs containing pharmaco-

phores derived from natural products (Cragg and

Newman 2000).

The use of the mouse tumour models has led to

concerns that the screening may have resulted in

preferential selection of drugs that were only active

against rapidly growing tumours. Indeed, most exist-

ing anticancer agents target DNA replication of

dividing cells, thus inhibiting tumour cell growth

but also normal cell growth. This causes many

undesired side effects such as nausea and vomiting,

alopecia, mucositis, myelosuppression and reproduc-

tive sterility. These side effects of chemotherapy can

have a devastating impact on a patient’s quality of

life. Particularly, the treatment of elderly patients

(older than 70 years) presents characteristics that

make the choice of the correct treatment more

difficult; for this reason, these patients are often

undertreated (Pasetto et al. 2007).

Smart anticancer drugs

The main aim of conventional cancer treatment was

systemic, non-specific, high-dose chemotherapy

(Abou-Jawde et al. 2003). The success of these

treatments was outweighed by the systemic toxicity.

However, as more was discovered about the mech-

anisms underlying the initiation and progression of

human cancers, anticancer drug discovery moved

away from the development of classic cytotoxic

agents to the rational design of small molecule

anticancer therapeutics. This development has

prompted a transition from empirical compound-

orientated preclinical screening to target-orientated

drug screening (Suggit and Bibby 2005). The so-

called smart drugs, which have been developed as a

result of the new screening methods, are small

molecules targeted at specific growth signalling

Table 1 Overview of traditional cytotoxic agents, their mode of action and side effects

Class of drug Examples Mode of action Side effects

Alkylating agents Cyclophosphamide

Chorambucil

Melphalan

Mitomycin

Formation of covalent bonds with DNA,

thus impeding DNA replication

Myelosuppression

GI disturbances, hair loss,

Reduced fertility, increased

risk of secondary malignancies

Antimitotic agents Vincristine

Vinblastine

Docetaxel

Paclitaxel

Interference with microtubule assembly

of cells, thereby blocking cell division.

As for alkylating agents;

Neurotoxicity

Antimetabolites Methotrexate

5-Fluoroacil

6-Mercaptopurine

Inhibition of metabolic pathways

involved in DNA synthesis.

As for alkylating agents:

Nephrotoxicity

DNA intercalators Doxorubicin

Mitoxantrone

Prevent DNA replication As for alkylating agents;

Doxorubicin can cause cardiotoxicity

Topoisomerase inhibitors Etoposide

Teniposide

Camptothecin

Irinotecan

Prevent unwinding of DNA strands,

thereby blocking DNA replication.

As for alkylating agents

Heavy metal derivatives Cisplatin

Carboplatin

As metallating agents. As for alkylating agents;

Nephrotoxicity;

Neurotoxicity and potention loss

of hearing

432 Phytochem Rev (2008) 7:431–443

123

pathways. These new drugs are expected to dominate

clinical trials in the years to come either as single-

drug modality or as combination treatment (van der

Poel 2004). The new generation of chemotherapeu-

tics act on specific molecular targets that are

responsible for malignant phenotypes, and thus

provide potentially tumour specific therapy (Table 2).

Most of the compounds are synthetic molecules that

are made to fit particular well-defined targets; very

few of this second generation of anticancer drugs are

based on natural products. Theoretically, the use of

these putatively tumour specific chemotherapeutic

agents, whether alone or in combination, should lead

to an improved clinical response. However, the

expression of the target proteins is not limited to

tumour cells alone; the proteins are over-expressed in

tumour cells, but also expressed at a lower frequency

in normal cells. Thus, problems remained with

unacceptable damage to normal cells and organs, a

narrow therapeutic index, or a relatively poor selec-

tivity for neoplastic cells. A potential strategy to

overcome the limitations of chemotherapeutic agents

is the use of prodrugs, i.e. compounds that need to be

transformed before exhibiting their pharmacological

action (Rooseboom et al. 2004). Prodrugs are often

divided into two groups:

1. prodrugs designed to increase the bioavailability

of antitumour agents, e.g. when the activated

drug has a low chemical stability or is rapidly

broken down in vivo, more stable prodrugs can

be administered.

2. prodrugs designed to locally deliver antitumour

agents (targeting)

CYP1-activated anticancer prodrugs

A recent discovery in the field of oncology was that

the cytochrome P450 enzyme CYP1B1 is expressed

at a high frequency in a wide range of human cancers,

including tumours of the breast, colon, lung, oesoph-

agus, skin, lymph node, brain, and testis (Table 3, 4).

The protein could not be detected in the surrounding

normal tissues (Murray et al. 1997, 2001). Prelimin-

ary data by Stanley et al. (2001a, b) have indicated

that the CYP1B1 protein is also over expressed in

premalignant tissue, which suggests that CYP1B1 has

a role in early tumour development. However, other

authors (Gibson et al. 2003) call for more caution

since they could detect CYP1B1 protein in patient

samples of human colonic tissue specimens, which

were excised C30 cm from the tumour, and did not

show any of the characteristics of premalignant

tissue. A related cytochrome P450, CYP1A1, is

abundant in tumour tissue, but various xenobiotics

(notably aryl hydrocarbons) can induce its expression

in healthy tissue, notably in the epithelial lining of the

lungs (Zhang et al. 2006).

