Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=iijf20

International Journal of Food Sciences and Nutrition

ISSN: 0963-7486 (Print) 1465-3478 (Online) Journal homepage: http://www.tandfonline.com/loi/iijf20

Apoptotic effect of a phytosterol-ingredient andits main phytosterol (β-sitosterol) in human cancercell lines

Andrea Alvarez-Sala, Alessandro Attanzio, Luisa Tesoriere, GuadalupeGarcia-Llatas, Reyes Barberá & Antonio Cilla

To cite this article: Andrea Alvarez-Sala, Alessandro Attanzio, Luisa Tesoriere, Guadalupe Garcia-Llatas, Reyes Barberá & Antonio Cilla (2018): Apoptotic effect of a phytosterol-ingredient and itsmain phytosterol (β-sitosterol) in human cancer cell lines, International Journal of Food Sciencesand Nutrition, DOI: 10.1080/09637486.2018.1511689

To link to this article: https://doi.org/10.1080/09637486.2018.1511689

Published online: 07 Sep 2018.

Submit your article to this journal

View Crossmark data

RESEARCH ARTICLE

Apoptotic effect of a phytosterol-ingredient and its main phytosterol(b-sitosterol) in human cancer cell lines

Andrea Alvarez-Salaa, Alessandro Attanziob, Luisa Tesoriereb, Guadalupe Garcia-Llatasa , Reyes Barber�aa

and Antonio Cillaa

aNutrition and Food Science Area, Faculty of Pharmacy, University of Valencia, Valencia, Spain; bDepartment of Biological, Chemicaland Pharmaceutical Sciences and Technologies (STEBICEF), University of Palermo, Palermo, Italy

ABSTRACTDietary interventions may effectively control cancer development, with phytosterols (PS) being aclass of cancer chemopreventive dietary phytochemicals. The present study, for the first time,evaluates the antiproliferative effects of a PS-ingredient used for the enrichment of several foodsand its main PS, b-sitosterol, at physiological serum levels, in the most prevalent cancer cells inwomen (breast (MCF-7), colon (HCT116) and cervical (HeLa)). In all three cell lines, these com-pounds induced significant cell viability reduction without a clear time- and dose-dependentresponse. Moreover, all treatments produced apoptotic cell death with the induction of DNAfragmentation through the appearance of a sub-G1 cell population. Thus, the use of PS as func-tional ingredients in the development of PS-enriched foods could exert a potential preventiveeffect against human breast, colon and cervical cancer, although further in vivo studies arerequired to confirm our preclinical findings.

ARTICLE HISTORYReceived 29 May 2018Revised 9 August 2018Accepted 10 August 2018

KEYWORDSAntiproliferation; apoptosis;breast cancer; cervicalcancer; colon cancer;plant sterols

Introduction

The global burden of cancer has gained great atten-tion, as it is among the leading causes of morbidityand mortality worldwide. It has been estimated thatthere were 14.1 million new cancer cases and 8.2 mil-lion deaths in 2012, and the numbers are expected torise significantly over the coming decades. In women,the most prevalent cancers are breast cancer (25.2%),followed by colorectal cancer (9.2%) and cervical can-cer (7.9%) (Ferlay et al. 2015). Bioactive food compo-nents such as phytosterols (PS) could constitute a newalternative as cancer chemopreventive and therapeuticagents against these kinds of cancers, intervening inseveral regulatory pathways. Different reviews on theanticancer effects of PS (mainly b-sitosterol) havebeen published (Woyengo et al. 2009; Bradford andAwad 2010; Shahzad et al. 2017). Moreover, it hasbeen reported that PS do not alter the viability andthe normal cell growth of non-cancerous cells such asdifferentiated Caco-2 (a model of normal small intes-tinal cells) (Awad et al. 2005), COS-1 (fibroblast cells)(Jayaprakasha et al. 2007), VERO (monkey kidneycells) (Baskar et al. 2010) or MRC-5 cells (fibroblast

derived from lung cells) (Rahman et al. 2013) in arange of concentrations (0.6–1000mM) in which areour own tested concentrations (13–52 mM).

The antiproliferative activity of b-sitosterol stand-ards has been studied in breast cancer (MCF-7)(Awad et al. 2007, 2008; Rubis et al. 2010), colon can-cer (HT-29, Caco-2 and/or HCT116) (Awad et al.1996, 1998; Choi et al. 2003; Daly et al. 2009;Montserrat-de la Paz et al. 2015) and cervical cancer(HeLa) cells (Cheng et al. 2015). Likewise, the anticar-cinogenic effects of isolated PS from plant extractshave also been evaluated in MCF-7 (Chai et al. 2008;Malek et al. 2009; Rahman et al. 2013; Yaacob et al.2015; Tahsin et al. 2017), in HCT116 (Malek et al.2009; Rahman et al. 2013) and in HeLa cells (Blocket al. 2004; Csupor-Loffler et al. 2011; Hamdan et al.2011; Han et al. 2013).

Since dietary PS alone are unable to offer the rec-ommended daily doses (1.5–3 g/day) for loweringLDL-cholesterol (Gylling et al. 2014), several foods arecurrently enriched with PS. The main sources of PSused for this purpose are vegetable oil deodoriser dis-tillate and tall oil (a byproduct of the kraft pulping of

CONTACT Alessandro Attanzio alessandro.attanzio@unipa.it Department of Biological, Chemical and Pharmaceutical Sciences and Technologies(STEBICEF), University of Palermo, Via Archirafi 28, Palermo 90123, Italy; Guadalupe Garcia-Llatas guadalupe.garcia@uv.es Nutrition and FoodScience Area, Faculty of Pharmacy, University of Valencia, Avda. Vicente Andr�es Estell�es s/n, Burjassot, Valencia 46100, Spain� 2018 Taylor & Francis Group, LLC

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITIONhttps://doi.org/10.1080/09637486.2018.1511689

wood) (Garc�ıa-Llatas and Rodr�ıguez-Estrada 2011;Gonz�alez-Larena et al. 2011).

