Food Chemistry 138 (2013) 1682–1688

Contents lists available at SciVerse ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Cyto-genotoxic and oxidative effects of a continuous UV-C treatment of liquidegg products

Poliana Mendes de Souza a,b,⇑, Karlis Briviba c, Alexandra Müller b, Avelina Fernández a, Mario Stahl b

a Instituto de Agroquímica y Tecnología de Alimentos, Consejo Superior de Insvestigaciones Cientificas, Department of Conservation and Quality,Avda. Agustín Escardino 7, 46980 Paterna, Spainb Max Rubner-Institut, Federal Research Institute of Nutrition and Food, Department of Food Technology and Bioprocess Engineering, Haid-und-Neu-Str. 9, 76131 Karlsruhe, Germanyc Max Rubner-Institut, Federal Research Institute of Nutrition and Food, Department of Physiology and Biochemistry of Nutrition, Haid-und-Neu-Str. 9, 76131 Karlsruhe, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Received 23 May 2012Received in revised form 9 November 2012Accepted 21 November 2012Available online 5 December 2012

Keywords:UV-CNon-thermal processingLiquid egg productsCytotoxicityGenotoxicityLipid and protein oxidation

0308-8146/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.foodchem.2012.11.105

⇑ Corresponding author at: Instituto de AgroquímicConsejo Superior de Insvestigaciones Cientificas, DepQuality, Avda. Agustín Escardino 7, 46980 Paterna, Spa+34 963636301.

E-mail address: poli.souza@iata.csic.es (P. Mendes

UV-C treatment of food is a promising non-thermal processing technology to improve food safety andpreservation. Most of the chemical constituents of food absorb UV-C light that can lead to chemical mod-ifications and quality changes. This work investigated the effects of UV-C treatment of liquid egg productson lipid, protein oxidations and potential cyto- and genotoxic effects on intestinal epithelial cells in vitro.Egg preparations (egg white, yolk, liquid whole egg) were treated with UV-C (254 nm, volumetric dosesbetween 0 and 115,619 J L�1) using a commercial UV-C processing unit equipped with a Dean Flow reac-tor. UV-C treatment at high doses (from 32,181 J L�1, about 2 times higher than that needed to inactivate5log of relevant microorganisms) showed an increased lipid oxidation in egg yolk and slight effects inliquid whole eggs; this was confirmed by slightly but not statistically significant increased peroxide val-ues. UV-C induced also slight protein damage, characterised by the total sulfhydryl group reduction.These UV-C-induced oxidative modifications in egg preparations however did not cause any increasein the cyto- or genotoxic (DNA strand breaks) effects in intestinal Caco-2 cells.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Food safety is one of the most important issues facing the foodmanufacturing and catering industries at any time. Actual consum-ers demand an increasing variety of minimally processed andready-to-eat meals. In this context, the growth potential of food-borne pathogens, such as Salmonella, Campylobacter, Listeria andEscherichia coli 0157:H7 needs to be minimised, therefore theimplementation of Food Safety Management Systems is mandatory(Luning, Bango, Kussaga, Rovira, & Marcelis, 2008). Many technol-ogies that could improve the actual processing methods are beingdeveloped, one such procedure could involve their radiation offood contact surfaces and liquid food with short-wave ultraviolet(UV-C) light. Standard UV equipments are relatively inexpensive,besides some safety precautions the technique is easy to use, andthe radiation is lethal to most types of microorganism. UV radia-tion at 254 nm is known to be germicidal against bacteria, viruses,protozoa, mycelial fungi, yeasts and algae (Koutchma, Keller,Chirtel, & Parisi, 2004; Falguera, Pagán, Garza, Garvín, & Ibarz,

ll rights reserved.

a y Tecnología de Alimentos,artment of Conservation andin. Tel.: +34 963900022; fax:

de Souza).

2011a). The damage induced by UV-C involves specific target mol-ecules leading to lethality by directly altering microbial DNA, e.g.through dimmers formation. Our previous data show that UV-Ctreatment of cloudy apple juice at a dose 3390 J L�1 (Franz, Specht,Cho, Graef, & Stahl, 2009) reduced the numbers of inoculated E. coliDH5a and Lactobacillus brevis LMG 11438 from an initial concen-tration of 106 and 104 to below detectable limits (Franz et al.,2009). Once the DNA has been damaged to a certain magnitude,the microorganisms can no longer reproduce and the risk ofdisease arising from them is minimized or eliminated (Bintsis,Litopoulou-Tzanetaki, & Robinson, 2000).

Besides the improvement in microbiological safety, food pro-cessing needs to pay attention to the effects on food quality andthe potential generation of toxic by-products, and to do not impairundesired reactions. Moreover, it has been proved that this tech-nology can be useful to decompose some toxins that are not af-fected by thermal processing (Falguera, Pagán, Garza, Garvín, &Ibarz, 2011b). For example, furan, a potential carcinogen, can be in-duced by heat from sugars, ascorbic acid, and fatty acids (Fan &Geveke, 2007), while at UV-C doses that could inactivate 5log ofE. coli, very low concentrations (<1 ppb) of furan are induced.