CYP1 enzymes are known to play a role in the

metabolism of estrogens. Notably, CYP1B1 converts

the oestrogen estradiol into 4-hydroxyestradiol. The

Table 2 Overview of targeted cancer treatments, their mode of action and side effects

Class of drug Examples Mode of action Side effects

Hormonal therapies Tamoxifen

Letrozole

Arimidex

Bicalutamide

Prevent biosynthesis of oestrogen,

or block the action of hormones

tumour development.

Endometrial cancer, blood clots, hot

flushes, fatigue, GI disturbances

Signalling inhibitors Imatinib

Gefitinib

Rituximab

Block activation of secondary

messenger molecules; inhibit tyrosine

kinase activity

Fluid retention, skin rashes,

diarrhoea, GI disturbances

Monoclonal antibodies Trastuzumab

Gemtuzumab

Ozogamicin

Antibodies bind to proteins that are

overexpressed in tumours, e.g. CD20 antigen,

or HER2 receptor.

Allergic response

Differentiating agents Tretinoin Induces differentiation and death

of leukaemia cells

Fever, severe respiratory

distress, weight gain

Phytochem Rev (2008) 7:431–443 433

123

latter compound plays a key role in the development

of breast and endometrial tumours. Of particular

interest was the finding that human CYP1B1 activity

is regulated by estradiol via the oestrogen receptor

(For a recent review see: Tsuchiya et al. 2005).

Taken together the findings made CYP1B1 an ideal

molecular target for the development of new anti-

cancer drugs, where the estradiol core could be taken

as a lead structure for drug design. Potter et al.

designed DMU-212 (3,4,5,40-tetramethoxy-stilbene)

which showed promise in in vitro human tumour

assays, and in mouse models (Sale et al. 2004). The

authors noted the close structural similarity of DMU-

212 with resveratrol, a naturally occurring stilbene

found in red wine, which had been shown to have

cancer preventative properties (Jang et al. 1997; Jang

and Pezzuto 1999). Resveratrol is classified as a

phytoestrogen because of its similarity to the endog-

enous oestrogen estradiol. It appeared that resveratrol

is a substrate for the CYP1 family of enzymes

(Fig. 1), and that the enzymes catalyse the conversion

of resveratrol into piceatannol and a second

compound, putatively identified as 3,4,5,40-tetrahydr-

oxystilbene (Potter et al. 2002; Piver et al. 2004).

Both CYP-1 conversion products of resveratrol are

known to inhibit cell proliferation (Ferrigni et al.

1984; Lu et al. 2001).

The role of diet in prevention of cancer

The notion that a diet rich in fresh fruit and

vegetables protects from the risk of most common

epithelial cancers—including those of the digestive

tract, and also several nondigestive neoplasms—is

well-established (e.g. La Vecchia et al. 2001).

When dietary habits of large cohorts were com-

pared and correlated with the occurrence of cancer,

a number of molecules were identified as the most

Table 3 Expression of

CYP1B1 in different types

of tumours and normal

tissue (after Murray et al.

1997)

Tissue Normal (no positive/no tested) Tumour (no positive/no. tested)

Bladder 0/8 8/8

Brain 0/12 11/12

Breast 0/10 12/12

Colon 0/10 11/12

Connective tissue 0/9 8/9

Oesophagus 0/8 8/8

Kidney 0/11 11/11

Liver 0/8 not tested

Lung 0/8 7/8

Lymph node 0/5 9/9

Ovary 0/5 7/7

Skin 0/6 6/6

Small intestine 0/5 not tested

Stomach 0/10 9/10

Testis 0/8 8/8

Uterus 0/7 7/7

Total 0/130 122/127

Table 4 Substitution patterns of selected dietary flavonols

O

O

R1

OH

R2HO

OH

OH

3'

4'

5'

A

B

C

Flavonols R1 R2

Kaempferol H H

Quercetin OH H

Myricetin OH OH

Isorhamnetin OMe H

434 Phytochem Rev (2008) 7:431–443

123

likely chemopreventive agents. At present, there is a

general consensus that plant polyphenols, flavonoids,

catechins, and lignans are the key constituents in

cancer prevention (Kinghorn et al. 2003, 2004). The

latter notion is wide-spread, also in the popular press,

and a number of preparations containing the com-

pounds mentioned above are available commercially

as health-enhancing supplements. Advertisements for

the commercially available preparations often refer to

scientific papers to underpin their health claims.

Roughly, two models that may explain the cancer

preventive properties of are cited time and time

again: the antioxidant model and the phytoestrogen

model. However, neither model can fully explain why

certain naturally occurring dietary compounds can

prevent the occurrence of cancer.

