9th International Food data Conference, Norwich, UK, September 14-16, 2011

Variability of glucosinolates and phenolics in local populations from Turkey, Italy and

Federico Ferioli, Manuela Manco, Elisa Giambanelli, Luigi Filippo D’Antuono Department of Agroenvironmental Science and Technology, Food Science University Campus, University of Bologna, Italy

Tânia G. Albuquerque, Ana Sanches-Silva Departamento de Alimentação e Nutrição, Instituto Nacional de Saúde Doutor Ricardo Jorge, I.P. (INSA), Lisboa, Portugal

Bike Koçaoglu, Osman Hayran TC Yeditepe University, Istanbul, Turkey 9th International Food data Conference, Norwich, UK, September 14-16, 2011

From: Gladis and Hammer, KALES 1994, modified Kales, collards “Leafy cabbages”  Non-heading forms of Brassica oleracea L. var. polycephala  Plants not forming any head: nor vegetative, var. palmifolia conv.costata var. gemmifera like cabbage or reproductive, like cauliflower or var. sabellica sabellica x capitata broccoli. var. selenisia conv. gemmifera var. viridis  Taxonomy: still under review var. costata var. sabauda conv. acephala  They are considered primitive types of B. var. capitata oleracea, the closest to wild types. var. medullosa conv. capitata  The leaves are used, at different developmental var. gongylodes var. alboglabra var. italica stages. var. botrytis conv. fruticosa var. ramosa subsp.capitata  In Europe, substantially neglected by conv. botrytis breeding: no modern varieties do exist. Brassica oleracea

Local types, Turkey Old and young crops, Portugal Commercial crop, Italy 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Local kale populations are still cultivated in several European areas

 Also popular in southern United States, India, East Africa  A cool season , among the few able to grow during winter

Local types, in winter, Italy

Local types, under snow, Portugal

 Also used as a famine or war time crop (Dig for Victory campaign ) Sean Poulter World War Two vegetable comes back as 'superfood' Mail on line, 3 October 2007 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Kales represent basic ingredients of local traditional dishes

boerenkhol grøn lankgål

kailkenny gruenkhol und pinkel

kara lahana chorbasi

caldo verde prebugiun ribollita toscana kara lahana dolma 9th International Food data Conference, Norwich, UK, September 14-16, 2011 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Putative health-promoting compounds of kales

Glucosinolates

 Kales represent the more relevant potential dietary source of the glucobrassicin / indole-3- carbinol system

glucobrassicin (% total glucosinolates) of kale populations north 1 Turkey 2 Italy 3 19.3-36.9 78.9-84.1 47.7-90.7

 Indole-3-carbinol is being investigated as a protective agent against cancer development and other immune and hormone-related diseases 4

1. Velasco P., Cartea M.E., Gonzales C., Vilar M., Ordas A., 2007. Factors affecting the glucosinolate content of kale (Brassica oleracea acephala group). J. Agric. Food Chem., 55, 955-962 2. Sar ıkam ış G., Balkaya A., Yanmaz R., 2008. Glucosinolates in kale genotypes from the Black sea region of Turkey. Biotechn. & Biotechn. Equip., 22, 942-946 3. Baldacci G., D’Antuono L.F., Elementi S., Neri R., 2007. Componenti nutrizionali del cavolo nero di Toscana (Brassica oleracea L., ssp. acephala DC., var. sabellica L.). Graduation thesis, Food Science and Technology, University of Bologna

4. Higdon J., Drake V.J., Williams D.E., 2008. Indole-3-carbinol. Linus Pauling Institute. Oregon State University. http://lpi.oregonstate.edu/infocenter/phytochemicals/i3c/ 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Putative health-promoting compounds of kales

Phenolics  Phenolic components presently represent the main focus as dietary antioxidant agents  Kales have been characterised for their phenolic content and profile 1-5  They proved to be comparatively rich sources of these components, and associated antioxidant capacity, with respect to other 4-5 Carotenoids  Kales have been recognised as the richest sources of chlorophyll-associated carotenoids (βββ-carotene and lutein) among leafy vegetables. 5-6

