Accepted Manuscript

Title: Combined therapy using bevacizumab and turmeric

ethanolic extract (with absorbable curcumin) exhibited
beneficial efficacy in colon cancer mice

Author: Grace Gar-Lee Yue Hin-Fai Kwok Julia Kin-Ming

Lee Lei Jiang Eric Chun-Wai Wong Si Gao Hing-Lok Wong
Lin Li Kar-Man Chan Ping-Chung Leung Kwok-Pui Fung
Zhong Zuo Clara Bik-San Lau

PII: S1043-6618(16)30168-2
DOI: http://dx.doi.org/doi:10.1016/j.phrs.2016.05.025
Reference: YPHRS 3185

To appear in: Pharmacological Research

Received date: 3-3-2016

Revised date: 11-5-2016
Accepted date: 25-5-2016

Please cite this article as: Yue Grace Gar-Lee, Kwok Hin-Fai, Lee Julia Kin-Ming,
Jiang Lei, Wong Eric Chun-Wai, Gao Si, Wong Hing-Lok, Li Lin, Chan Kar-
Man, Leung Ping-Chung, Fung Kwok-Pui, Zuo Zhong, Lau Clara Bik-San.Combined
therapy using bevacizumab and turmeric ethanolic extract (with absorbable curcumin)
exhibited beneficial efficacy in colon cancer mice.Pharmacological Research
http://dx.doi.org/10.1016/j.phrs.2016.05.025

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Combined therapy using bevacizumab and turmeric ethanolic extract (with absorbable

curcumin) exhibited beneficial efficacy in colon cancer mice

Grace Gar-Lee Yue1,2, Hin-Fai Kwok1,2, Julia Kin-Ming Lee1,2, Lei Jiang1,2,3, Eric Chun-Wai Wong1,2,

Si Gao1,2, Hing-Lok Wong1,2, Lin Li1,2, Kar-Man Chan1,2, Ping-Chung Leung1,2, Kwok-Pui Fung1,2,3,

Zhong Zuo4 and Clara Bik-San Lau1,2,*

1
Institute of Chinese Medicine; 2State Key Laboratory of Phytochemistry and Plant Resources in

West China (CUHK), The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong;
3
School of Biomedical Sciences, 4School of Pharmacy, The Chinese University of Hong Kong,

Shatin, New Territories, Hong Kong

* Corresponding author

Clara Bik-San LAU

Address: Institute of Chinese Medicine, E305, Science Centre East Block, The Chinese University

of Hong Kong, Shatin, New Territories, Hong Kong.

Tel: +852 3943 6109 Fax: +852 2603 5248

E-mail address: claralau@cuhk.edu.hk

Running title: Combined efficacies of turmeric and bevacizumab

G.G.L. Yue et al.

Abstract

Turmeric is commonly used as a medicinal herb and dietary supplement. Its active ingredient,

curcumin, has been shown to possess antitumor effects in colorectal cancer patients. However,

poor absorption of curcumin in intestine impedes its wide clinical application. Our previous

findings showed that the presence of turmerones increased the accumulation of curcumin inside

colonic cells. Hence, we hypothesized that curcumin with turmerones or present in turmeric

ethanolic extract would augment its anti-tumor activities in tumor-bearing mice. The

pharmacokinetics of curcumin in different preparations (containing same amount of curcumin)

were studied in mice. The anti-tumor efficacies of curcumin or turmeric extract (with absorbable

curcumin) in combination with bevacizumab were further investigated in HT29 colon

tumor-bearing mice. Pharmacokinetic results showed that the plasma curcumin level of turmeric

extract-fed mice was the highest, suggesting turmeric extract had the best bioavailability of

curcumin. Besides, combined turmeric extract plus bevacizumab treatment significantly inhibited

the tumor growth. Such inhibitory effects were stronger than those of curcumin plus bevacizumab

or bevacizumab alone and were comparable with those of 5-fluorouracil + leucovorin + oxaliplatin

(FOLFOX) plus bevacizumab. Notably, there was no observable side effect induced by turmeric

extract treatment while significant side effects were found in FOLFOX-treated mice. In conclusion,

combination of turmeric extract with bevacizumab possessed potent anti-tumor effects without

observable side effects, strongly suggesting the adjuvant use of turmeric extract in colorectal

cancer therapy. Our current findings warrant the confirmation regarding the benefits arising from

the combined use of bevacizumab and turmeric in colorectal cancer patients in the near future.

Keywords

Bevacizumab; colorectal cancer; curcumin; Turmeric; turmerone

2
G.G.L. Yue et al.

Abbreviations: 5-FU, 5-fluorouracil; ALT, alanine aminotransferase; AST, aspartate

aminotransferase; AUC, area under concentration–time curve; Cmax, maximum concentration; CK,

creatine kinase; FBS, fetal bovine serum; FOLFOX, 5-fluorouracil plus leucovorin plus oxaliplatin;

Tmax, maximum time; PBS, phosphate-buffered saline; UPLC, ultra performance liquid

chromatography; VEGF, vascular endothelial growth factor

3
G.G.L. Yue et al.

1. Introduction

Colorectal cancer is the top three most common cancers in both sexes worldwide. It is the 4th

and 3rd leading cause of cancer-related mortality in men and women, respectively.1 Dietary

supplements are widely used in cancer patients, especially those of plant-derived. Turmeric, the

dried rhizome of the plant Curcuma longa Linn. (family Zingiberaceae), is a commonly used spice

and traditional medicine in Indian and Chinese culture. Turmeric is recommended for prevention of

cancer and other diseases.2 Previous epidemiological studies suggested that turmeric contributes

to the lower incidence of large-bowel cancers in Indians.3,4 The active ingredient, curcumin, is

frequently described as a chemopreventive agent for colorectal cancer.5,6

However, the poor bioavailability of curcumin is still regarded as a major problem and hence

not being approved as a therapeutic agent.7 A previous clinical trial reported that in colorectal

cancer patients ingested curcumin, both normal and malignant colorectal tissues were found to

have taken up curcumin.8 Pharmacokinetic studies in human demonstrated that oral administration

of curcumin was well tolerated without any toxicity, but only trace amount of curcumin was

detected in plasma.9 A commercially available formulation containing unknown ratio of curcumin

and non-curcuminoid components of turmeric was shown to have enhanced curcumin absorption

in human subject when compared to normal curcumin.10 Recently, the curcumin-free turmeric

components have also been shown to exhibit numerous biological activities.11

Our previous in vitro study using colonic Caco-2 cell monolayers has demonstrated that the

presence of turmerones, the non-polar constituents in turmeric ethanolic extract, could increase

the accumulation of curcumin inside colonic cells.12 We thus hypothesized that more curcumin

could be absorbed into the colonic cells with turmerones so that curcumin could take advantage of

its anti-tumor and anti-inflammatory properties towards the colorectal malignant cells. Our recent

study revealed that in HT29 tumor xenograft-bearing mice fed with curcumin alone or turmeric

ethanolic extract (in which the dose of curcumin was the same), the tumor burden was lower in

turmeric extract-fed mice than in curcumin-fed mice, suggesting turmeric extract provided better

4
G.G.L. Yue et al.

anti-tumor activities than the same amount of curcumin alone did.13

Clinically, the commonly used chemotherapeutics for colorectal cancer is 5-fluorouracil (5-FU),

leucovorin plus oxaliplatin (FOLFOX).14 In USA, for those patients receiving 5-FU-containing

regimen as first-line chemotherapy, majority (64%) were treated with FOLFOX plus

bevacizumab.15 Bevacizumab, which is a monoclonal antibody targeting on vascular endothelial

growth factor, was approved as first line therapy with any fluorouracil-based chemotherapy by US

FDA since 2004.16 Emerging findings from clinical trials suggested that the co-administration of

bevacizumab and neoadjuvant chemotherapy increased overall survival and/or the pathologic

complete responses.17 On the other hand, the inclusion of curcumin in conventional

chemotherapeutic (FOLFOX) was shown to reduce the survival of chemo-surviving colon cancer

cells.18 Other in vivo studies also showed that curcumin could enhance the apoptotic effects of
19
oxaliplatin and sensitize colorectal cancer cells to capecitabine.20 Curcumin is in fact safe,
6,9,21,22
affordable and it can induce tumor cell death in different pathways. Nevertheless, the

interaction between natural products and bevacizumab has rarely been reported in colon cancer

cells or animal studies.

It is possible that the anti-tumor efficacy of curcumin in combination with bevacizumab could

be enhanced in colon cancer animal model. In addition, the bioavailability of curcumin could also

be enhanced, according to our hypothesis, by combination with turmerones or present in turmeric

extract. Hence, the aims of the present study were to evaluate the pharmacokinetics of curcumin

(alone, in the presence of turmerones or in turmeric extract) in nude mice model so that the most

bioavailable curcumin preparation could be determined. Then, the anti-tumor efficacies of

curcumin or turmeric extract combined with bevacizumab were investigated and compared with

those of FOLFOX plus bevacizumab in colon tumor-bearing mice. The side effects attenuation

properties of herbal preparations plus bevacizumab treatment were also compared with those of

FOLFOX plus bevacizumab treatment.