Fig. 1 CYP1 catalysed

conversions of estradiol and

resveratrol

Phytochem Rev (2008) 7:431–443 435

123

Antioxidants and cancer prevention

The antioxidant model was proposed in an influential

review by Ames (1983). To date, the paper has been

cited over 2,000 times. In short, reactive oxygen

species, formed as a by-product of oxidative catab-

olism—e.g. singlet oxygen and various radicals—are

possibly involved in DNA damage and tumour

promotion. Antioxidants would prevent the accumu-

lation of reactive oxygen species, and consequently

any cell damage caused by these radicals. Ames lists

vitamin E (tocopherol), b-carotene, and vitamin C

(ascorbic acid) as dietary natural products with

antioxidant activity, but he is reticent about the role

of plant phenols. Nevertheless, plant phenols with

antioxidant activity, e.g. stilbenes, chalcones, and

flavonoids, have been a major focus of attention.

Plant polyphenols have been shown to be highly

effective scavengers of most types of oxidizing

molecules, and the antioxidant properties of these

molecules have often been cited as an explanation for

their cancer preventive properties. Over 50 reviews

have been written on this topic over the past five

years (for a recent example, see Valko et al. 2006).

However, there is a paradox here; supplements of

ascorbate, vitamin E, or b-carotene—known antiox-

idants—do not decrease DNA damage in most studies

(Halliwell 2000, 2007; Halliwell et al. 2005). In

addition, in animal models antimutagens and anti-

carcinogens have been given in unrealistically high

doses, which do not usually reflect the human

exposure situation (Knasmuller et al. 2002).

Intriguingly, Ames’ review also highlights the role

of dietary carcinogens in the onset of cancer. In

particular benzo[a]pyrene, a compound that occurs in

charred meat and in cigarette smoke, has been well-

investigated since. It is now widely accepted that

benzo[a]pyrene is actually a pro-carcinogen which is

activated by the human cytochrome P450 enzyme

CYP1A1. It is the activated compound, benzo[a]pyr-

ene diol epoxide, that causes cell damage (see review

by Alexandrov et al. 2002).

Phytoestrogens and cancer prevention

Oestrogens are hormones responsible for the female

sex characteristics; they are not restricted to females

since small amounts are produced in the male testes.

The principal and most potent example is estradiol.

The aromatic ring (Fig. 2) makes the oestrogen

molecule almost planar and is essential for activity.

Estradiol has been linked to breast cancer etiology,

but the molecular mechanisms underlying the devel-

opment of breast cancer are not completely

understood (Dietel et al. 2005; Russo and Russo

2004).

A variety of plant compounds have been shown to

interact with human oestrogen receptors. Unsurpris-

ingly maybe, the molecules share some structural

Fig. 2 Structural

similarities between the

human hormone estradiol

(1), a steroid, and

phytoestrogens belonging to

the chemical groups of

flavones (2), flavonols (3),

and stilbenes (4). In an

attempt to highlight the

similarities, some structures

are not presented in the

conventional way

436 Phytochem Rev (2008) 7:431–443

123

similarity with estradiol, e.g. the almost planar

configuration and the aromatic rings. Several of these

so-called phytoestrogens have been linked with

cancer prevention, and have been selected as candi-

date drugs in various studies (Usui 2006). Although

to date no clear model exists that explains how

phytoestrogens affect the development of hormone-

dependent tumours, epidemiological data suggest that

the antitumour activity of phytoestrogens has little to

do with their activity as endocrine disruptors. For

example, genistein can stimulate estrogen receptor-

positive (ER+) breast cancer growth and interfere

with the antitumour activity of the endocrine disrup-

tor tamoxifen at low levels. Thus, women who have

ER+ tumours are advised not to increase their

phytoestrogen intake. Several studies suggest an

inhibitory effect on ER- breast cancer cell growth,

and it may be reasonable for women with

ER- tumours to safely consume soy and possibly

other phytoestrogens (Duffy and Cyr 2003).

An alternative explanation for the anticancer

properties of phytoestrogens may be that they prevent

the activation of pro-carcinogens like benzo[a]pyrone

(Tsuji and Walle 2006) since they inhibit the activity

of the enzymes CYP1A1 and CYP1B1. However, at

the same time it has been observed that phytoestro-

gens induce CYP1A1 and CYP1B1 gene expression

(Moon et al. 2006). These two observations seem to

be contradictory.

Dietary prodrugs

In the classical search for anticancer drugs from

botanical sources, plant extracts were fractionated by

various means and the fractions tested for anticancer

properties in a variety of in vitro assays. However,

the compounds that are present in the plant are not

necessarily the compounds that have a pharmacolog-

ical effect in the human body. This has often been

cited as one of the weaknesses of in vitro testing. A

few examples may demonstrate that several plant-

derived compounds act as prodrugs rather than drugs:

1. Contrary to popular belief, the bark of the willow

(Salix alba) does not contain aspirin (acetyl

salicylic acid). Nevertheless, like aspirin, extract

of willow bark has been effectively used to treat

pains and fevers. We now know that the gluco-

side salicin from the bark extract is converted in

the digestive tract and liver (by CYP enzymes) to

form the active ingredient salicylic acid (Fig. 3).