The autoxidation of unsaturated lipids and the lipid peroxida-tion of membrane lipids result in a mixture of hydroperoxidesand chain-cleavage products which are a serious concern for foodquality and consumer health (Berliner, Subbanagounder, Leitinger,

P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688 1683

Watson, & Vora, 2001). Thus, for example, 4-hydroxynonenal, achain-cleavage product from x6 fatty acids, disturbs gap-junctioncommunications in endothelial cells and induces genotoxic effectsin hepatocytes and lymphocytes (Esterbauer, 1993). The productsof fatty acid oxidations and peroxidations are also able to modifylipoproteins, rendering more atherogenic compounds. Fatty acidcomposition is the predominant factor affecting lipid oxidativestability in both bulk oil and emulsion systems (McClements &Decker, 2000; Osborn & Akoh, 2003). Processing parameters, suchas temperature, or exposition to UV light, are also relevant to accel-erate lipid autoxidation and peroxidation reactions. The oxidativestress generated by oxygen free radicals is also able to catalysethe oxidation of proteins. As a consequence, relevant characteris-tics related to protein functionality can be altered. The oxidativedamage can be hindered by natural antioxidants; but due to thehealth risks associated, the derived potential cytotoxic or geno-toxic phenomena should be taken into account.

A number of methods have been developed to measure thein vitro cyto- and genotoxicity. Some of them have already beentested in food matrices; also effects of non-thermal food processinghave been evaluated, finding encouraging results about the reten-tion of antimutagenic activity in high pressure treated foods (Butzet al., 2003), or the innocuous effect of UV-C on the mutagenicity ofmeat products (Sommers, Geveke, Pulsfus, & Lemmenes, 2009).Those tests are accepted by the scientific community and providecomplementary information that approximates realistic scenarios.An extreme treatment may help to estimate the risk potential of awrong commercial processing. Therefore, the oxidative effects ofpasteurising UV-C treatments are evaluated in liquid egg products,and the potential subsequent cytotoxic or genotoxic effects of theoxidation by-products generated by UV-C are assessed.

2. Materials and methods

2.1. Materials

Fresh eggs, from in cage hens laid between March and June2011, were purchased from Gutshof-Ei GmbH (Schackendorf,Germany). They were of yellow shell, and weighed between 55and 61 g. After reception, eggs were inspected for shell integrityand stored under refrigeration at 4 �C. Just before experimentswere carried out, the egg content (separately, egg whites and eggyolks) were removed under aseptic conditions, and collected insterile containers. The chalaza was removed and the separatedegg fractions were then homogenised for 1 min using a commercialblender (31BL44, Waring, USA), at the maximum speed. To preparethe whole egg samples, 13.3 ml of egg yolk were mixed with26.7 ml of egg white; this proportion is the normal average compo-sition of whole egg and was used to attain a constant white to yolkrelationship.

Unless otherwise described, all chemicals were purchased fromSigma–Aldrich (Steinheim, Germany). Eagle’s minimal essentialmedium (EMEM), not essential amino acids (NEAA), phosphatebuffered saline (PBS) without calcium or magnesium (KH2PO4

144 mg l�1; NaCl 9 mg l�1; Na2HPO4 795 mg l�1) physiological pHrange of 7.0–7.6, penicillin/streptomycin were from Lonza (Ver-

Table 1UV-C doses used on the experiments, per cycle.

Egg fraction Flow rate (L h�1) Electrical energy input (Eel) (J L�1)

Egg white 14.48 2238Whole egg 9.48 3417Egg yolk 2.58 12558

viers, Belgium), low melting point agaroseCambrex (Rockland,USA), foetal calf serum (FCS) from PAA (Pashing, Austria).

2.2. Methods

2.2.1. UV-C treatment of egg samplesIn this study, the patented UV-C reactor UVivatec Lab� was pro-

vided by the manufacturer, Bayer Technology Services GmbH (BTS,Leverkusen, Germany). The essential part of the reactor is a helicallywound Teflon tubing (PTFE) wrapped around a quartz glass tubecontaining a 9 Watt UV-C low-pressure mercury lamp with an en-ergy maximum at 253.7 nm (standard UV-C-lamp). Flow rates ofliquids through the reactor can be adjusted from 2 to 20 L h�1 bya peristaltic pump (Schmidt & Kauling, 2007). Flow was adjustedto 14.48 L h�1 for egg white, 9.48 L h�1 for whole egg and2.58 L h�1 for the egg yolk after preliminary studies taking into ac-count the pressure admitted by the pump. The product dose expo-sure depends on the flow rate, irradiation intensity, turbidity, andflow field (Franz et al., 2009; Huch et al., 2010; Schmidt & Kauling,2007). Here, we used the doses of 0, 4214, 7491, 14982, 20133 J L�1

for egg white; 0, 7151, 14303, 28606 and 32181 J L�1 for whole eggand 0, 31533, 47299, 94598 and 115619 J L�1 for egg yolk, based onthe required doses to inactivate 5log of Salmonella subterranea DSM16208, E. coli DH5a and Listeria innocua WS 2258 (8000, 16,000,63,000 J L�1 for egg white, whole egg and egg yolk respectively).To achieve the doses, egg products were recirculated throw the de-vice, being samples homogenised between passes. On the eggwhite, 0, 9, 16, 32 and 43 passes were necessary to process eggwhite at the cited doses. Whole egg was processed for 0, 10, 20,40 and 45, egg yolk were 0, 12, 18, 36 and 44 passes. Table 1 givesan overview of the doses used in this study.