5
G.G.L. Yue et al.

2. Materials and methods

2.1 Preparation of turmeric ethanolic extract, curcumin and turmerones

Turmeric ethanolic extract was prepared as described in our previous studies.12,23 Dried

rhizome of Curcuma longa Linn. (also known as Turmeric) were authenticated morphologically and

chemically in accordance with the Chinese Pharmacopoeia 2010.24 Authenticated voucher

specimen (Number: 3353) was deposited in the museum of the Institute of Chinese Medicine, The

Chinese University of Hong Kong. The dried rhizome of Curcuma longa was powdered and then

extracted twice under reflux using 95 % ethanol for 1 h. The crude ethanolic extract was

evaporated under reduced pressure at 60 °C to dryness. The percentage yield of the turmeric

ethanolic extract was 12.7 % (w/w). The turmeric ethanolic extract (named “turmeric extract” used

throughout this manuscript) was stored at 4 °C and protected from light. For the pure compounds

(Figure 1A), curcumin was purified from the commercially available crude curcumin (purity ≥ 65%,

CAS: 458-37-7, Sigma, USA), while the /-turmerone and aromatic-turmerone (Ar-turmerone)

were isolated from turmeric extract as described in previous study.23 The quantification of curcumin

and Ar-turmerone were determined using an Agilent HPLC system (CA, USA). The column used

was Agilent Poroshell 120 EC-C18, 4.6 mm x 100 mm packed with 2.7 µm hydrophobic bonded

C18 phase, accompanied with an Agilent UHPLC Guard Poroshell 120 EC-C18, 4.6 mm x 5 mm, 2.7

µm. The mobile phase consisted of acetonitrile (A) and 0.1% acetic acid in water (B). The

extracted sample was eluted on a gradient mobile phase starting from 38% of A at zero time to

43% A in 4 min in a linear gradient, 43% to 80% A in 4 min and then to 100% A in 2 min at a flow

rate of 1 mL/min. The injection volume for all samples was 5 mL. Detection wavelength was set at

240 nm (for Ar-turmerone) and 425 nm (for curcumin). Standard curves were constructed for

calibration. The confirmation of curcumin and Ar-turmerone in turmeric ethanolic extract were also

conducted using an Agilent 1290 UHPLC with 6530 QTOF system (CA, USA). The column used

was Agilent ZORBAX Eclipse Plus C18 RRHD, 2.1 x 150 mm, 1.8 µm, accompanied with a guard

column (Agilent ZORBAX Eclipse Plus C18 UHPLC Guard, 2.1 x 5 mm, 1.8 µm). The

6
G.G.L. Yue et al.

chromatographic separation was conducted at 40°C under gradient conditions at a flow rate of 0.4

mL/min. The HPLC system was as follows: Mobile phase: (A) 0.1% formic acid, in deionized and

distilled H2O and (B) 0.1% formic acid in methanol; Gradient (in buffer A): 0–1 min, 2% B; 1-30 min,

2–80% B; 30-35 min 80% B; 35-38 min, 80-100% B; 38-40min, 100% B; 40-41 min, 100-2% B;

41-43 min, 2% B. High purity nitrogen was used as the curtain and collision gas with a flow rate of

9 L/min. The drying gas temperature was set at 350°C, and the nebulizer pressure was set at

60psi. Spectra were recorded in positive ion mode at spray voltage of 3500V. The mass scan

range was between 50-1000 m/z. Data analysis was performed using Agilent MassHunter

Workstation Qualitative Analysis Software (CA, USA, version B.06.00). Curcumin was determined

at 369m/z [M+H]+ and Ar-turmerone was determined at 217m/z [M+H]+.

2.2 Chemicals and reagents

Human colon cancer cells (HT29) were purchased from American Type Culture Collection

(MD, USA). Cells were maintained in McCoy's 5A medium (ATCC, USA), containing 10 % (v/v)

heat-inactivated fetal bovine serum, 100 units/mL penicillin, and 100 g/mL streptomycin. The cells

were incubated at 37 C in a humidified atmosphere of 5% CO2 and those cells in the exponential

growth phase were used for experiments. Cell culture supplements and Trizol were obtained from

Life technologies (NY, USA). -glucuronidase, sulfatase, hesperetin as well as chemotherapeutic

drugs, 5-fluorouracil, oxaliplatin, leucovorin were purchased from Sigma-Aldrich (MO, USA) and

bevacizumab was purchased from F. Hoffmann-La Roche Ltd. (Germany). Plasma enzymes,

aspartate aminotransferase (AST), alanine aminotransferase (ALT) and creatine kinase (CK)

quantification kits were obtained from Stanbio Laboratory (TX, USA). Plasma troponin-I ELISA kit

was obtained from Life Diagnostics (PA, USA). In situ cell death detection kit was obtained from

Roche (Germany). For immunohistochemical staining, anti-mouse CD31 antibodies were from

Dianova (Germany), while anti-mouse Factor VIII antibodies, peroxidased 1 blocking reagent and

7
G.G.L. Yue et al.

Diva Declocker reagent were obtained from Biocare Medical (CA, USA). The secondary

horseradish peroxidase-conjugated antibodies were from Life Technologies (NY, USA). QuantiFast

SYBR Green RT-PCR kit was from Qiagen (Germany).

2.3 Experimental animals

Male Balb/c nude mice (aged 6-8 weeks) were supplied by Laboratory Animal Services Centre,

The Chinese University of Hong Kong. The animal experiments were approved by Animal

Experimentation Ethics Committee of The Chinese University of Hong Kong (Ref no. 10/091/MIS

and Ref no. 11/071/MIS). Mice were fed on a standard laboratory diet with free access to water

under a controlled temperature at 20 – 22 °C and relative humidity of 50% with 12/12 hour

light/dark cycle prior to the study.

2.4 Pharmacokinetic studies of curcumin preparations

Nude mice from five treatment groups (Table 1) received oral administrations of various curcumin

and turmerones combinations at a dose equivalent to 75 mg/kg curcumin. All the preparations

were suspended in distilled water containing 1% methylcellulose as described before.25 The freely

soluble curcumin doses in each curcumin preparations were further verified by UPLC (Table 1). In

each treatment group, mice were terminated at different time intervals of 0.25, 0.5, 1, 2, 4, 8, 12

and 24 h (n = 3 per time point) post dosing followed by collection of blood via intra-cardiac

puncture and various tissues including small intestine, colon, liver. Plasma was obtained by

centrifugation of blood and stored at -80 °C until analysis. The collected tissue samples were

rinsed in phosphate buffered saline (PBS) and immediately snap frozen in liquid nitrogen and

stored at -80 C before sample preparation for UPLC analysis of curcumin.

2.5 Plasma sample preparations for curcumin content analyses

Considering the potential extensive glucuronidation and sulfation of curcumin, each plasma

8
G.G.L. Yue et al.

sample was analyzed in absence and presence of glucuronidase and sulfatase.26,27 Curcumin was
26
extracted from plasma as previously described with modification with hesperetin as an internal

standard.25 Briefly, an aliquot of plasma (150 L) was mixed with water (80 L) and internal

standard (40 L, 250 g/mL) and vortex mixed for 30 s. The solution was then extracted twice with

900 L extraction reagent (95 % ethyl acetate/ 5 % methanol) and vortex mixed for 30 s. The

samples were centrifuged at 13,000 rpm for 5 min. The upper organic layer was transferred to a

clean tube. The organic phases from the two extractions were pooled and evaporated under a

stream of nitrogen gas at room temperature. For those samples undergone hydrolysis, an aliquot

of plasma (150 L) was mixed with water (80 L) and internal standard (40 L, 250 g/mL) and

vortex mixed for 30 s. The samples were then mixed with -glucuronidase (50 L, 446 units) in 0.1

mol/L phosphate buffer (pH 6.8) and sulfatase (45 L, 52 units) in 0.1 mol/L sodium acetate buffer

(pH 5.0) and incubated at 37 C for 3.5 h. The hydrolysed samples were extracted with extraction

reagent as described above.

2.6 Tissue sample preparations for curcumin content analyses

The liver, small intestine and colon tissue samples were homogenized by a TissueRuptor (Qiagen,

USA) in PBS. PBS (400 µL) was added to liver tissue (2 mL/g tissue) and intestinal tissue (4 mL/g

tissue). Samples were prepared for two sets of experiments as described below. In the first set of

experiment, all tissue homogenate (~450 L) was mixed with internal standard (40 L, 250 g/mL)

and extracted twice with 900 L chloroform, followed by centrifugation at 13,000 rpm for 5 min at 4

C. The organic layer was pooled and evaporated under a stream of nitrogen gas at room

temperature. In order to measure the concentration of both parent curcumin and its conjugate

metabolites, part of the tissue samples were hydrolysed in the second set of experiment. One third

of the whole homogenate (~150 L) was mixed with water (100 L) and internal standard (40 L)

and vortex for 30 s. The samples were mixed with the enzymes mixture (described above) and

incubated at 37 C for 3.5 h. The samples were extracted twice with chloroform at 1:3 (v/v),

9
G.G.L. Yue et al.

followed by centrifugation at 13,000 rpm for 5 min at 4 C. The organic layer was pooled and

evaporated to dryness. The remaining homogenate (~300 L) was mixed with internal standard

and extracted in the same way as those hydrolysed samples.