2. Epidemiological investigations showed a corre-

lation between reduced risk of chronic diseases

and increased concentration of the so-called

‘human’ lignans enterodiol and enterolactone

enterolactone in blood plasma (Fig. 4). The

‘human’ lignans play a role in the prevention of

colorectal cancer (Kuijsten et al. 2006). How-

ever, the anticancer mechanism is still unknown.

In fact, the ‘human’ lignans are not formed by

human metabolism at all, but by bacterial

fermentation of plant-derived lignans in the

colon (Rickard et al. 1996).

3. Another example of pharmaceutically active

compounds that arise as the result of fermentation

by gut bacteria are equal and O-demethylango-

lensin (Fig. 5) which are derivatives of the dietary

isoflavones formononetin and daidzein (Atkinson

et al. 2005).

Based on data obtained from literature and observa-

tions in our own lab, we propose that a variety of

naturally occurring compounds (notably phytoestro-

gens belonging to the compound class of the

flavonoids) act as prodrugs with relatively little

biological activity. These compounds are substrates

Fig. 3 Bioconversion of

salicin

Phytochem Rev (2008) 7:431–443 437

123

for enzymes that are highly expressed in tumour cells

(i.e. CYP1A1 and CYP1B1), and the conversion

products inhibit cell growth (i.e. selectively inhibit

growth of tumour cells).

This mechanistic model that may explain the

epidemiological finding that regular consumption of

fruit and vegetables can prevent the development of

tumours. The model also explains why flavonoid-type

antioxidants, but not other antioxidants like ascorbic

acid or tocopherol, seem to play a role in cancer

prevention. In addition, the model explains the

prominence of phytoestrogens in the lists of plant-

derived compounds with possible anticancer

properties.

Flavonoids are substrates for CYP1 enzymes

In an attempt to identify whether or not dietary

flavonoids interact with CYP1 enzymes, the effect of

the flavonoids on CYP1-catalysed 7-ethoxyresorufin

dealkylation (Burke and Mayer 1974) was monitored.

Most of the flavonoids tested inhibited the dealkyla-

tion in a dose dependent manner (Table 5). All the

Fig. 4 Bioactivation of

dietary lignan

secoisolariciresinol

Fig. 5 Bioactivation of

some isoflavones

438 Phytochem Rev (2008) 7:431–443

123

flavonoids tested, except cirsiliol and genkwanin,

inhibited CYP1B1 activity more effectively than

CYP1A1 and CYP1A2. In particular the flavonols

quercetin, myricetin and kaempferol were potent

CYP1B1 inhibitors, whereas they seemed to be less

effective inhibitors for CYP1A1 and CYP1A2.

The flavonoids which were shown to be potent

inhibitors in this study, might well be potential

substrates for CYP1 enzymes, e.g. resveratrol has

been reported to be a substrate (Potter et al. 2002;

Piver et al. 2004) and an inhibitor (Chun et al. 1999).

The 7-ethoxyresorufin-O-deethylase activity assay

(EROD assay) does not differentiate between the

different types of inhibition, but a molecule which is

a strong competitive inhibitor, would be a likely

substrate, since it will fit in the active site of the

enzymes.

Enzyme assays, where selected flavonoids were

incubated with CYP1 enzymes, confirmed that the

CYP1 enzymes accept a range of flavones and

flavonols as substrates. The reaction products were

analysed by HPLC-DAD, or HPLC-MS. In a number

of cases, the reaction products could be identified by

comparison with authentic reference compounds. We

can confirm that apigenin is hydroxylated to form

luteolin, and that kaempferol is hydroxylated to form

quercetin (Fig. 6, see also Breinholt et al. 2002;

Gradolatto et al. 2004). With other flavonoids, in

most cases reaction products could be observed.

However, due to the absence of authentic references

and the low amounts of product, a conclusive

chemical identification has not been possible. Thus

we can only assign putative structures to these

reaction products. Control incubations, either without

Table 5 Substitution patterns of selected dietary flavones

O

O

R4

R5

R1

R3

B

CA

3'

4'

56

7

R2

Flavones R1 R2 R3 R4 R5

Chrysin OH H OH H H

Baicalein OH OH OH H H

Apigenin OH H OH H OH

Luteolin OH H OH OH OH

Scutellarein OH OH OH H OH

Diosmetin OH H OH OH OMe

Eupatorin OH OMe OMe OH OMe

Cirsiliol OH OMe OMe OH OH

Genkwanin OH H OMe H OH

Fig. 6 CYP1-catalysed

conversion of the flavone

apigenin (R1 = H).