Previous studies conducted by our research grouphave reported a positive effect on cardiovascular riskand bone turnover markers after the regular con-sumption of milk-based fruit beverages enriched withPS from tall oil (Granado-Lorencio et al. 2014); anti-carcinogenic effects of PS standards (b-sitosterol, cam-pesterol and stigmasterol, alone or combined) uponCaco-2 cells, at concentrations compatible withphysiological serum levels after the intake of such bev-erages (Cilla et al. 2015), and cytoprotective effects ofthe bioaccessible fractions obtained after simulatedgastrointestinal digestion of these beverages in Caco-2cells (L�opez-Garc�ıa et al. 2017a). Moreover, due to thefact that PS undergo less absorption (0.5–2%) (Gyllinget al. 2014), these compounds may reach the colonand exert local actions at gastrointestinal level. In thiscontext, a recent study carried out by our researchgroup (L�opez-Garc�ıa et al. 2017b), considering theestimated colonic concentrations of PS capable ofreaching the colon after the intake of the aforemen-tioned kind of beverage, revealed antiproliferativeeffects upon the Caco-2 cell line. However, the anti-carcinogenic activity of the tall oil PS-ingredient usedfor food enrichment has not been evaluated so far.Thus, the present work, for the first time, studies theantiproliferative effect of a PS-ingredient from tall oilused for the enrichment of milk-based fruit beverageand its main PS (b-sitosterol) at concentrations com-patible with physiological serum levels, following itsregular consumption, upon human breast (MCF-7),colon (HCT116) and cervical cancer (HeLa) cell lines.In this regard, the present study could complementthe well-known hypocholesterolemic effect of PS andmoreover contribute to extend their use as functionalingredients in the development of PS-enriched foodsto maximise their functionality.

Materials and methods

Reagents

Dimethyl-sulfoxide (DMSO), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylthiazolium bromide (MTT), (24R)-ethylcholest-5-en-3b-ol (b-sitosterol), RPMI-1640medium, propidium iodide (PI) and RNase A werepurchased from Sigma Chemical Co. (St. Louis, MO).Annexin V apoptosis detection kit FITC was fromeBioscience (San Diego, CA). Antibiotic solution(penicillin–streptomycin), nonessential amino acids,N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid(HEPES), phosphate buffered solution (PBS) and

trypsin–EDTA solution (2.5 g/l trypsin and 0.2 g/lEDTA) were obtained from Gibco (Scotland, UK).

Phytosterol powder free microcrystalline ingredientfrom tall oil, previously characterised by our researchgroup (b-sitosterol 78.86%; sitostanol 11.95%; campes-terol 7.13%; campestanol 1.20% and stigmasterol0.82%; Gonz�alez-Larena et al. 2011), containing malto-dextrin, sucrose ester and inulin was purchased fromLipofoods, Barcelona, Spain (LipohytolVR 146MEDispersible).

Cell culture and treatments

The human breast (MCF-7), colon (HCT116) and cer-vical cancer (HeLa) cells (American Type CultureCollection, LGC Promochem, Italy) were usedbetween passages 33 and 47, and were grown in75 cm2 Falcon flasks in RPMI-1640 medium contain-ing 4.5 g/l glucose and supplemented with 10% (v/v)foetal bovine serum, 1% (v/v) nonessential aminoacids, 1% (v/v) HEPES and antibiotic solution (peni-cillin–streptomycin). Cells were incubated in ahumidified atmosphere (37 �C, 5% CO2).

In all experiments, cells were seeded at a density of5� 104 cells/cm2 after trypsin treatment (2.5 g/l tryp-sin and 0.2 g/l EDTA). For viability assays, cell lineswere seeded during 24 h onto 96-well plates with0.2ml of RPMI medium, followed by treatment for24 h or 48 h with the tall oil PS-ingredient or b-sitos-terol at different concentrations (13, 26 and 52mM).These concentrations are included in the range of PSphysiological serum levels (5–30mM), reaching valuesof 50 mM after the consumption of PS-enriched foods(Daly et al. 2009; Gylling et al. 2014). The lowest PSconcentration assayed in the present work (13mM), issimilar to the serum b-sitosterol concentration(15mM) obtained after the consumption of milk-basedfruit beverages enriched with the tall oil PS-ingredientpreviously evaluated by our research group (Garcia-Llatas et al. 2015). In the case of the PS-ingredient,the selected concentrations were based on the mainPS (b-sitosterol).

For cell cycle and apoptosis assays, cells wereseeded onto 24-well plates with 1ml of RPMImedium. Following 24 h from seeding, cell lines weretreated with PS-ingredient (13 mM) or b-sitosterol atdifferent concentrations (13, 26 and 52mM) for 48 h,based on the cell viability results. Control cultures andPS-ingredient were prepared with RPMI medium andb-sitosterol in tetrahydrofuran (0.1%) for all assays.Under the above-mentioned conditions, tetrahydro-furan did not affect cell viability (data not shown).

2 A. ALVAREZ-SALA ET AL.

Cell viability assay

Cells were seeded onto 96-well plates at a density of18,000 cells per well, and were maintained underappropriate culture conditions (37 �C, 5% CO2) for24 h. After 24 or 48 h of treatment with the three con-centrations of PS-ingredient or b-sitosterol (13, 26and 52 mM), 4 ll of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma) (5mg/ml)was added, followed by incubation at 37 �C for 2 h.Then, the MTT reaction agent was removed and100 ll of DMSO was added to allow the dissolution ofpurple formazan products, as previously described(Girasolo et al. 2014).

