Epidemiological studies have shown that regular consumption of Brassica vegetables is associated with a reduced risk of some chronic diseases; in particular, gastric or lung cancers and CVD(Reference Mukherjee, Gangopadhyay and Das1, Reference Kim and Park2). Brassica vegetables contain a wide range of natural chemopreventive agents, such as folate, fibre, vitamin C, tocopherols, polyphenols, carotenoids and chlorophylls(Reference Lynn, Collins and Fuller3–Reference Podsedek6). Furthermore, cruciferous vegetables are a rich source of glucosinolates, sulphur-containing compounds, whose hydrolysis results in the formation of biologically active compounds, including indoles and isothiocyanates(Reference Higdon, Delage and Williams7–Reference Song and Thornalley9). All these molecules can retard the development and progression of precancerous cells into malignant cells(Reference Surh10) acting through different mechanisms and in different compartments, but they are thought to be mainly free radical scavengers(Reference Berger11).
It is known that cooking induces significant changes in chemical composition, influencing the concentration and availability of bioactive compounds in vegetables. Both positive and negative effects have been reported depending upon the differences in process conditions and morphological and nutritional characteristics of the vegetable species(Reference Podsedek6, Reference Nicoli, Anese and Parpinel12–Reference Miglio, Chiavaro and Visconti16). The cooking of vegetables can decrease water-soluble and heat-sensitive nutrients, such as vitamin C, but also can lead to disruption of the food matrix and dissociation of some compounds from the plant matrix components. This determines the release of plant components, which, in turn, can improve their digestion and absorption, increasing their bioavailability(Reference Link and Potter17–Reference Bugianesi, Salucci and Leonardi19).
Although some of the vegetables consumed in the human diet are processed to be eaten, there are few works that correlate their biological activity on human cell lines in vitro to cooking treatments. Recently, the effect of grilling and boiling processes on eggplant showed a positive effect on antioxidant concentrations and chemopreventive activities of the human polymorphonuclear neutrophils(Reference Lo Scalzo, Fibiani and Mennella20).
There is an increasing attention in the evaluation of the antiproliferative effects induced by vegetable extracts on human cancer cell lines, performed mainly by cytotoxicity tests such as 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assays(Reference Heo, Chon and Park21, Reference Sharif, Auger and Alhosin22). Furthermore, an interesting method for studying the effects of foods on the risk of cancer is the Comet assay, a single-cell gel electrophoresis technique for measuring DNA breakage in individual cells(Reference Cemeli, Baumgartner and Anderson23). The comet assay is performed mainly to evaluate the effects of antioxidants in vitro (Reference Wasson, McKelvey-Martin and Downes24, Reference Wong, Szeto and Collins25) or in intervention studies(Reference Riso, Martini and Visioli26).
The goal of the present study was to evaluate the effect of common cooking practices (i.e. boiling, microwaving and steaming) on the biological activities of broccoli, cauliflower and Brussels sprouts. The anti-proliferative activity and chemoprotective effect towards DNA oxidative damage of fresh and cooked vegetable extracts were evaluated by MTS and Comet assays.
Experimental methods
Chemicals
All chemicals were of analytical grade or they complied with the standards required for tissue culture experiments. Reagents for electrophoresis, normal melting point (1 %) and low melting point (0·7 %) agarose, dimethylsulfoxide, ethidium bromide, ascorbic acid and general laboratory chemicals were from Sigma-Aldrich Company Limited (Milan, Italy). The cell culture medium and reagents were from BioWhittaker (Lonza, Milan, Italy). Stabilised H2O2 (3 %) was from Zeta Farmaceutici (Vicenza, Italy).
Vitamin C: stock solutions of 100 mm ascorbic acid were prepared fresh in a modified Krebs buffer, containing 137 mm-NaCl, 5·4 mm-KCl, 2·8 mm-CaCl2, 1 mm-MgSO4, 0·3 mm-NaH2PO4, 0·3 mm-KH2PO4, 10 mm-glucose, 10 mm-HEPES and 10 mm-Tris, pH 7·4. Dilutions of these stock solutions were made directly in the cell culture media.
Preparation of vegetables
Freshly harvested broccoli (Brassica oleracea L. cv. Italica), Brussels sprouts (B. oleracea L. cv. gemmifera) and white cauliflower (B. oleracea L. cv. botrytis cauliflora) of a single batch were analysed. Fresh Brussels sprouts were purchased from a local market on the day of processing after road transport under refrigerated conditions within 24 h of harvest and stored at 7°C for up to 1 d before they were taken. Fresh broccoli and cauliflower were obtained by a local purchaser and analysed within 3 d of harvest. The samples were stored at 4°C before analysis.
