Nestle (1997) listed the following potentially anticarcinogenic attributes and components of broccoli and other cruciferous vegetables: low fat, low energy, macronutrients, micronutrients, vitamin A, vitamin C, vitamin E, folic acid, selenium, fibre, carotenoids, coumarins, dithiolthiones, flavonoids, glucosinolates, indoles, isothiocyanates, phenols and terpenes.

The proximate nutrient value per 100 g edible portion of raw broccoli (USDA 2012) was reported as follows: water, 89.30 g; energy, 34 kcal (141 kJ); protein, 2.82 g; total lipid, 0.37 g; ash, 0.87 g; carbohydrate, 6.64 g; total dietary fibre, 2.6 g; total sugars, 1.7 g; sucrose, 0.10 g; glucose, 0.49 g; fructose, 0.68 g; lactose, 0.21 g; maltose, 0.21 g; minerals Ca, 47 mg; Fe, 0.73 mg; Mg, 21 mg; P, 66 mg; K, 316 mg; Na, 33 g; Zn, 0.41 mg; Cu, 0.049 mg; Mn, 0.210 mg; Se, 2.5 μg); vitamins vitamin C, 89.2; thiamine, 0.071 mg; riboflavin, 0.117 mg; niacin, 0.639 mg; pantothenic acid, 0.573 mg; vitamin B 6, 0.175 mg; total folate, 63 μg; total choline, 18.7 mg; betaine, 0.1 mg; vitamin A, 623 IU; vitamin A, 31 μg RAE; β-carotene, 361 μg; lutein + zeaxanthin, 1,403 μg; vitamin E (α-tocopherol), 0.78 mg; β-tocopherol, 0.01 mg; γ-tocopherol, 0.17 mg; vitamin K (phylloquinone), 101.6 μg; total saturated fatty acids, 0.039 g; 14:0 (myristic acid), 0.001 g; 16:0 (palmitic acid), 0.029 g; 18:0 (stearic acid), 0.006 g; 20:0 (arachidic acid), 0.002 g; 22:0 (behenic acid), 0.002 g; total monounsaturated fatty acids, 0.011 g; 17:1 (heptadecenoic acid), 0.001 g; 18:1 undifferentiated, 0.010 g; total polyunsaturated fatty acids, 0.038 g; 18:2 undifferentiated (oleic acid), 0.017 g; 18:3 undifferentiated (linolenic acid), 0.021 g; and amino acids tryptophan, 0.033 g; threonine, 0.088 g; isoleucine, 0.079 g; leucine, 0.129 g; methionine, 0.038 g; cystine, 0.028 g; phenylalanine, 0.117 g; tyrosine, 0.050 g; valine, 0.125 g; arginine, 0.191 g; histidine, 0.059 g; alanine, 0.104 g; aspartic acid, 0.325 g; glutamic acid, 0.542 g; glycine, 0.089 g; proline, 0.110 g; and serine, 0.121 g.

Maximum mean vitamin C (52.9 mg/100 g), β-carotene (0.81 mg/100 g), lutein (0.68 mg/100 g), dl-α-tocopherol content (0.47 mg/100 g) and phenol content (63.4 mg/100 g) were recorded in broccoli (Singh et al. 2007). The following carotenoids were found in broccoli: neoxanthin, neochrome, violaxanthin, luteoxanthin, auroxanthin, lutein-5,6-epoxide, flavoxanthin, lutein and β-carotene (Khachik et al. 1986, 1991). Studies found substantial variation in α-carotene, β-carotene, α-tocopherol, γ-tocopherol and ascorbate contents within and between subspecies of Brassica oleracea (50 broccoli and 13 cabbage, kale, cauliflower and Brussels sprouts accessions) (Kurilich et al. 1999). Kale had the highest levels of vitamins, followed by broccoli and Brussels sprouts with intermediate levels and then by cabbage and cauliflower, with comparatively low concentrations. Variability in vitamin content among the broccoli accessions suggested potential health benefits associated with consumption were genotype dependent. Broccoli and other vegetables and fruit were found to contain glucaric acid, the content of which varied from a low of 1.12–1.73 mg/100 g for broccoli and potatoes to a high of 4.53 mg/100 g for oranges (Dwivedi et al. 1990). The predominant glucosinolates in broccoli were 4-methylsulphinylbutyl glucosinolate (glucoraphanin), 3-butenyl glucosinolate (gluconapin) and 3-indolylmethyl glucosinolate (glucobrassicin) (Kushad et al. 1999). Glucoraphanin concentration in broccoli ranged from 0.8 μmol/g⁻ DW in cv. EV6-1 to 21.7 μmol/g DW in cv. Brigadier. Concentrations of the other glucosinolates in broccoli varied similarly over a wide range.

Glucosinolates identified in broccoli inflorescences included the following: 3-methylsulphinylpropyl-glucosinolate (glucoiberin), 2-hydroxy-3-butenyl-glucosinolate (progoitrin), 4-methylsulphinylbutyl-glucosinolate (glucoraphanin), 5-methylsulphinylpentyl-glucosinolate (glucoalyssin), 3-butenyl-glucosinolate (gluconapin), 4-hydroxy-3-indolylmethyl-glucosinolate (4-hydroglucobrassicin), 4-pentenyl-glucosinolate (glucobrassicanapin), 3-indolylmethyl-glucosinolate (glucobrassicin), 2-phenylethyl-glucosinolate (gluconasturtin), 4-methoxy-3-indolylmethyl-glucosinolate (4-methoxyglucobrassicin) and 1-methoxy-3-indolylmethy-glucosinolate (neoglucobrassicin) (Rosa and Rodrigues 2001; Vallejo et al. 2003a, b; Moreno et al. 2006). The predominant glucosinolates in all broccoli cultivars were 4-methylsulphinylbutyl-glucosinolate (glucoraphanin), 3-indolylmethyl-glucosinolate (glucobrassicin) and 1-methoxy-3-indolylmethyl-glucosinolate (neoglucobrassicin) (Rosa and Rodrigues 2001; Vallejo et al. 2003a). A new glucosinolate cinnamoyl derivative 6′-O-trans-(4″- hydroxy cinnamoyl)-4-(methylsulphinyl) butyl glucosinolate was isolated from broccoli florets (Survay et al. 2010). Two glucosinolates, glucoiberin and 3-hydroxy, 4(α-l-rhamnopyranosyloxy) benzyl glucosinolate, were identified in aqueous Broccoli extract (Hashem et al. 2012b). Further, two compounds were isolated after enzymatic hydrolysis of the aqueous extract by myrosinase; one of them was identified as 4-vinyl-3-pyrazolidinone and the second compound (sulphoraphane) 1-isothiocyanate-4-methyl-sulphinyl butane was converted to the most stable form of thiourea (sulphoraphane thiourea).