2.2.2. Lipid oxidationTo evaluate the extension of the lipid oxidation on the samples,

the determination of the amount of the formed 2-thiobarbituricacid-reactive substances (TBARS) was undertaken, according toVynckel (1970) and Ramanathan and Das (1992), previouslyadapted and described by Souza and Fernandez (2011). In brief,the samples were freeze-dried in a Genesis Freeze Dryer (SP Virtis,35EL Genesis SQEL85, New York, USA). The TBARS index was esti-mated in a spectrophotometer (Agilent, St. Claire, USA) at 530 nm.The concentrations of TBARS were determined using a standardcurve prepared using malondialdehyde (MDA) and expressed asmg MDA kg�1 of dry sample.

2.2.3. Peroxides valueThe initial peroxide value (PV) of egg yolk and whole egg was

determined by using a modified ferrous iron method as describedby Wang, Jiang, and Hammond (2005). The absorbance was mea-sured at 515 nm with a Unicam UV/VIS spectrometer UV2 (Tokyo,Japan), and the PV was calculated based on a standard curve. Thestandard curve was established as a linear plot of the absorbanceof a series of dilutions of an oxidised oil sample against the micro-equivalents of hydroperoxide in each tube. The PV of the originaloxidised oil was determined by using AOCS Official Method Cd 8-53 (USDA, 1994).

UV-C surface dosage (D) (mJ cm�2) UV-C energy input per volume (J L�1)

43.3 467.9666.1 715.13243.0 2626.38

(a)

(b)

Fig. 1. Influence of UV-C radiation on lipid oxidation of (a) whole egg (b) egg yolk.Results are the mean of three replicates. Bars with different superscripts aresignificantly different (P < 0.05), as determined by the Tukey’s test at the 95%confidence level.

1684 P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688

2.2.4. Protein oxidation: sulfhydryl contentThe concentration of sulfhydryl (SH) groups of the egg fractions

was determined using Ellman’s reagent (5,5-dithiobis (2-nitroben-zoic acid), DTNB) following the method described by Van derPlancken, Van Loey, and Hendrickx (2005) adapted from Beveridge,Toma, and Nakai (1974). Exposed SH groups were determinedusing the Ellman’s reagent. For the determination of the total SHgroups before the addition of the DTNB the sample was dilutedwith SDS–Tris–glycine and kept at 40 �C in a water bath for15 min to allow the protein to unfold and all sulfhydryl groups tobe accessible to DTNB. A molar extinction coefficient of13,600 M�1 cm�1 at 412 nm was used to calculate the amount ofexposed and total sulfhydryl groups (Beveridge et al., 1974). Theamount of buried sulfhydryl groups was calculated by subtractingthe amount of exposed SH groups from the total amount of SHgroups and the sulfhydryl contents are expressed as the percentageof total sulfhydryl groups present in the untreated egg solution.Changes in the sulfhydryl group contents were measured intriplicate.

2.2.5. Cell cultureThe human colon adenocarcinoma cell line Caco-2 was obtained

from the German Collection of Microorganisms and Cell Cultures(Braunschweig, Germany). Cells were routinely cultivated in75 cm2 cell culture flasks from Corning (Corning, USA) in EMEM con-taining 10% (v/v) FCS, 1% NEAA, 1% glutamine, 50 units ml�1penicillinG/50 lg ml�1 streptomycin. Cells were maintained in a humidifiedatmosphere with 5% CO2 at 37 �C. Cell culture medium was replacedthree times a week.

2.2.6. Cytotoxicity: Calcein assayCaco-2 cells were seeded into 96-well cell culture plates

(4500 cells/well) and grown in EMEM containing 10% (v/v) FCS,1% NEAA, 1% glutamine, 50 units ml�1 penicillin G/50 lg ml�1

streptomycin until they reached confluence. Subsequently, cellswere incubated in a 96-well plate (in a humidified atmospherewith 5% CO2 at 37 �C) with egg preparations in the following con-centrations: 2.5%, 5%, 10% and 20% (v/v) in cell culture medium for24 h. The effect of egg preparations on cell viability was deter-mined by the calcein assay according to the manufacturer’s proto-col (Trevigen) with slight modifications. Incubation of cells withhydrochloric acid/methanol in the medium was used as positivecontrol.

