Introduction

Sulforaphane is considered as the most powerful anticancer compound found naturally in food (Elbarbry and Elrody 2011). Besides, it acts as indirect antioxidant at cellular level by inducing nuclear factor (erythoid-derived 2)—like 2 (Nrf2) and stimulating the synthesis of antioxidant enzymes (Zhou et al. 2016). Its precursor is glucoraphanin, which is the most abundant glucosinolate in some broccoli cultivars. Glucoraphanin has no bioactive effect when ingested. The hydrolysis of glucoraphanin to yield sulforaphane proceeds through the action of myrosinase (EC 3.2.1.147), which is found in Brassicaceae plants as well as in the intestine microbiota (Shapiro et al. 2001). Unfortunately, the chemical conditions in the intestine disfavor sulforaphane formation since the hydrolysis privileges the formation of other compounds. Then, the bioavailability of sulforaphane can be improved if it is ingested directly in its active form.

In the intact plant tissue, glucoraphanin is physically segregated from the enzyme. When the vegetal tissue is disrupted through chewing or processing, glucoraphanin interacts with myrosinase and the hydrolysis proceeds, yielding sulforaphane as well as other compounds such as nitriles, thiocyanates and epithionitriles. The formation of these products is determined by the chemical conditions, i.e. pH, temperature, presence of Fe+2, presence and activity of epithiospecifier protein (ESP) (Gu et al. 2012) and the concentration of ascorbic acid, which acts as a co-factor of myrosinase (Bones and Rossiter 2006). Hence, selecting the processing conditions of broccoli that favor the formation of sulforaphane seems attractive for developing a broccoli-based functional food, naturally enriched in the anticancer compound sulforaphane.

Some authors investigated broccoli processing under different conditions in order to favor sulforaphane formation and at the same time minimize the formation of nitriles and other products (van Eylen et al. 2009; Jones et al. 2010; Wang et al. 2012). Specific time–temperature combinations in hydrothermal treatments of broccoli increase the sulforaphane content (Matusheski et al. 2004; Wang et al. 2012), most likely due to denaturation of ESP at temperatures above 40 °C. Matusheski et al. (2004) reported that heating broccoli at 60 °C during 5 or 10 min produced an increase in sulforaphane content. Guo et al. (2013) maximized the formation of isothiocyanates in broccoli sprouts by incubation at 40 °C during 4 h in 4 mL citrate–phosphate buffer pH 4.0 containing 0.02 mmol EDTA and 0.16 mg ascorbic acid. Recently, Pérez et al. (2014) reported that sulforaphane content of broccoli was maximized when subjecting the vegetable to a blanching step consisting of immersion in water at 57 °C during 13 min. In that work, the authors also showed that despite increasing 3.3-fold the sulforaphane content with respect to the fresh vegetable, there was still a considerable amount of glucoraphanin (approximately 8 µmol g−1 dry weight) that was not hydrolyzed. Then, incubating the blanched broccoli under optimal conditions would allow the complete conversion of glucoraphanin and hence the maximum sulforaphane formation.

The aim of this work was to determine the incubation conditions that maximize the content of sulforaphane in pre-processed broccoli. The effect of processing conditions on sulforaphane content and myrosinase activity was also investigated.

Materials and methods

Plant material

Broccoli (Brassica oleracea var Italica cv. Avenger) heads (3 days after harvesting) were purchased at the local market (Santiago, Chile) from a single supplier. Broccoli heads were washed and cut into single florets of 5-cm length and 0.7–0.9 cm width (stem) immediately after washing and before blanching.

Blanching

In order to denature ESP and maximize sulforaphane formation, broccoli was subjected to blanching using the optimal conditions determined by Pérez et al. (2014). Broccoli florets (300 g) were immersed in 1.5 L deionized water in a thermostatic bath (Stuart, United Kingdom, Great Britain) at 57 °C for 13 min. After blanching, broccoli florets were immediately put in an ice-water bath and then crushed to obtain 0.5-cm broccoli pieces.

Experimental design

Optimization of the incubation conditions was performed through a Box–Behnken design with three central points. The experimental factors (and levels) were temperature (22 and 38 °C), incubation time (3 and 7 h) and addition of ascorbic acid (0.03 and 1 mg g−1 fresh broccoli). The levels in this design were chosen based on preliminary incubation experiments made to define the experimental region. The response variables were sulforaphane concentration and myrosinase activity. Table 1 shows the experimental matrix. Results are presented as average ± standard deviation.

Table 1 Experimental matrix of the optimization design, sulforaphane content and myrosinase activity obtained in each run
Full size table

Incubation procedure

Samples of crushed blanched broccoli were put in empty, hermetically closed flasks and were immersed in a thermostatic water bath (Trilab, Mexico City, Mexico) at the conditions stipulated by the experimental design. Two independent replicates were carried out for each condition, and each sample was analyzed in triplicate. Incubation was performed in darkness. After the treatments, the samples were kept at −20 °C until analysis.

