Saffron: A potential candidate for a novel anticancer drug against hepatocellular carcinoma


  • Potential conflict of interest: Nothing to report.


Saffron has been proposed as a promising candidate for cancer chemoprevention. The purpose of this investigation was to investigate the chemopreventive action and the possible mechanisms of saffron against diethylnitrosamine (DEN)-induced liver cancer in rats. Administration of saffron at doses of 75, 150, and 300 mg/kg/day was started 2 weeks prior to the DEN injection and was continued for 22 weeks. Saffron significantly reduced the DEN-induced increase in the number and the incidence of hepatic dyschromatic nodules. Saffron also decreased the number and the area of placental glutathione S-transferase–positive foci in livers of DEN-treated rats. Furthermore, saffron counteracted DEN-induced oxidative stress in rats as assessed by restoration of superoxide dismutase, catalase, and glutathione-S-transferase levels and diminishing of myeloperoxidase activity, malondialdehyde and protein carbonyl formation in liver. The results of immunohistochemical staining of rat liver showed that saffron inhibited the DEN-mediated elevations in numbers of cells positive for Ki-67, cyclooxygenase 2, inducible nitric oxide synthase, nuclear factor-kappa B p-65, and phosphorylated tumor necrosis factor receptor. Saffron also blocked the depletion in the number of cells positive for TUNEL (terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling) and M30 CytoDeath in liver tissues of DEN-treated rats. In vitro experiments carried out using HepG2 cells also confirmed these findings and showed inhibition of nuclear factor-kappa B activation, increased cleavage of caspase-3, as well as DNA damage and cell cycle arrest upon saffron treatment. Conclusion: This study provides evidence that saffron exerts a significant chemopreventive effect against liver cancer through inhibition of cell proliferation and induction of apoptosis. This report also shows some evidence that saffron protects rat liver from cancer via modulating oxidative damage and suppressing inflammatory response. (HEPATOLOGY 2011;)

Hepatocellular carcinoma (HCC) is the fifth most common cancer and the third leading cause of cancer mortality in the world. Chronic infection with hepatitis B and C are the major risk factors for HCC worldwide. Other factors that contribute to the formation of HCC include exposure to environmental carcinogens, iron overload, fatty liver disease, and alcohol abuse.1 Diethylnitrosamine (DEN) is one of the most important environmental carcinogens and is present in tobacco smoke, cosmetics, gasoline, and various processed foods such as milk and meat products.1 DEN is also commonly used to induce lesions in rats that mimic different types of benign and malignant tumors in humans.2

Given the limited treatments available, preventive control approaches have been considered among the best strategies to protect against cancer. Cancer chemoprotection is based on the use of exogenous phytochemicals to enhance endogenous mechanisms against various stages of cancer development.3 Lately, there has been a lot of interest in exploring the chemopreventive properties of natural herbs and plants. Saffron is a naturally derived plant product from the dried stigma of the Crocus sativus flower (family Iridaceae) that may have biologically useful properties. In fact, saffron extract and its biologically active compounds, including crocin, crocetin, carotene, and safranal, have been shown both in vitro and in vivo to possess antioxidant, anticancer, anti-inflammatory, and memory-improving properties.4-6 Saffron is also used in folk medicine as an antispasmodic, antidepressant, and aphrodisiac.4 Furthermore, it is one of the most commonly used species around the world for flavoring and coloring foods.4

Saffron has recently gained considerable interest for its capacity to interfere with cancer at initiation and promotion stages as well as for cancer treatment.6 Although saffron and its constituents have been shown to have antitumorigenic and proapoptotic activities in different cancer cell lines,4 to date, the exact mechanism of anticancer effect of saffron is not clear.

In order to understand the mechanisms of chemopreventive action of saffron, this study used a well-described model of HCC to study the mechanism of the anticancer action of saffron by evaluating its antioxidant, proapoptotic, antiproliferative, and anti-inflammatory effects.


2-AAF, 2-acetyl aminofluorene; ABTS, 2,2-azino-bis(3-ethylbenzothiazoline-6-sulfonate); ALT, alanine aminotransferase; ANOVA, analysis of variance; AST, aspartate aminotransferase; CAT, catalase; COX-2, cyclooxygenase 2; DEN, diethylnitrosamine; DPPH, 1,1-diphenyl-2-picrylhydrazyl; FAH, foci of altered hepatocytes; FRAP, ferric reducing antioxidant power; GGT, gamma glutamyl transpeptidase; GSH, glutathione; GST, glutathione S-transferase; HCC, hepatocellular carcinoma; IL-8, interleukin-8; iNOS, inducible nitric oxide synthase; MDA, malondialdehyde; MPO, myeloperoxidase; NF-κB-p65, nuclear factor-kappa B p-65; P.Carbonyl, protein carbonyl; PI, propidium iodide; p-TNFR1, phosphorylated tumor necrosis factor receptor 1; ROS, reactive oxygen species; SOD, superoxide dismutase; TUNEL, terminal deoxynucleotidyl transferase–mediated dUTP nick-ending labeling.