2.2.7. Genotoxicity: Comet assayCaco-2 cells were seeded into a 6-well cell culture plates

(105 cells well�1) and grown in EMEM containing 10% (v/v) FCS,1% NEAA, 1% glutamine, 50 units ml�1 penicillin G/50 lg ml�1

streptomycin until they reached confluence. Thereafter, cells wereincubated with egg samples at a concentration of 5% (v/v) in cellculture medium for 24 h. Control cells were incubated in mediumonly. Incubation of cells with hydrogen peroxide (30 lM) in med-ium for 60 min was used as positive control. The trypsinized cellswere then analysed by the Comet assay (single cell gel electropho-resis assay) as described by Briviba et al. (2008). The percentage offluorescence in the tail (tail intensity, %) was assessed fluorimetri-cally (DM 400 B, Leica Microsystems) and then quantified using theimaging software of Perceptive Instruments (Halstead, UK). Resultsare given as mean ± SD.

2.3. Statistical analysis

One-way analysis of variance (ANOVA) and the Tukey–Kramertest were performed with the software XLSTAT-Pro (Win) 7.5.3(Addinsoft, NY). Statistical analysis was run with a confidence levelof 95%.

3. Results and discussion

3.1. Lipid oxidation

A large number of toxic by-products are formed during lipidperoxidation and they have effects at a site away from their gener-ation area. The intermediate and end products of lipid peroxidationmay also be mutagenic and carcinogenic (Marnett, 1999). UV radi-ation is known to be an effective promotor of lipid peroxidationand UV-C treatment of egg products can cause quality deteriora-tion due to oxidation of unsaturated fatty acids and cholesterol.The evaluation of the formation of TBARS provides an estimate tothe extent of lipid oxidation originated by UV-C treatment.

The TBARS values of fresh egg products ranged approx.0.567 ± 0.070 mg MDA kg�1 for whole egg and 0.737 ± 0.047 mgMDA kg�1 for egg yolk, the tests for egg white were not performedsince this fraction do not have lipids. The TBARS values obtained inthis study are comparable to those obtained by Souza andFernandez (2011) for untreated whole egg (0.594 mg MDA kg�1)and egg yolk (0.791 mg MDA kg�1) showing that the sample prep-aration steps did not affect the quality of the initial product.

The TBARS values (expressed as mg MDA kg�1 of dry sample)for UV-C treated liquid egg products are shown in Fig. 1a and b.UV-C induced a slight, but not statistically significant, increase ofthe TBARS values in whole egg homogenates, and a statistically sig-nificant increase in the egg yolk preparations already at the lowestdose studied (31,533 J L�1). At the highest doses applied(32,181 J L�1, or 2.976 J cm�2, for whole egg and 115,619 J L�1, or10.693 J cm�2, on the egg yolk) the TBARS values increased from0.567 ± 0.070 to 0.705 ± 0.041 mg MDA kg�1 on the whole egg,and from 0.737 ± 0.070 to 1.077 ± 0.016 mg MDA kg�1 on the eggyolk. Oxidation is considerably reduced due to the minimization

P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688 1685

in retention time achieved by Dean vortex mixing. Thus, Souza andFernandez (2011) reported higher TBARS values for similar sam-ples in a bench UV equipment operating under dynamic conditionswith doses up to 3.645 J cm�2 (about 30 min UV-C treatment).There are only few reports in the literature about the effects ofthe treatments on the TBARS values of eggs. But it is known thatthe thermal treatment of eggs caused a remarkable increase inthe TBARS, e.g., scrambling and boiling generated 2 times higherTBARS levels in fresh eggs (Ren, 2009). TBARS values also increasedconsiderably in heat processed commercial egg samples; the val-ues increased from 0.17 mg MDA kg�1 to 2.24 mg MDA kg�1, asstudied in the work of Liu, Yang, Lin, and Lee (2005). Also, Souzaand Fernández (2011) reported an increment from 0.7 to 1.6 mgMDA kg�1 in egg yolk after pasteurisation at 61.1 �C for 3.5 min. Li-pid hydroperoxides are intermediates of lipid peroxidation derivedfrom unsaturated fatty acids, phospholipids, glycolipids, choles-terol esters and cholesterol itself. Their formation occurs in enzy-matic or non-enzymatic reactions by reactive oxygen or nitrogenspecies (ROS/RNS). The toxicity of lipid peroxides is well knownand the absence of an endogenous antioxidant enzyme such as glu-tathione peroxidase 4 (GPx4), which is responsible for the reduc-tion of lipid hydroperoxides, in GPx4 knockout mice leads to thedeath of animals on past embryonic day 8 indicating that the re-moval of lipid hydroperoxides is essential for mammalian life(Muller, Lustgarten, Jang, Richardson, & Van Remmen, 2007).

To assess the effects of the UV-C treatments on the oxidativestability of liquid egg products the peroxide value was measuredas an indirect measure of primary oxidation products (Fig. 2a andb). The maximum level for peroxide value of refined oil is10 meq O2 kg�1 (Joint FAO/WHO, 1989).

The UV-C treatment at a dose of 32,181 J L�1 of the whole egghomogenate caused an increase from 0.24 ± 0.09 before irradiationto 0.33 ± 0.05 meq O2 kg�1 after treatment and a UV-C dose of

(a)

(b)

Fig. 2. Peroxide value (a) whole egg (b) egg yolk. Results are the mean of threereplicates. Bars with different superscripts are significantly different (P < 0.05), asdetermined by the Tukey’s test at the 95% confidence level.