All the extraction procedures were performed under dim light to prevent degradation of curcumin.

The extracted dried product was reconstituted in 100 L of mobile phase reagent for UPLC

analysis.

2.7 UPLC analysis of curcumin

Curcumin concentrations in plasma and tissue lysates were determined using a Waters Acquity

UPLC system (MA, USA). The column used was Waters ACQUITY UPLC BEH C18, 2.1 mm  100

mm packed with 1.7 µm hydrophobic bonded C18 phase, accompanied with a guard column of 2.1

mm x 5 mm, 1.7 µm (Waters ACQUITY UPLC C18 VanGuard Pre-Column). The mobile

phase consisted of acetonitrile (A) and 0.1% acetic acid in water (B). The extracted

sample was eluted on a gradient mobile phase starting from 38% of A at zero time to 43% A in 4

min in a linear gradient, and then to 100% A in 0.5 min at a flow rate of 0.5 mL/min. The injection

volume for all samples was 5 L. Detection wavelength was set at 280 nm and 425 nm for

curcumin and hesperetin, respectively. Standard curves were constructed using plasma or tissue

homogenized lysate added to curcumin. Blank plasma samples or tissue lysates were added to

varying amount of standard curcumin to yield final curcumin concentration over the range of 0.05

to 25 g/mL. Spiked samples were extracted as described above. This UPLC method offered a

detection limit of 0.05 g/mL. Each sample was analysed in triplicate. Method validation for

curcumin was conducted according to the “Guidance for industry: bioanalytical method validation”

of US Food and Drug Administration. The mean recovery of samples containing curcumin was

87.6% and the intra-day and inter-day error were 2.3 % and 2.7 %, respectively.

2.8 Pharmacokinetic analysis

10
G.G.L. Yue et al.

The pharmacokinetic parameters of curcumin, including area under concentration–time curve

(AUC), maximum concentration and time (Cmax and Tmax) were calculated by WinNonlin™ with

non-compartmental approach. The software was purchased from Phar-sight, Mountain View (CA,

USA). AUC0→∞ obtained from different treatment groups were normalized by the free soluble

curcumin in different preparations and the relative bioavailability (F) was calculated as the ratio of

AUC0→∞ after oral administration of curcumin with /-turmerone, or Ar-turmerone, or

/-turmerone + Ar-turmerone, or curcumin in turmeric extract, versus that after oral administration

of curcumin only.

2.9 Colon tumor-bearing mouse model

The turmeric ethanolic extract, which was shown to be the most absorptive in nude mice, was

orally administered to the colon tumor-bearing mice to assess its anti-tumor efficacies. The efficacy

of curcumin alone was also evaluated for comparison in the same model. For experiment, male

nude mice were subcutaneously inoculated on the right flank with HT29 colonic cancer cells (5 

106 cells per 200 L PBS). Tumors were allowed to grow for 7 days.28 Mice with tumors in

appropriate size were randomly divided into different treatment groups (Table 2) and treatments

were started on day 8. Curcumin and turmeric ethanolic extract were dissolved in 95% ethanol to

give 100 mg/mL and reconstituted in distilled water containing 1% methylcellulose for animal

studies. The vehicle control groups received distilled water containing 1% methylcellulose.

The dose of bevacizumab was modified from previous study,29 in which 3 doses (0.4, 1.2 and 2.5

mg/kg) of bevacizumab could inhibit the growth of human colorectal cancer xenografts. Our results

from pilot studies showed that all three doses of bevacizumab could significantly decrease the

tumor size when compared to untreated control, and no significant difference in tumor size among

the 3 dosage groups. Hence, the minimum effective dose (0.4 mg/kg) was chosen for the

combination with turmeric extract or FOLFOX in the subsequent experiments. The dose,

administration route and time interval of FOLFOX were adopted from another previous study.30

11
G.G.L. Yue et al.

The treatments lasted for 30 days. Body weights were monitored and the size of tumors were

measured with a caliper every 3-4 days and were calculated using the formula:31

(length × width)2/2 (1)

At the end of treatment period, mice were anaesthetized and blood was collected in heparinized

tube by intracardiac puncture. After terminated with cervical dislocation, tumors, hearts, livers of

the mice were collected for further analysis. In another set of experiment, the survival of untreated

control group, treatment groups (curcumin plus bevacizumab, turmeric extract plus bevacizumab

and FOLFOX plus bevacizumab) (n = 20) has been recorded during the experiment until 60 days

after tumor inoculation.

2.10 Determination of plasma enzyme activities and haematological parameters

Blood collected from the mice was centrifuged at 4,000 rpm for 10 min. Plasma was collected and

stored at -80 °C prior to use. The levels of plasma enzymes, aspartate transaminase (AST) and

alanine transaminase (ALT), creatine kinase (CK) and troponin levels were measured using ELISA

kits. The assay was carried out according to the procedures recommended by the manufacturer.

The hemoglobin levels as well as the complete blood count (including blood platelets, leukocytes

and neutrophils, etc.) were also measured.

2.11 Histological analysis

Hearts, livers and tumors were excised from the nude mice and fixed in 10 % buffered formalin,

embedded in paraffin wax and cut into 5-m thick sections. The heart and liver sections were

stained with haematoxylin and eosin using standard protocol. Histological changes were examined

under light microscope. Apoptotic cells in the heart, liver and tumor sections were assessed using

a direct TUNEL labelling assay (In situ cell death detection kit). The area of TUNEL labelled in

tumor sections were quantified in a double-blind manner using Image J software (version 1.43, U.

S. National Institutes of Health, MD, USA).13 In addition, tumor sections were subjected to

12
G.G.L. Yue et al.

immunohistochemical staining for vascular endothelium marker CD31 and angiogenic marker

Factor VIII.32,33 Detailed procedures were described in supporting information.

2.12 Real time-PCR analysis for tumor tissues

Tumors were harvested with TRIzol Reagent. Total RNA was extracted from cells using TRIzol

Reagent according to the manufacturer’s protocols. The RNA concentration was

spectrophotometrically determined using a BioPhotometer (Eppendorf, Hauppauge, NY, USA). To

quantify the amount of mRNA of COX-2, AKT1, cyclin D1, survivin and 18 S, RT-PCR were

performed as described in supporting information. The specific gene mRNA levels were

normalized to that of the internal control endogenous human 18 S and then expressed as the fold

change compared to the control group. All the experiments were performed in triplicates.

2.13 Statistical analyses

Data were expressed as mean + S.E.M. Statistical analyses and significance were analyzed by

one-way ANOVA with Tukey’s post-hoc test using GraphPad PRISM software version 6.0

(GraphPad Software, CA, USA). In all comparisons, p < 0.05 was considered as statistically

significant.

3. Results

3.1 Standardization of turmeric ethanolic extract and quantification of chemical markers

The turmeric ethanolic extract was subjected to standardization in which the UPLC chemical

profiles were registered (Figures 1B and 1C). Quantification of two commercially available

chemical markers, curcumin and Ar-turmerone (Sigma, USA), in turmeric ethanolic extract was

performed using by HPLC. The contents of curcumin and Ar-turmerone present in turmeric

ethanolic extract were 18.7 % (w/w) and 5.3 % (w/w), respectively. Curcumin and Ar-turmerone

were shown in the 3-D chromatograms in Figure 1D and 1E. Besides, the LC-MS study has been

conducted to confirm curcumin and Ar-turmerone in the turmeric ethanolic extract. The extracted
13
G.G.L. Yue et al.

ion chromatograms (EIC) and MS spectra of these compounds were shown in Figure 1F and 1G.

3.2 Curcumin present in turmeric extract is more bioavailable than curcumin on its own

The pharmacokinetics of curcumin was characterized in nude mice fed with different curcumin

preparations. The plasma concentration-time profiles of curcumin in mice following an oral

administration of different curcumin preparations were illustrated in Figure 2A. Pharmacokinetic

parameters in plasma obtained from the concentration-time data are summarized in Table 3. At all

sampling points, the plasma concentrations in mice treated with turmeric extract were higher than

those treated with curcumin alone or curcumin plus turmerones. The Cmax values of 2.9 g/mL for

turmeric extract, about 8 times higher as compared to 0.35 g/mL for curcumin alone, indicating

that turmeric extract can achieve a higher in vivo plasma level of curcumin. Meanwhile, the

AUC0→∞ increased by about 4-fold for turmeric extract compared to that of curcumin alone.