Fresh broccoli and cauliflower were cleaned by removing of inedible parts and then chopped into homogeneous pieces, leaving a stem of 2·5 cm. Brussels sprouts were deprived of the external leaves.
To obtain more homogeneous samples, each vegetable was prepared in batches of 500 g. Each batch was then divided into five equal portions. One portion was retained raw; the others were cooked in four different methods in triplicate, as given below.
Cooking treatments
Cooking conditions were optimised by preliminary experiments carried out for each vegetable. For all cooking treatments, the minimum cooking time to reach tenderness for adequate palatability and taste, according to Italian eating habits, were used.
Boiling
Brassica vegetables were added to boiling tap water in a covered stainless steel pot (1:5, food/water) and cooked on a moderate flame. Cooking time was 8, 10 and 10 min for broccoli, Brussels sprouts and cauliflower, respectively. For each cooking trial, ten samples were boiled. Then, the samples were drained off for 30 s.
Steaming
Two types of steaming equipment were employed: an air/steam impingement oven and a domestic cooker equipped with a mesh basket. Air/steam oven treatments were carried out in a Combi-Steal SL oven (V-Zug, Zurich, Switzerland). Nine specimens of vegetables were placed in the oven equilibrated to room temperature before each cooking trial. Eight samples were arranged in a circle, and one put at the centre to ensure uniform heating conditions in all samples for each cooking trial. Cooking time was 13, 17 and 13 min for broccoli, Brussels sprouts and cauliflower, respectively. The samples were put into the oven when a temperature of 100°C was reached (displayed by the apparatus).
A single layer of nine specimens of Brassica vegetables was steamed in a domestic closed vessel using a stainless steel steam basket suspended above a small amount of boiling water. Cooking time was 15, 18 and 11 min for broccoli, Brussels sprouts and cauliflower, respectively.
Microwaving
Microwave treatments were carried out in a domestic microwave oven (Samsung Electronics Company Limited, Paldal-Gu Suwon Kyungki-Do, Korea) without water. Ten specimens of vegetables were exposed at a frequency of 2450 Hz at low power (300 W) on the rotatory turntable plate of the oven. Cooking time was 30, 18 and 30 min for broccoli, Brussels sprouts and cauliflower, respectively.
After all cooking experiments, the samples were cooled rapidly on ice and then freeze-dried utilising a Brizzio-Basi instrument (Milan, Italy). The dried sample material was finely ground, kept in sealed bags, and stored at − 20°C until the analysis.
Vegetable extract preparation
Chemical analyses of the phytochemical compound content in fresh and cooked samples were performed before the preparation of the extracts. For a complete description see Pellegrini et al. (Reference Pellegrini, Chiavaro and Gardana27). A synthesis of the results is reported in Fig. 1.
Fig. 1 Phytochemical compounds of raw and cooked Brassica specimens, values are expressed as mg/100 g of dry weight (dw; modified from Pellegrini et al. (Reference Pellegrini, Chiavaro and Gardana27)). (a) Glucosinolates; (b) ascorbic acid and (c) phenol compounds. 
, Cauliflower; 
, Brussels sprouts; 
, broccoli.
Aqueous extracts were prepared by mixing the dried samples with Dulbecco's modified Eagle's medium (DMEM; supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum). The suspensions were further subjected to Stomaker treatment for 6 min, a centrifugation step (20 min, 4°C and 15 000 g) and the supernatants were sterile-filtered (0·20 μm syringe filter; SARSTEDT, Nümbrecht, Germany) and divided into aliquots (stored at − 20°C). The relative concentration of each sample, equivalent of wet weight g/ml (g eq ww/ml), was calculated by dividing the wet weight for the volume (ml) of the extract obtained after the centrifugation and filtration steps(Reference Laky, Knasmüller and Gminski28, Reference Kassie, Laky and Gminski29).
Cell lines
The colon is the major cancer site thought to be protected by vegetables, so for in vitro studies we employed a cell line, HT-29, used widely as a model for colon cancer.