Materials and Methods

Plant Material and Preparation.

Stigmas of pure saffron were obtained from Golpeech Saffron (register 82558; Mashhad, Iran), and a voucher sample was preserved for reference in the herbarium of United Arab Emirates University. The stigma materials (500 g) were extracted with water or 80% (vol/vol) aqueous ethanol and the mixture was macerated for 5 days at 4°C. The resulting mixture was then filtered and dried under reduced pressure in a rotary evaporator at 40°C to give water and ethanol crude extracts. Photochemical analysis of crocin and safranal derived from saffron was determined by high-performance liquid chromatography. The final powder form of saffron extract contained 73.25 mg/g crocin and 33.21 μg/g safranal. The ethanol extract was then administered to animals at a dose of 75, 150, and 300 mg/kg body weight in a volume of 5 mL/kg body weight. The highest dose of saffron used (300 mg/kg) in this study contained 22 mg crocin/kg body weight and 9.96 μg safranal/kg body weight.

Hepatocarcinogenesis Model.

The experimental hepatocarcinogenesis was initiated by DEN and promoted by 2-AAF, according to the protocol described by Espandiari et al.7 with some modifications. In this model, 96-hours fasting rats were refed as mitotic proliferative stimuli. Then, 24 hours after refeeding, rats were injected intraperitoneally with a single dose of DEN (200 mg/kg body weight) dissolved in saline. Two weeks after DEN treatment, rats received six daily intragastric doses of 2-AAF (30 mg/kg in 1% Tween 80) for promoting hepatocarcinogenesis. All animals received humane care according to the guidelines of the Animal Research Ethics Committee of the UAE University.

Treatment Regimen.

The experimental design is depicted in Fig. 1. Rats were randomly divided into six groups (n = 8) and were subjected to the following treatments: Group 1 (Control): Rats were treated with distilled water (5 mL/kg body weight) throughout the experimental period and injected with single dose of saline. Group 2 (Saffron): Rats were administered orally 300 mg/kg saffron alone throughout the experimental period. Group 3 (HCC): Rats were induced with DEN and 2-AAF as described before. Whereas animals of the protective groups (groups 4-6) were orally fed 300 mg/kg (+Saffron HD [high dose]), 150 mg/kg (+Saffron MD [medium dose]), and 75 mg/kg (+Saffron LD [low dose]), saffron suspension, respectively, 2 weeks prior to HCC initiation and continued for 22 weeks. Doses of saffron were selected based on previously reported pharmacological studies of this plant. Saffron at doses varying from 100-200 mg/kg has been reported to suppress chemically-induced skin cancer.4, 6 No adverse effect has been reported for saffron treatment up to 2000 mg/kg.8 Hepatocarcinogenesis was then induced as detailed in the previous section. Group 1 was treated with equal volume of vehicle. After 22 weeks of DEN administration, all animals were anesthetized 24 hours after the last treatment. Following anesthesia, blood was collected from the retro-orbital plexus.

Figure 1.

Experimental design. b.wt., body weight; i.p., intraperitoneal; p.o., per os.

Sample Preparations.

Collected blood samples were centrifuged at 1500×g for 20 minutes at 4°C to obtain serum. For biochemical determination, frozen liver samples were homogenized in ice-cold Tris-HCL buffer (150 mM, pH 7.4). The wt/vol ratio of the tissue to the homogenization buffer was (1:10 wt/vol). Aliquots were prepared and used for determination of different biochemical markers.

Serum Markers of Liver Damage and Cancer.

Activities of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma glutamyl transpeptidase (GGT) were measured spectrophotometrically using Shimadzu recording spectrophotometer (UV-160). The assay was performed using BioMerieux reagent kit, according to the protocol supplied with the kit.

Alpha-fetoprotein (AFP) is produced by the fetal liver and is used as marker for hepatic carcinoma.9 AFP was determined in serum by enzyme-immunoassay method using a commercial kit (Calbiotech) and following their protocol.

In Vivo Antioxidant Status.

Determination of MDA in liver homogenate was carried out based on its reaction with thiobarbituric acid to form a pink complex with absorption maximum at 535 nm.10 Protein carbonyl (P.carbonyl) contents were determined according to the method of Reznick and Packer.11 This method is based on spectrophotometric detection of the end product of reaction of 2,4-dinitophenylhydrazine with P.carbonyl to form protein hydrazones at 370 nm. The results were expressed as nanomoles of carbonyl group per milligram of protein with molar extinction coefficient of 22,000 M/cm. Myeloperoxidase (MPO) activity in homogenate was determined as described in Hillegass et al.12 One unit of MPO was defined as the amount of MPO present that degrades 1 μM peroxide per minute. Catalase (CAT) activity was determined by measuring the exponential disappearance of H2O2 at 240 nm and expressed in units per milligram protein as described by Aebi.13

Superoxide dismutase (SOD) activity in liver homogenate was determined according to the method described by Nandi and Chatterjee.14 This method is based on the ability of SOD to inhibit the auto-oxidation of pyrogallol at an alkaline pH. One unit of SOD is described as the amount of enzyme required to cause 50% inhibition of pyrogallol auto-oxidation.