Analytical determinations

Sulforaphane content

Sulforaphane content was assessed by reverse phase HPLC, using the method proposed by Liang et al. (2006) with some minor modifications. Samples of incubated broccoli were homogenized in a mortar, until obtaining a homogeneous meal. One gram of the sample was extracted two times with 10 mL methylene chloride (J.T. Baker, Center Valley, PA, USA), which was combined with 0.5 g anhydrous sodium sulfate (Sigma-Aldrich, Schnelldorf, Germany). The equipment was a HPLC–DAD (Shimadzu, Kyoto, Japan), and a C18 column (5 μm particle size, 250 × 4.6 mm) (Agilent Technologies, Santa Clara, CA, USA) was used. Quantification was carried out by comparison with a sulforaphane standard curve (0.056–6.75 μg). Organic solvents (HPLC grade) were purchased from Merck (Darmstadt, Germany). Sulforaphane content was expressed in μmol g−1 (dry weight). All determinations were made in triplicate.

Myrosinase activity

Myrosinase extraction was made following the protocol described by Guo et al. (2011). Samples of incubated broccoli were homogenized in a mortar, until obtaining a homogeneous meal. After that, 0.25 g were massed and mixed with 1 mL sodium phosphate buffer (100 mM pH 6.5). The mixture was vortexed and centrifuged at 15,000×g during 20 min at 4 °C. The supernatant was recovered. Enzyme activity was assessed spectrophotometrically through the method described by Rakariyatham et al. (2006). One mL of substrate buffer (33 mM sodium phosphate buffer, pH 7.4 containing 10 mM sinigrin, 3 mM MgCl, 0.55 ATP, 0.72 mM NADP, 3.5 U hexokinase, and 1.75 U glucose-6-phosphate dehydrogenase) (Sigma–Aldrich, Schnelldorf, Germany) was pre-incubated for 3 min at 30 °C. Then, 100 µL of sample were added. After mixing, the reaction was monitored at 346 nm. Myrosinase activity was calculated from increasing absorbance rate, which was due to the formation of NADPH. Protein concentration was measured by the method of Bradford (1976) using bovine serum albumin as a standard. One unit of myrosinase activity was defined as the amount of enzyme that catalyzed the liberation 1 μmol of glucose per min from sinigrin under the conditions described above. Results were expressed in units per mg protein. All determinations were made in triplicate.

Statistical analyses

The significance of the statistical effects was assessed by ANOVA followed by the least significant difference (LSD) test, at 95% confidence level. The optimization design was analyzed by response surface methodology, using a second order polynomial model to describe the experimental behavior (Eq. 1).

$$ \hat{Y} = \beta_{0} + \mathop \sum \limits_{i = 1}^{k} \beta_{i} x_{i} + \mathop \sum \limits_{i = 1}^{k} \beta_{ii} x_{i}^{2} \mathop \sum \limits_{i < j = 1}^{k} \beta_{ij} x_{i} x_{j} $$
(1)

In Eq. 1, Ŷ is the predicted response; β0, βi, βii, and βij are the regression coefficients for interception, linear, quadratic and interaction effects, respectively; k is the number of independent parameters (k = 3 in this study), and Xi, Xj are the coded levels of the experimental conditions. The model quality was assessed by the determination coefficient (R2). The optimum incubation conditions predicted by the regression model were validated experimentally. Statistical analyses were performed with JMP 9.0.1 software (SAS Institute).

Results and discussion

The antioxidant activity of broccoli after incubation, determined through the method of Brand-Williams et al. (1995), was significantly lower (p < 0.001) than that of the fresh vegetable. Fresh broccoli had an anti-radical power (ARP) equal to 26.4 ± 4.4 µmol g−1, while incubated broccoli had an average ARP of 21.5 ± 4.2 µmol g−1. This difference may be attributed to thermal decomposition of ascorbic acid and to the loss of some antioxidant compounds such as polyphenols through leaching in the blanching step.

The sulforaphane content and myrosinase activity obtained in each condition are given in Table 1. The superscripts letters indicate the membership to a homogeneous group, determined through the LSD multiple range test. Temperature had a significant positive effect on the response (p < 0.05), as well as the interaction of temperature with time (p < 0.05), temperature with ascorbic acid (p < 0.05) and the interaction of ascorbic acid with itself (p < 0.05). Accordingly, these factors and interactions were considered in the regression model for the optimization. The model that describes sulforaphane content as a function of the significant factors in coded levels is given in Eq. 2. Here, [S] is sulforaphane content, T is temperature, t is time and [AA] is ascorbic acid concentration.

$$ \left[ S \right] = 6.194 + 1.086T - 1.843Tt - 1.858T\left[ {AA} \right] - 1.997\left[ {AA} \right]^{2} $$
(2)

The adjustment of the model to the experimental data gave a determination coefficient (R2) equal to 89.88%, i.e. the model is able to explain about 90% of the data variability.