The human colon adenocarcinoma cell line HT-29 was kindly obtained from the Northern Ireland Centre for Food and Health. Prior to the experiments, the cells were thawed and grown in tissue culture flasks as a monolayer in DMEM, supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum at 37°C in a humidified (95 %) CO2 (5 %) incubator. The cultured cells were trypsinised with trypsin/EDTA for a maximum of 5 min and seeded with a subcultivation ratio of 1:3–1:8. The cell numbers and viabilities were determined with the trypan blue exclusion test.
Antiproliferative activity
Cell viability was detected by CellTiter96R AQueous One Solution Cell Proliferation Assay (Promega Corporation, Madison, WI, USA). This test contains a tetrazolium compound (MTS, inner salt) and an electron coupling reagent (phenazine ethosulfate). The MTS tetrazolium compound is bioreduced by cells into a coloured formazan product that is soluble in a tissue culture medium. This conversion is accomplished by NADPH or NADH produced by dehydrogenase enzymes in metabolically active cells.
In order to determine cell viability, in the exponential phase of growth, the cells were seeded at 5 × 104/ml in ninety-six-well plates, in DMEM supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 5 % fetal bovine serum. After seeding (24 h), HT-29 cells were treated, in quadruplicate, with increasing concentrations of the vegetable extracts, using ranges of concentrations 0·001–10 g eq ww/ml and incubated for 24 h. The cytotoxicity assay was performed by adding a small amount of the CellTiter96R AQueous One Solution Cell Proliferation Assay directly to culture wells, incubating for 4 h and then recording the absorbance at 450 nm with a ninety-six-well plate reader (MULTISKAN EX; Thermo Electron Corporation, Vantaa, Finland).
Genotoxicity studies
DNA damage was measured using single-cell gel electrophoresis (Comet assay). The alkaline version (pH >13) of the assay was performed to detect single-strand breaks and alkali-labile sites, such as apyrimidinic and apurinic sites that are formed when bases are lost and oxidised.
The cells were seeded at 1 × 105/ml in six-well plates in DMEM supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum. After seeding (24 h), the HT-29 cells were treated with the vegetable extracts, using the higher non-toxic concentrations as detected by the MTS assay for the uncooked vegetables. After incubation, the cells were trypsinised and re-suspended in DMEM supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum at a concentration of 5 × 104 cell/ml.
After this, the suspensions were centrifuged (1 min, 800 g) to recover the cells. The cell pellets were re-suspended in 90 μl low melting agarose 0·7 %, transferred onto degreased microscope slides previously dipped in 1 % normal melting agarose for the first layer. The agarose was allowed to set for 15 min at 4°C before the addition of a final layer of low melting agarose. Cell lysis was carried out at 4°C overnight by exposing the cells to a buffer containing 2·5 m-NaCl, 100 mm-Na2EDTA, 8 mm-Tris–HCl, 1 % Triton X-100 and 10 % dimethylsulfoxide, pH 10.
In an alkaline buffer, pH>13, (1 mm-Na2EDTA, 300 mm-NaOH, 0°C) the electrophoretic migration was performed (DNA unwinding: 20 min; electrophoresis: 20 min, 0·78 V/cm, 300 mA).
DNA was stained with 75 μl ethidium bromide (10 μg/ml) before examination at a 400 × magnification under a Leica DMLS fluorescence microscope (excitation filter BP 515-560 nm, barrier filter LP 580 nm), using an automatic image analysis system (Comet Assay III; Perceptive Instruments Limited, Haverhill, Suffolk, UK).
The total percentage of fluorescence in the tail (TI, tail intensity) provided representative data on genotoxic effects. For each sample, coded and evaluated blind, 100 cells were analysed and the median value of TI was calculated. At least three independent experiments were performed for each extract and the mean of the median TI values was used for statistical analyses.
Antioxidant activity
The Comet assay was used to study the antioxidant effect of the test extracts. In that case, the amount of DNA damage caused by an oxidative damage-inducing agent on the cells pre-treated with the vegetable extracts was evaluated.
The cells were seeded at 1 × 105/ml in six-well plates in DMEM supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum. After seeding (24 h), the HT-29 cells were incubated with the vegetable extracts, using the higher non-toxic concentrations as detected by the MTS assay for the fresh vegetables, or with a known antioxidant agent, vitamin C (1 mm), and incubated for 24 h. After incubation, the cells were trypsinised and re-suspended in DMEM supplemented with 1 % glutamine, 0·5 % penicillin/streptomycin and 10 % fetal bovine serum at a concentration 1 × 105 cell/ml for a further treatment in suspension, before performing the Comet assay, with H2O2 (100 μm) on ice for 5 min. After this, the suspensions were centrifuged twice (1 min, 800 g) to wash and recover the cells. The slides were prepared and analysed as reported above.