Cell cycle analysis

For cell cycle assay, each cell line was treated andincubated during 48 h (37 �C, 5% CO2) with samples.Aliquots of 1� 106 cells were harvested by centrifuga-tion (125� g, 5min), and the cell pellet was washedwith PBS and incubated in the dark (30min at 4 �C)in a 5mM sodium phosphate buffer solution contain-ing 20 mg/ml PI, Triton (0.1%, v/v) and 200 mg/mlRNase. DNA fluorescence was measured by cytofluor-ometry using an Epics XLTM flow cytometer withExpo32 software (Beckman Coulter, Miami, FL). Therelative distribution of 1� 104 events was analysed foreach sample as described by Cilla et al. (2015).

Assessment of apoptosis throughphosphatidylserine exposure

The apoptosis measurement was carried out accordingto Cilla et al. (2015) through flow cytometry by dou-ble staining with Annexin V/PI to detect externalisa-tion of phosphatidylserine to the cell surface. Each cellline was treated and incubated during 48 h (37 �C, 5%CO2) with samples, and then adjusted to 1� 106 cells/ml with binding buffer. Cell suspension (100 ml) wasadded to a new tube and incubated with 5 ml AnnexinV and 10 ml of 20mg/ml PI solution at room tempera-ture in the dark (15min). Then, for each sample1� 104 events were analysed by flow cytometry withthe appropriate two-dimensional gating method.

Statistical analysis

The analysis of all samples was performed in tripli-cate. One-way analysis of variance (ANOVA), fol-lowed by LSD post hoc testing were applied todetermine differences between treated and controlcells on the same day of treatment. A paired t-test

was used for the MTT assay to detect statistically sig-nificant differences between different time periods forone same treatment. A significance level of p< .05was adopted for all comparisons, and theStatgraphicsVR Centurion XVI.I statistical package(Statpoint Technologies Inc., Warrenton, VA) wasused throughout.

Results

The effects of PS treatment with tall oil PS-ingredientor its main PS (b-sitosterol) upon cell viability, cellcycle and apoptosis were evaluated at physiologicalconcentrations in cell lines corresponding to the mostprevalent cancers in women: breast (MCF-7), colon(HCT116) and cervical (HeLa).

MCF-7 cells

The effect of PS-ingredient or b-sitosterol (13, 26,52mM) after 24 and 48 h of incubation upon MCF-7cell growth, assessed by MTT assay, is shown inFigure 1(A). Only the phytosterol-ingredient at 13mMand the b-sitosterol standards showed a significant(p< .05) decrease in cell viability, being greater at48 h. The decrease in cell viability produced by thePS-ingredient (13 lM) was 17% or 53%, which wassimilar to the results for b-sitosterol (13, 26 and52lM), ranging from 19–22% or 50–67% at 24 and48 h, respectively – with no statistically significant(p< .05) dose-response effect among all the concen-trations tested (Figure 1(A)). For this reason, the fol-lowing assays (cell cycle and apoptosis studies) (Figure1(B and C)) were carried out only with the physio-logical PS-ingredient concentration (13 mM) andb-sitosterol (13, 26 and 52mM) at 48 h.

MCF-7 cell staining with PI was used to evaluatethe cell cycle at 48 h by flow cytometry (Figure 1(B)).The percentage of cells in the sub-G1 phase (consid-ered as a marker of DNA fragmentation) (Choi et al.2003) increased with all treatments versus MCF-7control cells, following the order: b-sitosterol 13lM(7.1-fold)>b-sitosterol 26 lM (6.3-fold)>PS-ingredi-ent 13 lM (5.4-fold)> b-sitosterol 52 lM (3.5-fold).This was accompanied by a concomitant and statistic-ally significant (p< .05) decrease, in the same order,for the other cell cycle phases (20–58%, 15–51% and21–55% in phases G1, S and G2/M, respectively)(Figure 1(B)). These results suggest that the PS-ingre-dient and b-sitosterol could exert an antiproliferativeeffect through apoptosis involving the modulation ofcell cycle progression.

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 3

Figure 1. Effect of PS-ingredient (PS-Ingr.) or b-sitosterol (b-sito) on (A) viability (MTT assay), (B) cell cycle and (C) apoptosis ofMCF-7 cells. Data are expressed as mean± SD; MTT assay: different lowercase letters (a–c) or different capital letters (A–B) indicatestatistically significant differences (p< .05) among the treatments after 24 or 48 h of incubation, respectively. #Denotes statisticallysignificant differences between 24 and 48 h for a same treatment; Cell cycle assay: different lowercase letters (a–e) indicate statis-tical significant differences (p< .05) among the different treatments after 48 h of incubation; Apoptosis assay: percentage ofAnnexin V/propidium iodide (PI) double-stained cells presented are representative images of three experiments in triplicate.

4 A. ALVAREZ-SALA ET AL.

The best marker for evaluating cell death throughapoptosis is phosphatidylserine externalisation, anevent occurring in the early phase of apoptotic celldeath when the cell membrane is still intact. For thispurpose, Annexin V-FITC and PI double labellingwas carried out by flow cytometry. As shown inFigure 1(C), all treatments induced an increase inearly apoptosis versus control MCF-7 cells, with thesame behaviour of samples (PS-ingredient and b-sitos-terol standard) as evaluated in the cell cycle assay.These results again indicated the absence of adose–response effect upon MCF-7 cells.

HCT116 cells

The effect of PS-ingredient or b-sitosterol (13, 26,52 mM) upon HCT116 cell growth after 24 and 48 h ofincubation is shown in Figure 2(A). All treatments(except PS-ingredient at 26mM after 24 h) induced astatistically significant (p< .05) decrease in cell viabil-ity with respect to control cells. The greatest viabilityreductions were observed with the PS-ingredient at13 mM (36 or 46% at 24 and 48 h, respectively) andb-sitosterol at 13, 26 and 52 mM (27–42% or 44–49%at 24 and 48 h, respectively), though with no cleartime- and dose-response effect (Figure 2(A)).