115,619 J L�1 from and 0.42 ± 0.17 to 0.59 ± 0.10 meq O2 kg�1 foregg yolk; the tests for egg white were not performed since thisfraction does not have lipids. However, the observed differenceswere not statistically significant. A sample pumped in the absenceof UV-C showed peroxide values equivalent to the control samples(0.24 ± 0.11 for whole egg and 0.42 ± 0.14 meq O2 kg�1 for eggyolk). In accordance, UV-C treatment caused a slight but statisti-cally significant increase of the TBARS values which is in agree-ment with a slight non statistically significant increase of theperoxides values. The primary oxidation processes in lipids derivemainly form hydroperoxides, which are measured by the PV. Ingeneral, the lower the PV, the better the quality of the lipids. How-ever PV decreases as secondary oxidation products appear, whilethe TBARS value is a method to investigate secondary oxidativealdehyde products, usually in PUFA (polyunsaturated fatty acids).

3.2. Protein oxidation

Protein oxidation plays a phytopathological role and may affectthe protein functions relevant in pathological processes. Mainlymethionine residues are oxidised, being also important the oxida-tion of cysteines because of the physiological and phytopathologi-cal role of those amino acids. For example, oxidative modificationsof enzymes have been shown to inhibit a wide array of enzymeactivities (Fucci, Oliver, Coon, & Stadtman, 1983; Stadtman,1990). Oxidative modification of enzymes can have either mild orsevere effects on the cellular or systemic metabolism, dependingon the amount of modified molecules and the chronicity of themodification (Shacter, 2000). Changes in the antioxidative activityof enzymes such as superoxide dismutase might induce oxidativestress and increase the risks associated to pathologies. One of themost commonly measured markers of protein oxidation in biolog-ical samples is the decrease in the sulfhydryl (SH) content.

The concentration of total SH groups in egg white was68 lM g�1. This value is in the range found by Van der Planckenet al. (2005) of 58.5 lM g�1 and the value of 50.7 lM g�1 dryweight observed by Beveridge et al. (1974). In the whole egg, a to-tal amount of SH groups of 47.7 ± 4.2 lM g�1 was obtained, and onegg yolk this value was 3.3 ± 0.6 lM g�1. In egg products the SHgroups are often buried in the protein core and therefore inacces-sible for 5,50-dithio-bis(2-nitrobenzoic acid) (DTNB) as shown bythe low content of exposed SH groups for the untreated samples.As with heat treatments, denatured egg proteins appear (Beveridgeand Arntfield, 1979), and UV-C treatments resulted in a slight de-crease of the total SH concentration. The relative values decreasedto 80.19% on egg white, 91.39% on whole egg and to 95.04% on eggyolk at the highest doses studied (Fig. 3). The amount of exposedSH groups increased only slightly, with differences being more pro-nounced after the application of higher doses, producing dosesabove 20,133 J L�1 and statistically significant effects in egg white.

In egg white, buried SH groups also decreased slightly due toUV-C treatments likely due to a certain protein unfolding and thesubsequent increase in exposed SH groups. One possible mecha-nism for the further reaction of exposed SH groups is the sulfhy-dryl–disulfide (SS) exchange reaction, in which the exposed SHgroups react with the molecule’s own disulfide bonds or that of an-other molecule, or even that of another SS-containing egg protein.Massive SH–SS exchange reactions and SH oxidations would resultin the formation of aggregates and a considerable decrease in pro-tein solubility. This can result in the formation of a protein networkand, depending on the protein, salt concentration and pH, in a gel(Croguennec, Nau, & Brulé, 2002). However, changes induced byUV-C in the concentration of buried and exposed SH groups areminimal if compared to the ones typically observed for heat orpressure treated ovalbumin (Van der Plancken et al., 2005). Conse-quently, UV-C treatments at microbiological reduction levels

(a)

(b)

(c)

Fig. 3. Effect of UV-C treatment on the exposed, masked and total sulfhydrylcontent of (a) egg white (b) whole egg and (c) egg yolk solutions (10% v/v) (Data areshown as mean of 3 repetitions, standard deviation is less than 0.9%).

(a)

(b)

(c)

Fig. 4. Effect of liquid (a) egg white (b) whole egg and (c) egg yolk treated with UV-C on the viability of Caco-2 cells. Confluent Caco-2 cells were incubated with eggpreparations used in the following concentrations: 20, 10, 5, and 2.5% (v/v). Controlcells were incubated in medium only. Data are given as means of 3 repetitions.

1686 P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688

equivalent to heat pasteurizations (data not show) are relativelymild for protein oxidation.

As on the egg white, the SH groups from the whole egg and eggyolk were mainly buried in the protein core and therefore inacces-sible for DTNB when no denaturant (SDS) was applied. The UV-Cinduced unfolding of the whole egg and egg yolk proteins also re-sulted in an increased exposure of buried sulfhydryl groups (Fig. 3band c) which also may take part of SH–SS exchanges.