The glutathione (GSH) content in the liver homogenate was determined using the method of Van Dooran et al.15 The basis of the GSH determination method is the reaction of Ellman's reagent (5,5′-dithiobis-[2-nitrobenzoic acid]) with thiol groups of GSH at pH 8.0 to produce the yellow 5-thiol-2-nitrobenzoate anion. Glutathione S-transferase (GST) activity was determined according to the method of Habig et al.16 In this assay, GST catalyzes the conjugation of GSH with 1-chloro-2,4-dinitrobenzene, producing a chromophore at 340 nm.

The total protein contents of liver tissues were determined according to the Lowry method as modified by Peterson.17 Absorbances were recorded using a Shimadzu recording spectrophotometer (UV-160) in all measurements.

Cell Culture and Treatment.

Liver cancer cell line HepG2 were maintained in Roswell Park Memorial Institute-160 medium with 10% fetal bovine serum and 1% of 100 U/mL penicillin and 100 μg/mL streptomycin at 37°C inside a humidified incubator with 5% CO2 and 95% room air. Cells were subcultured every 4-7 days with trypsin/ethylenediamine tetraacetic acid (1:250; PAA Laboratory, Germany). Cells were treated with several concentrations of saffron extract for several time points.

MTT Cell Proliferation and Cytotoxicity Assays.

The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, a yellow tetrazole) proliferation assay was used in the HepG2 cell line to assess the effect on cell proliferation of a range of concentrations of saffron extract. Cells (104) were plated and grown in 200 μL of growth medium in 96-well microtiter plates. After an overnight attachment period, cells were treated with varying concentrations of saffron extract (1.0, 2.0, 4.0, and 6.0 mg/mL) prepared from a 100 mg/mL stock solution dissolved in water. All studies were performed in triplicate and repeated three times independently. Cell growth was quantified by the ability of living cells to reduce the yellow dye MTT to a purple formazan product. Cells were incubated with MTT (Sigma) at 37°C in a humidified 5% CO2 atmosphere for 2 hours. The MTT formazan product was then dissolved in dimethylsulfoxide, and absorbance was measured at 570 nm in a microplate reader.

Flow Cytometric Analysis of DNA Content.

One day before treatment, cells were seeded at a density of 1.2 × 106 cells per plate. After the indicated times, the cells were harvested by trypsin release, washed twice with phosphate-buffered saline, fixed with 70% ethanol, treated with 1% ribonuclease, and finally stained with propidium iodide (100 μg/mL final concentration). Distribution of cell cycle phases with different DNA contents was determined using a flow cytometer LSR1 (Becton-Dickinson). Sub-G1 cells in flow cytometric histograms were considered apoptotic cells. Analysis of cell cycle distribution and the percentage of cells in the G1, S, and G2/M phases of the cell cycle were determined using the software FlowJo (Tree Star, Ashland, OR).

Quantification of Apoptosis by Annexin-Propidium Iodide staining.

To detect apoptosis, the Annexin V–FLUOS kit (Roche Diagnostics) was used. Cells were treated for 6, 24, or 48 hours with saffron extract. After washing twice in phosphate-buffered saline, 1 × 106 cells were stained with 100 μL annexin V staining solution, consisting of 20 μL fluorescein isothiocyanate–conjugated annexin V reagent (20 μg/mL), 20 μL isotonic propidium iodide (PI; 50 μg/mL), and 1000 μL of 1 mol/L HEPES buffer, for 15 minutes at room temperature. Cells were analyzed on a FACSCalibur flow cytometer (Becton-Dickinson) using a 488 nm excitation and 530/30 nm band-pass filter for fluorescein detection and a long-pass filter 2P670 nm for PI detection after electronic compensation. Because positive annexin V staining indicates apoptotic and necrotic cells, PI-positive cells were used to measure late apoptotic cells and necrotic cells, whereas annexin V–positive and PI-negative cells were counted as early apoptotic cells.

Western Blotting.

Whole-cell lysates were prepared from HepG2 tumor cells. Protein concentration of lysates was determined with a Bio-Rad DC Protein Assay (Bio-Rad Laboratories, Hercules, CA), and 30 μg proteins were loaded onto 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis gels. The gels were transferred to nitrocellulose membranes before immunodetection processing with anti-phospho-H2AX (Millipore), anti-caspase-3 (Cell Signaling Technology), anti-IκB (Abcam, Cambridge, UK), anti-TNFR1 (Santa Cruz Biotechnology, Santa Cruz, CA), and with secondary antibodies (anti-mouse or anti-rabbit IgG peroxidase conjugated; Pierce, Rockford, IL). Bound antibodies were detected by incubating the blots in West Pico chemiluminescent substrate (Pierce). The level of immunoreactivity was measured as peak intensity using an image-capture and analysis system (GeneGnome; Syngene, UK). Hybridization with anti-GAPDH was used to control equal loading and protein quality.