After 48 h of exposure, an increase in the percent-age of cells in sub-G1 phase was observed with alltreatments (Figure 2(B)), with a significant (p< .05)higher value for PS-ingredient at 13 mM and b-sitos-terol at 13 and 26 mM (6.7–8.8 fold) versus b-sitosterol52 mM (3.3-fold). At the same time, we observed a sig-nificant (p< .05) decrease only in phase G1(19.4–23.9%), with the exception of b-sitosterol at52 mM (Figure 2(B)). This behaviour was also observedafter Annexin V/PI assay (Figure 2(C)), where alltreatments significantly (p< .05) induced an increasein early apoptosis (12.9–20.9%) versus control cells(1%), following the order: PS-ingredient13 lM�b-sitosterol 13 lM� b-sitosterol 26 lM>b-sitosterol 52lM. These results again indicated theabsence of a dose–response effect upon HCT116 cells,and that the effect of PS-ingredient on this cell linewas due to the apparent predominant action of itsmain PS (b-sitosterol).

HeLa cells

In the HeLa cell line after 24 h of incubation, only PS-ingredient (52 mM) and b-sitosterol (13 mM) showed asignificant (p< .05) decrease of 20.5–21.4% versuscontrol cells (Figure 3(A)). However, at 48 h, the PS-

ingredient at concentrations of 13 and 26mM (22.6and 34.1%, respectively) also induced a decrease inHeLa cell viability. Moreover, a greater decrease pro-duced by PS-ingredient at 26 and 52mM and b-sitos-terol (13mM) at 48 h versus 24 h of exposure wasobserved (Figure 3(A)). Regarding cell cycle assay(Figure 3(B)), and in contrast to the results obtainedin the other studied cell lines, the greatest increase inthe percentage of cells in the sub-G1 phase corre-sponded to PS-ingredient 13mM (3.4-fold), followedby b-sitosterol 52mM (1.9-fold)� 26 mM (1.5-fold)� 13 mM (1.3-fold) versus control cells, with aconcomitant significant (p< .05) decrease in phase G1(28, 8.4 and 4.6% for PS-ingredient 13 mM, b-sitosterol52 and 13 mM, respectively). Moreover, treatment withPS-ingredient (13 mM) also induced a significant(p< .05) reduction in phases S and G2/M of 20.6 and23.7%, respectively (Figure 3(B)). This behaviour wasalso observed after Annexin V/PI assay (Figure 3(C)),where PS-ingredient 13 mM showed the highest pro-portion of early HeLa cell apoptosis (24.5%) versuscontrol cells, and a dose–dependent effect of b-sitos-terol standards was observed (14.3–20.4%), with agreater apoptotic effect induced by b-sitosterol 52 mM.

Discussion

In general, as it will be shown below, the differencesobserved in comparison with our study may be dueto: (i) the kind of bioactive compound studied(b-sitosterol standard with or without cyclodextrincomplexation, PS mix instead of PS alone, or isolatedPS from plant extracts); (ii) the use of different incu-bation times (1–9 days); (iii) different concentrationof PS (0–400mM); and (iv) the use of differentcells lines.

Breast cancer cells

Regarding the b-sitosterol standards, the growthinhibition percentages observed at 13mM for 48 h arewithin levels recorded in previous studies (29–81%)with b-sitosterol at 16 mM tested between 1 and 5days of incubation (Awad et al. 2007, 2008; Rubiset al. 2010). On the other hand, it has been observedthat b-sitosterol intake is associated to a greater prob-ability of oestrogen receptor-positive (related to MCF-7) than oestrogen receptor-negative tumours, possiblyexplaining that MCF-7 cells may be more resistant tothis compound (Touillaud et al. 2005; Grattan 2013).These authors suggested biological effects of PS uponoestrogen receptor regulation, indirectly influencing

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 5

cellular processes such as oestrogen metabolism, oes-trogen receptor function and expression, since PS canmodify the fluidity of cholesterol-rich cell membranes(without altering their integrity) and inhibit mem-brane-bound molecules. However, Ju et al. (2004)reported that the reduction of tumour size in ovariec-tomized athymic mice injected with MCF-7 cells afterthe addition of b-sitosterol glycoside or b-sitosterol:b-sitosterol glycoside (99:1) to the diet (0.2 and 10 g/kgdiet, respectively) was independent of oestrogen sig-nalling. These authors, therefore, suggested that PSmay be beneficial for women with breast cancer,though the mechanism involved is unclear. Overall,based on the epidemiological data and studies carriedout with cells or animals, it can be concluded thatcontroversy exists regarding the role of PS as a medi-ator of breast cell growth through oestrogen pathways.

Several mechanisms have been proposed for clarify-ing the action of b-sitosterol (16 mM) in relation tobreast cancer cell (MCF-7) proliferation, such as theactivation of the extrinsic apoptotic pathway throughincreased activity of caspase 8 (1.9-fold increase versuscontrol), and a 30% increase in first apoptosis signalreceptor (Fas) (Awad et al. 2007). Moreover, anincrease in ceramide (an intracellular modulator ofcell growth), which suggests that b-sitosterol activatesde novo ceramide synthesis by stimulating serine pal-mitoyl transferase activity (Awad et al. 2008) is alsoobserved. Globally, these results confirm the apoptoticaction of b-sitosterol, in agreement with our ownresults. Regarding cell cycle distribution, and in con-trast to our findings, b-sitosterol (1 and 5mM) did notinduce significant changes in the cell cycle of MCF-7cells at 24 h (Rubis et al. 2010), possibly because ofthe lower concentrations and shorter incubationtimes involved.

