Investigation of the cytotoxicity of food-grade nanoemulsions in Caco-2 cell monolayers and HepG2 cells

Download Investigation of the cytotoxicity of food-grade nanoemulsions in Caco-2 cell monolayers and HepG2 cells

Post on 31-Dec-2016

216 views

Category:

Documents

3 download

TRANSCRIPT

<ul><li><p>Food Chemistry 141 (2013) 2933Contents lists available at SciVerse ScienceDirect</p><p>Food Chemistry</p><p>journal homepage: www.elsevier .com/locate / foodchemInvestigation of the cytotoxicity of food-grade nanoemulsions in Caco-2cell monolayers and HepG2 cells0308-8146/$ - see front matter 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.foodchem.2013.03.009</p><p> Corresponding author. Tel.: +1 732 932 7193; fax: +1 732 932 6776.E-mail address: qhuang@aesop.rutgers.edu (Q. Huang).</p><p>1 Current address: CP Kelco, 8225 Aero Drive, San Diego, CA 92123 USA.Hailong Yu 1, Qingrong Huang Department of Food Science, Rutgers, The State University of New Jersey, 65 Dudley Road, New Brunswick, NJ 08901, USA</p><p>a r t i c l e i n f oArticle history:Received 14 September 2012Received in revised form 15 January 2013Accepted 4 March 2013Available online 14 March 2013</p><p>Keywords:NanoemulsionCytotoxicityCaco-2 cell monolayersHepG2 cellsa b s t r a c t</p><p>Nanoemulsions represent one of the emerging formulations for nutraceutical delivery. However, the pos-sible toxicity associated with the small droplet size (diameter </p></li><li><p>30 H. Yu, Q. Huang / Food Chemistry 141 (2013) 2933for in vitro assessment of hepatic toxicity (Bort, Ponsoda, Jover,Gomez-Lechon, &amp; Castell, 1999; Lu &amp; Cederbaum, 2006).</p><p>In the present study, the possible cytotoxicity of food-gradenanoemulsions was investigated by using two in vitro cell culturesystems and comparing with micron-sized emulsions with thesame composition. To mimic the exposure of the small intestineto the nanoemulsions, Caco-2 cell monolayers were treated withnanoemulsions and the cell membrane leakage and tight junctionintegrity were examined. To investigate the toxicity on the liver,HepG2 cells were treated with nanoemulsions and micron-sizedemulsions and the cell proliferation/viability rate were compared.These studies may provide the first set of evidence for the possibletoxicity of food-grade nanoemulsions.2. Materials and methods</p><p>2.1. Materials</p><p>Medium chain triacylglycerol (MCT) was provided by StepanCompany (NEOBEE 1053). Modified starch was obtained from Na-tional Starch (HiCap 100). Tween 20 was purchased from FisherScientific. Whey protein isolate (WPI) was from Davisco FoodsInternationals (BiPro).</p><p>Dulbeccos Modified Eagle Medium (DMEM), Minimum Essen-tial Medium Eagle (MEM), Hanks Buffered Salt Solution (HBSS),foetal bovine serum (FBS), 100 non-essential amino acids, 100penicillin and streptomycin and bovine serum albumin (BSA) wereall purchased from Fisher Scientific. Transwell permeable polycar-bonate inserts (0.4 lm) were obtained from Corning.2.2. Preparation of micron-sized and nanoemulsions</p><p>Ten grammes of emulsions, consisting of 8.5 g dH2O, 1.0 g MCTand 0.5 g emulsifiers (modified starch, Tween 20 or WPI), werebriefly mixed by magnetic stirring. To prepare micron-sized emul-sions, the mixture was homogenised at 6500 rpm for 5 min with ahigh speed homogenizer (ULTRATURRAX T-25 basic, IKA Works).Nanoemulsions with the same composition were generated byultrasonication at about 175W for (accumulatively) 5 min. ThepH of modified starch-emulsion was acidic (between 4 and 5),and that of WPI and Tween 20 emulsions was nearly neutral. Sincethe emulsions were greatly diluted in a buffer system during celltreatment, the pH of emulsion did not affect the pH of the cell cul-ture system.2.3. Measurement of the oil droplet size of emulsions</p><p>Photographs of diluted micron-sized emulsions and nanoemul-sion were taken with a Nikon TE-2000U inverted microscope. Thesurface-averaged diameters of micron-sized emulsions were deter-mined by using ImageJ software, to analyse nine photographs. Par-ticle