Analogous conversions

have been found for the

flavonol kaempferol

(R1 = OH)

Phytochem Rev (2008) 7:431–443 439

123

CYP1 enzymes or without co-factors, did show

detectable conversion of the flavonoid substrates. In

addition, incubations with a range of other recombi-

nant human cytochromes P450—CYP2A6, CYP2B6,

CYP2C8, CYP2C9, CYP2C19, CYP2D6, and

CYP3A5—did not give detectable conversion of the

flavonoids we tested. Only CYP2E1 (Piver et al.

2004) and CYP3A4 (our data) appeared to be minor

catalysts.

Interestingly, the isoflavonoid daidzein, a phyto-

estrogen from soy beans that has long been linked

with prevention cancer, does not inhibit CYP1A1 or

CYP1B1 activity (Chan and Leung 2003), and in our

assays we could not detect any bioconversion prod-

ucts from daidzein.

Phytoestrogens as natural prodrugs in cancer

prevention

Recent developments in smart anticancer drug dis-

covery have led to synthetic agonists of the AhR that

are prodrugs activated by CYP1A1. Phortress (Mac-

Fadyen et al. 2004; Brantley et al. 2006) and

aminoflavone (Kuffel et al. 2002; Pobst and Ames

2006) are both in Phase I clinical trials at the

moment. Flavones have been found to induce

CYP1A1 and CYP1B1 gene expression in animals,

and in in vitro cultivated human cells (Durgo et al.

2007; Hodek et al. 2006), possibly through modula-

tion of the aryl hydrocarbon receptor (AhR) (Pohl

et al. 2006).

Dietary flavonoids naturally occur as glycosides.

During digestion of these conjugated flavonoids, the

sugar moiety is split off allowing absorption of the

aglycone. However, the aglycone is rapidly reconju-

gated and one common characteristic of flavonoids is

that they occur in the bloodstream as glucuronide and

sulfate conjugates (Shimoi et al. 1998; Dupont et al.

2004; Erlund 2004; Chen et al. 2007). The concen-

tration of the conjugates of the more common dietary

flavonoids in the human blood stream is in the

micromolar range, and this is the concentration that

reaches the tumour tissue. Around tumour tissue,

significant extracellular levels of b-glucuronidase can

be found, due to the liberation of this enzyme from

the lysosomes of inflammatory cells (Dodds et al.

2002). Thus, tumour cells are likely to be met by

flavonoid aglycones at micromolar concentrations.

In order to screen for the presence of CYP-1

activated prodrugs, a test panel of human breast cell

cultures had been designed (Potter and Butler 2004).

This panel was used to test our hypothesis that

phytoestrogens are selectively activated in CYP-1

expressing tumour cells. The panel consists of the

human breast adenocarcinoma cell line MCF-7—

which normally does not express CYP-1 enzymes but

can be induced with the dioxin TCDD, the human

breast adenocarcinoma cell line MDA-MB-468—

which constitutively expresses CYP1s-, and the non-

tumorigenic epithelialbreast cell line MCF-10A—

which was used a model for normal breast cells. The

efficacy of a pro-drug can be expressed by determin-

ing its IC50 value (concentration needed to inhibit cell

proliferation by 50%) for the different cell lines, and

then calculating its activation factor or its tumour

selectivity factor. When the activation factor is

greater than one, cells expressing CYP1 enzymes

are inhibited more by the test substances than cells

not expressing CYP1 enzymes. When the tumour

selectivity factor is greater than one, tumour cells are

more inhibited than non-tumour cells (Tables 6, 7)

shows that some common dietary flavonoids have

activation factors and tumour selectivity factors

ranging from 2 to 10. The activation and tumour

Table 6 Inhibition of recombinant human CYP1 enzymes by

dietary flavonoids. CYP1 (CYP1A1, CYP1A2, CYP1B1)

microsomes (5 pmol/ml) were incubated with various con-

centrations of flavonoids

Compound CYP1B1 CYP1A1 CYP1A2

Flavonols

Quercetin 0.7 ± 0.3 7.0 ± 0.0 12 ± 3.5

Myricetin 0.9 ± 0.2 4.0 ± 0.0 13 ± 0.7

Kaempferol 1.3 ± 0.2 8.0 ± 1.4 9.0 ± 2.1

Flavones

Apigenin 2.0 ± 0.0 4.2 ± 0.0 6.0 ± 0.0

Luteolin 1.8 ± 0.4 10 ± 0.7 15 ± 1.4

Diosmetin 0.5 ± 0.1 1.2 ± 0.3 18 ± 4.0

Baicalein 7.0 ± 1.1 20 ± 0.0 20 ± 0.0

Chrysin 0.5 ± 0.1 0.5 ± 0.1 2.0 ± 0.0

Eupatorin 1.1 ± 0.1 2.3 ± 0.1 25

Genkwanin [25 [25 [25

Cirsiliol [25 [25 [25

Results are calculated from percentage of inhibition of

7-ethoxyresorufin-O-deethylase activity compared to control,

and expressed as IC50 (lM). Each value corresponds to

mean ± standard deviation for n = 4 determinations

440 Phytochem Rev (2008) 7:431–443

123

selectivity factors presented in Table 7 indicate that

the dietary flavonoids tested cannot be used in cancer

therapy; the dose that stops tumour growth is too

close to the dose that stops normal cell development.