To the best of our knowledge, no other similarantiproliferative studies have been made with PS-ingredients; however, different studies have evaluatedthe antiproliferative effect of PS (mainly b-sitosterolor PS mixtures) isolated from plant extracts uponMCF-7 cells, mainly through the determination ofIC50 (the PS concentration causing 50% inhibition ofcell growth), which is directly dependent upon thedose and exposure time involved. There is controversyas to when a plant extract can be considered to havean active cytotoxic effect based on IC50. According toTahsin et al. (2017), a plant extract is active withIC50< 500 mg/ml, while Malek et al. (2009) indicatethat a plant extract is regarded as active with IC50 �20 mg/ml following 48 to 72 h of incubation.Nevertheless, it is recognised that the significant or

nonsignificant cytotoxicity reflected by a given IC50 isconditioned to the sensitivity of the cell line involved.

Generally, it has been reported that treatment withmixtures of PS (mainly b-sitosterol, campesterol andstigmasterol) from different plant extracts showed noantiproliferative effects upon MCF-7 cells at concen-trations of 0.1–100 mg/ml and incubation times of 24,48 and 72 h (Malek et al. 2009; Rahman et al. 2013;Yaacob et al. 2015). Unlike, the PS-ingredient used inour study, mainly comprising b-sitosterol, showed anantiproliferative effect (17 and 53% at 24 and 48 h,respectively) at the lowest concentration used(5.39 mg/ml, 13 mM). The antiproliferative effectobserved with the PS-ingredient could be mainlyattributable to the presence of b-sitosterol, since cellviability inhibition values consistent with our ownresults, have been reported with similar (10mM) orhigher (174 mM) concentrations of b-sitosterol at 72 h(Chai et al. 2008; Malek et al. 2009).

Colon cancer cells

As far as we are aware, only one other study (Choiet al. 2003) has been carried out with b-sitosterolstandard (2.5–20 mM) in the HCT116 cell line, induc-ing slightly greater growth inhibition in a dose-dependent manner (50 and 75% with 7.5 and 20 mM,respectively) compared with our results referred tob-sitosterol standard at 48 h (44–49% at 13, 26 and52mM, with no clear dose-response). Using differentcolon cancer cell lines as HT-29, b-sitosterol standard(16mM) complexed with cyclodextrin, and involvinglonger treatment periods (3–9 or 5 days), showedsimilar (�35–67%) (Awad et al. 1996) or lower (19%)(Awad et al. 1998) antiproliferative responses than inour own study (27–49% at 13 mM and at 24 and 48 h).Our results are also within the range of antiprolifera-tive values (�10–50%) reported by Montserrat-de laPaz et al. (2015) with b-sitosterol standard(0–100 mM) at 24 and 48 h upon HT-29 cells.Regarding Caco-2 cells, Daly et al. (2009) found thathigher doses of b-sitosterol standard (200 and400mM) are necessary in order to obtain similar anti-proliferative values (39–47%) at 48 h to our ownresults. In the same cell line, using serum (6, 12 and24mM) (Cilla et al. 2015) or estimated colonic concen-trations (115 mM) (L�opez-Garc�ıa et al. 2017b) ofb-sitosterol standard or PS mix standards (b-sitosterol,campesterol and stigmasterol) (serum: 6.6, 13.2,26.5mM or colonic: 132 mM, respectively), wasrecorded a decrease in cell viability (21–44% or57–59%, respectively), similar to that found in our

6 A. ALVAREZ-SALA ET AL.

Figure 2. Effect of PS-ingredient (PS-Ingr.) or b-sitosterol (b-sito) on (A) viability (MTT assay), (B) cell cycle and (C) apoptosis ofHCT116 cells. Data are expressed as mean± SD; MTT assay: different lowercase letters (a–d) or different capital letters (A–D) indi-cate statistically significant differences (p< .05) among the treatments after 24 h or 48 h of incubation, respectively; Cell cycleassay: different lowercase letters (a–c) indicate statistical significant differences (p< .05) among the different treatments after 48 hof incubation; Apoptosis assay: percentage of Annexin V/propidium iodide (PI) double-stained cells presented are representativeimages of three experiments in triplicate.

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 7

study at 24 h. Moreover, in agreement with our ownfindings, no clear dose-response relationship wasobserved between the different concentrations studied,with no evident additive or synergistic effects for thePS mix.

In general, HT-29 and Caco-2 cells seem to bemore resistant to antiproliferative PS action thanHCT116 cells, since the former needed longer timesor higher PS concentrations than in our study toreach the same effects in the latter cells. This may be

Figure 3. Effect of PS-ingredient (PS-Ingr.) or b-sitosterol (b-sito) on (A) viability (MTT assay), (B) cell cycle and (C) apoptosis ofHeLa cells. Data are expressed as mean± SD; MTT assay: different lowercase letters (a–b) or different capital letters (A–D) indicatestatistically significant differences (p< .05) among the different treatments after 24 or 48 h of incubation, respectively; #Denotesstatistically significant differences between 24 and 48 h for a same treatment. Cell cycle assay: different lowercase letters (a–d) indi-cate statistical significant differences (p< .05) among the different treatments after 48 h of incubation; Apoptosis assay: percentageof Annexin V/propidium iodide (PI) double-stained cells presented are representative images of three experiments in triplicate.

8 A. ALVAREZ-SALA ET AL.

due to the fact that HT-29 and Caco-2 cells havemutated the tumour suppressor protein p53, whileHCT116 cells have functional p53 (p53 wild type)(Ahmed et al. 2013). Accordingly, tumours withmutated p53 can have a higher proportion of prolifer-ating cells, and may be more metastatic than similartumours with wild-type p53 (Brown andWouters 1999).