size of nanoemulsions was measured with a DLS-based BIC90plus particle size analyzer equipped with a Brookhaven BI-9000ATdigital correlator (Brookhaven Instrument) at a fixed scattering an-gle of 90 at ambient temperature.2.4. Maintenance of Caco-2 and HepG2 cell cultures</p><p>Caco-2 cells (passage 3545) were maintained in DMEM with10% FBS, 1 non-essential amino acids and 1 penicillin and strep-tomycin. HepG2 cells (passage 717) were cultured in MEM with10% FBS, 1 penicillin and streptomycin. Cells were incubated at37 C with 5% CO2.2.5. Generation and treatment of Caco-2 cell monolayers</p><p>To generate Caco-2 cell monolayers, 0.5 ml of 6 105 cell/mlCaco-2 cells was plated onto the inserts (apical compartment) of12-well plates. Subsequently, 1.5 ml of culture media were addedto the lower compartment of each well. Media were changed everytwo days. Cytotoxicity experiments were performed after2129 days of plating.</p><p>Before the experiments, Caco-2 cell monolayers were washedwith and kept in HBSS (0.5 ml in apical compartment and 1.5 mlin basolateral compartment) for 2030 min at 37 C before thetreatment.</p><p>Five microlitres of micron-sized emulsions or nanoemulsionswere directly added in the apical compartment to make 1100dilution. Then the Caco-2 cell monolayers were kept in a platformshaker set at 100 rpm at 37 C for 2 h.2.6. Lactate dehydrogenase leakage assay</p><p>After treatment of Caco-2 cell monolayers with nano- and mi-cron-sized emulsions, 50 ll of media from the apical chamber wereremoved and the lactate dehydrogenase (LDH) leakage was deter-mined using a CytoTox-ONE Homogeneous Membrane IntegrityAssay kit (Promega). One-hundred-time diluted emulsions in HBSSwere used as the negative control. Cell lysates of the Caco-2 cellmonolayers were used as the positive control. Percentage of rela-tive LDH leakage was calculated as:</p><p>% relative LDH leakage FItreatment FInegative controlFIpositive control FInegative control 100%</p><p>1</p><p>where FI stands for fluorescence intensity with excitation wave-length at 560 nm and emission wavelength at 590 nm, as describedin the kit.2.7. Measurement of transepithelial electrical resistance (TEER)</p><p>Immediately before the emulsion treatment, the TEER acrossCaco-2 cell monolayers was measured, using the Evom2 epithelialvoltmeter (World Precision Instrument). After treatment, Caco-2cell monolayers were washed with and incubated in HBSS for2030 min, before the post-treatment TEER measurement.</p><p>Changes in TEER were expressed as % relative TEER.</p><p>% relative TEER post-treatment TEERpre-treatment TEER</p><p> 100% 22.8. MTT assay on HepG2 cells</p><p>The MTT assays on HepG2 cells were performed using the pre-viously published procedures (Yu &amp; Huang, 2010). Briefly, cellswere plated at the density of 1000 cell/well and treated with differ-ent dilutions of nanoemulsion or micron-sized emulsions (from1:100 to 1:1600) for about 24 h, followed by incubation withMTT and UVVis absorption measurements.2.9. Statistics analysis</p><p>One-way and two-way analyses of variance (ANOVA) were per-formed, using SigmaPlot 10.0 with SigmaStat integration (Systatsoftware).</p></li><li><p>Table 1Summary of the particle size of prepared nanoemulsions and micron-sized emulsions.</p><p>Micron-sized emulsion Nanoemulsion</p><p>D2,0 (lm) Diameter (nm) Polydispersity</p><p>Modified starch 10.3 1.0 155 1.4 0.272 0.002Tween 20 5.7 0.8 168.2 0.7 0.303 0.014WPI 8.7 0.9 172.7 1.8 0.271 0.009</p><p>Data for micron-sized emulsion are presented as means standard deviation, n = 9.Data for nanoemulsion are shown as means standard error, n = 3.</p><p>H. Yu, Q. Huang / Food Chemistry 141 (2013) 2933 313. Results</p><p>3.1. General</p><p>In this work, the possible cytotoxicity of nanoemulsions wasexamined by comparing the cellular response to the treatment ofnanoemulsions and micron-sized emulsions, assuming that mi-cron-sized emulsions are not toxic.