Total and individual glucosinolate levels varied significantly between seasons, among cultivars and between inflorescences. The cv. ‘Shogun’ contained the highest total glucosinolate levels (between 35.2 mmol/kg dry weight in primary inflorescences of the early crop and 47.9 mmol/kg in secondary inflorescences of the late crop) (Rosa and Rodrigues 2001). Total and individual glucosinolate levels were generally higher in the late than in the early crop. Similarly total glucosinolates were detected more significantly in the late than in the early season and all broccoli cultivars showed a higher content of indolic glucosinolates than aliphatic glucosinolates (Vallejo et al. 2003a). Broccoli florets were characterized by particularly high glucoraphanin content, 17.95 μmol/g dry weight on average, which comprised about 50 % of total glucosinolates (Borowski et al. 2008). High individual variation was observed for several bioactive compounds, as well as for DPPH^• and OH^• radical-scavenging activity. Among glucosinolates, the highest coefficient of variation (CVs) was noted for progoitrin (34.22 %), 4-hydroxyglucobrassicin (27.32 %) and neoglucobrassicin (24.44 %). High CVs were also observed for vitamin C (29.11 %), including dehydroascorbic acid (26.72 %), and for OH^• (25.76 %) and DPPH (21.77 %) radical-scavenging activities. Smaller variations were found for glucoraphanin (CV = 14.84 %) and polyphenols (CV = 14.95 %). The highest concentration of glucoraphanin occurred in young broccoli seedlings and seeds (Rangkadilok et al. 2002a). The glucoraphanin concentration decreased from the start of seed germination to the flowering stages. The lowest concentration was also found at the flowering stage. A higher concentration of glucoraphanin was detected in the green broccoli heads and flower heads than in other reproductive tissues. Each kilogram of extracted broccoli seed yielded approximately 4.8 g of sulforaphane and 3.8 g of sulforaphane nitrile (Matusheski et al. 2001). Vitamin C was not detected in dormant broccoli seeds, and its content increased with the germination, reaching values ranging from 53 (cv. Nubia) to 64 (cv. Marathon) mg/100 g FW, at the end of the monitored period (14 days) (Pérez-Balibrea et al. 2011). The total glucosinolate content in seeds and 3-day-old sprouts was higher in cv. Marathon (1,005 and 556 mg/100 g FW, respectively); however, cv. Viola sprouts registered the highest glucosinolate content 7 and 14 days after sowing (235 and 208 mg/100 g FW, respectively). Glucoraphanin was the predominant glucosinolate in cv. Nubia and cv. Marathon, whereas glucoiberin was the major glucosinolate in cv. Viola. The flavonoid and total phenolic content was significantly higher in cv. Viola. Also, seeds of this cultivar showed the highest antioxidant capacity (2.7 mg Trolox/g FW).

Broccoli cvs. was found to have a predominance of 4-(methylsulfinyl)butyl-glucosinolate and 4-(methylthio)butyl-glucosinolate in the head and 2-hydroxy-3-butenyl-glucosinolate as secondary component; other glucosinolates present included allyl-glucosinolate, 3-(methylthio)propyl-glucosinolate, 3-(methylsulfinyl)-propyl- glucosinolate, 3-butenyl- glucosinolate, 4-pentyl- glucosinolate, 2-phenyl-ethyl- glucosinolate and 3-indolyl-methyl- glucosinolate (Carlson et al. 1987). Eight aliphatic glucosinolates, four indole glucosinolates and one aromatic glucosinolate were identified and quantified in the florets of 5 main Chinese broccoli and 143 parent materials (Wang et al. 2012b). Glucoraphanin, glucobrassicin, 4-methoxyglucobrassicin and neoglucobrassicin were present in all samples. The anticancer glucoraphanin concentration ranged from 0.06 to 24.17 μmol/g in pure lines and from 1.57 to 5.95 μmol/g in commercial cultivars. The progoitrin concentration in commercial cultivars varied from 1.77 to 6.07 μmol/g with a mean value of 3.20 μmol/g. Significant variations were observed in the concentration of individual glucosinolates and in each class of glucosinolates among broccoli populations.