1. Cartea M.E., Francisco M., Soengas P., Velasco P., 2011. Phenolic compounds in Brassica vegetables. Molecules, 16, 251-280. 2. Lin L.Z., Harnly J.M., 2009. Identification of the phenolic components of collard greens, kale, and chinese broccoli. J. Agric. Food Chem., 57, 7401-7408. 3. Olsen H., Aaby K., Borge G.I.A., 2009. Characterisation and quantification of flavonoids and hydroxyconnamic acids in curly kale (Brassica oleracea L. convar. acephala var. sabellica ) by HPLC-DAD-ESI-MS. 4. Ayaz F.A., Hay ırl ıoglu-Ayaz S., Aplay-Karaoglu S., Grúz J., Vanentová K., Ulrichová J., Strnad M., 2008. Phenolic acid contents of kale Brassica oleracea L. var. acephal a DC.) extracts and their antioxidant and antibacterial activities. Food chemistry, 107, 19- 25. 5. Nilsson J., Olsson K., Engqvist G., Ekvall J., Olsson M., Nyman M., Akesson B., 2006. Variation in the content of glucosinolates, hydroxycinnamic acids, carotenoids, total antioxidant capacity and low-molecular weight carbohydrates in Brassica vegetables. J. Sci. Food Agric., 86, 528-538.

6. Lefsrud M. Kopsell D., Wenzel A., Sheehan J., 2007. Changes in kale (Brassica oleracea L. var. acaphala) carotenoids and chlorophyll pigment concentrations during leaf ontogeny. Sci. Hortic., 112, 136-141. 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Kales in BaSeFood

Kales have been targeted as one of the investigation topics of BaSeFood (Sustainable exploitation of bioactive components from the Black Sea Area traditional foods), because of the following characters:  they are local crops, widely used as food plants, highly appreciated in the native areas  they are potential sources of health promoting substances and functional foods

A targeted research has been started in three BaSeFood participant countries: Turkey, Italy and Portugal, aimed at:

 comparing local populations and uses, in a cross country, cross-cultural perspective

 comparing phytochemical content of some populations

 determining bioactive retention in some traditional food preparation schemes

The comparison of glucosinolate and phenolic content of local populations from native areas, or grown in a common environment, is the subject of this contribution 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Materials and methods

Local populations

 25 samples taken from local crops or markets, winter 2010 • Turkey: 7 (kara lahana - black cabbage) • Portugal: 9 (couve galega - Galician kale) • Italy: 9 (cavolo nero - black cabbage)

 mature leaves random sampled, in triplicate  refrigerated and transported to the lab  frozen, freeze dried and kept at –20 °C until extraction 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Materials and methods

Experimental trial

 15 seeds samples obtained from local sources (local seeds companies, farmers) • Turkey: 6 • Portugal: 2 • Italy: 7

 seedlings produced in nursery, sown August 8, 2010  transplant in the field, Martorano 5 experimental farm, September 2, 2010  3 replication, randomised block  mature leaves harvested December 9, 2010.  frozen, freeze dried and kept at –20 °C until extraction 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Materials and methods Analytical determination of glucosinolates (GLS) and phenols

 inactivation of endogenous mirosinase in freeze-dried material at 70°C mixtures  simultaneous extraction of phenolics and glucosinolates by methanol/water  purification of GLS from phenols by solid phase extraction (stationary phase: DEAE Sephadex anion exchanger)  desulphation of GLS on SPE cartridge by sulphatase (from Helix pomatia )  elution of desulpho-GLS by water  quantification of GLS and phenols (on crude extract) by high performance liquid chromatography (HPLC) by external standard method  identification of compounds by a coupled system HPLC-mass spectrometry

HO OH HO OH Sinigrin O HO OH HO O HO O O O HO OH O O S OH HO OH OH O Kaempferol-3-O-(2-synapoyl- HO O OH O N Quercetin-3-O-(2-caffeoyl- sophoroside)-7-O-glucoside OH O OH sophoroside-7-O-glucoside) HO HO O HO O HO HO Glucobrassicin HO HO O O HO O HO O OH HO O HO S OH HO HO O HO O OCH3 N O OH N OH O H OH OH OCH3 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Total glucosinolates