In general, UV-C induced small changes in SH groups (total, ex-posed, and buried) in liquid egg fractions, but changes were mostpronounced in egg whites, being the characteristic compositionof the egg yolk able to decrease the impact of UV-C on egg proteins.Under those conditions, no gel formation occurred; only turbidsuspensions of protein aggregates were obtained after treatmentat the highest doses, probably due to a certain degree of proteinaggregation induced by the exposed SH groups. UV-C induced low-er protein oxidation levels than the traditional pasteurisationmethod employed.

3.3. Effects on the vitality of the intestinal Caco-2 cells

Ultraviolet radiations at oxidising wavelengths increase the oxi-dative stress due to the formation of ions and free radicals, andmight also accelerate the oxidation of important food components.Regarding the oxidative damage in liquid egg products, the impactof dynamic UV-C treatments is relatively low, but oxidised residuesmight have physiological implications which can be assessedin vitro. The production of furan in other food systems, such asUV-C treated apple juice and cider (Bule et al., 2010; Fan & Geveke,2007) is of main concern, being mandatory to discard cytotoxic ef-fects in UV-C decontaminated food products.

The viability of Caco-2 cells decreased at increasing egg concen-trations. Cell viability was however not significantly affected byincubation for 24 h with 10%, 5% and 2.5% (v/v) (Fig. 4a–c), and sta-tistically significant differences were only found when the Caco-2

Fig. 5. Effect of liquid egg fractions on DNA strand breaks in Caco-2 cells. Cells wereincubated with egg fractions at a concentration of 5% (v/v) in cell culture mediumfor 24 h. Control cells were incubated in medium only. Incubation of cells withhydrogen peroxide (30 lM) in medium for 60 min was used as positive control.(NT) not treated egg fractions; (UV-C) treated with ultraviolet, 29,992 J L�1 for eggwhite, 30,508 J L�1 or whole egg and 149,149 J L�1 for egg yolk. Data are shown asmeans of 3 repetitions. Bars with different superscripts are significantly different(P < 0.05), as determined by the Tukey’s test at the 95% confidence level.

P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688 1687

cells were incubated with egg preparations at a concentration of20% (v/v). Egg yolk showed a slightly higher effect on cell viabilitythan the other liquid egg fractions, probably due to the higher lipidconcentration. Therefore, egg components increasingly affectedcell metabolism, resulting in cellular death. Several mechanismscould be implied. For example, raw egg white contains conalbuminwhich binds iron, additionally, avidin binds to biotin and can im-pair the metabolism of B-vitamins (Pollack, 1958). Low cytotoxiceffects have also been reported for pork meat patties in contactwith Caco-2 cells after a simulated gastric digestion (Kenny,O’Callaghan, & O’Brien, 2008), in agreement with the modesteffects observed in this study for liquid egg products.

Remarkably, differences in cytotoxicity between non-treatedand UV-C-treated egg preparations were not statistically significant(P < 0.5). Consequently, the presence of new cytotoxic compounds,or variations in the concentration of essential food components,cannot be confirmed at the investigated treatment levels inin vitro systems. This is in agreement with the low oxidative defectsfound for lipids and proteins after dynamic UV-C treatments, con-firming the UV-C decontamination of liquid egg products as a feasi-ble application. Further chemical analyses are howeverrecommended to completely discard the presence of contaminants.

3.4. Effects on DNA damage

The influence of diet on carcinogenesis is complex; the cometassay is a relatively simple biomarker to evaluate DNA damageand repair which has been used to study the role of micronutrientsand food components, nutrients and secondary metabolites, in car-cinogenesis (Wasson, 2008). The antioxidant or prooxidant effectsof whole foods have been exhaustively evaluated (Aruoma, 1994;Palozza, Serini, Nicuolo, Piccioni, & Calviello, 2003; Puddey, Zilkens,& Croft, 2003, chap. 2). Some food components, such as the hetero-cyclic amines and polycyclic aromatic hydrocarbons generated incooked meat, increase the levels of strand breaks in a variety of celltypes (Wasson, 2008; Kenny, O’Callaghan, & O’Brien, 2008). Thecomet assay is a standard method to identify food genotoxic effectsand can provide useful data on the effects of UV-C treatments infood.

For liquid egg products, the incubation of cells with non-treatedor UV-treated (20,133 J L�1 for egg white, 32,181 J L�1 for wholeegg and 115,619 J L�1 for egg yolk) egg preparations at a concentra-tion of 5% (v/v) for 24 h has not significantly increased the DNAstrand breaks in Caco-2 cells. Confirming the low effects of UV-Con liquid egg matrices, UV-C-treated egg preparations are non sig-nificantly different (P < 0.05) from non-treated (Fig. 5). However, anumber of biologically active compounds are responsible for anti-oxidant or prooxidant effects, also in eggs. Also, UV-C increasedslightly the amount of peroxides and the TBARS values. Conse-quently, a complete chemical characterization is advisable to dis-card the genotoxic effects of potentially modified individualcomponents in different cell lines, but results in food matrices al-ready suggest that UV-C decontamination treatments produce neg-ligible cyto- or genotoxic effects in Caco-2 cells. This is inagreement with Sommers et al. (2009), who found no UV-C in-duced mutagenesis in Frankfurters containing potassium lactateand sodium diacetate.