3.3 Differential curcumin tissues distributions from various preparations

The pharmacokinetics of curcumin in colon, small intestine and liver tissues were further

characterized in nude mice fed with different curcumin preparations. As shown in Figure 2B and

2C, most of curcumin was found in the colon and small intestine at 1 and 0.5 h after oral

administrations, respectively. The Cmax values of curcumin plus /-turmerone in colon, curcumin

alone and curcumin in turmeric extract in small intestine were around 2.7 g/g tissue (Table 4). In

colon tissues, the AUC0→∞ of curcumin in curcumin + Ar-turmerone group was the highest; while in

small intestine tissues, curcumin alone group was the highest. Figure 2D showed that curcumin

could be detected in liver tissue from mice fed with turmeric extract and curcumin alone in both

parent form and its glucuronide and/or sulfate at 0.5 h after oral administration.

Our previous in vitro study using intestinal Caco-2 cell monolayers demonstrated that the presence

of turmerones could increase the accumulation of curcumin inside intestinal cells. Meanwhile, the
12
decreased solubility of curcumin in the presence of turmerones was observed. Therefore, in the

14
G.G.L. Yue et al.

present study, the freely soluble curcumin concentration in different curcumin preparations has

been taken into account when the bioaccessibility of curcumin was compared after feeding with

different preparations. Although the curcumin concentrations in each preparation were assumed to

be the same (i.e 75 mg/kg), the freely soluble curcumin concentrations varied as shown in (Table 1)

due to the presence of other substances. In the presence of turmerones, the freely soluble

curcumin concentrations decreased 21-35% as compared to curcumin alone. In turmeric extract,

the freely soluble curcumin was one-fifth of the curcumin alone preparation. In this regards, the

relative bioavailability (F) was calculated as mentioned in methodology section.

As shown in Figure 3A, the dose-normalized plasma AUC0→∞ after oral administration of turmeric

extract group was 0.057 while that of curcumin alone group was 0.0025. The relative bioavailability

for the turmeric extract group was 22.4-fold higher than that of curcumin alone group. Statistical

analysis showed that the dose-normalized AUC0→∞ of turmeric extract group was significantly

different from all the other groups (p < 0.005). Similarly, the relative bioavailability in tissues was

calculated. In colon tissues, there was no significant difference of the dose-normalized AUC0→∞

among different groups (Figure 3B). Nevertheless, in small intestine and liver, the

dose-normalized AUC0→∞ of turmeric extract group was the highest (Figures 3C and 3D),

suggesting the efficiency of turmeric extract for delivery of curcumin to the intestinal tissues. The

relative bioavailability in small intestine for the turmeric extract group was almost 3-fold higher than

the curcumin alone group.

These results revealed that the relative bioavailability of curcumin in plasma and intestinal tissues

were higher in turmeric ethanolic extract-fed group than those in curcumin alone or curcumin plus

turmerones-fed groups. Hence, the turmeric ethanolic extract was also chosen for the subsequent

anti-tumor studies using tumor-bearing mouse model.

3.4 Combined curcumin/ turmeric extract with bevacizumab treatments inhibit tumor

growth

15
G.G.L. Yue et al.

The adjuvant effect of curcumin and turmeric extract to the chemotherapy (bevacizumab) was

explored in the HT29 tumor-bearing mice. The tumor volumes were recorded during the

experiments. Tumor volume was smaller, when mice treated with curcumin plus bevacizumab,

turmeric extract plus bevacizumab, as well as FOLFOX plus bevacizumab, than that of the

untreated control group (Figure 4A). The combination of turmeric extract and bevacizumab was

effective and achieved a significant regression of the tumor from day 18 and till the end of

experiment (Figure 4A and 4B). Its efficacy was almost comparable to FOLFOX plus

bevacizumab treatment as there was no significant difference of final tumor weights between these

two groups (Figure 4C). Furthermore, treatments of curcumin alone, turmeric extract alone or

curcumin plus bevaiczumab reduced tumor weights without significant difference when compared

with untreated group (p > 0.05). The bevacizumab alone (0.4 mg/kg) treatment showed inhibitory

activities on tumor volume and weight; however, the difference compared with untreated control

group was not statistically significant.

3.5 The herbal treatments do not induce toxicity in tumor-bearing mice

Among all herbal treatment groups, no significant body weight loss was found in curcumin or

turmeric extract plus or minus bevacizumab-treated groups or bevacizumab alone group (Figure

5A). However, the body weight of FOLFOX plus bevacizumab-treated group was significantly

decreased after 23 days of treatment. Furthermore, to evaluate the treatment-induced systemic

toxicity, several tissue damage-related enzyme activities in plasma, including AST, ALT, CK and

troponin were determined. As shown in Figure 5B, no significant difference was shown in all

herbal treatment groups on plasma activities of liver-related (AST, ALT), muscle-related (CK), and

heart-related (troponin) enzymes (p > 0.05). Nevertheless, the AST and ALT activities in FOLFOX

plus bevacizumab, and bevacizumab alone treatment groups were increased.

On the other hand, the histological analysis of hearts and livers of mice were performed to

examine any pathological change induced by the treatments. As shown in Figure 5C, there were

16
G.G.L. Yue et al.

no significant histological changes in the H & E-stained sections of hearts and livers in all

treatment groups. Terminal deoxytransferase-catalyzed DNA nick-end labeling (TUNEL) assay

was also performed in heart and liver sections and no apoptotic cells could be found. Together with

the plasma enzyme activity data, all the tested herbal treatments did not cause damage to hearts

and livers.

3.6 Combined turmeric extract with bevacizumab treatment do not cause side effects on

haematological aspect

Experiments of complete blood count were performed to evaluate the side effects of the treatments.

As shown in Figure 6A-6D, significant decreases in RBC and platelet count, haemogloblin

concentration and haematocrit were observed in FOLFOX plus bevacizumab treatment group as

compared to control (p < 0.01). Nonetheless, the curcumin and turmeric extract plus or minus

bevacizumab treatments did not affect these parameters. In addition, the count of WBC,

neutrophils, lymphocytes, monocytes and eosinophils were found to be lower in FOLFOX plus

bevacizumab treatment group than those in untreated control group (Figure 6E-6I). In contrast,

the WBC, neutrophils and lymphocytes counts were increased in mice treated with turmeric extract,

curcumin plus bevacizumab and turmeric extract plus bevacizumab. The counts of these crucial

immune cells in turmeric extract-treated mice were significantly different from those in FOLFOX

plus bevacizumab-treated mice, suggesting that the herbal treatments plus bevacizumab may not

induce severe side effects as conventional chemotherapeutics do.

3.7 Combined turmeric extract with bevacizumab treatment inhibit blood vessel growth

and induce apoptosis in tumors

To determine whether the reduced growth of tumors after treatments was associated with the

synergistic or additive anti-angiogenesis to bevacizumab, the formation of neovasculature in tumor

xenografts was examined. Figure 7A showed that the changes in number of cells stained with

17
G.G.L. Yue et al.

antibody against endothelial cell markers CD31 or Factor VIII. Most of the treatments significantly

reduced the numbers of CD31 positive-stained cells (p < 0.05), except curcumin alone group.

Combination of curcumin and bevacizumab could reduce the CD31 positive-stained cells (Figure

7B). Another critical angiogenic factor, Factor VIII, was also determined by immunohistochemical

staining. Turmeric extract with or without bevacizumab treatments could sharply reduce the Factor

VIII positive-stained cell numbers, whereas FOLFOX plus bevacizumab or bevacizumab alone

treatments did not show potent inhibitory effects (Figure 7C).

On the other hand, to further characterize the anti-tumor effects of curcumin and turmeric extract

treatments, the tumor sections for apoptosis were analyzed. The apoptotic area in all treatment

groups was increased while the increases were significant in turmeric extract plus bevacizumab

and FOLFOX plus bevacizumab-treated groups (p < 0.05, Figure 7D). The induction of apoptosis

in tumors by single treatments (curcumin, turmeric extract or bevacizumab) was not as potent as

the combined treatments mentioned above.

3.8 Combined turmeric extract with bevacizumab treatment interfere the growth of tumor

and prolong survival of tumor-bearing mice

34
Although the in vivo anti-tumor effects of curcumin have been reported, only few studies

demonstrated the inhibitory effects of curcumin in colon cancer xenografts. Hence, in the present

study, the mRNA expressions of COX-2, AKT1, cyclin D1 and survivin have been determined in

the tumor xenografts of turmeric extract-treated mice. Results showed that the combined turmeric

extract plus bevacizumab (T400 + B) as well as FOLFOX + B treatments reduced the mRNA

expressions of AKT1, cyclin D1 and survivin (Figure 8B-8D), which may account for the inhibitory

effects on the tumor growth. The down-regulation of COX2 induced by T100 + B treatment was

observed whereas FOLFOX + B treatment did not cause the alteration of COX2 expression

(Figure 8A).