Phenolics found in broccoli inflorescences included: caffeoylquinic derivatives (3-O-caffeoylquinic (neochlorogenic acid), 5-O-caffeoylquinic (chlorogenic acid)); sinapic derivatives (1,2-disinapoylgentiobiose, 1-sinapoyl-2-feruloylgentiobiose, 1,2,2-trisinapoylgentiobiose, 1,2-disinapoyl-2-feruloylgentiobiose, 1-sinapoyl-2,2-diferuloylgentiobiose, 1,2,2-trisinapoylgentiobiose); and feruloyl acid derivatives (1,2-diferuloylgentiobiose) (Vallejo et al. 2003b). The main hydroxycinnamic acids (sinapic, ferulic, p-coumaric and caffeic acids) were isolated from broccoli and broccolini by capillary zone electrophoresis (Lee et al. 2011). Eight minor glucosinolates, viz., 1-methylpropyl-glucosinolate, 2-methylpropyl-glucosinolate, 2-methylbutyl-glucosinolate, 3-methylbutyl-glucosinolate, n-pentyl-glucosinolate, 3-methylpentyl-glucosinolate, 4-methylpentyl-glucosinolate and n-hexyl-glucosinolate, were identified in crude sample extracts of broccoli, cauliflower and rocket salad (Lelario et al. 2012). The occurrence of these glucosinolates belonging to the saturated aliphatic side chain families C4, C5 and C6, presumably formed by chain elongation of leucine, homoleucine and dihomoleucine as primary amino acid precursors, was described. Two nitrogenous compounds were uridine and uridine 9-acetate isolated from the ethyl acetate broccoli extract (Hashem et al. 2012a).

A large number of hydroxycinnamic acid esters of kaempferol and quercetin glucosides were characterized in broccoli inflorescences (Vallejo et al. 2004b) The structures of the flavonoid glycosides were analyzed after alkaline hydrolysis and were identified as 3-sophoroside/sophorotrioside-7-glucoside/sophoroside of kaempferol, quercetin and isorhamnetin. Additionally, several less complex glucosides based on the same aglycones were identified. Flavonoids acylated with hydroxycinnamic acid derivatives included sinapoyl/feruloyl gentiobioside compounds; 1,2-disinapoylgentiobiose; 1-sinapoyl-2-feruloylgentiobiose; 1,2-diferuloylgentiobiose; 1,2,2′-trisinapoylgentiobiose; 1,2′-disinapoyl-2-feruloylgentiobiose; 1-sinapoyl-2,2′-diferuloylgentiobiose; 1,2,2′-trisinapoylgentiobiose; and 1,2,2′-triferuloylgentiobiose. Some of the flavonoids identified included: quercetin-3-O-sophorotrioside-7-O-glucoside; quercetin-3-O-sophoroside-7-O-glucoside; quercetin-3-O-sophorotrioside-7-O-sophoroside; kaempferol-3-O-sophorotrioside-7-O-glucoside; kaempferol-3-O-sophorotrioside-7-O-sophoroside; kaempferol-3-O-sophoroside-7-O-glucoside; isorhamnetin-3-O-sophorotrioside-7-O-glucoside; kaempferol-3-O-sophoroside-7-O-sophoroside; isorhamnetin-3-O-sophorotrioside-7-O-sophoroside; quercetin-3,7-di-O-glucoside; quercetin-3-O-glucoside-7-O-sophoroside; kaempferol-3,7-di-O-glucoside; kaempferol-3-O-glucoside-7-O-sophoroside; isorhamnetin-3,7-di-O-glucoside; isorhamnetin-3-O-glucoside-7-O-sophoroside; quercetin-3-O-sophoroside; kaempferol-3-O-sophoroside; quercetin-3-O-glucoside; kaempferol-3-O-glucoside; quercetin-3-O-(feruloyl/sinapoyl)-sophorotrioside-7-O-glucoside; and kaempferol-3-O-(caffeoyl/sinapoyl)-sophorotrioside-7-O-sophoroside. Using liquid chromatography-photodiode array-solid-phase extraction-nuclear magnetic resonance/mass spectrometry, five related glycosylated phenolic acids were identified in broccoli: 1,2-di-O–E-sinapoyl-β-gentiobiose; 1-O–E-sinapoyl-2-O–E-feruloyl-β-gentiobiose; 1,2-di-O–E-feruloyl-β-gentiobiose; 1,2,2′-tri-O E-sinapoyl-β-gentiobiose; and 1,2′-di-O–E-sinapoyl-2-O–E-feruloyl-β-gentiobiose (Moco and Vervoort 2012).

Major volatile components identified in broccoli included nonanal, octanol, 3-methylthiobutyl cyanide, 2-phenylethyl cyanide and 2-phenylethyl isothiocyanate (Buttery et al. 1976). The major volatiles identified in blanched, cooked broccoli florets were: n-pentanal, 3-methyl-2-pentanone, n-hexanal, n-heptanal, ethyl acetate, cyclopentanecarboxaldehyde, 3-methylbutanal, 3-butenenitrile, 2-methylbutanal, dimethyl trisulfide and dimethy disulfide (Kallio et al. 1999). In broccoli and cabbage, exogenous methyl jasmonate induced the emission of the volatiles sesquiterpene (E,E)-α-farnesene, the homoterpene (E)-4,8-dimethyl-1,3,7-nonatriene (DMNT) and green leaf volatiles (Z)-3-hexenyl acetate and octanal (Ibrahim et al. 2005). Volatiles produced by the action of endogenous cystine lyase on S-methyl-cysteine sulphoxide present in Broccoli florets showed three sulphur components: dimethyl disulphide, dimethyl trisulphide and 3,5-dimethyl-1,2,4-trithiolane (Motawea et al. 2010). Four isothiocyanate compounds were also present. Cystine lyase enzymatic fission of endogenous myrosinase yielded the unsaturated glucosinolates ethenyl isothiocyanate and allyl isothiocyanate, together with the saturated 4-methylthiobutyl isothiocyanate (erucin) and the aromatic 2-phenylethyl isothiocyanate (PEITC).