To better compare the ability of the extracts in reducing the DNA damage induced by H2O2 treatment, a TI variation percentage was calculated as follows:
Statistical evaluation
The data were analysed using the statistical and graphical functions of SPSS 15 (SPSS, Inc., Chicago, IL, USA). Differences were assessed using ANOVA, followed by Bonferroni's post test as appropriate, for parameters normally distributed such as means of optical density values and of median TI values. In particular, for the Comet assay, statistical differences between controls and treated samples were first determined with the non-parametric Wilcoxon rank-sum test for each experiment. Significance was accepted at P < 0·05 level.
Results
Anti-proliferative activity
Anti-proliferative activity was detected by the MTS assay, a colorimetric method for determining the number of viable cells in proliferation (Fig. 2). In the range of the doses used (10− 3/101g eq ww/ml), the raw vegetable extracts that were analysed showed high anti-proliferative/toxic activity on the HT-29 colon carcinoma cells. The lowest effective dose (LED, P < 0·001) was 0·25 g eq ww/ml for broccoli and 0·50 g eq ww/ml for cauliflower and Brussels sprouts.
Fig. 2 Antiproliferative effects, detected by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay, induced by increasing concentrations (equivalent of wet weight g/ml; g eq ww/ml) of raw and cooked (a) broccoli; (b) cauliflower; (c) Brussels sprouts extracts (24 h-treatment) on HT-29 cells. OD, optical density. 
, Raw; 
, boiled; 
, microwaved; 
, basket steamed; 
, oven steamed; 
, cell seeding.
The strong anti-proliferative activity detected with the raw broccoli extract was maintained with the microwave cooking method (LED: 0·25 g eq ww/ml, P < 0·001), while the boiled and steamed broccoli extracts showed no cytotoxic activity up to 4·00 g eq ww/ml.
All the cooking methods, except basket steaming up to 4·00 g eq ww/ml, partially maintained the anti-proliferative activity detected in the raw cauliflower extract: microwaving LED: 1·00 g eq ww/ml (P < 0·01); boiling LED: 2·00 g eq ww/ml (P < 0·01); oven steaming LED: 1·00 g eq ww/ml (P < 0·05).
In the case of Brussels sprouts, all the cooking methods applied decreased the anti-proliferative activity of the raw extract and no cytotoxic activity was traceable up to 4 g eq ww/ml.
Genotoxicity
On the basis of MTS values of raw and cooked vegetables we determined the sub-toxic concentrations to adopt for the DNA migration analysis (alkaline Comet assay) on HT-29 cells. To better compare the results between the extracts from fresh and cooked vegetables, which presented different antiproliferative activity, we analysed the genotoxicity induced by the highest non-toxic concentration identified with the raw extracts for all samples (i.e. broccoli: 0·01 g eq ww/ml; cauliflower and Brussels sprouts: 0·10 g eq ww/ml) and a concentration ten times higher only for cooked samples (broccoli: 0·10 g eq ww/ml; cauliflower and Brussels sprouts: 1·00 g eq ww/ml).
None of the vegetable extracts was able to induce DNA damage at the concentrations tested (Table 1).
Table 1 Genotoxicity evaluated by the Comet assay on HT-29 cells treated (24 h) with raw and cooked vegetable extracts at different concentrations*
(Mean values and standard deviations)
TI, tail intensity; g eq ww/ml, equivalent of wet weight g/ml.
* Negative controls: HT-29 untreated and vitamin C treated (1 mm); positive control: H2O2 (100 μm).
Antioxidant activity
The cells were treated for 24 h with vegetable extracts, after discharging the medium plus the exhausted extract; they were subjected to 5 min shock with H2O2 (one of the most common oxidative damage-inducing agents). The ability to reduce the DNA damage induced by H2O2 treatment in HT-29 cells was chosen as an indicator of the chemopreventive capability of the vegetable extracts. Among the antioxidants, ascorbic acid (vitamin C) plays a central role as it contributes to the regeneration of vitamin E and constitutes a strong line of defence in retarding free radical-induced cellular damage(Reference Chan30). The antioxidant activity of ascorbic acid (1 mm) was determined. Vitamin C pre-treatment produced a significantly strong decrease (approximately 50 %) of DNA damage induced by H2O2 at 100 μm (P < 0·05). All the raw vegetable extracts had significant antioxidant activity (P < 0·05) comparable with vitamin C, in particular the raw broccoli extract showed its defensive capability at a ten-fold lower concentration than the other two raw vegetable extracts (Fig. 3).