Since the tumor growth was suppressed in curcumin plus bevacizumab, turmeric extract plus

18
G.G.L. Yue et al.

bevacizumab, FOLFOX plus bevacizumab treatment groups, the survival rate of mice in these

groups were further studied. The death number of mice was recorded in either natural death from

tumor-bearing or euthanasia due to the large tumor (diameter exceeded 15 mm). As shown in

Figure 8E, the mice-treated with turmeric extract plus bevacizumab lived for the longest period as

there were still 2 mice alive on day 60. The survival curve of turmeric extract plus bevacizumab

treatment group was significantly different from that of untreated control group (p < 0.05). Median

survival days of mice in each treatment groups were listed in Table 5. Results showed that both

treatments of turmeric extract plus bevacizumab and FOLFOX plus bevacizumab could prolong

the median survival time (or increased the life span) of the tumor-bearing mice. When the slopes of

the survival curves were compared between turmeric extract plus bevacizumab and untreated

control groups, the rate of deaths in turmeric extract plus bevacizumab treatment group was at

almost half of the rate in the control group.

4. Discussion

Turmeric is the dried rhizome of the plant Curcuma longa Linn., which is commonly used as

medicinal herb in Chinese and Indian Ayurveda medicines. The medicinal values of turmeric have

been demonstrated in many preclinical studies.34 Previous epidemiological studies also suggested

that turmeric may contribute to the lower incidence of large-bowel cancers in Indians.3,4 Curcumin,

the active ingredient of turmeric, is frequently described as chemopreventive agent for colorectal

cancer.5,6 Colorectal cancer is the second and third common cancer worldwide in male and female,

respectively. The use of health supplements including herbal medicines, for prevention and

adjuvant treatment of colorectal cancer by patients draw attention to the clinicians and pharmacists.

In Asian countries, it is very common in cancer patients to take health supplements during or after

conventional therapies. Therefore, the present pre-clinical study aimed at evaluating the combined

efficacy of a reputed anti-colorectal cancer herb, turmeric, with the chemotherapy (bevacizumab).

Findings from this study would provide scientific evidence on the use of turmeric as an adjuvant for

colorectal cancer.
19
G.G.L. Yue et al.

Our previous studies on turmeric demonstrated that curcuminoids and turmerones inhibited

the growth of several human cancer cell lines,35 whereas the polysaccharides isolated from

turmeric exerted immunomodulatory effect on human lymphocytes.36 Due to the poor

bioavailability of curcumin, we have made attempts to increase the absorption of curcumin in

colonic Caco-2 cell monolayer model. Our findings showed that the presence of turmerones could

increase the accumulation of curcumin inside colonic cells.12 Hence, we hypothesized that more

curcumin could be absorbed into the colonic cells with turmerones so that curcumin could take

advantage of its anti-tumor and anti-inflammatory properties towards the colorectal malignant cells.

Our recent study demonstrated that turmeric ethanolic extract possessed stronger inhibitory

effects on the growth of colonic cancer cells, the tube formation ability of endothelial cells as well

as the growth of colon tumors in mice, when compared with the same amount of curcumin applied

alone.13 The enhanced anti-tumor effects of turmeric extract would be the results of combination of

several active components, such as curcumin, /-turmerone and Ar-turmerone. Besides, our

study also showed that Ar-turmerone possessed potent suppressive effect on angiogenesis in vitro

and in vivo.23 Based on our previous findings, we suspected that turmeric extract would exert

stronger anti-tumor activities than curcumin alone since the absorption of curcumin is enhanced in

the presence of turmerones. Hence, the in vivo pharmacokinetic analyses in nude mice fed with

different curcumin preparations (administered alone or with turmerones, or in turmeric extract)

have been performed in the present study.

The tested dose of 75 mg/kg in pharmacokinetic studies was determined from our pilot

studies, in which 750 mg/kg (a mouse daily dose calculated from human efficacious dose of 60

mg/kg curcumin in colon cancer patients)8 have been fed to mice. The feeding suspension of

curcumin alone preparation at 750 mg/kg and turmeric extract at 4010 mg/kg was further diluted by

10-times to 75 mg/kg and 401 mg/kg, respectively, since the curcumin powder and the extract at

the former concentrations were not fully suspended in 1% methylcellulose solution. On the other

hand, due to the poor solubility and extensive metabolism of curcumin, hydrolysis of plasma and

20
G.G.L. Yue et al.

tissue samples was performed in order to detect both free curcumin and its conjugates.26,27 The

UPLC data of plasma samples after hydrolysis with -glucuronidase and sulfatase suggested that

almost 99% of curcumin in plasma was conjugated with glucuronide and sulphate (data not

shown). The plasma levels of curcumin in mice or rats after oral administration have been reported

in a few studies with diverse values.7 Our data showed that the Cmax of plasma curcumin was 0.35

g/mL after curcumin administration at 75 mg/kg to nude mice, while Pan et al. reported Cmax of

0.22 g/mL after curcumin administration at 1 g/kg to Balb/c mice.37 Another study reported that

after curcumin administration at 197 mg/kg to CD-1 mice, the Cmax was 1.6 g/mL.38

Furthermore, our study showed for the first time the superior bioavailability of curcumin

provided by the crude turmeric ethanolic extract (without conjugation and structural modifications

of curcumin).6,38 The plasma level of curcumin was 8-times higher in turmeric extract-fed mice than

that in curcumin-fed mice (Cmax: 2.90 vs 0.35 g/mL). As the free soluble curcumin concentration

was lower in turmeric extract than that in curcumin alone preparation, the dose-normalized

AUC0→∞ has been calculated to reflect the actual bioavailable distribution of curcumin in plasma

and organs. Our data showed that the dose-normalized AUC0→∞ of plasma was 22-fold higher in

turmeric-fed group than that in curcumin-fed group. In a previous study, rats were fed with

turmeric- or curcumin-containing diet for 3 weeks; the serum level of turmeric diet group was 2-fold

higher than that of curcumin diet group.39 Nevertheless, our results of intestine tissues were in-line

with the findings of this group. They demonstrated that the distribution of curcumin in the rat

intestine was greater in the turmeric-containing diet-fed animals, with nearly 10-fold increase in

mean curcumin levels.39 In our study, after single oral administration, a 3-fold increase in the

dose-normalized AUC0→∞ of small intestine was observed in turmeric ethanolic extract-fed group.

In regard to our hypothesis, /- and Ar-turmerone may play a role in the absorption of curcumin in

intestinal tissues since the Cmax values of curcumin plus /-turmerone-fed group and curcumin

plus Ar-turmerone-fed group were higher than that of curcumin alone. However, other components

in turmeric extract may also contribute to the absorption of curcumin as the plasma curcumin

21
G.G.L. Yue et al.

concentration was the highest in turmeric extract-fed group. Summarizing the above, the most

absorptive among the tested curcumin preparations was curcumin present in turmeric ethanolic

extract, which has been subsequently evaluated for its anti-tumor effects.

The superior anti-tumor effects of turmeric extract than curcumin alone in colon tumor-bearing

mouse model have been shown in our recent study,13 while the combined used of bevacizumab

with turmeric extract was demonstrated hereby for the more potent anti-tumor efficacies in colon

tumor-bearing mice. Despite bevacizumab (at 0.4-2.5 mg/kg) was previously shown to have
29
anti-tumor effect, the anti-tumor effect of bevacizumab (at 0.4 mg/kg) used this study was not

apparent, suggesting the additive/synergistic effects of the combined treatments. Furthermore, the

combined uses of bevacizumab with curcumin (at the same dose as it was present in the turmeric

ethanolic extract) and bevacizumab with FOLFOX were also compared in this model. FOLFOX,

which consist of 5-fluorouracil, leucovorin and oxaliplatin, was served as positive control treatment.

For the FOLFOX plus bevacizumab regimen, more than 75% growth of HT29 colonic tumor was

inhibited, while comparable findings were observed in gastric tumor in other study.30 The inhibition

in tumor growth in two herbal treatments (curcumin and turmeric extract) with bevacizumab was

greater than that without bevacizumab. The inhibitory effect of turmeric extract plus bevacizumab

was comparable to that of FOLFOX plus bevacizumab, while that of curcumin plus bevacizumab

was weaker than the two treatments mentioned above. The increased bioavailability of curcumin in

turmeric extract as well as additional anti-tumor and anti-angiogenic components present in

turmeric extract could account for the differences of efficacies. On the other hand, oral treatments

with curcumin or turmeric ethanolic extract (with or without bevacizumab) did not cause change in

body weights. However, the mice in FOLFOX-treated group started losing weights at around 14

days after treatment. Besides, the observable well-being of mice was better in turmeric

extract-treated group than those of the FOLFOX-treated group (data not shown).