Statistical Analysis.

SPSS (version 10) statistical program (SPSS Inc., Chicago, IL) was used to carry out a one-way analysis of variance (ANOVA) on our data. When significant differences by ANOVA were detected, analysis of differences between the means of the treated and control groups were performed by using Dunnett's t test.

Other Experimental Procedures.

Other experimental procedures are described in detail in the supporting information. These include animal housekeeping and treatment, in vitro antioxidant properties, terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-ending labeling (TUNEL) assay, immunohistochemical analyses, morphology and histopathology analysis, as well as enzyme-linked immunosorbent assay (ELISA).


In Vitro and In Vivo Antioxidant Properties of Saffron.

Because saffron has been previously reported to be rich in antioxidants, we wanted to confirm that our specific preparations also contained antioxidants. Comparison of the ethanolic and water extracts of saffron showed that the ethanolic extract had the highest phenolic content (67.62 mg gallic acid/g) and the best antioxidant activity. The ascorbic acid equivalent antioxidant capacities of the ethanolic extract were 0.253 and 0.304 mmol/g in the FRAP and ABTS assays, respectively. Correspondingly, the best free radical scavenging activity, as measured by the DPPH assay, was exerted by ethanolic extract (median inhibitory concentration [IC50] = 2.0 mg/mL) (Table 1).

Table 1. In Vitro Antioxidant Capacities of Ethanol and Water Extracts of Saffron
Saffron ExtractPercent of YieldTotal Phenolic ContentFRAP AssayDPPH Scavenging IC50ABTS+ Assay
  • Percent of yield expressed as percent of weight/g dried extract. Total phenolic content was expressed as milligrams gallic acid equivalents per gram dried extract. Total antioxidant by ABTS and FRAP assay was expressed as mmol ascorbic acid equivalents/g dried extract. IC50 is the concentration of extract that is able to scavenge DPPH radicals by 50%.

  • *

    P < 0.05.

  • ***

    P < 0.001 by t test.

Ethanol 80%58.467.62 ± 0.22***0.253 ± 0.001*2.0 ± 0.05***0.304 ± 002*
Water71.064.25 ± 0.290.237 ± 0.0033.72 ± 0.020.275 ± 0.002

The antioxidant activity of ethanolic saffron extract was also measured in vivo and is summarized in Table 2, which shows the antioxidant status of liver of control and experimental animals. Group 3 (HCC) animals exhibited significant increase (P < 0.001) in MDA, P.Carbonyl, GSH levels, as well as in MPO and GST activities (P < 0.001) and decrease in CAT (P < 0.01) and SOD (P < 0.001) activities compared to group 1 (control). These findings are consistent with hepatic oxidative stress and damage caused by DEN-2-2AAF. However, as can be seen in Table 2, pretreatment with saffron (groups 4, 5 and 6) significantly attenuated the changes in these oxidative stress markers compared to control. Both medium and high doses of saffron abolished DEN-2-2AAF-induced oxidative stress more effectively than the lower dose. Surprisingly, GSH levels remained elevated in groups pretreated with saffron and did not return to normal levels as opposed to other oxidative stress markers. It is well established that GSH synthesis is up-regulated during oxidative damage, inflammation and cell proliferation. It is possible that even in saffron-treated animals, DEN-2-2AAF causes a persistent low level of oxidative stress as well as low levels of inflammation/proliferation, which in turn causes GSH levels to remain, elevated.

Table 2. Effect of Saffron on the Hepatic Antioxidant Parameters in Rats
GroupsGroup 1 ControlGroup 2 Saffron HDGroup 3 HCCGroup 4 HCC+ Saffron HDGroup 5 HCC+ Saffron MDGroup 6 HCC+ Saffron LD
  • Values are expressed as mean ± SEM of eight rats per group. Concentration is expressed as nmol/mg protein for MDA, P.Carbonyl, and GSH. Activity is expressed as unit/mg protein for CAT, SOD, and GST. Activity is expressed as m unit/mg protein for MPO. Significance was determined by one-way analysis of variance followed by Dunnett's t test.

  • *

    P < 0.001.

  • P < 0.01.

  • P < 0.05 versus control group 1.