However, micromolar concentrations of these dietary

compounds in the blood stream are sufficient to

selectively inhibit the proliferation of tumorigenic

cells expressing CYP1-enzymes, and thus prevent

development of tumours at the very earliest stages.

Thus, our hypothesis of phytoestrogens as natural

prodrugs explains how micromolar amounts of die-

tary flavones may prevent the onset of a wide range

of cancers. More research is needed to identify

dietary prodrugs and characterise their activated

products, and to establish the exact mechanisms by

which these compounds selectively inhibit the pro-

liferation of tumour cells.

Conventional screening methods for anticancer

compounds have led to a number of clinically used

drugs for cancer therapy. Although, new compounds

will still be discovered using conventional screening

methods, it now seems to be past its peak. Most of the

smart drugs, and probably none of the pro-drugs,

would have been picked up by conventional screen-

ing. Many drug companies have abandoned major

conventional screening programmes in favour of

novel methods for smart drug development using

tools that have become available the post-genomic

era, i.e. specific in vitro enzyme assays or even

in silico data mining.

References

Abou-Jawde R, Choueiri T, Alemany C, Mekhail T (2003) An

overview of targeted treatments in cancer. Clin Ther

25:2121–2137

Alexandrove K, Cascorbi I, Rojas M, Bouvier G, Kriek E,

Bartsch H (2002) CYP1A1 and GSTM1 genotypes affect

benzo[a]pyrene DNA adducts in smokers’ lung: compar-

ison with aromatic/hydrophobic adduct formation.

Carcinogenesis 23:1969–1977

Ames BN (1983) Dietary carcinogens and anticarcinogens—

oxygen radicals and degenerative diseases. Science

221:1256–1264

Atkinson C, Frankenfeld CL, Lampe JW (2005) Gut bacterial

metabolism of the soy isoflavone daidzein: Exploring the

relevance to human health. Exp Biol Med 230:155–170

Brantley E, Antony S, Kohlhagen G, Meng LH, Agama K,

Stinson SF, Sausville EA, Pommier Y (2006) Anti-tumor

drug candidate 2-(4-amino-3-methylphenyl)-5- fluor-

obenzothiazole induces single-strand breaks and DNA-

protein cross-links in sensitive MCF-7 breast cancer cells.

Cancer Chemother Pharmacol 58:62–72

Breinholt VM, Offord EA, Brouwer C, Nielsen SE, Brøsen K,

Friedberg (2002) In vitro investigation of cytochrome

P450-mediated metabolism of dietary flavonoids. Food

Chem Toxicol 40:609–616

Burke MD, Mayer RT (1974) Ethoxyresorufin—direct fluori-

metric assay of a microsomal O-dealkylation which is

preferentially inducible by 3-methylcholanthrene. Drug

Metab Dispos 2:583–588

Chan HY, Leung LK (2003) A potential protective mechanism

of soya isoflavones against tumour initiation. Brit J Nutr

90:457–465

Chen T, Li LP, Lu XY, Jiang HD, Zeng S (2007) Absorption

and excretion of luteolin and apigenin in rats after oral

administration of Chrysanthemum morifolium extract.

J Agri Food Chem 55:273–277

Chun YJ, Kim MY, Guengerich FP (1999) Resveratrol is a

selective human cytochrome P450 1A1 inhibitor. Bio-

chem Biophys Res Comm 262:20–24

Cragg GM, Newman DJ (2000) Antineoplastic agents from

natural sources: achievements and future directions.

Expert Opin Investig Drugs 9:2783–2797

Dietel M, Lewis MA, Shapiro S (2005) Hormone replacement

therapy: pathobiological aspects of hormone-sensitive

cancers in women relevant to epidemiological studies on

HRT: a mini-review. Hum Repr 20:2052–2060

Dodds HM, Tobin PJ, Stewart CF, Cheshire P, Hanna S,

Houghton P, Rivory LP (2002) The importance of tumor

glucuronidase in the activation of irinotecan in a mouse

xenograft model. J Pharmacol Exp Ther 303:649–655

Table 7 Effect of selected

flavonoids on the

proliferation of in vitro

human cell cultures

Compound MCF-7 IC50 (lM) MCF-7 + TCDD IC50 (lM) Activation factor

Apigenin 200 20 10

Kaempferol 70 35 2

Scutellarein 30 15 2

Compound MCF-10A IC50 (lM) MDA-MB-468 IC50 (lM) Tumour selectivity

Apigenin 200 50 4

Kaempferol 60 30 2

Scutellarein 40 10 4

Phytochem Rev (2008) 7:431–443 441

123

Duffy C, Cyr M (2003) Phytoestrogens: Potential benefits and

implications for breast cancer survivors. J Women Health

12:617–631

DuPont MS, Day AJ, Bennett RN, Mellon FA, Kroon PA

(2004) Absorption of kaempferol from endive, a source of

kaempferol-3-glucuronide, in humans. Eur J Clin Nutr

58:947–954

Durgo K, Vukovic L, Rusak G, Osmak M, Colic JF (2007)