In another aspect, the overall survival of mice treated with turmeric extract plus bevacizumab

was higher than those treated with FOLFOX plus bevacizumab, although the median survival was

22
G.G.L. Yue et al.

the same in these two groups. The results suggested the promising adjuvant role of turmeric

extract in the overall anti-tumor effect in tumor-bearing mice. Regarding the adjuvant effects (e.g.

increase survival rate, reduce side effects) of curcumin/ turmeric extract on chemotherapy in

tumor-bearing mice model, all the tested treatments did not cause damage to hearts and livers, as

reflected by plasma enzymes activities and histological analysis of the organs. Other common side

effects of FOLFOX are neutropenia and neurotoxicity.14 Our data also showed that the counts of

neutrophil, eosinophils, RBC and platelet were decreased in FOLFOX-treated mice. However,

such decreases were not found in turmeric extract- or curcumin-treated mice. Taken together, the

treatments of turmeric ethanolic extract with bevacizumab in colonic tumor-bearing mice was

effective and safe. These findings strongly suggested the beneficial interaction between curcumin/

turmeric and bevacizumab in treating colon cancer, which has never been reported. Most recent

study demonstrated that the combination of curcumin and bevacizumab synergistically inhibited

the VEGF signalling pathways in hepatocellular carcinoma,40 which further supports our findings

that turmeric extract containing curcumin (which act synergistically with bevacizumab) and

Ar-turmerone (which itself has potent anti-angiogenic effects) work well with bevacizumab in colon

cancer xenograft model.

The antitumor effects of turmeric extract, curcumin and FOLFOX treatments were further

elucidated by immunohistochemical analysis of tumors. The apoptotic area in the tumors of

FOLFOX-treated groups was the largest among all treatment groups as the chemotherapeutics

are known to be cytotoxic to tumor cells. For the turmeric extract- and curcumin-treated groups,

the apoptotic area was larger than that in control group, with turmeric extract plus bevacizumab

treatment the most potent. It is possible that other components (other than curcumin), such as

furanodiene41 and -sesquiphellandrene,42 present in turmeric extract were responsible for the

pro-apoptotic effect. Furthermore, the anti-angiogenic effects of turmeric extract were further

confirmed since the vascular endothelium marker CD31 and angiogenic marker factor VIII

expressed cells were significantly decreased in tumors of turmeric extract-treated groups. In fact,

23
G.G.L. Yue et al.

when the tumor-bearing mice treated with bevacizumab alone (at 0.4 mg/kg), the CD31, but not

Factor VIII, positive-stained cell numbers were decreased (data not shown). In the meantime,

when turmeric extract was combined with bevacizumab, there was further decrease in Factor VIII

expressed cells, suggesting that turmeric extract and bevacizumab may have synergistic effects

on endothelial cells for regulating angiogenesis.

To further extend the application of the above-mentioned beneficial herb-drug interaction,

turmeric extract could be supplemented to FOLFOX plus bevacizumab treatment in colon cancer.

A recent phase I clinical study using a combination of curcumin with FOLFOX demonstrated that

the combination enhanced anti-proliferative effects in patient-derived explants and also suggested

that curcumin was well-tolerated chemotherapy adjunct.43 Nevertheless, as our present study

demonstrated the potent anti-tumor effects of turmeric extract plus bevacizumab treatment. For the

future clinical trial, the doses of chemotherapeutics could be reduced, which would result in

diminished side effects.

In conclusion, the present study demonstrated for the first time that an increase in

bioavailable curcumin in nude mice could be achieved by oral administration of turmeric ethanolic

extract instead of curcumin alone. The anti-tumor efficacy of the combination of turmeric extract

with bevacizumab was comparable to that of conventional chemotherapeutics in HT29 colonic

tumor. Our current findings warrant the confirmation regarding the benefits arising from the

combined use of bevacizumab and turmeric in colorectal cancer patients in the near future.

Authors’ Contributions

G. Y. participated in the design of the study, carried out the animal studies, performed the statistical

analysis and drafted the manuscript. H. K., J. L., L. J. carried out the animal studies and performed

the statistical analysis. E. W. and K. C. carried out the phytochemical analyses. S. G., H. W. and L.

L. carried out the histological analyses. P. C. and K. F. participated in the design of the study. Z. Z.

participated in the design of pharmacokinetic study. C. L. conceived of the study, participated in its

design and coordination and helped to draft the manuscript. All authors read and approved the
24
G.G.L. Yue et al.

final manuscript.

Acknowledgements

This work was financially supported by Food and Health Bureau, Hong Kong Special

Administrative Region, Health and Medical Research Fund, No. 09100341. The authors would like

to thank Mr. Ching-Po Lau, Ms. Ling Cheng and Mr. Kwok-Yung Wong for their technical support.

Conflict of interest

The authors declare that they have no conflict of interest.

25
G.G.L. Yue et al.

References

1. GLOBOCAN, Cancer Fact Sheets. [Online] Available from

http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx. [Accessed: 1 Nov 2015 2015].
2. Aggarwal, B. B., Shishodia, S., Molecular targets of dietary agents for prevention and therapy
of cancer. Biochem Pharmacol 2006, 71, 1397-1421.
3. Mohandas, K. M., Desai, D. C., Epidemiology of digestive tract cancers in India. V. Large and
small bowel. Indian J Gastroenterol 1999, 18, 118-121.
4. Chauhan, D. P., Chemotherapeutic potential of curcumin for colorectal cancer. Curr Pharm
Des 2002, 8, 1695-1706.
5. Goel, A., Kunnumakkara, A. B., Aggarwal, B. B., Curcumin as "Curecumin": from kitchen to
clinic. Biochem Pharmacol 2008, 75, 787-809.
6. Prasad, S., Tyagi, A. K., Aggarwal, B. B., Recent developments in delivery, bioavailability,
absorption and metabolism of curcumin: the golden pigment from golden spice. Cancer Res
Treat 2014, 46, 2-18.
7. Anand, P., Kunnumakkara, A. B., Newman, R. A., Aggarwal, B. B., Bioavailability of curcumin:
problems and promises. Mol Pharm 2007, 4, 807-818.
8. Garcea, G., Berry, D. P., Jones, D. J., Singh, R., et al., Consumption of the putative
chemopreventive agent curcumin by cancer patients: assessment of curcumin levels in the
colorectum and their pharmacodynamic consequences. Cancer Epidemiol Biomarkers Prev
2005, 14, 120-125.
9. Gupta, S. C., Patchva, S., Aggarwal, B. B., Therapeutic roles of curcumin: lessons learned
from clinical trials. AAPS J 2013, 15, 195-218.
10. Antony, B., Merina, B., Iyer, V. S., Judy, N., et al., A Pilot Cross-Over Study to Evaluate
Human Oral Bioavailability of BCM-95CG (Biocurcumax), A Novel Bioenhanced Preparation
of Curcumin. Indian J Pharm Sci 2008, 70, 445-449.
11. Aggarwal, B. B., Yuan, W., Li, S., Gupta, S. C., Curcumin-free turmeric exhibits
anti-inflammatory and anticancer activities: Identification of novel components of turmeric.
Mol Nutr Food Res 2013, 57, 1529-1542.
12. Yue, G. G. L., Cheng, S. W., Yu, H., Xu, Z. S., et al., The role of turmerones on curcumin
transportation and P-glycoprotein activities in intestinal Caco-2 cells. J Med Food 2012, 15,
242-252.
13. Yue, G. G. L., Jiang, L., Kwok, H. F., Lee, J. K. M., et al., Turmeric ethanolic extract
possesses stronger inhibitory activities on colon tumor growth than curcumin — the

26
G.G.L. Yue et al.

importance of turmerones. J Funct Foods 2016, 22, 565-577.