Glucosinolates found in broccoli purchased from local Singapore markets were in nmol/g wet weight: glucoraphanin, 522 nmol; gluconasturtin, NQ (not quantified); sinigrin, NQ; glucobrassicin, 4,250 nmol; 4-hydroxy glucobrassicin, 198 nmol; 4-methoxy glucobrassicin, 432 nmol; 1-methoxy glucobrassicin, 1,100 nmol; glucoiberin, 40.9; progoitrin (or epiprogrotrin), NQ; glucoalyssin, 18.4 nmol; gluconapoleiferin, NQ; gluconapin, NQ; glucobrassicanapin, NQ; glucoerucin, NQ; 7-methylthioeptyl, NQ; 8-methylthioeptyl, NQ; total glucosinolates, 6,570 nmol; and glucobrassicans, 91 % of total glucosinolates (Hecht et al. 2004). Glucobrassicins (glucobrassicin, 1-methoxyglucobrassicin, 4-methoxyglucobrassicin and 4-hydroxyglucobrassicin), precursors to indole-3-carbinols, were the predominant glucosinolates in seven of the nine vegetables (including cauliflower and broccoli) studied, accounting for 70.0–93.2 % of all glucosinolates. Six individual desulfo-glucosinolates, including progoitrin, glucoraphanin, sinigrin, gluconapin, glucobrassicanapin and glucobrassicin, were commonly identified in the flower head of 95 Korean broccoli assessions (Lee et al. 2012). The total glucosinolate contents varied from 4.2 to 29 μmol/g DW, and the glucoraphanin (1.6–13.9 μmol/g DW) was confirmed as a major constituent in the total glucosinolate profile; in contrast, progoitrin was only detected in 13 accessions. Glucobrassicanapin, glucoraphanin, glucobrassicin, and gluconapin were the major glucosinolates.

Two novel glucosinolates identified as 2-mercaptomethyl sulfinyl glucosinolate [(Z)-4-(methylsulfinyl)-N-(sulfooxy)-2-((2′S,3′R,4′S,5′S,6′R)-3′,4′,5′-trihydroxy-6′(hydroxylmethyl)-2′-mercapto tetrahydro-2H-pyran-2-yl) butane amide] and (Z)-1-((2S,5S)-5-hydroxytetra-hydro-2H-pyran-2-ylthio)-2-(1H-indol-3-yl) ethylidene amino sulfate along with one known cinnamoyl [6′-O-trans-(4″-hydroxy cinnamoyl)4-(methylsulphinyl)butyl glucosinolate] were isolated from broccoli florets (Survay et al. 2012).

Six organoselenium compounds including selenium analogues of known myrosinase-derived Brassica volatiles, 4-(methylseleno)butanenitrile, 5-(methylseleno) pentanenitrile, 3-(methylseleno)propylisothiocyanate, 4-(methylseleno) butylisothiocyanate and 5-(methylseleno) pentylisothiocyanate, were identified in the pentane/ether extracts of broccoli and cauliflower florets and roots of forage rape, all obtained from plants treated with sodium selenate (Matich et al. 2012). LC–MS analysis of ethanolic extracts identified three selenoglucosinolates: 3-(methylseleno) propylglucosinolate (glucoselenoiberverin), 4-(methylseleno) butylglucosinolate (glucoselenoerucin) and 5-(methylseleno)pentylglucosinolate (glucoselenoberteroin). In broccoli, concentrations of the selenoglucosinolates and their aglycones (mainly nitriles) were up to 60 and 1,300 %, respectively, of their sulphur analogues.

A DNA topoisomerase I with an 80 kDa monomer was isolated from broccoli (Kieber et al. 1992). Three isoforms (a, b and c) of cystine lyase were found in broccoli inflorescence tissues (Ukai and Sekiya 1999). Cystine lyase b, the most abundant isoform, had a molecular weight of 160,000 and composed of four identical subunits with a molecular weight of 40,000. The purified cystine lyase b utilized l-cystine and S-alkyl l-cysteine sulfoxide as substrates. Cystine lyase a and b were localized in cytosolic and/or vacuole fraction. Three peroxidase (POD) isoenzymes were purified from a soluble extract of broccoli stems (Thongsook and Barrett 2005). The neutral and basic PODs had molecular masses of approximately 43 kDa, and the acidic POD had a molecular mass of 48 kDa. All three of the purified isoenzymes were glycosylated. Two cDNAs encoding proteins with homocysteine S-methyltransferase (HMT) activity were isolated from broccoli and functionally characterized (Lyi et al. 2007). While one gene product exhibited only HMT activity, the other displayed both HMT and selenocysteine Se-methyltransferase (SMT) activities, indicating the gene had evolved to include an additional function in sulfur/selenium accumulating species. The presence of acidic metalloproteases, serine proteases and cysteine proteases was found in broccoli florets (Rossano et al. 2011). Post-harvest senescence of broccoli florets was characterized by decrease in protein and chlorophyll contents and increase of protease activity, particularly cysteine protease. Pheophytinase (PPH) activity in broccoli florets increased concomitantly with a decline in chlorophyll a and b, suggesting that PPH may be involved in chlorophyll degradation (Aiamla-or et al. 2012). PPH activity in broccoli flowers treated with a UVB dose of 19 kJ/m² was repressed for the first 2 days of storage at 15 °C, whereas it increased gradually with senescence of broccoli florets. UVB treatment delayed upregulation of chlorophyll-degrading enzyme chlorophyllase genes resulting in the suppression of floret yellowing in stored broccoli.