In agreement with the present work, several studieshave demonstrated that the antiproliferative effects ofPS may be due to their ability to modulate cell cycleprogression and apoptosis; however, variability ofresponse was reported. In coincidence with our find-ings, Choi et al. (2003) observed in HCT116 cells thatb-sitosterol (2.5–20mM) induces apoptosis by increas-ing the sub-G1 cell population (3.8–16.5%) after 48 h,this effect being more accentuated from 7.5 mM. Inthis regard, Cilla et al. (2015) also observed apoptosisof Caco-2 cells through an increase in the number ofcells in sub-G1 phase (145%) with exposure to stand-ard PS mix at serum concentrations (13.2 mM), whileb-sitosterol standard (12mM) produced no significantchange in this phase – though arrest in G2/M (16%)was recorded at 24 h. In contrast, other authorsobserved different behaviour of the PS upon cell cycle.Montserrat-de la Paz et al. (2015), using only b-sitos-terol at 100mM, have reported statistically significantirreversible arrest in phase G0/G1 of the cell cycle inHT-29 cells – this effect not being significant forb-sitosterol at 50 mM. In this same regard, L�opez-Garc�ıa et al. (2017b) showed that standard PS mix(132 mM) and b-sitosterol (115 mM) also increase thenumber of cells in phase G0/G1. This suggests thatb-sitosterol or PS mix modulates Caco-2 cell growthby blocking phase G0/G1 at high concentra-tions (100–132 mM).

Other mechanisms involved in the inhibitory effectof PS upon colon cancer cells have been reported,mainly associated with the apoptosis pathway, as inour own study. In the cell line used in the presentwork (HCT116), Choi et al. (2003) previously showedthat the apoptosis mechanism of b-sitosterol(2.5–20mM) could be mediated by caspase-3 and cas-pase-9 activation, and can be associated withdecreased expression of the anti-apoptotic Bcl-2 pro-tein, with a concomitant increase in pro-apoptotic Baxprotein as well as with cytochrome c release from themitochondria into the cytosol (Choi et al. 2003). InHT-29 cells, b-sitosterol (16mM) was found to altermembrane lipids – suggesting that the inhibition ofcell growth could be due to activation of the sphingo-myelin cycle (involved in physiological parameters of

apoptosis and cell growth), increasing ceramide pro-duction (50%) (Awad et al. 1996, 1998). Moreover,b-sitosterol at a higher concentration (100 mM), upre-gulated LXR-a and LXR-b gene expression (nuclearfactors crucial for colon cancer progression)(Montserrat-de la Paz et al. 2015). On the other hand,PS anticarcinogenic activity against Caco-2 cells hasbeen reported mediated by the mitochondrial pathwayof apoptosis, secondary to an increase in cytosolicCa2þ and oxidative stress (Cilla et al. 2015), orthrough necrotic cell death (L�opez-Garc�ıa et al.2017b). In this context, it has been suggested byL�opez-Garc�ıa et al. (2017b) that PS exhibit a biphasiceffect in Caco-2 cells, activating different molecularpathways depending on the concentration used, asobserved on establishing comparisons with the resultsobtained in a previous report (Cilla et al. 2015). Itseems that apoptosis is the main pathway implicatedin cell death at low concentrations, while at high con-centrations the apoptotic pathway may be suppressed,and cell death through necrosis prevails (L�opez-Garc�ıaet al. 2017b). In this sense, in our own work, abiphasic effect is also suggested, since we observedgreater apoptosis response at lower concentrations (13and 26 mM) than at higher concentration (52mM)of b-sitosterol.

The cell line involved in this study (HCT116) hasalso been used to evaluate the antiproliferative effectof PS isolated from plant extracts. In this regard, ithas been reported that b-sitosterol and/or a mixtureof PS (campesterol, stigmasterol and b-sitosterol)extracted from Pereskia bleo or Curcuma zedoariadoes not display cytotoxic actions (IC50> 100 mg/ml)(Malek et al. 2009; Rahman et al. 2013).

Cervical cancer cells

To the best of our knowledge, only one study (Chenget al. 2015) has shown b-sitosterol (20mM) present inPinellas tuber to exert an antiproliferative effect (40%)upon HeLa cells at 24 h. A comparatively lesser anti-proliferative effect was observed in our study (20.5%)for b-sitosterol (13 mM) at 24 h – the effect being simi-lar (44.8%) at 48 h. Cheng et al. (2015) demonstratedthat the b-sitosterol antiproliferative effect was due toreduction of the expression of proliferating cellnuclear antigen (PCNA), since b-sitosterol probablyinhibits DNA synthesis in cells of this kind.Alterations in cell morphology were also observed(loss of cell surface microvilli, increased electron dens-ity of the cell membrane, and decreased organellepresence), suggesting that these cells gradually lose

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 9

their malignant tumour characteristics upon treatmentwith b-sitosterol. The authors concluded that elevatedlevels of p53 mRNA (2.9-fold) and reduced levels ofHPVE6 viral oncogenes (1.6-fold) observed upontreatment with b-sitosterol, could explain the anti-cancer activity against HeLa cells.

Most of the PS antiproliferative studies have beencarried out with PS isolated from plant extracts uponHeLa cells. Han et al. (2013) observed that b-sitosterolisolated from Benicasa hispida at different concentra-tions (2.5–50mM) did not show cytotoxic activity(inhibitory rate <35%) at 24 h. This behaviour is inagreement with our own work, since at 24 h, b-sitos-terol at 26 and 52 mM showed no statistically signifi-cant (p< .05) antiproliferative effect, and b-sitosterolat 13 mM only produced 20.5% inhibition of cellgrowth. Several studies (Block et al. 2004; Csupor-Loffler et al. 2011) have observed that mixtures ofb-sitosterol and stigmasterol isolated from Crotonzambesicus or Conyza Canadensis, respectively, exertedno cytotoxic effect (IC50> 30 mg/ml) at 72 h.Moreover, with shorter exposure times (48 h),Hamdan et al. (2011) reported that a mixture ofb-sitosterol and stigmasterol isolated from Citrusjambhiri Lush exhibited an IC50 at 114.2 mM. Theauthors suggested that the cytotoxic activity could bedue to the lipophilic character of these compounds,since they may interact with the lipophilic side chainsof phospholipids or cholesterol, and in turn affectmembrane fluidity and disrupt cell membrane func-tion by modifying the three-dimensional conformationof membrane proteins. These observations are inagreement with our own results, since PS-ingredient(containing b-sitosterol and stigmasterol) at the con-centrations tested (13, 26 and 52mM) did not reach50% proliferation inhibition in HeLa cells after 24or 48 h.