Total glucosinolates Analysis of variance Significance 12000 Local samples Origin ** Field experiment 10000 Source ** B A C 8000 Origin × Source ** Samples ** d.m. -1 6000 A ** p < 0.01; * p < 0.05; NS, not significant

4000 A A mg kg B 2000 B B

0 Italy Portugal Turkey Bars = overall variability range 9th International Food data Conference, Norwich, UK, September 14-16, 2011

HPLC trace of desulpho-GLS extracted from a Turkish kale sample

1 Identified GLS (wavelength: 229 nm) 1 1: glucoiberin 2 + 3 2 + 3 : glucoraphanin + epi-progoitrin 4 + 5 : glucolepidin + sinigrin 6: 2-hydroxyethyl GLS 1 3 7: glucoalyssin 8: gluconapin 9: 4-hydroxybrassicin 10 : glucotropaeolin (internal standard) 11 : glucobrassicin 1 12 : glucobrassicanapin 0 13 : 4-methoxyglucobrassicin 6 7 1 14 : neoglucobrassicin 2 1 4 + 5 8 1 9 4 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Total phenolics

Total phenolics Analysis of variance Significance Local samples 400000 Field experiment Origin ** B A C Source NS 300000 Origin × Source ** d.m.

-1 A A Samples ** 200000 A ** p < 0.01; * p < 0.05; NS, not significant

mg kg B A A

100000

0 Italy Portugal Turkey Bars = overall variability range 9th International Food data Conference, Norwich, UK, September 14-16, 2011

HPLC trace of phenolics extracted from an Italian kale sample 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Discriminant analysis of relative content of glucosinolate and phenolic compounds Distribution of the experimental cases (origin x source) in the Correlations between relative component space of the five discriminant factors (DF) content and discriminant factors (DF)

DF1 DF2 DF3 DF4 DF5 10 6 1, Glucoiberin -0.512 -0.148 -0.071 0.282 0.179 Local samples DF 2 Local samples DF 3 2. Glucoraphanin 0.122 -0.136 0.242 -0.052 -0.309 Field experiment Field experiment 3. Sinigrin -0.910 -0.152 0.045 0.005 -0.006 Italy Italy 4. Gluconapin 0.035 0.279 -0.421 -0.050 0.145 Portugal Portugal 5. 4-Hydroxy-glucobrassicin -0.048 0.173 -0.181 -0.147 0.173 Turkey Turkey 6. Glucobrassicin 0.894 0.052 0.054 -0.044 -0.032 7. 4-Methoxy-glucobrassicin 0.234 0.367 0.400 -0.223 -0.220 8. Neoglucobrassicin -0.456 -0.363 -0.201 -0.063 -0.218 0.298 -0.187 0.244 -0.123 0.139 -16 8 1. 3-Caffeoylquinic acid 2. 3-p-Coumaroylquinic acid 0.236 -0.302 -0.104 -0.050 -0.041 DF 1 -16 8 3. 4-p-Coumaroylquinic acid 0.202 -0.303 -0.195 0.026 -0.020 4. 3-Feruoylquinic acid -0.158 -0.211 -0.141 0.050 -0.025 DF 1 5. Quer-3-diglu-7-glu 0.041 -0.407 -0.062 0.098 0.012 6. Quer-3-caf-diglu-7-glu + kaemp-diglu 0.113 -0.366 -0.049 -0.004 -0.035 7. Kaemp-3-diglu-7-glu -0.764 0.214 0.168 0.062 0.007 -6 -6 8. Sin-glu 0.414 -0.230 0.480 -0.115 0.126 9. Kaemp-3-hydrfer-diglu-7-diglu 0.264 -0.112 -0.012 -0.035 0.234 6 6 Local samples DF 4 Local samples DF 5 10. Quer-3-sin-diglu-7-glu 0.090 -0.473 -0.184 0.295 -0.119 Field experiment Field experiment 11. Quer-3-fer-diglu-7-glu + quer-3-diglu-7-sin-diglu -0.179 0.104 -0.560 0.317 -0.152 12. Kaemp-3-caf-diglu-7-glu + quer-3-diglu-7-fer-diglu -0.011 -0.080 -0.225 0.083 -0.118 Italy Italy 13. Kaemp-3-sin-diglu-7-glu + kaemp-3-sin-diglu-7-diglu 0.248 0.157 0.125 0.155 -0.158 Portugal Portugal 14. Kaemp-3-fer-diglu-7-glu -0.032 0.490 0.542 -0.243 -0.016 Turkey Turkey 15. Kaemp-3-fer-diglu-7-diglu 0.409 -0.228 -0.122 -0.110 0.077 16. Kaemp-3-fer-triglu-7-diglu -0.336 0.435 0.196 0.081 0.128 17. Keamp-3-cum-diglu-7-glu 0.005 -0.194 -0.020 0.193 0.078 -16 8 -16 8 18. Quer-3-sin-triglu-7-sin-diglu -0.168 -0.114 -0.206 -0.104 0.382 DF 1 DF 1 19. Quer-3-disin-triglu-7-glu -0.306 -0.030 -0.335 0.010 -0.174 20. Kaemp-3-disin-triglu-7-diglu 0.517 -0.001 -0.645 -0.118 0.133 21. Kaemp-3-fer,sin-triglu-7-diglu 0.129 0.002 0.127 -0.194 -0.116 22. Disin-diglu 0.296 -0.375 -0.441 -0.023 0.039 23. Sin-fer-diglu 0.373 -0.352 0.055 -0.470 0.138 -6 -6 24. Trisin-diglu 0.122 -0.246 -0.497 0.214 0.027 25. Disin-fer-diglu 0.040 0.124 -0.188 -0.233 -0.064