MDA0.66 ± 0.020.73 ± 0.030.84 ± 0.03*0.71 ± 0.010.69 ± 0.010.73 ± 0.02
P.Carbonyl0.97 ± 0.040.87 ± 0.021.78 ± 0.09*0.88 ± 0.021.09 ± 0.081.04 ± 0.06
CAT116.0 ± 0.92116.25 ± 0.37101.9 ± 0.92*112.75 ± 1.75116.88 ± 0.19116.63 ± 0.11
SOD6.23 ± 0.066.23 ± 0.055.26 ± 0.23*5.92 ± 0.166.19 ± 0.076.41 ± 0.09
MPO18.15 ± 0.5117.35 ± 0.6428.95 ± 1.76*23.13 ± 1.9921.17 ± 1.1525.33 ± 1.27
GSH57.72 ± 0.4856.00 ± 1.0468.45 ± 4.2968.70 ± 0.1264.47 ± 2.5868.61 ± 1.31
GST17.10 ± 0.1917.12 ± 0.2021.91 ± 1.06*16.81 ± 0.3517.89 ± 0.2618.11 ± 0.68

Saffron Inhibited DEN-Induced Hepatic Nodule Formation.

Macroscopically, rats in group 3 (HCC) revealed enlarged liver with multiple nodules on the liver surface, the nodule incidence was 75% and the number of nodules per nodule-bearing animal (nodule multiplicity) was 2.33 in this group (Table 3). Groups 4-6, where animals treated with saffron and DEN-2-2AAF, showed a decrease in liver enlargement, nodule incidence and multiplicity. The protective effect of saffron was most dramatic in group 4 rats (highest dose of saffron), where saffron completely inhibited the appearance of hepatic nodule altogether.

Table 3. Inhibition of DEN-2-AAF–Induced Liver Cancer in Male Rats by Oral Administration of Saffron
GroupsNumber of Rats with NodulesNodule Incidence (%)Total Number of NodulesNodule* Multiplicity
  • *

    Number of nodules per nodule-bearing liver. Rats of the protective groups (groups LD, MD, and HD) were orally fed a saffron suspension at doses of 75, 150, and 300 mg/kg body weight, respectively, 2 weeks before HCC initiation and continued for 22 weeks.

+ Saffron HD0/80%00
+ Saffron MD2/825%21
+ Saffron LD2/825%21

Saffron Restored Liver Function.

Serum activities of GGT, ALT, and AFP were significantly increased in group 3 (HCC) as compared to control rats (group 1), thus indicating liver damage. However, pretreatment with saffron (groups 4-6) caused a significant decrease in the elevation of these proteins (Table 4). Interestingly, higher doses of saffron caused a lesser decrease in the DEN-induced GGT and ALT levels, as opposed to the lowest doses almost reversed these DEN-induced enzyme increase. This is apparently consistent with the proapoptotic effect of higher concentrations of saffron, which would in fact cause more of the hepatocarcinogenic cells to die and release these enzymes in the serum.

Table 4. Effect of Saffron on Serum Markers
GroupsGroup 1 ControlGroup 2 Saffron HDGroup3 HCCGroup 4 HCC+ Saffron HDGroup 5 HCC+ Saffron MDGroup 6 HCC+ Saffron LD
  • Values are expressed as mean ± SEM of eight rats per group. Level of AFP is expressed as ng/mL. Activity is expressed as U/L for ALT, AST, and GGT. Significance was determined by one-way analysis of variance followed by Dunnett's t test.

  • *

    P < 0.001.

  • P < 0.05, versus group 1.

  • §

    P < 0.05 versus group 3.

AFP1.03 ± 0.070.96 ± 0.082.26 ± 0.07*1.14 ± 0.111.22 ± 0.152.05 ± 0.07*
GGT1.78 ± 0.081.83 ± 0.0815.42 ± 1.20*11.32 ± 1.9*§2.83 ± 0.54§2.64 ± 0.34§
ALT24.22 ± 0.1522.35 ± 0.3829.39 ± 1.9924.07 ± 0.56§23.93 ± 2.33§20.73 ± 0.74§
AST56.25 ± 3.6155.75 ± 2.8863.51 ± 3.6955.87 ± 4.3167.75 ± 5.1053.50 ± 2.23

Saffron Inhibited DEN-Induced Foci of Altered Hepatocyte Formation and GST-p Expression.

Histological examination shows nodular liver in animals treated with DEN-2-2AAF alone (Supporting Fig. 1). These nodules are composed of large, irregular, and pale hepatocytes with large hyperchromatic nuclei and represent the classical foci of altered hepatocytes (FAH). However, in saffron-treated groups, a significant reduction in the number and size of these nodules were observed and a larger number of regular hepatocytes were observed. Most dramatically, in group 4 (rats exposed to highest dose of saffron), the nodular architecture was completed suppressed. These findings show the dramatic protection offered by saffron against hepatocellular carcinoma.

Induction of GST-p is considered as an early biomarker of hepatocarcinogenesis. GST-p foci larger than 15 cells were measured using color image processor. The number and areas of foci per square centimeter of liver sections were calculated. In animals treated with DEN-2-2AAF, the number of GST-p positive foci and area per square centimeter were dramatically increased. Saffron treatment caused significant decrease both in the number of GST-p positive foci and in the area per square centimeter (groups 4-6) as compared to rats that received the carcinogen alone (Fig. 2; Supporting Fig. 2).

Figure 2.