Brassica plants (e.g., broccoli and cauliflower) contain substantial quantities of isothiocyanates (mostly in the form of their glucosinolate precursors), some of which (e.g., sulforaphane or 4-methylsulfinylbutyl isothiocyanate) are very potent inducers of phase 2 enzymes (Fahey et al. 1997). Three-day-old sprouts of cultivars of certain crucifers including broccoli and cauliflower were found to contain 10–100 times higher levels of glucoraphanin (the glucosinolate of sulforaphane) than do the corresponding mature plants. Broccoli seedlings cultivated using a 30/15 °C (day/night) temperature regime had significantly higher glucosinolate levels (measured at six consecutive days postemergence) than did sprouts cultivated at lower temperatures (22/15 and 18/12 °C). Both higher (33.1 °C) and lower (11.3 °C) constant temperatures induced higher glucosinolate levels in sprouts grown to a uniform size (Pereira et al. 2002). Glucosinolate levels were highest in cotyledons and lowest in roots of sprouts dissected both early and late in the 11-day developmental span investigated. Nongerminated seeds have the highest glucosinolate levels and concordantly greater induction of mammalian phase 2 detoxication enzymes. Levels declined as sprouts germinated and developed, with consistently higher glucosinolate content in younger developmental stages, independent of the temperature regime. Temperature stress or its associated developmental anomalies induced higher glucosinolate levels, specific elevations in glucoraphanin content and parallel induction of phase 2 chemoprotective enzymes. Pérez-Balibrea et al. (2008) found broccoli (flower head) to be a rich source of phytochemicals (glucosinolates and phenolic compounds) and micronutrients (vitamins and minerals), but germinated broccoli sprouts were found to contain much higher levels (10–100 times) of aliphatic (glucoraphanin) and indolic glucosinolates than the inflorescences. Broccoli sprouts grown in the light were found to have much higher concentrations of vitamin C (by 83 %), glucosinolates (by 33 %) and phenolic compounds (by 61 %) than those grown in the dark. Among the different organs studied (seeds, cotyledons, stems and roots), the cotyledons contained the highest levels of bioactive compounds, while the roots contained the lowest. Glucosinolate concentration in broccoli sprouts was strongly influenced by germination, causing a rapid increase during the first 3 days after sowing, and decreasing afterwards (Pérez-Balibrea et al. 2010). Fertilization with potassium sulphate 15, 30 and 60 mg/l at 9 and 12 days after sowing enhanced glucosinolate content.

Using liquid chromatography–mass spectrometry, Maldini et al. (2012) validated the presence of eight glucosinolates: glucobrassicin, glucoraphanin, glucoiberin, glucoerucin, progoitrin, gluconapin, sinigrin and glucocheirolin in broccoli sprouts with an overall recovery of 99 % for the eight glucosinolates. Using high-performance liquid chromatography–electrospray ionization–tandem mass spectrometry, Tian et al. (2005) quantified ten individual glucosinolates in broccoli, broccoli sprouts, Brussels sprouts and cauliflower. Detection limits for glucoiberin, sinigrin, progoitrin, glucoerucin and glucotropaeolin were 1.75, 1.38, 1.36, 0.6 and 0.63 pmol, respectively. Aliphatic glucosinolates (glucoraphanin, glucoiberin and glucoerucin) and a group of indole glucosinolates including 4-hydroxy-glucobrassicin and folates preformed in broccoli seeds were detected in broccoli sprouts (Rychlik and Adam 2008). In the early stage of sprouting, a reduction (approx. 20 %) of the aliphatic glucosinolates was determined, and levelling off up to the 12th day while 4-hydroxy-glucobrassicin declined continuously, three minor indole derivatives increased steadily, but remained at a comparatively low level. During germination, the contents of total folates increased to 72 μg/100 g fresh mass and 546 μg/100 g dry mass on the 4th day decreased again to 13 μg/100 g fresh mass until the 8th day of germination and remained at this low level. 5-Methyltetrahydrofolate was found as the predominant vitamer at each stage. Vitamin C, phenolic compounds, and glucosinolates in these purple sprouting varieties, EEP (Extra Early), EP (Early) and LP (Late) grown in Spain, were higher than in traditionally grown green broccolis and other purple broccolis grown under different climate conditions (Rodríguez-Hernández et al. 2012).

The following acylated anthocyanins were identified in broccoli sprouts: cyanidin-3-O-diglucoside-5-O-glucoside acylated and double acylated with p-coumaric, sinapic, caffeic, ferulic or sinapic acids, with at least three predominant anthocyanins isomers of cyanidin 3-O-(acyl)diglucoside-5-O-glucoside, cyanidin 3-O-(acyl 1)(acyl 2)diglucoside-5-O-glucoside and cyanidin 3-O-(acyl 1)(acyl 2)diglucoside-5-O-(malonyl)glucoside (Moreno et al. 2010). The early purple sprouting broccoli sprouts (‘Viola’) showed significantly higher concentrations of cyanidin 3-O-(sinapoyl) diglucoside-5-O-glucoside, cyanidin 3-O-(feruloyl) diglucoside-5-O-glucoside, cyanidin 3-O-(sinapoyl) (sinapoyl) diglucoside-5-O-glucoside and cyanidin 3-O-(sinapoyl) (feruloyl) diglucoside-5-O-(malonyl) glucoside. The total concentration of anthocyanins identified in ‘green head’ broccoli cultivars (Marathon, Nubia) and the green variety-for-sprouts (Intersemillas) was similar (0.23–0.29 mg/100 g) and significantly lower than that in cv. ‘Viola’ (0.64 mg/100 g).