Figure 1a, b show the response surface obtained for sulforaphane concentration. Figure 1a shows that there is an optimal combination of incubation time and ascorbic acid concentration, which corresponds to 3 h of incubation and 0.22 mg ascorbic acid per g of raw broccoli. In this analysis we considered that temperature was kept at 38 °C. Figure 1b shows that low temperature and long incubation time favor sulforaphane formation. However, in the opposite directions, there was an increase of sulforaphane content. This agrees with the fact that at the low temperature level (22 °C), an increase from the low (3 h) to the high (7 h) level of immersion time results in an increase of 113% in sulforaphane content, from 3.3 to 7.0 μmol g−1 (dry weight). At the high level of temperature (38 °C) an increase in incubation time results in a significant decrease in sulforaphane equal to 40% [from 9.1 to 5.4 μmol g−1 (dry weight)]. The maximum sulforaphane content was achieved at 38 °C. According to the regression model, the optimal incubation conditions are 3 h of incubation at 38 °C in presence of 0.22 mg ascorbic acid per g fresh broccoli. In these conditions, the maximum sulforaphane content predicted by the model was 8.0 µmol g−1 (dry weight). This value was verified experimentally, resulting in a sulforaphane concentration equal to 8.1 ± 0.3 µmol g−1 (dry weight), thus agreeing with the prediction of the model. This value represents a 585% increase with respect to fresh broccoli and a 119% increase with respect to blanched broccoli, whose sulforaphane content was 1.2 ± 0.2 and 3.7 ± 0.3 μmol g−1 (dry weight), respectively.

Fig. 1
figure 1

Response surfaces for sulforaphane concentration obtained from the optimization design. a Effect of time and ascorbic acid concentration; b Effect of time and temperature

Full size image

Our results partially agreed with some related studies. Guo et al. (2013) reported the maximum formation of isothiocyanates by incubating homogenized broccoli sprouts at 40 °C for 3 h at pH 4.0 with the addition of 0.02 mmol EDTA and 0.16 mg ascorbic acid. Shen et al. (2010) found the maximum sulforaphane formation by incubating broccoli seeds powder mixed with distilled water and ascorbic acid at 25 °C for 8 h. The authors found a glucosinolates conversion equal to 35%, while in the present work we achieved about 94% conversion of glucoraphanin, equivalent to 75% conversion of glucosinolates. The optimal ascorbic acid concentration derived from our results is higher than the reported by Shen et al. (2010), probably because that study was performed in broccoli suspensions or homogenates, where mass transfer has less hindrances.

Figure 2a, b show the response surface obtained for myrosinase activity. The high level of temperature favored enzyme activity at short incubation time and low ascorbic acid concentration. At low temperature, high concentration of ascorbic acid and long incubation times led to a higher myrosinase activity. The maximum activities were observed at the high temperature level (38 °C). The statistical analysis showed that the interaction of temperature with time and ascorbic acid had significant negative effects on myrosinase activity, with p values equal to 0.028 and 0.041, respectively. Myrosinase activity exhibited a similar behavior to that of sulforaphane concentration. Both responses were maximized at the high temperature level and short incubation time. However, despite myrosinase showed the highest activity in these conditions, it is not possible to ensure that these are the optimal conditions with certainty, since the best incubation conditions are too close to the limits of the experimental region. Nevertheless, as shown in Fig. 2a, the optimal incubation conditions should be close to 38 °C and 3 h incubation. These conditions agree with the conditions that maximize sulforaphane content.

Fig. 2
figure 2

Response surfaces for myrosinase activity obtained from the optimization design. a Effect of time and temperature; b Effect of ascorbic acid concentration and temperature

Full size image

Conclusion

The optimal incubation conditions predicted by the regression model were temperature equal to 38 °C during 3 h in the presence of 0.22 mg ascorbic acid per g fresh broccoli. The predicted maximum sulforaphane concentration (8.0 µmol g−1) was confirmed experimentally yielding a value of 8.1 ± 0.3 µmol g−1 (dry weight). This represents an increase of 585% with respect to fresh broccoli and of 119% with respect to blanched broccoli, equivalent to a conversion of 94% of glucoraphanin. This is, to our knowledge, the first study about the effects of the incubation conditions of broccoli on sulforaphane formation. These findings could be used as basis for designing a process to significantly increase sulforaphane content avoiding artificial chemical synthesis. The compound could probably be isolated from broccoli in the future, and be delivered as a nutraceutical or functional ingredient in further elaborated foods.