Saffron inhibits DEN-induced induction of AHF and of GST-p expression in liver. Quantitative analysis of GST-p–positive foci on GST-p immune-stained liver sections (Supporting Fig. 2). The number and area of GST-p–positive foci was reduced saffron-treated rats. Significance was determined by one-way ANOVA followed by Dennett's t test: a P < 0.001 versus HCC group 3.

Effect of Saffron on Cell Proliferation and Apoptosis in DEN-Treated Rats.

The nuclear Ki-67 is an established marker of cellular proliferation.18 Liver sections from group 3 (HCC) were significantly higher in the number of Ki-67–positive cells than the control group. The treatment with saffron resulted in a dramatic decrease in the number of Ki-67–positive cells (groups 4-6) compared to rats received the carcinogen alone (Fig. 3; Supporting Fig. 3).

M30 CytoDeath antibody which detects the caspase-cleaved fragment of cytokeratin 18 during early apoptotic changes was used as an apoptotic marker. DEN did not induce significant increase in the number of TUNEL-positive cells and M30 CytoDeath–positive cells compared to control (group 1). However, the number of cells positive for TUNEL and M30 CytoDeath were significantly increased in groups treated with saffron and DEN-2-2AAF suggesting an up-regulation of apoptosis by saffron administrations in DEN-2-2AAF–exposed rats. There were no significant differences in the number of Ki-67-, TUNEL-, and M30 CytoDeath–positive cells between control and saffron-only rats (groups 1 and 2) (Fig. 3; Supporting Figs. 3–5). DEN-2-2AAF exposure also caused an increase in the number of p-TNFR1–positive cells, which were significantly decreased in saffron-treated groups compared to HCC group (Fig. 3; Supporting Fig. 6). Additionally, DEN exposure significantly increased the expression of COX-2, iNOS, NF-κB-p65 and ED-2, which were expressed mostly in hepatocytes around the central vein and in Kupffer cells (Fig. 4; Supporting Figs. 7-10). This increase in COX-2, iNOS, NF-κB-p65, and ED-2–positive cell numbers were significantly inhibited in the saffron-protected groups (groups 4, 5, 6) compared to rats that received DEN-2-2AAF alone (group 3). This indicates that the anti-inflammatory effect of saffron in the HCC model system could be due to blocking of NF-κB signaling.

Figure 3.

Effects of saffron on cell proliferation, apoptotic cell death and p-TNFR1 expression. Quantitative analyses of Ki-67-, TUNEL-, M30 CytoDeath-, and p-TNFR1–positive cells in different experimental groups. The positive expression of Ki-67, TUNEL, M30, and p-TNFR1 in each section was calculated by counting the number of brown stained cells in 10 fields at 400× magnification then the number of positive cells per field. Values expressed as mean ± SEM for eight animals in each group. Significance was determined by one-way ANOVA followed by Dennett's t test: a, P < 0.001 versus control group; b, P < 0.001; c, P < 0.01 versus HCC group 3.

Figure 4.

Saffron inhibits DEN-induced up-regulation of COX-2, iNOS, NF-κB-p65, and ED-2 expression. Quantitative analyses of cells positive for COX-2, iNOS, NF-κB-p65, and ED-2. The number of positive cells was reduced in saffron-treated rats. The positive expression of COX-2, iNOS, NF-κB-p65, and ED-2 in each section was calculated by counting the number of brown staining in 10 fields at 400× magnification, then the number of positive cells per field. Values are expressed as mean ± SEM for eight animals in each group. Significance was determined by one-way ANOVA followed by Dennett's t test: a P < 0.001 versus control group 1; b P < 0.001 versus HCC group 3.

Saffron Induces Growth Arrest and Apoptosis In Vitro.

To better understand the anticancer effects of saffron, more detailed in vitro analyses were carried out. HepG2 cells were treated with various concentrations of saffron (1-6 mg/mL) for 6, 24 and 48 hours. The MTT test showed that saffron significantly reduced the viability of HepG2 cells in a time- and dose-dependent manner (Fig. 5A). For further studies, a saffron concentration of 6 mg/mL was used. IL-8 level was also shown to be reduced upon saffron treatment of HepG2 cells (Fig. 5C).

Figure 5.

The effect of saffron extract on liver cancer cell line HepG2. (A) MTT viability test showing the HepG2 cells untreated (control) or treated with different concentrations of saffron (1 to 6 mg/mL) for 24 and 48 hours. The MTT data shown are performed in triplicate (*P < 0.001). (B) Effect of saffron on cell cycle progression. Cells were untreated (control) or treated with 6 mg/mL saffron for 6, 24, and 48 hours, harvested, and DNA was stained with PI for flow cytometric analysis of DNA content with FACScan flow cytometry. (C) IL-8 secretion after saffron treatment. HepG2 cells were treated with 6 mg/mL saffron for 6, 24, and 48 hours, and subsequently, the supernatants were analyzed by IL-8 ELISA (*P < 0.001). (D) Lysates prepared from HepG2 cells treated with saffron for 6, 24, and 48 hours were analyzed by anti-caspase-3, anti-IκB, anti-TNFR1, anti-pH2AX, and anti-GAPDH western blotting. GAPDH served as an internal control for equal loading. (E) Apoptosis measurement after saffron treatment. Annexin-PI measurements of untreated cells (control) and HepG2 cells treated with 6 mg/mL saffron for 6, 24, and 48 hours the profile represents annexin V–fluorescein isothiocyanate (FITC) staining in the x axis and PI in the y axis.