Broccoli offers many health-promoting properties owing to its content of antioxidant and anticarcinogenic compounds (Mahn and Reyes 2012). The concentration and bioavailability of polyphenols, glucosinolates, sulforaphane and selenium depend on plant biochemistry, cultivation strategy and type of processing. Steaming and drying had been reported to result in an apparent increment of sulforaphane content as well as antioxidant activity, most likely due to an increase of the extractability of antioxidants and sulforaphane. Freezing and boiling were found to diminish polyphenols concentration, mainly due to volatilization and leaching into the cooking water. The two main flavonol glycosides quercetin 3-O-sophoroside and kaempferol 3-O-sophoroside and three minor glucosides of quercetin and kaempferol, namely, isoquercitrin, kaempferol 3-O-glucoside and a kaempferol diglucoside, were identified in broccoli florets (Price et al. 1998). The sophorosides of quercetin and kaempferol were present in raw florets at a level of 65 and 166 mg/kg fresh weight, respectively. The total content of quercetin and kaempferol glycosides expressed as aglycone was 43 and 94 μg/g fresh weight, respectively, and these agree with other recently published data. During the cooking process, only 14–28 % of the individual glucosides were retained in the cooked tissue, the remainder being largely leached into the cooking water with only a small loss attributed to the formation of the respective aglycones. Studies showed a general decrease in the levels of glucosinolates, phenolic compounds, and vitamin C compounds except for mineral nutrients which were stable under all microwave cooking conditions (López-Berenguer et al. 2007). Vitamin C showed the greatest losses mainly because of degradation and leaching, whereas losses for phenolic compounds and glucosinolates were mainly due to leaching into water. In general, the longest microwave cooking time and the higher volume of cooking water should be avoided to minimize losses of nutrients. Studies showed that during stir-frying of broccoli, phenolics and vitamin C were more affected than glucosinolates and minerals (Moreno et al. 2007). Stir-fry cooking with extra virgin olive, soybean, peanut or safflower oil did not reduce the total glucosinolate content of the cooked broccoli compared with that of the uncooked sample. The vitamin C content of broccoli stir-fried with extra virgin olive or sunflower oil was similar to that of the uncooked sample, but greater than those samples stir-fried with other oils.

The predominant sinapic acid derivatives in broccoli inflorescence after cooking were identified as 1,2,2′-trisinapoylgentiobiose and 1,2′-disinapoyl-2-feruloylgentiobiose; 1,2-diferuloylgentiobiose and 1-sinapoyl-2,2′-diferuloylgentiobiose were also identified (Vallejo et al. 2003c). The results showed large differences in flavonoid and hydroxycinnamoyl derivative contents in broccoli among the four cooking treatments. Clear disadvantages were detected when broccoli was microwaved, namely, high losses of flavonoids (97 %), sinapic acid derivatives (74 %) and caffeoylquinic acid derivatives (87 %). Conventional boiling led to a significant loss of flavonoids (66 %) from fresh raw broccoli, while high-pressure boiling caused considerable leaching (47 %) of caffeoylquinic acid derivatives into the cooking water. In contrast, steaming had minimal effects, in terms of loss, on both flavonoid and hydroxycinnamoyl derivative contents. Isothiocyanate content in broccoli juice treated by freezing, pasteurization and high pressure was found to range from 1.45 to 10.68, 1.94 to 5.54 and 1.87 to 11.27 μmol/l, respectively (Totušek et al. 2011). Sulforaphane content in broccoli juice treated by freezing, pasteurization and high pressure was found to range from 7.77 to 14.25, 8.87 to 15.67 and 8.24 to 14.64 μg/ml, respectively.

Production of nitrile during hydrolysis of unheated broccoli cultivars varied among cultivars from 91 to 52 % of hydrolysis products (Pinnacle > Marathon > Patriot > Brigadier) (Wang et al. 2012a). Boiling and microwave heating caused an initial loss of nitrile, with a concomitant increase in sulforaphane, followed by loss of sulforaphane, all within a minute. In contrast, steaming enhanced sulforaphane yield between 1 and 3 minutes in all but cv. Brigadier. The data indicated that steaming for 1–3 minutes provided less nitrile and more sulforaphane yield from a broccoli meal. Processing was found to reduce the glucosinolate content of broccoli, among other aspects due to thermally induced degradation (Hanschen et al. 2012). In broccoli sprouts, methylsulfanylalkyl glucosinolates were more susceptible to degradation at high temperatures, whereas methylsulfinylalkyl glucosinolates were revealed to be more affected in aqueous medium under alkaline conditions. Besides small amounts of isothiocyanates, the main thermally induced breakdown products of sulfur-containing aliphatic glucosinolates were nitriles. Although nitriles were most rapidly formed at comparatively high temperatures under dry heat conditions, their highest concentrations were found after cooking in acidic medium, conditions being typical for domestic processing.

The highest content of glucoraphanin and quinone reductase activity was found in broccoli florets stored under controlled atmosphere storage of 21 % O₂ + 10 % CO₂ at 5 °C (Xu et al. 2006). These conditions were able to maintain the visual quality, glucoraphanin content and quinone reductase activity of the broccoli florets for 20 days.

Vapor cooking of fresh broccoli did not generate 2,3,5-trithiahexane (TTH), but cutting broccoli into pieces followed by cooking in water allowed the detection of this odorant (Spadone et al. 2006). Cutting broccoli activates cysteine sulfoxide lyase transforming methylcysteine sulfoxide into methylsulfenic acid, which upon heating gives rise to dimethylsulfide and dimethyl trisulfide that react to TTH. The formation of TTH was enhanced upon frozen storage of cut broccoli pieces for a few weeks. Visual quality of broccoli heads declined significantly with increasing temperature and length of storage, caused primarily by increasing yellowing and loss of turgor (Winkler et al. 2007). Glucoraphanin, quercetin and kaempferol contents were not significantly affected by storage and marketing temperature and time. The results suggested that current transport and marketing practices were not likely to have a deleterious effect on the levels of aliphatic glucosinolates and flavonols in broccoli.

Studies showed significant differences among broccoli and cauliflower cultivar groups for the glucosinolate proportions as well as the contents of health-promoting and flavour-influencing alkyl, alkenyl and indole glucosinolates (Schonhof et al. 2004). These differences impacted on differences in sensory properties such as colour; taste properties such as bitter and sweet; and flavour such as green/grassy, spicy, broccoli-like, cabbage-like, cauliflower-like, kohlrabi-like, leek-like and mouthfeel pungent. Consumers were seen to prefer cultivars with a bright colour, a lower level of bitter-tasting glucosinolates (alkenyl and indole glucosinolates) and a higher sucrose content.