Fig. 3 Antioxidant activity evaluated by Comet assay on HT-29 cells treated (24 h) with vitamin C (Vit C; 1 mm) or raw (R) or cooked vegetables extracts at different concentrations (0·01, 0·1, 1·0 g eq wet weight/ml) and treated with H2O2 (100 μm, 5 min). (a) Broccoli; (b) Cauliflower; (c) Brussels sprouts. It is reported the tail intensity (TI) variation (%): (
) × 100. M, microwaved; B, boiled; BS, basket steamed; OS, oven steamed.
Cooking broccoli with microwaving and steaming methods determined a slight lowering of the antioxidative properties shown by the raw vegetable. In fact, the same significant DNA migration reduction (P < 0·05) induced by the raw extract was detected for cooked vegetables extracts at a concentration that was ten-fold higher. Otherwise, no antioxidant activity was detected in the boiled broccoli at the extract concentrations tested (Fig. 3), in agreement with literature data(Reference Knutson, Marth and Wagner31).
A significant antioxidant activity (P < 0·05) was detected in the cauliflower extracts, independently from the cooking method performed (Fig. 3). In particular, the microwaved vegetable extract seems more effective than the raw extract and vitamin C.
The microwaved Brussels sprouts extract showed an antioxidant activity comparable with the raw extract and vitamin C, whereas the extracts of boiled and steamed Brussels sprouts showed the same protective activity at ten-fold higher concentrations (Fig. 3).
Discussion
In this study, the effect of cooking practices on the anti-proliferative and/or antioxidant properties of three fresh Brassica species (broccoli, cauliflower and Brussels sprouts) was evaluated by MTS and Comet assays. Data demonstrated a significant dose-dependent anti-proliferative effect when the human colon cancer cells (HT-29) were treated with uncooked vegetable extracts. The highest anti-proliferative activity was found with the broccoli extract with the LED twofold lower than the Brussels sprouts and cauliflowers extracts. This anti-proliferative rank order could be linked to the higher content of bioactive compounds, mainly glucosinolates, present in broccoli than in the other Brassica vegetables analysed (Fig. 1)(Reference Pellegrini, Chiavaro and Gardana27), even though a difference in single compounds present cannot be ruled out. It is known that some chemopreventive vegetable components could induce cell cycle arrest in HT-29 cells; in particular the hydrolysis of glucosinolates by myrosinase could induce the formation of chemopreventive molecules such as isothiocyanates(Reference Keum, Jeong and Kong32, Reference Juge, Mithen and Traka33). Even though we did not measure the myrosinase activity in the extracts, during their preparation the Stomaker treatment could have mimicked the mastication process, determining the activation of myrosinase and, hence, the formation of hydrolysis products, as during the Brassica processing before cooking (i.e. chopping, treading, etc.)(Reference Rungapamestry, Duncan and Fuller34). Among the vegetables analysed, broccoli has the highest concentration of the glucosinolate glucoraphanin, which could be hydrolysed by myrosinase in the reactive isothiocyanate sulforaphane with chemopreventive properties. Sulforaphane exerts its anti-proliferative effect by arresting the cell cycle; this arrest has been documented in the colon, prostate, breast, bladder and T cells(Reference Juge, Mithen and Traka33).
The release of active molecules from the Brassicaceae tissues, following the glucosinolate hydrolysis by myrosinase, is strongly influenced by the cooking practice adopted(Reference Rungapamestry, Duncan and Fuller34). The lowering or disappearance of the Brassica anti-proliferative effects detected after cooking practices, particularly boiling and steaming, could be ascribed to an alteration of the glucosinolate–myrosinase system(Reference Dekker, Verkerk and Jongen35). Conversely, the higher retention of the anti-proliferative effect of microwaved Brassica samples with respect to other cooking practices could be linked to a reduced inactivation of myrosinase due to the mild cooking conditions applied (i.e. low energy inputs with consequent low temperature cooking), as demonstrated by Verkerk & Dekker(Reference Verkerk and Dekker36).