UV-C is able to modify a number of chemical food constituentsleading to the formation of new compounds with unknown biolog-ical activities. Not all products of such chemical modifications areknown. In vitro tests allow the investigation of whether new com-pounds with cyto-genotoxic properties are generated during UV-Ctreatment of food at considerable concentrations. Here we showedthat UV-C treatment of egg preparations treated with high UV-Cdoses up to 115,619 J L�1 (10,693 mJ cm�2) did not cause measur-able changes in the cyto- and genotoxic effects of egg preparations.

4. Concluding remarks

Results obtained confirm that UV-C treatment is a promisingand safe technology for the treatment of liquid eggs. TBARS valuesincreased as the radiation doses increased; similar results were ob-tained for protein oxidation. Only for the egg yolk the lipid oxida-tion attained at the higher doses was significantly different fromthe untreated control, although these differences were lower thanto heat treated egg yolks. No significant changes were observed onsulfhydryl groups (protein oxidation). No changes were observedon the whole egg. UV-C did not cause measurable changes in thecytotoxic and genotoxic effects of egg preparations at the applieddose levels.

Acknowledgements

P.M. de Souza thanks the Generalitat Valenciana for Grant num-ber BFPI/2008/230 and BEFPI/2011/027. The authors’ thanks Dr. H.Brod, Dr. M. Poggel, Dr. Pastor and Ms. M. Wübben from BayerTechnology Services for their kind support and supply of the UV-C reactor (UVivatec�). The authors are also grateful for the financialsupport, JC2010-0261.

References

Aruoma, O. I. (1994). Nutrition and health aspects of free radicals and antioxidants.Food and Chemical Toxicology, 32(7), 671–683.

Berliner, J. A., Subbanagounder, G., Leitinger, N., Watson, A. D., & Vora, D. (2001).Evidence for a role of phospholipid oxidation products in atherogenesis. Trendsin Cardiovascular Medicine, 11(3–4), 142–147.

Beveridge, T., Toma, S. J., & Nakai, S. (1974). Determination of SH- and SS-groups insome food proteins using Ellman’s reagent. Journal of Food Science, 39(1), 49–51.

Bintsis, T., Litopoulou-Tzanetaki, E., & Robinson, R. K. (2000). Existing and potentialapplications of ultraviolet light in the food industry – A critical review. Journalof the Science of Food and Agriculture, 80(6), 637–645.

Briviba, K., Bub, A., Möseneder, J., Schwerdtle, T., Hartwig, A., Kulling, S., et al.(2008). No differences in DNA damage and antioxidant capacity betweenintervention groups of healthy, non-smoking men receiving 2, 5 or 8 servings/day of vegetables and fruit. Nutrition and Cancer, 60(2), 164–170.

Bule, M. V., Desai, K. M., Parisi, B., Parulekar, S. J., Slade, P., Singhal, R. S., et al. (2010).Furan formation during UV-treatment of fruit juices. Food Chemistry, 122(4),937–942.

Butz, P., Garcia, A. F., Lindauer, R., Dieterich, S., Bognar, A., & Tauscher, B. (2003).Influence of ultra high pressure processing on fruit and vegetable products.Journal of Food Engineering, 56(2), 233–236.

Croguennec, T., Nau, F., & Brulé, G. (2002). Influence of pH and salts on egg whitegelation. Journal of Food Science, 67(2), 608–614.

1688 P. Mendes de Souza et al. / Food Chemistry 138 (2013) 1682–1688

Esterbauer, H. (1993). Cytotoxicity and genotoxicity of lipid-oxidation products.American Journal of Clinical Nutrition, Bethesda, 57(5), S779–S786 [Supplement].

Falguera, V., Pagán, J., Garza, S., Garvín, A., & Ibarz, A. (2011a). Ultraviolet processingof liquid food: A review. Part 1: Fundamental engineering aspects. Food ResearchInternational, 44(6), 1571–1579.

Falguera, V., Pagán, J., Garza, S., Garvín, A., & Ibarz, A. (2011b). Ultraviolet processingof liquid food: A review: Part 2: Effects on microorganisms and on foodcomponents and properties. Food Research International, 44(6), 1580–1588.

Fan, X., & Geveke, D. J. (2007). Furan formation in sugar solution and apple ciderupon ultraviolet treatment. Journal of Agricultural and Food Chemistry, 55(19),7816–7821.

Franz, M. A. P. C., Specht, I., Cho, G. S., Graef, V., & Stahl, M. (2009). UV-C-inactivation of microorganisms in naturally cloudy apple juice using novelinactivation equipment based on dean vortex technology. Food Control, 20(12),1103–1107.

Fucci, L., Oliver, C. N., Coon, M. J., & Stadtman, E. R. (1983). Inactivation of keymetabolic enzymes by mixed-function oxidation reactions: Possible implicationin protein turnover and ageing. Proceedings of the national academy of sciencesUSA, 80(6), 1521–1525.