14. Yu, Y., Kanwar, S. S., Patel, B. B., Nautiyal, J., et al., Elimination of Colon Cancer Stem-Like
Cells by the Combination of Curcumin and FOLFOX. Transl Oncol 2009, 2, 321-328.
15. Hess, G. P., Wang, P. F., Quach, D., Barber, B., Zhao, Z., Systemic Therapy for Metastatic
Colorectal Cancer: Patterns of Chemotherapy and Biologic Therapy Use in US Medical
Oncology Practice. J Oncol Pract 2010, 6, 301-307.
16. Zafar, S. Y., Marcello, J. E., Wheeler, J. L., Rowe, K. L., et al., Longitudinal patterns of
chemotherapy use in metastatic colorectal cancer. J Oncol Pract 2009, 5, 228-233.
17. Alishekevitz, D., Bril, R., Loven, D., Miller, V., et al., Differential therapeutic effects of
anti-VEGF-A antibody in different tumor models: implications for choosing appropriate tumor
models for drug testing. Mol Cancer Ther 2014, 13, 202-213.
18. Patel, B. B., Gupta, D., Elliott, A. A., Sengupta, V., et al., Curcumin targets FOLFOX-surviving
colon cancer cells via inhibition of EGFRs and IGF-1R. Anticancer Res 2010, 30, 319-325.
19. Li, L., Ahmed, B., Mehta, K., Kurzrock, R., Liposomal curcumin with and without oxaliplatin:
effects on cell growth, apoptosis, and angiogenesis in colorectal cancer. Mol Cancer Ther
2007, 6, 1276-1282.
20. Kunnumakkara, A. B., Diagaradjane, P., Anand, P., Harikumar, K. B., et al., Curcumin
sensitizes human colorectal cancer to capecitabine by modulation of cyclin D1, COX-2,
MMP-9, VEGF and CXCR4 expression in an orthotopic mouse model. Int J Cancer 2009, 125,
2187-2197.
21. Kantara, C., O'Connell, M., Sarkar, S., Moya, S., et al., Curcumin promotes autophagic
survival of a subset of colon cancer stem cells, which are ablated by DCLK1-siRNA. Cancer
Res 2014, 74, 2487-2498.
22. Ravindran, J., Prasad, S., Aggarwal, B. B., Curcumin and cancer cells: how many ways can
curry kill tumor cells selectively? AAPS J 2009, 11, 495-510.
23. Yue, G. G. L., Kwok, H. F., Lee, J. K. M., Jiang, L., et al., Novel anti-angiogenic effects of
aromatic-turmerone, essential oil isolated from spice turmeric. J Funct Foods 2015, 15,
243-253.
24. Commission, C. P., Chinese Pharmacopoeia, Commercial Press, Shanghai 2015.
25. Zhongfa, L., Chiu, M., Wang, J., Chen, W., et al., Enhancement of curcumin oral absorption
and pharmacokinetics of curcuminoids and curcumin metabolites in mice. Cancer Chemother
Pharmacol 2012, 69, 679-689.
26. Vareed, S. K., Kakarala, M., Ruffin, M. T., Crowell, J. A., et al., Pharmacokinetics of curcumin
conjugate metabolites in healthy human subjects. Cancer Epidemiol Biomarkers Prev 2008,

27
G.G.L. Yue et al.

17, 1411-1417.
27. Asai, A., Miyazawa, T., Occurrence of orally administered curcuminoid as glucuronide and
glucuronide/sulfate conjugates in rat plasma. Life Sci 2000, 67, 2785-2793.
28. Priego, S., Feddi, F., Ferrer, P., Mena, S., et al., Natural polyphenols facilitate elimination of
HT-29 colorectal cancer xenografts by chemoradiotherapy: a Bcl-2- and superoxide
dismutase 2-dependent mechanism. Mol Cancer Ther 2008, 7, 3330-3342.
29. Yanagisawa, M., Fujimoto-Ouchi, K., Yorozu, K., Yamashita, Y., Mori, K., Antitumor activity of
bevacizumab in combination with capecitabine and oxaliplatin in human colorectal cancer
xenograft models. Oncol Rep 2009, 22, 241-247.
30. Jiang, J., Zhang, Y., Chuai, S., Wang, Z., et al., Trastuzumab (herceptin) targets gastric
cancer stem cells characterized by CD90 phenotype. Oncogene 2012, 31, 671-682.
31. Yue, G. G. L., Lee, J. K., Kwok, H. F., Cheng, L., et al., Novel PI3K/AKT targeting
anti-angiogenic activities of 4-vinylphenol, a new therapeutic potential of a well-known styrene
metabolite. Sci Rep 2015, 5, 11149.
32. Subramaniam, D., May, R., Sureban, S. M., Lee, K. B., et al., Diphenyl difluoroketone: a
curcumin derivative with potent in vivo anticancer activity. Cancer Res 2008, 68, 1962-1969.
33. Wang, D., Stockard, C. R., Harkins, L., Lott, P., et al., Immunohistochemistry in the evaluation
of neovascularization in tumor xenografts. Biotech Histochem 2008, 83, 179-189.
34. Bordoloi, D., Roy, N. K., Monisha, J., Padmavathi, G., Kunnumakkara, A. B. Multi-Targeted
Agents in Cancer Cell Chemosensitization: What We Learnt from Curcumin Thus Far. Recent
Pat Anticancer Drug Discov 2016, 11, 67-97.
35. Yue, G. G. L., Chan, B. C., Hon, P. M., Lee, M. Y., et al., Evaluation of in vitro anti-proliferative
and immunomodulatory activities of compounds isolated from Curcuma longa. Food Chem
Toxicol 2010, 48, 2011-2020.
36. Yue, G. G. L., Chan, B. C., Hon, P. M., Kennelly, E. J., et al., Immunostimulatory activities of
polysaccharide extract isolated from Curcuma longa. Int J Biol Macromol 2010, 47, 342-347.
37. Pan, M. H., Huang, T. M., Lin, J. K., Biotransformation of curcumin through reduction and
glucuronidation in mice. Drug Metab Dispos 1999, 27, 486-494.
38. Yu, H., Huang, Q., Improving the oral bioavailability of curcumin using novel organogel-based
nanoemulsions. J Agric Food Chem 2012, 60, 5373-5379.
39. Martin, R. C., Aiyer, H. S., Malik, D., Li, Y., Effect on pro-inflammatory and antioxidant genes
and bioavailable distribution of whole turmeric vs curcumin: Similar root but different effects.
Food Chem Toxicol 2012, 50, 227-231.
40. Gao, J. Z., Du, J. L., Wang, Y. L., Li, J., et al., Synergistic effects of curcumin and

28
G.G.L. Yue et al.

bevacizumab on cell signaling pathways in hepatocellular carcinoma. Oncol Lett 2015, 9,

295-299.
41. Zhong, Z., Tan, W., Chen, X., Wang, Y., Furanodiene, a natural small molecule suppresses
metastatic breast cancer cell migration and invasion in vitro. Eur J Pharmacol 2014, 737,
1-10.
42. Tyagi, A. K., Prasad, S., Yuan, W., Li, S., Aggarwal, B. B., Identification of a novel compound
(beta-sesquiphellandrene) from turmeric (Curcuma longa) with anticancer potential:
comparison with curcumin. Invest New Drugs 2015, 33, 1175-1186.
43. James, M. I., Iwuji, C., Irving, G., Karmokar, A., et al., Curcumin inhibits cancer stem cell
phenotypes in ex vivo models of colorectal liver metastases, and is clinically safe and
tolerable in combination with FOLFOX chemotherapy. Cancer Lett 2015, 364, 135-141.

29
G.G.L. Yue et al.

Figure legends

Figure 1 (A) Chemical structures of -turmerone, -turmerone, aromatic-turmerone (Ar-turmerone)

and curcumin. HPLC profiles of the ethanolic extract of turmeric showing (B) curcumin and (C)

Ar-turmerone. 3-D chromatograms showing curcumin and Ar-turmerone in the mixture of two

standards (D) and in turmeric ethanolic extract (E). Extracted ion chromatograms (EIC) and MS

spectra of curcumin (F) and Ar-turmerone (G) in turmeric ethanolic extract were shown.

Figure 2 (A) Plasma concentration-time, (B) colon, (C) small intestine and (D) liver tissue

level-time profiles after single oral administration of different curcumin preparations. Data were

expressed as mean + S.E.M. (n = 3 per time point, 8 sampling point per treatment).

Figure 3 Dose-normalized AUC0 of curcumin in (A) plasma, (B) colon, (C) small intestine and (D)

liver tissue after single oral administration of different curcumin preparations. Data were expressed

as mean + S.E.M. (n = 3 per time point, 8 sampling point per treatment). Differences among

groups were determined by one-way ANOVA followed by post-hoc Tukey's multiple comparison

test. In (A) and (C), *** p < 0.005 as compared among different treatment groups. In (B) and (D), no

significant difference was observed among treatment groups.

Figure 4 Effects of bevacizumab alone, curcumin plus bevacizumab, turmeric extract plus

bevacizumab and FOLFOX plus bevacizumab treatments on the tumor growth of HT29

subcutaneous xenografts. Mice were injected subcutaneously with HT29 cells. (A) The tumor

growth in different treatment groups. Each point represented mean ± S.E.M. of 12-15 mice. Effects

of different treatments on (B) tumor volume measured on day 30 and (C) final tumor weight were

compared. Results were expressed as mean + S.E.M. of 12-15 mice each group. ** p < 0.01, *** p

< 0.005 when treatment groups compared with untreated control group. # p <0.05, ## p < 0.01, ### p

< 0.005 as compared among treatment groups by one-way ANOVA followed by post-hoc Tukey's

multiple comparison test. Treatment groups: C75 = Curcumin alone 75 mg/kg; T400 = Turmeric

ethanolic extract 400 mg/kg; B = bevacizumab 0.4 mg/kg; FOLFOX = 5-fluorouracil 15 mg/kg +

leucovorin 5 mg/kg + oxaliplatin 5 mg/kg; B alone = bevacizumab 0.4 mg/kg.