Conclusions

The antiproliferative effect of b-sitosterol or PS-ingre-dient upon MCF-7, HTC116 and HeLa cells was gen-erally evident at the concentration of 13mM – thisbeing compatible with physiological serum levels afterthe regular intake of a beverage containing these com-pounds. A possible biphasic/hormetic effect (with agreater response at lower concentrations versus higherconcentrations) and/or a threshold of optimal actionat the lowest assayed concentrations cannot be ruledout. Anticarcinogenic activity was observed throughthe induction of DNA fragmentation and apoptosis,with no clear time- and dose-dependent effect. With

all the caution called for in drawing conclusions fromin vitro assays, our results suggest that the PS-ingredi-ent used for enrichment of functional beverages andits main PS (b-sitosterol), may be regarded as naturalanticarcinogenic compounds against breast, colon andcervical cancer in these preclinical models, though fur-ther research is warranted. These observations may becomplementary to the decrease in risk of cardiovascu-lar disease and osteoporosis previously reported afterthe intake of a PS-enriched functional beverage con-taining the PS-ingredient assayed in this study; more-over, they should be taken into account by the foodindustry for using this PS-ingredient in the develop-ment of PS-enriched food products, with a view tomaximising their functionality.

Disclosure statement

No potential conflict of interest was reported bythe authors.

Funding

Andrea �Alvarez-Sala Mart�ın holds grants for PhD study(ACIF/2015/251) and research stay (BEFPI/2016/046) fromthe Generalitat Valenciana (Spain).

ORCID

Guadalupe Garcia-Llatas http://orcid.org/0000-0002-4681-4208

References

Ahmed D, Eide PW, Eilertsen IA, Danielsen SA, Eknaes M,Hektoen M, Lind GE, Lothe RA. 2013. Epigenetic andgenetic features of 24 colon cancer cell lines.Oncogenesis. 2:e71–e79.

Awad AB, Barta SL, Fink CS, Bradford PG. 2008. Beta-Sitosterol enhances tamoxifen effectiveness on breast can-cer cells by affecting ceramide metabolism. Mol NutrFood Res. 52:419–426.

Awad AB, Chen YC, Fink CS, Hennessey T. 1996. Beta-Sitosterol inhibits HT-29 human colon cancer cell growthand alters membrane lipids. Anticancer Res.16:2797–2804.

Awad AB, Chinnam M, Fink CS, Bradford PG. 2007. B-Sitosterol activates Fas signaling in human breast cancercells. Phytomedicine. 14:747–754.

Awad AB, Fink CS, Trautwein EA, Ntanios FY. 2005. Beta-Sitosterol stimulates ceramide metabolism in differenti-ated Caco2 cells. J Nutr Biochem. 16:650–655.

Awad AB, von Holtz RL, Cone JP, Fink CS, Chen YC.1998. Beta-Sitosterol inhibits growth of HT-29 humancolon cancer cells by activating the sphingomyelin cycle.Anticancer Res. 18:471–479.

10 A. ALVAREZ-SALA ET AL.

Baskar AA, Ignacimuthu S, Paulraj GM, Al Numair KS.2010. Chemopreventive potential of b-sitosterol in experi-mental colon cancer model – an in vitro and in vivostudy. BMC Complement Altern Med. 10:24.

Block S, Baccelli C, Tinant B, Van Meervelt L, Rozenberg R,Habib Jiwan J-L, Llabr�es G, De Pauw-Gillet MC, Quetin-Leclercq J. 2004. Diterpenes from the leaves of Crotonzambesicus. Phytochemistry. 65:1165–1171.

Bradford PG, Awad AB. 2010. Modulation of signal trans-duction in cancer cells by phytosterols. Biofactors.36:241–247.

Brown JM, Wouters BG. 1999. Apoptosis, p53, and tumorcell sensitivity to anticancer agents. Cancer Res.59:1391–1399.

Chai JW, Kuppusamy UR, Kanthimathi MS. 2008. Beta-sitosterol induces apoptosis in MCF-7 cells. Malays JBiochem Mol Biol. 16:28–30.

Cheng D, Guo Z, Zhang S. 2015. Effect of b-sitosterol onthe expression of HPV E6 and p53 in cervical carcinomacells. Contemp Oncol (Pozn). 19:36–42.

Choi YH, Kong KR, Kim Y-A, Jung K-O, Kil J-H, Rhee S-H, Park K-Y. 2003. Induction of Bax and activation ofcaspases during beta-sitosterol-mediated apoptosis inhuman colon cancer cells . Int J Oncol. 23:1657–1662.

Cilla A, Attanzio A, Barber�a R, Tesoriere L, Livrea MA.2015. Anti-proliferative effect of main dietary phytoster-ols and b-cryptoxanthin alone or combined in humancolon cancer Caco-2 cells through cytosolic Caþ2 – andoxidative stress-induced apoptosis. J Funct Foods.12:282–293.

Csupor-Loffler B, Hajd�u Z, Zupk�o I, Moln�ar J, Forgo P,Vasas A, Kele Z, Hohmann J. 2011. Antiproliferative con-stituents of the roots of Conyza canadensis. Planta Med.77:1183–1188.

Daly TJ, Aherne SA, O’Connor TP, O’Brien NM. 2009.Lack of genoprotective effect of phytosterols and conju-gated linoleic acids on Caco-2 cells. Food Chem Toxicol.47:1791–1796.

Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C,Rebelo M, Parkin DM, Forman D, Bray F. 2015. Cancerincidence and mortality worldwide: sources, methods andmajor patterns in GLOBOCAN 2012. Int J Cancer.136:E359–E386.

Garcia-Llatas G, Cilla A, Alegr�ıa A, Lagarda MJ. 2015.Bioavailability of plant sterol-enriched milk-based fruitbeverages: in vivo and in vitro studies. J Funct Foods.14:44–50.