</p><p>3.2. Generation of micron-sized emulsions and nanoemulsions</p><p>To investigate the size effect of food-grade emulsions on thecytotoxicity, three micron-sized emulsions and nanoemulsionswith the same compositions were prepared by high speed homog-enisation and ultrasonication, respectively. Modified starch, Tween20 and whey protein isolates (WPI) were chosen to representdifferent types of emulsifiers. As shown in Fig. 1 and Table 1, forall the micron-sized emulsions, the particle size (diameter) of theoil droplets was between 5 and 10 lm. In comparison, the averageoil droplet size in the nanoemulsions was less than 200 nm, repre-senting about 50-fold reduction. In Fig. 1, the photographs for thenanoemulsions only show the oil droplets in the tail region of theparticle size distribution, and the majority of nano-sized oil drop-lets were not possible to visualise with optical microscopy.</p><p>3.3. Examination of the cell membrane integrity on Caco-2 cellmonolayers</p><p>Caco-2 cell monolayers were treated with micron-sized emul-sions and nanoemulsions to mimic the exposure of the small intes-Fig. 1. Micrographs of micron-sized emulsions and nanoemulsionstine epithelium to the emulsions. After treatment, the cellmembrane integrity was examined by detecting LDH leakage(Fig. 2). For untreated Caco-2 cell monolayers, less than 3% of cellshad cell membrane rupture throughout the experimental proce-dures. For the monolayers treated with nanoemulsions andmicron-sized emulsions, there were about 36% of cells with LDHleakage. From one-way ANOVA, it was revealed that there wasno significant difference in the LDH leakage between untreatedcells and most of the emulsion treatment, except for the nano-emulsion made with modified starch. More importantly, no signif-icant difference was detected between all three pairs of themicron-sized emulsions and nanoemulsions, suggesting that nano-emulsions did not possess more cytotoxicity on the Caco-2 cellmonolayers than did regular micron-sized emulsions.</p><p>3.4. Investigation of the tight junction integrity in the Caco-2 cellmonolayers</p><p>After well differentiation, tight junctions form between adja-cent Caco-2 cells, similar to those in the small intestine epithelium.made with modified starch, Tween 20 and WPI, respectively.</p></li><li><p>1:100a 1:200a 1:4 00a 1:8 00a 1:16 00a</p><p>% re</p><p>lativ</p><p>e ce</p><p>ll vi</p><p>abili</p><p>ty </p><p>0</p><p>20</p><p>40</p><p>60</p><p>80</p><p>100</p><p>120</p><p>140micron-sized emulsionA </p><p>nanoemulsionB modified starch</p><p>ll vi</p><p>abili</p><p>ty </p><p>100</p><p>120</p><p>140micron-sized emulsionA </p><p>nanoemulsionA tween-20</p><p>untreated Modified starch Tween 20 WPI</p><p>% r</p><p>elat</p><p>ive </p><p>LD</p><p>H le</p><p>akag</p><p>e</p><p>0</p><p>2</p><p>4</p><p>6micron-sized emulsionnanoemulsion</p><p>a,a',a"</p><p>a,b</p><p>b</p><p>a'</p><p>a'</p><p>a"</p><p>a"</p><p>Fig. 2. LHD leakage of Caco-2 cell monolayers untreated, or treated with differentmicron-sized emulsions or nanoemulsions. Data are presented as means standarddeviation (n = 3). Different letters indicate significant difference (p &lt; 0.05).</p><p>32 H. Yu, Q. Huang / Food Chemistry 141 (2013) 2933Disruption of the tight junction may make the cell monolayer morepermeable to water-soluble (toxic) compounds, which should beavoided in the normal physiological condition. In this study, theTEER values of Caco-2 cell monolayers treated with nanoemulsionswere compared to that with micron-sized emulsions and un-treated. As shown in Fig. 3, one-way ANOVA revealed that, com-pared with no treatment, modified starch and WPI emulsionscaused TEER decrease, and that Tween 20 emulsion had no appar-ent effect on the tight junction integrity. These observations areconsistent with other groups finding that some types of emulsifi-ers and surfactants may interrupt the tight junctions (Ohno, Naga-numa, Ogawa, &amp; Muramoto, 2006; Yamashita et al., 2000). Moreimportantly, comparing the three nanoemulsions with the corre-sponding micron-sized emulsions, respectively, there was no sig-nificant difference in terms of the TEER change. With theassumption that regular micron-emulsions are considered safe,nanoemulsions did not display additional disruption of the tightjunctions.