30
G.G.L. Yue et al.

Figure 5 Effects of bevacizumab alone, curcumin plus bevacizumab, turmeric extract plus

bevacizumab, and FOLFOX plus bevacizumab treatments on body weight, plasma enzyme

activities, heart and liver tissues damage of HT29 subcutaneous xenograft-bearing mice. (A) The

body weight change during treatments. Results were expressed as mean + S.E.M. of 12-15 mice

each group. * p < 0.05, ** p < 0.01 when treatment groups compared with untreated control group

by one-way ANOVA followed by post-hoc Tukey's multiple comparison test. (B) Plasma AST, ALT,

CK and troponin activities were determined at the end of treatments. No significant difference was

observed among treatment groups. (C) Representative microphotographs showed H&E-stained

sections and TUNEL-labelled sections of hearts and livers of HT29 tumor-bearing mice treated

with different curcumin, turmeric extract or FOLFOX plus or minus bevacizumab. Eight samples of

each treatment group were detected and the images were randomly taken from four visual fields

per sample; magnification 40X.

Figure 6 Effects of curcumin plus bevacizumab, turmeric extract plus bevacizumab, and FOLFOX

plus bevacizumab treatments on blood cells count and haematological parameters of mice.

Results were expressed as mean + S.E.M. of 9 mice each group. * p < 0.05, ** p < 0.01 when
# ## ###
treatment groups compared with untreated control group. p <0.05, p < 0.01, p < 0.005 as

compared among treatment groups by one-way ANOVA followed by post-hoc Tukey's multiple

comparison test. In (H), there was no significant difference in monocytes number among treatment

groups.

Figure 7 Effects of bevacizumab alone, curcumin plus bevacizumab, turmeric extract plus

bevacizumab, and FOLFOX plus bevacizumab treatments on blood vessel growth and apoptosis

in tumors. Tumor samples were collected at the end of experiment, fixed in formalin and mounted

on glass slides. Immunohistochemical staining with anti-CD31 or anti-Factor VIII was performed.

(A) Representative microphotographs showed CD31- or Factor-VIII-stained sections of tumors

from different treatment groups, of magnification 40X. (B-C) Quantification of (B) CD31 or (C)

Factor-VIII positive-stained cells in tumor sections was conducted in a double-blind manner.

31
G.G.L. Yue et al.

Results were expressed as mean + S.E.M. of 12-15 mice each group. *p < 0.05, **p < 0.01, ***p <

0.005 when treatment groups compared with untreated control group. #p < 0.05, ##
p < 0.01 as

compared among treatment groups by one-way ANOVA followed by post-hoc Tukey's multiple

comparison test. (D) Treatment-induced tumor cell death was assessed by TUNEL assay.

Quantification of apoptotic area in tumor sections according to apoptotic index (%), defined as the

percentage of apoptotic cell area over total tumor area. Results were expressed as mean + S.E.M.

of 12-15 mice each group. * p < 0.05, *** p < 0.005 as treatment groups compared with untreated
# ##
control group. p <0.05, p < 0.01, as compared among treatment groups by one-way ANOVA

followed by post-hoc Tukey's multiple comparison test.

Figure 8 Effects of curcumin plus bevacizumab, turmeric extract plus bevacizumab, and FOLFOX

plus bevacizumab treatments on the growth of tumor and survival of HT29 subcutaneous

xenograft-bearing mice. (A-D) Quantitative RT-PCR analyses of COX2, AKT1, cyclin D1 and

survivin mRNA expressions in tumor tissues of each treatment groups. Tumor tissues were

collected for RNA extraction. Data were normalized to corresponding human 18 S expressions in

control cells. mRNA expressions results were expressed as fold of control (mean fold of control +

S.E.M. from 7-8 mice each group). Differences among the treated and vehicle-treated control

groups were determined by one-way ANOVA. There was no significant difference among all

groups. (E) Survival curves of mice in each treatment group (each group consisted of 20 mice).

Data were shown in Kaplan–Meier survival curves. * p < 0.05 when treatment groups compared

with untreated control group using Mantel–Cox log-rank test.

32
G.G.L. Yue et al.

Tables

Table 1 Grouping and dosages of compounds/extract for pharmacokinetic experiments

Treatment groups Calculated dose for Dose of freely soluble
(n = 3 per time point, 8 mice curcumin a)
sampling points per (mg/kg) (g/kg)
treatment)
Curcumin 75 308.0
Curcumin + 75
241.2
/-turmerone 7.5b)
Curcumin + 75
200.1
Ar-turmerone 7.5 b)
Curcumin + 75
/-turmerone + 7.5 b) 226.9
b)
Ar-turmerone 7.5
Turmeric ethanolic extract 400 c) 58.3
a)
Freely soluble curcumin in distilled water containing 1% methylcellulose
b)
The dose of turmerones was 1/10 of the curcumin dose according to our previous study 35
c)
The dose of turmeric extract (400 mg/kg) was equivalent to the dose of curcumin (75 mg/kg),
which was calculated based on the content of curcumin detected in the ethanolic extract (i.e.
18.7%).

33
G.G.L. Yue et al.

Table 2 List of extract and drugs and their dosages in each treatment groups
Treatment Extract/ drugs Doses Administration route and time
groups (mg/kg) interval
Control Distilled water -- Oral daily
C75 Curcumin 75 a) Oral daily
T400 Turmeric ethanolic 400 a) Oral daily
extract
C75 + B Curcumin 75 a) Oral daily
Bevacizumab 0.4 Intraperitoneal twice a week for 3
weeks
T400 + B Turmeric ethanolic 400 a) Oral daily
extract
Bevacizumab 0.4 Intraperitoneal twice a week for 3
weeks
FOLFOX + B 5-fluorouracil 15 Intraperitoneal daily for 3 weeks
leucovorin 5 Intraperitoneal daily for 3 weeks
oxaliplatin 5 Intraperitoneal once a week for 3
weeks
Bevacizumab 0.4 Intraperitoneal twice a week for 3
weeks
B alone Bevacizumab 0.4 Intraperitoneal twice a week for 3
weeks
a)
Optimized doses were obtained from the pharmacokinetic study.

34
G.G.L. Yue et al.

Table 3 The plasma pharmacokinetic parameters for curcumin after single oral administration of
different curcumin preparations in nude mice (mean  S.E.M.)
Treatment groups AUC0- Tmax Cmax
(n = 3 per time point for 8 sampling ((hg)/mL) (h) (g/mL)
points)
Curcumin 0.78  0.19 1.00  0.00 0.35  0.04
Curcumin + /-turmerone 0.43  0.07 0.83  0.17 0.30  0.05
Curcumin + Ar-turmerone 0.49  0.11 1.00  0.00 0.29  0.02
Curcumin + /-turmerone + 0.39  0.10 1.00  0.00 0.32  0.04
Ar-turmerone
Turmeric ethanolic extract 3.31  0.34 0.50  0.00 2.90  0.49

35
G.G.L. Yue et al.

Table 4 The colon, small intestine and liver tissue pharmacokinetic parameters for curcumin after
single oral administration in different curcumin preparations in nude mice (mean 
S.E.M.)
Treatment groups AUC0 Tmax Cmax
(n = 3 per time point for 8 sampling ((hg)/mL) (h) (g/g
points) tissue)
Curcumin 3.69  1.34 1.33  0.33 1.69  0.68
Curcumin + /-turmerone 2.98  1.16 1.00  0.00 2.70  0.97
Curcumin + Ar-turmerone 3.92  0.86 1.00  0.00 2.62  0.34
Colon
Curcumin + /-turmerone +
1.79  1.03 1.67  0.33 0.54  0.45
Ar-turmerone
Turmeric extract 0.73  0.11 1.67  1.17 0.49  0.07
Curcumin 2.84  0.62 0.50  0.25 2.77  0.26
Curcumin + /-turmerone 1.56  0.36 0.67  0.67 1.55  0.37
Small Curcumin + Ar-turmerone 1.52  0.32 0.83  0.67 1.85  0.63
intestine Curcumin + /-turmerone +
0.33  0.06 0.58  0.22 0.74  0.19
Ar-turmerone
Turmeric extract 1.51  0.45 0.50  0.00 2.70  0.55
Curcumin 0.12  0.05 0.42  0.083 0.24  0.09
Curcumin + /-turmerone 0.03  0.02 0.75  0.63 0.04  0.02
Curcumin + Ar-turmerone 0.03  0.03 0.67  0.67 0.02  0.01
Liver
Curcumin + /-turmerone +
0.00  0.00 0.00  0.00 0.00  0.00
Ar-turmerone
Turmeric extract 1.04  0.75 0.67  1.67 0.72  0.15

36
G.G.L. Yue et al.

Table 5 Summary of survival of different treatment groups

Treatment groups Median survival % Increase in life Survivors/ group
time (days) span a)
Control 26 - 0/20
C75 + B 29.5 13.5 0/20
T400 + B 33 26.9 2/20
FOLFOX + B 33 26.9 0/20
a)
% Increase in life span, calculated as [(median survival time in treated mice/median survival time
in control group)-1]  100%

37