Effect of flavonoids on glutathione level, lipid peroxida-

tion and cytochrome P450CYP1A1 expression in human

laryngeal carcinoma cell lines. Food Tech Biotech 45:

69–79

Erlund I (2004) Review of the flavonoids quercetin, hesperetin,

and naringenin. Dietary sources, bioactivities, bioavail-

ability, and epidemiology. Nutri Res 24:851–874

Ferrigni NR, Mclaughlin JL, Powell RG, Smith CR (1984) Use

of potato disk and brine shrimp bioassays to detect

activity and isolate piceatannol as the antileukemic prin-

ciple from the seeds of Euphorbia-lagascae. J Nat Prod

47:347–352

Gibson P, Gill JH, Khan PA, Seargent JM, Martin SW, Batman

PA, Griffith J, Bradley C, Double JA, Bibby MC, Load-

man PM (2003) Cytochrome P450 1B1 (CYP1B1) is

overexpressed in human colon adenocarcinomas relative

to normal colon: Implications for drug development. Mol

Cancer Ther 2:527–53

Gradolatto A, Canivenc-Lavier MC, Basly JP, Siess MH,

Teyssier C (2004) Metabolism of apigenin by rat liver

phase I and phase II enzymes and by isolated perfused rat

liver. Drug Metab Dispos 32:58–65

Halliwell B (2000) The antioxidant paradox. The Lancet

335:1179–1180

Halliwell B (2007) Dietary polyphenols: Good, bad, or indif-

ferent for your health? Cardiovascular Res 73:341–347

Halliwell B, Rafter J, Jenner A (2005) Health promotion by

flavonoids, tocopherols, tocotrienols, and other phenols:

direct or indirect effects? Antioxidant or not? Am J Clin

Nutr 81:268S–276S

Hodek P, Hanustiak P, Krizkova J, Mikelova R, Krizkova S,

Stiborova M, Trnkova L, Horna A, Beklova M, Kizek R

(2006) Toxicological aspects of flavonoid interaction with

biomacromolecules. Neuroendocrin Lett 27(2):14–17

Jang MS, Cai EN, Udeani GO, Slowing KV, Thomas CF,

Beecher CWW, Fong HHS, Farnsworth NR, Kinghorn

AD, Mehta RG, Moon RC, Pezzuto JM (1997) Cancer

chemopreventive activity of resveratrol, a natural product

derived from grapes. Science 275:218–220

Jang M, Pezzuto JM (1999) Cancer chemopreventive activity

of resveratrol. Drug Exp Clin Res 25:65–77

Kinghorn AD, Su BN, Lee D, Gu JQ, Pezzuto JM (2003) Cancer

chemopreventive agents discovered by activity-guided

fractionation: An update. Curr Org Chem 7:213–226

Kinghorn AD, Su BN, Jang DS, Chang LC, Lee D, Gu JQ,

Carcache-Blanco EJ, Powlus AD, Lee SK, Park EJ,

Cuendet M, Gills JJ, Bhat K, Park HS, Mata-Greenwood

E, Song LL, Jong MH, Pezzuto JM (2004) Natural

inhibitors of carcinogenesis. Planta Medica 70:691–705

Knasmuller S, Steinkellner H, Majer BJ, Nobis EC, Scharf G,

Kassie F (2002) Search for dietary antimutagens and an-

ticarcinogens: methodological aspects and extrapolation

problems. Food Chem Toxicol 40:1051–1062

Kuffel MJ, Schroeder JC, Pobst LJ, Naylor S, Reid JM,

Kaufmann SH, Ames MM (2002) Activation of the anti-

tumor agent aminoflavone (NSC 686288) is mediated by

induction of tumor cell cytochrome P450 1A1/1A2. Mol

Pharmacol 62:143–153

Kuijsten A, Arts ICW, Hollman PCH, van’t Veer P, Kampman

E (2006) Plasma enterolignans are associated with lower

colorectal adenoma risk. Cancer Epidemiol Biomarkers

Prev 15:1132–1136

La Vecchia C, Altieri A, Tavani A (2001) Vegetables, fruit,

antioxidants and cancer: a review of Italian studies. Eur

J Nutr 40:261–267

Lu J, Ho C, Ghai G, Chen KJ (2001) Resveratrol analogue

3,4,5,40-tetrahydroxystilbene differentially induces pro-

apoptotic p53/Bax gene expression and inhibits the

growth of transformed cells but not their normal coun-

terparts. Carcinogenesis 22:321–328

McFadyen MCE, Melvin WT, Murray GI (2004) Cytochrome

P450 enzymes: Novel options for cancer therapeutics.