Endogenous folate poly-γ-glutamates in broccoli florets were found predominantly as hepta- and hexa-γ-glutamates (Munyaka et al. 2009). Crushing raw broccoli, acidification and LTLT (low temperature long time, 60 °C/40 minutes) blanching enhanced folate deconjugation, resulting in monoglutamate, di- and tri-γ-glutamates. Compared to other treatments, HTST (high temperature short time, 90 °C/4 minutes) blanching preformed prior to crushing resulted in the highest concentration of long-chain poly-γ-glutamates. Acidification combined with LTLT blanching decreased folate poly-γ-glutamates stability, whereas HTST blanching combined with different sequences of blanching and crushing did not affect folate poly-γ-glutamates stability. It was concluded that crushing (prior to heating), acidification and blanching could be strategically applied to increase the folate monoglutamate content of broccoli. Crushing of raw broccoli resulted in significant poly-γ-glutamate profile changes in broccoli, indicating γ-glutamyl hydrolase catalyzed hydrolysis (Munyaka et al. 2010). During treatments at 25–140 °C, folate retention was higher at 80 and 100 °C than at the other temperatures. A similar trend in thermal stability was observed for folates, vitamin C, total phenols and Trolox equivalent antioxidant capacity (TEAC) value, an indication that conditions that resulted in endogenous antioxidants degradation might also result in folate degradation.

Green colour of broccoli florets was retained under N₂ flow (<0.01 % O₂ and <0.25 % CO₂) and no flow (down to 1.3 % O₂ and up to 30 % CO₂), but these same conditions led to increased sour and sulfurous odors of the fresh product (Hansen et al. 1993). The relative concentration of ethanol, 3-hydroxy-2-butanone, and 2,3-butanediol increased, and C₅–C₇ aldehydes and alcohols decreased in the cooked broccoli which was stored fresh under N₂ or no flow conditions as compared to fresh samples and samples stored under air flow (20.5 % O₂ and <0.5 % CO₂), or restricted air flow (down to 17.2 % O₂ and up to 3.7 % CO₂). Schonhof et al. (2007) found elevated atmospheric CO₂ concentration had a differing effect on individual glucosinolates and glucosinolate groups in broccoli grown in the green house. Total glucosinolate content increased at elevated atmospheric CO₂ concentration as a result of a strong increase in both methylsulphinylalkyl glucosinolates glucoraphanin and glucoiberin. In contrast, indole glucosinolates simultaneously decreased, predominantly because of a reduction of glucobrassicin and 4-methoxy-glucobrassicin contents. Elevated CO₂ concentration increased photochemical quenching coefficient values in broccoli leaves by up to 100 and 89 % in heads, while glucose and sucrose in leaves increased by about 60 % (Krumbein et al. 2010). Further, in broccoli heads, elevated CO₂ concentration induced approximately a twofold increase in concentrations of three fatty acid-derived C(7) aldehydes ((E)-2-heptenal, (E,Z)-2,4-heptadienal and (E,E)-2,4-heptadienal), two fatty acid-derived C(5) alcohols (1-penten-3-ol, (Z)-2-pentenol), and two amino acid-derived nitriles (phenyl propanenitrile, 3-methyl butanenitrile). Contrariwise, concentrations of the sulfur-containing compound 2-ethylthiophene and C(6) alcohol (E)-2-hexenol decreased. Finally, elevated CO₂ concentration increased soluble sugar concentrations due to enhanced photochemical activity in leaves and heads, which may account for the increased synthesis of volatiles. Both applied modified atmospheres (1 % O₂ + 21 % CO₂; 8 % O₂ + 14 % CO₂) provided by two different microperforated biaxial-oriented polypropylene films maintained aliphatic glucosinolates in cauliflower florets, whereas in broccoli florets, the aliphatic glucosinolate concentration decreased slightly in each modified atmosphere (Schreiner et al. 2007). In addition, total indole glucosinolate concentration for both broccoli and cauliflower florets was maintained. Thus, to simultaneously maintain glucosinolates and external appearance as well as to prevent off-odor, a modified atmosphere of 1 % O₂ + 21 % CO₂ provided a suitable environment for storage of Brassica floret medley for up to 7 days at 8 °C. Rangkadilok et al. (2002 b) found the best method for preserving glucoraphanin concentration in broccoli heads after harvest was storage of broccoli in modified atmosphere packaging treatments and refrigeration at 4 °C. Such condition maintained the glucoraphanin concentration for at least 10 days and also maintained the visual quality of the broccoli heads.

The organosulfur chemicals, namely, glucosinolates and the S-methyl cysteine sulphoxide, found in broccoli (Stoewsand 1995; Vasanthi et al. 2009) in concert with other constituents such as vitamins E, C and K and the minerals such as iron, zinc and selenium and the polyphenols, namely, kaempferol, quercetin glucosides and isorhamnetin, were reported to be responsible for various health benefits of broccoli (Vasanthi et al. 2009). Thus, broccoli consumption had been reported to mediate a variety of functions including providing antioxidants, regulating enzymes and controlling apoptosis and cell cycle. Thus, the cancer chemopreventive effects of Brassica vegetables (like broccoli) that had been shown in human and animal studies may be due to the presence of both types of sulfur-containing phytochemicals (i.e., certain glucosinolates and S-methyl cysteine sulfoxide) (Stoewsand 1995). The cancer protective effect of cruciferous vegetables had been attributed to the presence of isothiocyanates, and sulforaphane, present in broccoli and broccoli sprouts, was by far the most extensively studied (Juge et al. 2007; Clarke et al. 2008). Sulforaphane had proved to be an effective chemoprotective agent in cell culture, carcinogen-induced, and genetic animal cancer models, as well as in xenograft models of cancer. These protective effects of sulforaphane involved multiple mechanisms activated in response to sulforaphane, including suppression of cytochrome P450 enzymes, induction of apoptotic pathways, suppression of cell cycle progression and inhibition of angiogenesis and antiinflammatory activity. Sulforaphane from broccoli had been reported to be an important potent inducer of cytoprotective proteins (also known as phase 2 enzymes) and to act as antioxidants with a dual protective role by (i) scavenging hazardous oxidants directly and instantaneously and (ii) inducing cytoprotective enzymes involved in the detoxification of carcinogens in carcinogenesis (Dinkova-Kostova and Talalay 2008).