Garc�ıa-Llatas G, Rodr�ıguez-Estrada MT. 2011. Current andnew insights on phytosterol oxides in plant sterol-enriched food. Chem Phys Lipids. 164:607–624.

Girasolo MA, Attanzio A, Sabatino P, Tesoriere L, RubinoS, Stocco G. 2014. Organotin (IV) derivatives with 5,7-disubstituted-1,2,4-triazolo [1,5-a] pyrimidine and theircytotoxic activities: the importance of being conformers.Inorg Chim Acta. 423:168–176.

Gonz�alez-Larena M, Garc�ıa-Llatas G, Vidal MC, S�anchez-Siles LM, Barber�a R, Lagarda MJ. 2011. Stability of plantsterols in ingredients used in functional foods. J AgricFood Chem. 59:3624–3631.

Granado-Lorencio F, Lagarda MJ, Garc�ıa-L�opez FJ,S�anchez-Siles LM, Blanco-Navarro I, Alegr�ıa A, P�erez-Sacrist�an B, Garcia-Llatas G, Donoso-Navarro E,

Silvestre-Mardomingo RA, et al. 2014. Effect of b-cryp-toxanthin plus phytosterols on cardiovascular risk andbone turnover markers in post-menopausal women: arandomized crossover trial. Nutr Metab Cardiovasc Dis.24:1090–1096.

Grattan BJ. 2013. Plant sterols as anticancer nutrients: evi-dence for their role in breast cancer. Nutrients.5:359–387.

Gylling H, Plat J, Turley S, Ginsberg HN, Ellegard L, JessupW, Jones PJ, L€utjohann D, Maerz W, Masana L, et al.2014. Plant sterols and plant stanols in the managementof dyslipidaemia and prevention of cardiovascular disease.Atherosclerosis. 232:346–360.

Hamdan D, El-Readi MZ, Tahrani A, Herrmann F,Kaufmann D, Farrag N, El-Shazly A, Wink M. 2011.Chemical composition and biological activity of Citrusjambhiri Lush. Food Chem. 127:394–403.

Han XN, Liu CY, Liu YL, Xu QM, Li XR, Yang SL. 2013.New triterpenoids and other constituents from the fruitsof Benincasa hispida (Thunb.) Cogn. J Agric Food Chem.61:12692–12699.

Jayaprakasha GK, Mandadi KK, Poulose SM, Jadegoud Y,Gowda GN, Patil BS. 2007. Inhibition of colon cancercell growth and antioxidant activity of bioactive com-pounds from Poncirus trifoliata (L.) Raf. Bioorg MedChem. 15:4923–4932.

Ju YH, Clausem LM, Alfred KF, Almada AL, Helferich W.2004. b-Sitosterol, b-sitosterol glucoside and a mixture ofb-sitosterol and b-sitosterol glucoside modulate thegrowth of estrogen-responsive breast cancer cells in vitroand in ovariectomized athymic mice. J Nutr.134:1145–1151.

L�opez-Garc�ıa G, Cilla A, Barber�a R, Alegr�ıa A. 2017a.Protective effect of antioxidants contained in milk-basedfruit beverages against sterol oxidation products. J FunctFoods. 30:81–89.

L�opez-Garc�ıa G, Cilla A, Barber�a R, Alegr�ıa A. 2017b.Antiproliferative effect of plant sterols at colonic concen-trations on Caco-2 cells. J Funct Foods. 39:84–90.

Malek SNA, Shin SK, Wahab NA, Yaacob H. 2009.Cytotoxic components of Pereskia bleo (Kunth) DC.(Cactaceae) leaves. Molecules. 14:1713–1724.

Montserrat-de la Paz S, Fern�andez-Arche MA, Berm�udez B,Garc�ıa-Gim�enez MD. 2015. The sterols isolated fromevening primrose oil inhibit human colon adenocarcin-oma cell proliferation and induce cell cycle arrest throughupregulation of LXR. J Funct Foods. 12:64–69.

Rahman SNSA, Wahab NA, Malek SNA. 2013. In vitromorphological assessment of apoptosis induced by anti-proliferative constituents from the rhizomes of Curcumazedoaria. J Evid Based Complementary Altern Med.2013:1.

Rubis B, Polrolniczak A, Knula H, Potapinska O,Kaczmarek M, Rybczynska M. 2010. Phytosterols inphysiological concentrations target multidrug resistantcancer cells. J Med Chem. 6:184–190.

Shahzad N, Khan W, Shadab MD, Ali A, Saluja SS, SharmaS, Al-Allaf FA, Abduljaleel Z, Ibrahim IAA, Abdel-Wahab AF, et al. 2017. Phytosterols as a natural anti-cancer agent: current status and future perspective.Biomed Pharmacother. 88:786–794.

INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 11

Tahsin T, Wansi JD, Al-Groshi A, Evans A, Nahar L,Martin C, Sarker SD. 2017. Cytotoxic properties of thestem bark of Citrus reticulata Blanco (Rutaceae).Phytother Res. 31:1215–1219.

Touillaud MS, Pillow PC, Jakovljevic J, Bondy ML,Singletary SE, Li D, Chang S. 2005. Effect of dietaryintake of phytoestrogens on estrogen receptor status inpremenopausal women with breast cancer. Nutr Cancer.51:162–169.

Woyengo TA, Ramprasath VR, Jones PJH. 2009.Anticancer effects of phytosterols. Eur J Clin Nutr.63:813–820.

Yaacob NS, Yankuzo HM, Devaraj S, Wong JKM, LaiCS. 2015. Anti-tumor action, clinical biochemistryprofile and phytochemical constituents of a pharma-cologically active fraction of S. crispus in NMU-induced rat mammary tumour model. PLoS One.10:e0126426.

12 A. ALVAREZ-SALA ET AL.