Mol Cancer Ther 3:363–371

Moon YJ, Wang XD, Morris ME (2006) Dietary flavonoids:

Effects on xenobiotic and carcinogen metabolism. Toxicol

In Vitro 20:187–210

Murray GI, Taylor MC, McFadyen MCE, McKay JA, Greenlee

WF, Burke MD, Melvin WT (1997) Tumor specific expres-

sion of cytochrome P450 CYP1B1. Cancer Res 57:3026–3031

Murray GI, Melvin WT, Greenlee WF, MD Burke (2001) Regu-

lation, function, and tissuespecific expression of cytochrome

P450 CYP1B1. Annu Rev Pharmacol Toxicol 41:297–316

Pasetto LM, Falci C, Compostella A, Sinigaglia G, Rossi E,

Monfardini S (2007) Quality of life in elderly cancer

patients. Eur J Cancer 43:1508–1513

Piver B, Fer M, Vitrac X, Merillon JM, Dreano Y, Berthou F,

Lucas D (2004) Involvement of cytochrome P450 1A2 in

the biotransformation of trans-resveratrol in human liver

microsomes. Biochem Pharmacol 68:773–782

Pobst LJ, Ames MM (2006) CYP1A1 activation of aminof-

lavone leads to DNA damage in human tumor cell lines.

Cancer Chemother Pharmacol 57:569–576

Pohl C, Will F, Dietrich H, Schrenk D (2006) Cytochrome

p450 1A1 expression and activity in Caco-2 cells: Mod-

ulation by apple juice extract and certain apple

polyphenols. J Agric Food Chem 54:10262–10268

Potter GA, Butler PC (2004) 3,4-Methylenedioxy-substituted

chalcones as therapeutic agents. Patent Application

Publication US2004/0254149 A1

Potter GA, Patterson LH, Wanogho E, Perry PJ, Butler PC, Ijaz

T, Ruparelia KC, Lamb JH, Farmer PB, Stanley LA, Burke

MD (2002) The cancer preventative agent resveratrol is

converted to the anticancer agent piceatannol by the cyto-

chrome P450 enzyme CYP1B1. Brit J Cancer 86:774–778

Rickard SE, Orcheson LJ, Seidl MM, Luyengi L, Fong HHS,

Thompson LU (1996) Dose dependent production of

mammalian lignans in rats and in vitro from the purified

precursor secoisolariciresinol diglycoside in flaxseed.

J Nutr 126:2012–2019

Rooseboom M, Commandeur JNM, Vermeulen NPE (2004)

Enzyme-catalyzed activation of anticancer prodrugs.

Pharmacol Rev 56:53–102

Russo J, Russo IH (2004) Genotoxicity of steroidal estrogens.

Trend Endocrinol Metab 15:211–214

442 Phytochem Rev (2008) 7:431–443

123

Sale S, Verschoyle RD, Boocock D, Jones DJL, Wilsher N,

Ruparelia KC, Potter GA, Farmer PB, Steward WP,

Gescher AJ (2004) Pharmacokinetics in mice and growth-

inhibitory properties of the putative cancer chemopre-

ventive agent resveratrol and the synthetic analogue trans

3,4,5,40-tetramethoxystilbene. Brit J Cancer 90:736–744

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-L-glucoside in rats

and humans. FEBS Lett 438:220–224

Stanley LA, Ball MT, Eaden J, Potter GA, Rolls S, Burke MD

(2001a) Cytochrome P450 CYP1B1 in colon tumorigen-

esis. Drug Metab Rev 33(Suppl 1):62

Stanley LA, Ball MT, Butler PC, Hoang TTV, Potter GA &

Burke MD (2001b) CYP1B1 expression in human cervi-

cal abnormalities. Drug Metab Rev 33(Suppl 1):77

Suggitt M, Bibby C (2005) 50 Years of Preclinical anticancer

drug screening: Empirical to target-driven approaches.

Clin Cancer Res 11:971–981

Tsuchiya Y, Nakajima M, Yokoi T (2005) Cytochrome P450-

mediated metabolism of estrogens and its regulation in

human. Cancer Lett 227:115–124

Tsuji PA, Walle T (2006) Inhibition of benzo[a]pyrene-acti-

vating enzymes and DNA binding in human bronchial

epithelial BEAS-2B cells by methoxylated flavonoids.

Carcinogenesis 27:1579–1585

Usui T (2006) Pharmaceutical prospects of phytoestrogens.

Endocr J 53:7–20

Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M (2006)

Free radicals, metals and antioxidants in oxidative stress-

induced cancer. Chem-Biol Interact 160:1–40

van der Poel HG (2004) Smart drugs in prostate cancer. Eur

Urol 45:1–17

Zhang JY, Wang Y, Prakash C (2006) Xenobiotic-metaboliz-

ing enzymes in human lung. Curr Drug Metab 7:939–948

Phytochem Rev (2008) 7:431–443 443

123