Studies found that the antioxidant capacity of hydrophilic extracts of eight broccoli genotypes ranged from 65.8 to 121.6 μmol Trolox equivalents (TE)/g of tissue and the capacity of lipophilic extracts ranged from 3.9 to 17.5 μmol TE/g using the oxygen radical absorbance capacity (ORAC) assay (Kurilich et al. 2002). Ascorbic acid and flavonoid content of the hydrophilic extracts did not account for the total variation in antioxidant capacity of those extracts, suggesting either the presence of other antioxidant components that have yet to be identified or that the known antioxidants are producing synergistic effects. The carotenoids did correlate with antioxidant capacity of the lipophilic extracts and accounted for the majority of the variability in that fraction. Studies in noncellular assay systems indicated that broccoli extracts possessed a high capacity for scavenging free radicals or interrupting free radical reactions (Kurilich et al. 2003). Broccoli extracts were also found to protect against reactive oxygen species in HepG2 cells using the dichlorofluorescein–diacetate assay. In HepG2 cells, the level of protection against AAPH (2,2′-azobis (2-amidinopropane) dihydrochloride)-induced reactive oxygen species differed among broccoli genotypes tested. It was concluded that the HepG2 cell assay provided more information about the antioxidative lipid-soluble fraction than the commonly reported ORAC assay. Total antioxidant activity (as determined by ferric reducing antioxidant power) of North Indian broccoli cultivars ranged from 2.05 to 3.56 μmol Trolox/g fresh weight (Kaur et al. 2007). Free radical scavenging activity, as estimated by 2, 2-diphenyl-1-picrylhydrazyl, ranged from 57 to 74 %. Free phenolics ranged from 19.60 to 41.40 mg/100 g fresh weight and on an average constituted 73 % of total extractable phenolics. There was strong positive correlation between free phenolics and antioxidant activity.

For lipophilic broccoli extracts, oxygen radical absorbance capacity (ORAC-L) correlated with inhibition of cellular oxidation of dichlorofluorescein (dichlorofluorescein-L, R² = 0.596) (Eberhardt et al. 2005). Also, DNA damage in the presence of the lipophilic extract was negatively correlated with both chemical and cellular measures of antioxidant activity as measured by ORAC-L (R² = −0.705) and dichlorofluorescein-L (R² = −0.671), respectively. However, no correlations were observed for hydrophilic (−H) broccoli extracts, except between polyphenol content and ORAC (ORAC-H; R² = 0.778,). Inhibition of cellular oxidation by hydrophilic extracts (dichlorofluorescein-H) and ORAC-H was approximately 8- and 4-fold greater than dichlorofluorescein-L and ORAC-L, respectively. Studies showed that steam processing of broccoli elevated the total ORAC (hydrophilic, lipophilic) value by 2.3-fold (Roy et al. 2009). The hydrophilic part of a steam-processed broccoli had a significant reduction of 2,2′-azobis [2-amidinopropane] dihydrochloride (AAPH)-induced intracellular ROS level in comparison to that of fresh counterpart. Total phenolic content and total flavonoid content also increased in steam-processed broccoli.

The chloroform and ethanol extracts of broccoli florets showed 100 % antioxidant activity in the DPPH assay at 10 mg/1 ml concentration resembling that of vitamin C (ascorbic acid), and the crude extract gave 90.7 % (Motawea et al. 2010). Two novel glucosinolates identified as 2-mercaptomethyl sulfinyl glucosinolate (1) and (Z)-1-((2S,5S)-5-hydroxytetra-hydro-2H-pyran-2-ylthio)-2-(1H-indol-3-yl) ethylidene amino sulfate (2) and a known cinnamoyl [6′-O-trans-(4″-hydroxy cinnamoyl)4-(methylsulphinyl)butyl glucosinolate] (3) isolated from broccoli florets exhibited antioxidant activity (Survay et al. 2012). Compound 1 exhibited scavenging activity against DPPH with an inhibitory concentration IC₅₀ of 20 mM, whereas compound 3 was a weak antioxidant when compared to the standard quercetin (5 mM) as a positive control. Total phenolic content (TPC), total flavonoid content (TFC) and total glucosinolate content (TGsC) were almost higher in Calabrese cultivar than Southern star cultivar (Naguib et al. 2012). Calabrese cultivar showed higher 1, 1-diphenyl-2-picrylhydrazyl DPPH scavenging activity with IC₅₀ value of 16.56 μg/ml compared to Southern star 19.42 μg/ml. Additionally, Calabrese showed higher chelating power (75.36 μg/ml) than Southern star (72.43 μg/ml) at 30 μg/ml when the organic fertilizer was applied. The results indicated that there is a good margin for enhancing antioxidant compounds of broccoli for economic production using organic fertilization. Studies showed that combined hot air and UVC treatment of minimally processed broccoli may enhance protection against oxidative molecules not only by increasing levels of phenolics and ascorbic acid but also by enhancing the activity of enzymes (catalase and ascorbate peroxidase) involved in removing reactive oxygen species (Lemoine et al. 2010).