To examine whether DNA-damage mediates saffron's anticancer effect, protein level of p-H2AX, a sensor for DNA double strand breaks, was analyzed by western blotting. HepG2 cells showed a remarkable induction of p-H2AX starting at 24 hours of saffron's treatment (Fig. 5D). The effect of saffron on cell cycle progression was also assessed using flow cytometric analysis. Saffron-treated HepG2 cells displayed an accumulation of the cell population at the S phase starting from 6 hours (Fig. 5B).

Because NF-κB signaling was inactivated in our animal model, we tested whether or not a similar saffron-dependent NF-κB inactivation persists in vitro. Thus, the presence of the phosphorylated form of the I-kappa-B protein (p-IκB) was evaluated by western blotting. Once phosphorylated, IκB is known to be rapidly degraded thereby allowing activation of the NF-κB complex through its translocation into the nucleus. Indeed, we found an early decrease of p-IκB protein levels in cells treated with saffron, thus confirming an early inactivation of NF-κB (Fig. 5D). Moreover, in agreement with the in vivo results, saffron treatment reduced the TNF receptor 1 (TNFR1) protein expression (Fig. 5D). This is also in accordance with the notable increase in the active form of caspase-3 (Fig. 5D), thereby reflecting a strong proapoptotic effect of saffron. These findings were further supported by measuring the apoptotic cell fraction after saffron treatment using annexin-PI staining. Saffron induced apoptosis in HepG2 as early as 6 hours after treatment (Fig. 5E). The apoptosis induction further increased in a time dependent manner reaching 77.5% after 48 hours. These findings are in agreement with the observed pre-G1 cell population which showed a progression in the induced apoptosis by the accumulation of DNA in cells treated with saffron.


Administration of saffron to DEN-treated rats caused a dramatic reduction in the number and incidence of dyschromatic nodules as well as reduced the development of neoplastic FAH. This saffron-mediated reduction of nodular hepatocytes and FAH formation was closely associated by significant decrease in the number of GST-P positive foci and plasma AFP level, which are reliable and sensitive markers of preneoplasia and neoplasia.10, 19

Among the physiological alterations cancer cells undergo as they continue to grow are the increase in cell proliferation and the loss of apoptotic mechanisms.20, 21 In this study, saffron demonstrated significant antiproliferative activity by causing pronounced cell cycle arrest in vitro (Fig. 5) and reducing the number of proliferative cells (Ki-67–positive cells18) in DEN-treated animals (Fig. 3; Supporting Fig. 3). The antiproliferative activity of saffron was also associated with the induction of apoptosis as evidenced in vitro by caspase-3 cleavage and the pre-G predominant fraction in PI-FACS analysis. The apoptotic induction must have resulted from DNA damage as reflected by the up-regulation of the double-stranded DNA breakage marker, p-H2XA, (Fig. 5D) suggesting an additional role of saffron in sensitizing cancer cells to the effects of other chemotherapeutics. Consistently, saffron treatment increased the number of TUNEL- and M30 CytoDeath–positive cells in vivo (Fig. 3; Supporting Figs. 4 and 5). These results are in agreement with previous in vitro studies showing apoptosis and antiproliferative effect of saffron in various tumor cell lines.4, 22 These results seem to indicate that the inhibition of neoplastic development in rat liver was associated with a reduction of cell proliferation and an induction of apoptosis.

Increased oxidative stress can induce a wide spectrum of cellular damage and cellular signaling changes that has been associated with carcinogenesis.20, 21 Administration of saffron to DEN-treated rats in this study counteracted DEN-induced oxidative stress as shown by restoration of antioxidant levels of SOD, CAT, and GST in the liver and diminishing of important markers of oxidative stress, such as oxidized lipids (MDA) and proteins (P.Carbonyl). The antioxidant effect of saffron was also accompanied by a decrease in liver damage markers, namely, serum ALT and GGT levels, suggesting a concomitant protection against hepatic damage. The prevention of oxidative stress and hepatic toxicity by saffron might be attributed to its potent antioxidant capacity which was confirmed in this study. Saffron showed ABTS and DPPH radical scavenging activities and exhibited significant reducing power as indicated by the FRAP assay. The Antioxidant property of saffron could be credited to its phenolic content and to its active ingredients (such as safranal, crocin, crocetin, and carotene) (Table 1), all of which have been reported to have antioxidant properties.23 The association between decreased oxidative damage and reduced nodular and GST-P positive foci formations suggest that the antioxidant efficacy exhibited by saffron may be an important factor for its anticarcinogenic property.