Many in vitro and in vivo studies demonstrated that some molecules (i.e. polyphenols, catechins, etc.) present in food are anti-carcinogenic by inducing apoptosis and inhibiting cell growth(Reference Surh10, Reference Sharif, Auger and Alhosin22). Probable mechanisms of action include antioxidant and free-radical scavenging activity. On the other hand, vegetables, including Brassica, present a variable chemical complex mixture and could also contain compounds with genotoxic activity(Reference Kassie, Parzefall and Musk37, Reference Heres-Pulido, Dueñas-García and Castañeda-Partida38). The vegetable extracts, evaluated in this research, were analysed both for their genotoxic or anti-genotoxic (antioxidant) activities by the Comet assay. DNA migration analysis of HT-29 cells treated (24 h) with vegetable extracts alone, at sub-toxic concentrations determined on MTS values, did not show any induced DNA damage, whereas HT-29 cells pre-treated for 24 h with raw and cooked Brassica and treated with H2O2 (100 μm, 5 min at 0°C) showed different sensitivity to DNA oxidative damage, as revealed by the alkaline Comet assay.
The chemopreventive action that was detected was determined by an endogenous response, due to the fact that the vegetable extract was removed before H2O2 treatment. In particular, raw, microwaved, boiled (except broccoli) and steamed vegetable extracts, presented a protective antioxidative action comparable with vitamin C (1 mm), as demonstrated by the reduced DNA migration. It is noteworthy that only the cooked cauliflower extracts showed an antioxidant activity comparable to the raw extract (0·10 g eq ww/ml) with the same TI (%) variation coefficient.
Among the different cooking practices, the boiling treatment seems to determine a different antioxidant status depending on the kind of Brassica analysed; although the highest loss of activity was found with broccoli, their cooked extracts remained the more active. The steaming and microwaving practices showed a general good retention of the antioxidant activity. Based on the present results, the antioxidant activity of raw extracts could be probably ascribed to the glucosinolate hydrolysis products. On the contrary, the antioxidant activity of cooked Brassica vegetables could be due to the formation of new compounds following the cooking, for instance, polyphenols with a different oxidation state that could exhibit their antioxidant activity not being masked by the glucosinolates hydrolysis products. However, these hypotheses should be confirmed by the identification and quantification of phytochemical compounds present in the aqueous extracts as no clear relationships were found when data on biological activities were compared with the concentration of chemical compounds present in the vegetables, before the extraction process.
At a first glance, the comparative analysis of the raw extract toxicological data (anti-proliferative and antioxidant activities) seems to suggest that high anti-proliferative activity could be concomitant with high antioxidant activity, suggesting that the chemopreventive action could be related to the control of the redox cellular status. This correlation disappears when data of cooked vegetable extracts are observed; generally, the loss of anti-proliferative activity seems independent of the antioxidant activity, suggesting a more complex design involving several pathways differently regulated by the complex mixtures present in the extracts. These results are in agreement with Boivin et al. (Reference Boivin, Lamy and Lord-Dufour39), who found that the anti-proliferative effect of Brassica vegetables, among others, largely independent of their antioxidant properties evaluated in in vitro experiments.
In conclusion, the analyses of the present data on the human colon carcinoma cell line HT-29 indicate that Brassica extracts could positively alter cellular endogenous chemoprotective status.
The lack of a chemical characterisation of the water extracts prevents us from a clear correlation among the biological activities tested and the specific compounds. Nevertheless, the increased ability of cells, treated with the Brassica samples, to contrast the reactive oxygen species damaging activity seems to be related to the complex mixture of the protective substances present in the vegetables more than to a single class of molecules.
The cooking methods applied influence the anti-proliferative activity of the Brassica extracts but do not alter considerably the high antioxidant activity presented by the raw vegetables. These data provide a new insight into the influence of domestic treatment on the quality of food, which could support the recent epidemiological studies suggesting that the consumption of cruciferous vegetables, mainly cooked food, may be related to a reduced risk of developing cancer.
Acknowledgements
This study was supported by Regione Emilia Romagna; SIQUAL project (Laboratorio Regionale per la Sicurezza e la Qualità degli Alimenti). A. B., N. P., and C. R. designed the research, L. F., T. M., C. M., S. G., and F. M. conducted the research, A. B., L. F. and N. P. analysed the data, A. B. and N. P. wrote the paper, C. R. and C. M. provided critical revision of the manuscript, A. B. had primary responsibility for the final content. All authors read and approved the final manuscript. We thank Antonietta Cirasuolo for excellent technical assistance. La Linea Verde spa, Insalate Pronte (Manerbio, Brescia, Italy), is acknowledged for the gift of Brassica vegetables. There are no conflicts of interest.