</p><p>Together with the results of the LDH leakage in the previoussection, it was illustrated that, compared with micron-sized emul-sions, no elevated toxicity was detected for nanoemulsions onCaco-2 cell monolayers, suggesting that nanoemulsions may notbe toxic to the small intestine epithelium.untre</p><p>ated</p><p>Modif</p><p>ied sta</p><p>rch</p><p>Tween</p><p> 20 WPI</p><p>% r</p><p>elat</p><p>ive </p><p>TE</p><p>ER</p><p>0</p><p>20</p><p>40</p><p>60</p><p>80</p><p>100</p><p>120</p><p>140micron-sized emulsionnanoemulsion</p><p>a,a',a"</p><p>b b</p><p>a'a' b" b"</p><p>Fig. 3. Relative TEER of Caco-2 cell monolayers untreated, or treated with differentmicron-sized emulsions or nanoemulsions. Data are presented as means standarddeviation (n = 3). Different letters indicate significant difference (p &lt; 0.05).3.5. Examination of the cytotoxicity on HepG2 cells</p><p>To examine the possible toxicity of the nanoemulsions on theliver, HepG2 cells were used as a model and treated with mi-cron-sized emulsions and nanoemulsions of different dilutions(Fig. 4). From two-way ANOVA, it was revealed that nanoemulsionsmade with modified starch and WPI caused less cell proliferation/viability than did the corresponding micron-sized emulsions(p = 0.0105 and p = 3.21 105, respectively). On the other hand,Tween 20 nanoemulsion did not affect the cell proliferation/viabil-ity more than did its micron-sized counterpart (p = 0.231). Theseresults suggested that the toxicity of nanoemulsions may bedependent on the emulsifiers, and more in vivo studies arerequired to examine the hepatotoxicity of nanoemulsions.4. Discussion</p><p>In this work, three nanoemulsions were prepared and theirpossible toxicity was examined in vitro. It was found that, com-pared with micron-sized emulsions, nanoemulsions did not affectthe cell membrane and tight junction integrity of Caco-2 cell mon-olayers. However, nanoemulsions formed by modified starch andWPI were found to affect the proliferation/viability of HepG2 cells.1:100a 1:200a 1:4 00a 1:8 00a 1:16 00a</p><p>% re</p><p>lativ</p><p>e ce</p><p>0</p><p>20</p><p>40</p><p>60</p><p>80</p><p>dilution1:100a 1:200a,b</p><p>% re</p><p>lativ</p><p>e ce</p><p>ll vi</p><p>abili</p><p>ty </p><p>0</p><p>20</p><p>40</p><p>60</p><p>80</p><p>100</p><p>120</p><p>140micron-sized emulsionA </p><p>nanoemulsionBWPI</p><p>1:4 00a,b 1:8 00b 1:16 00b</p><p>Fig. 4. Cytotoxicity of nanoemulsions on HepG2 cells. Data are presented asmeans standard deviation (n = 4). Different letters indicate significant difference(p &lt; 0.05).</p></li><li><p>H. Yu, Q. Huang / Food Chemistry 141 (2013) 2933 33The aim of this work was focussed on food-grade nanoemul-sions. The exposure routes of the nanoemulsions were thus limitedto the digestive tract, especially the small intestine, considering itslarge surface area, and possibly the liver, to where digested and ab-sorbed food are carried for possible metabolism and storage. Basedon these considerations, Caco-2 cell monolayers and HepG2 cellswere chosen to mimic the small intestine epithelium and the liverhepatocytes, respectively. Meanwhile, micron-sized emulsionswere prepared in parallel and served as the control, since it is gen-erally accepted that regular (micron-sized) emulsions do not bearany toxicity to the human body.</p><p>In the two experiments on Caco-2 cell monolayers, compared tomicron-sized emulsions, nanoemulsions did not reveal apparenttoxicity on Caco-2 cell monolayers, suggesting that nanoemulsionsmay not bear significant toxicity to the small intestine. Actually, itis expected that food-grade nanoemulsions undergo digestion(once ingested) along with other food components and, becauseof their very large surface areas, nanoemulsions are digested muchfaster than regular micron-sized emulsions (Li &amp; McClements,2010; Yu &amp; Huang, 2012). In the in...</p></li></ul>

Recommended

View more >