Chronic inflammation is also known to lead to early changes associated with the development of cancer through unbalanced production and/or activation of proinflammatory mediators such as cytokines, prostaglandins (PGs), nitric oxide (NO) and transcription factors including NF-κB.21, 24 In this study, saffron displayed an efficacy to protect against DEN-induced liver inflammation by decreasing numbers of Kupffer cells (Fig. 4; Supporting Fig. 10) and levels of hepatic MPO (Table 2), a marker of neutrophil infiltration.25 This decrease of Kupffer cells and neutrophils seems to be associated with an early inactivation of NF-κB signaling pathway, as reflected in the early in vitro inhibition of p-IκB and IL-8 (Fig. 5). Saffron also inhibited the in vivo protein expressions of COX-2 and iNOS - both of which are key enzymes involved in producing proinflammatory signals. Additionally, saffron administration resulted in dramatic down-regulation of activation of TNFα Receptor in vivo (Figs. 3 and 4; Supporting Figs. 6-8) and of receptor's expression in vitro (Fig. 5D). Taken together, these results suggest that saffron-based protection against carcinogenesis could be mediated by its anti-inflammatory effects through down-regulating COX-2 and iNOS expressions and decreasing the numbers of active TNFα receptors in tumor cells. The NF-κB pathway mediates many of the protumoral effects of TNFα and has been targeted for anticancer therapy. For example, TNFα inhibitors such as infliximab and etanercept have been shown to reduce the level of NF-κB and lower the constitutive production of IL-8 in different cell lines.26 Tumor cells are normally exposed to TNFα delivered both by tumor-associated cells (infiltrating monocytes (Kupffer cells) and other stromal cells) and the tumor cells themselves. Given that saffron reduced the numbers of Kupffer cells and down- regulated p-TNFR1, it seems that saffron exerts its antitumoral action, at least in part, via short-circuiting the TNFα feedback loop between tumor cells and the Kupffer cells in their microenvironment. Similar strong effects have been reported where different tumor cell lines showed reduced tumor growth and a reduced number of liver metastases when the mice were repeatedly treated with anti-TNFα agents.26

Increased expressions of COX-2 and iNOS have been observed in several human tumor tissues and in chemically-induced animal tumors.23, 27 Interestingly, NF-κB (a key player in inflammation) has been shown to be activated by increased oxidative damage and is involved in up-regulation of COX-2 and iNOS.21, 24, 27 It is conceivable therefore that the DEN induces inflammation via an oxidative-dependent manner involving reactive oxygen species (ROS). This notion is supported by our observation that antioxidant containing saffron administration causes significant down-regulation of NF-κB (Fig. 4).

NF-κB is normally present in the cytoplasm bound to an inhibitory protein, IkB. In response to certain stimuli such as ROS or inflammatory cytokines, IκB is phosphorylated, which causes it to release NF-κB which is then transported into the nucleus, where it induces transcription of a large variety of target genes that induce inflammatory (COX-2 and iNOS) and antiapoptotic responses.21, 24 In this study, administration of saffron to DEN-treated rats reversed DEN-induced up-regulation of NF-κB-p65 subunit (Fig. 4). A similar result is reported in our in vitro studies, where saffron treatment caused an early decrease in the phosphorylation state of IκB (Fig. 5D). This anti-inflammatory effect of saffron against acute and chronic models of inflammation has been previously reported as well.8

In summary, the data presented here show that saffron dramatically inhibited both nodular and FAH formation in livers of DEN-treated rats. This inhibition was associated with induced apoptosis, reduced cell proliferation, decreased oxidative stress and down-regulation of inflammatory markers, such as COX-2, iNOS, NF-κB-p65 and p-TNFR1 expressions. Figure 6 incorporates our data into a model showing a possible mechanism of the anticancer protective effect of saffron by promoting apoptosis, inhibiting cell proliferation, and blocking inflammation in hepatocarcinomas. Further investigations are currently underway to investigate in more detail the mechanism of action of saffron extract.

Figure 6.

Schematic model showing the protective effect of saffron in DEN-induced HCC by preventing proliferation and inflammation and promoting apoptosis. This model incorporates the in vivo and in vitro data presented in this study. Saffron prevents activation of Kupffer cells leading to attenuation of hepatic TNFα receptor activation, which in turn leads to down-regulation of NF-κB signaling, including decrease in the levels of iNOS, COX-2, and IL-8 (preventing inflammation). Saffron also causes cell-cycle arrest (stopping proliferation and inducing apoptosis), as well as up-regulating caspases.


This work was financially supported by Emirates Foundation grant 2009-079 to A.A. Authors are grateful to Aktham Awwad (Tawam Hospital), Rkia Al-Kharrge (UAE University) for assessing hepatic nodules. Authors are also indebted to Hamdi Kandil (UAE University) and Moustafa Abdalla (Groves High School, MI USA) for their technical assistance.