Huch, M., Mueller, A., Strohhaecker, J., Vogt, S., Hanak, A., Graef, V., et al. (2010). Useof UV-C treatment for the inactivation of microorganisms. Fruit Processing,20(5), 196–201.

Joint FAO/WHO. (1989). Food standard program recommended internationalstandards. London: Joint FAO/WHO.

Kenny, O., O’Callaghan, Y., & O’Brien, N. M. (2008). Effects of ingredientincorporation into sausage meat on the micellarisation and uptake of a-tocopherol by Caco-2 human intestinal cells. Food Science and TechnologyInternational, 14(1), 79–86.

Koutchma, T. B., Keller, S., Chirtel, S., & Parisi, B. (2004). Ultraviolet disinfection ofjuice products in laminar and turbulent flow reactors. Innovative Food Scienceand Emerging Technologies, 5(2), 179–189.

Liu, L.-Y., Yang, M.-H., Lin, J.-H., & Lee, M.-H. (2005). Lipid profile and oxidativestability of commercial egg products. Journal of Food and Drug Analysis, 13(1),78–83.

Luning, P. A., Bango, L., Kussaga, J., Rovira, J., & Marcelis, W. J. (2008). Comprehensiveanalysis and differentiated assessment of food safety control systems: Adiagnostic instrument. Trends in Food Science & Technology, 19(10), 522–534.

Marnett, L. J. (1999). Lipid peroxidation-DNA damage by malondialdehyde.Mutation Research, 424(1–2), 83–95.

McClements, D. J., & Decker, E. A. (2000). Lipid oxidation in oil-in-water emulsions:Impact of molecular environment on chemical reactions in heterogeneous foodsystems. Journal of Food Science, 65(8), 1270–1282.

Muller, F. L., Lustgarten, M. S., Jang, Y., Richardson, A., & Van Remmen, H. (2007).Trends in oxidative aging theories. Free Radical Biology & Medicine, 43(4),477–503.

Osborn, H. T., & Akoh, C. C. (2003). Effect of natural antioxidants on iron catalyzedlipid oxidation of structured lipid-based emulsions. Journal of the American OilChemists’ Society, 80(9), 847–852.

Palozza, P., Serini, S., Nicuolo, F. D., Piccioni, E., & Calviello, G. (2003). Prooxidanteffects of b-carotene in cultured cells. Molecular Aspects of Medicine, 24(6),353–362.

Pollack, O. J. (1958). Serum cholesterol levels resulting from various egg diets:experimental studies with clinical implications. Journal of the AmericanGeriatrics Society, 6(8), 614–618.

Puddey, I. B., Zilkens, R. R., & Croft, K. D. (2003). Antioxidant and pro-oxidant effectsof alcoholic beverages: Relevance to cardiovascular disease. In R. R. Watson & V.R. Preedy (Eds.), Nutrition and alcohol: Linking nutrient interactions and dietaryintake (pp. 19–40). London: CRC Press.

Ramanathan, L., & Das, N. P. (1992). Studies on the control of lipid oxidation inground fish by some polyphenolic natural products. Journal of Agricultural andFood Chemistry, 40(1), 17–21.

Ren, Y. (2009). Oxidative stability of omega-3 polyunsaturated fatty acids enrichedeggs. Master’s thesis. Master of Science in food science and technology.Department of agricultural, food and nutritional sciences. University ofAlberta. Edmonton, Alberta, Canada.

Shacter, E. (2000). Quantification and signification of protein oxidation in biologicalsamples. Drug Metabolism Reviews, 32(3–4), 307–326.

Schmidt, S., & Kauling, J. (2007). Process and laboratory scale UV inactivation ofviruses and bacteria using an innovative coiled tube reactor. ChemicalEngineering and Toxicology, 30(7), 945–950.

Sommers, C. H., Geveke, D. J., Pulsfus, S., & Lemmenes, B. (2009). Inactivation ofListeria innocua on frankfurters by ultraviolet light and flash pasteurization.Journal of Food Science, 74(3), M138–M141.

Souza, P. M., & Fernandez, A. (2011). Effects of UV-C on physicochemical qualityattributes and Salmonella enteritidis inactivation in liquid egg products. FoodControl, 22(8), 1385–1392.

Stadtman, E. R. (1990). Metal ion-catalyzed oxidation of proteins: Biochemicalmechanism and biological consequences. Free Radical Biology and Medicine, 9(4),315–325.

Van der Plancken, I., Van Loey, A., & Hendrickx, M. E. G. (2005). Changes insulfhydryl content of egg white proteins due to heat and pressure treatment.Journal of Agricultural and Food Chemistry, 53(14), 5726–5733.

Vynckel, W. (1970). Direct determination of the thiobarbituric acid value intricloroacetic acid extracts of fish as a measure of oxidative rancidity. FetteSeifen Anstrichmittel, 72(12), 1084–1087.

Wang, T., Jiang, Y., & Hammond, E. G. (2005). Effect of randomization on theoxidative stability of corn oil. Journal of the American Oil Chemists’ Society, 82(2),111–117.

Wasson, R. R. (2008). Complementary and alternative therapies and the agingpopulation. San Diego: Academic Press, p. 597.