Variability explained 0.660 0.186 0.090 0.045 0.019 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Relative content of relevant glucosinolates

Glucoiberin Sinigrin 0.6 0.8 Local samples Local samples Field experiment Field experiment 0.6 B A B 0.4 B A C C total GLStotal ) total GLStotal ) -1 -1 A 0.4 A 0.2 A B A ANOVA Significance 0.2 A B A Origin ** A A Fraction ( Fraction mg mg Fraction Fraction ( mg mg Source ** A A 0.0 Origin × Source ** 0.0 Italy Portugal Turkey Samples ** Italy Portugal Turkey ** p < 0.01; * p < 0.05; NS, not significant

ANOVA Significance ANOVA Significance Glucobrassicin Local samples Origin ** Field experiment Origin ** 1.0 Source ** A B A Source ** Origin × Source ** A Origin × Source ** A Samples ** 0.8 B A Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant total GLStotal ) -1 0.6 A

0.4 B 0.2 Fraction ( ( Fraction mg mg 0.0 Bars = overall variability range Italy Portugal Turkey 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Relative content of relevant glucosinolates

4-Methoxyglucobrassicin Neoglucobrassicin 0.4 0.2 Local samples Local samples Field experiment Field experiment

B A C

B B A GLStotal ) total GLStotal ) -1 -1 0.2 0.1

A A A A A AA B B A ANOVA Significance AA Fraction ( Fraction mg mg Fraction ( ( Fraction mg mg 0.0 Origin ** 0.0 Italy Portugal Turkey Source ** Italy Portugal Turkey Origin × Source ** Samples ** ANOVA Significance ** p < 0.01; * p < 0.05; NS, not significant ANOVA Significance Origin ** Origin ** Gluconapin Source ** Source ** 0.5 Origin × Source ** Local samples Origin × Source ** Field experiment Samples ** Samples ** Samples ** 0.4 ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant B C A total GLStotal )

-1 0.3

0.2

A 0.1 A A B A B Fraction ( Fraction mg mg 0.0 Bars = overall variability range Italy Portugal Turkey A 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Relative content of relevant phenolics – correlated to DF1

Kaemp-3-diglu-7-glu Local samples Kaemp-3-sin-diglu-7-glu + Field experiment 0.4 kaemp-3-sin-diglu-7-diglu 0.6 C A B Local samples Field experiment a b A a 0.4

total total phenolics ) b total total phenolics ) -1

-1 A 0.2 A A A A A A A A A 0.2

0.0

Fraction Fraction ( mg mg 0.0 Fraction ( Fraction mg mg Ita ly Portugal Turke y Italy Portugal Turke y

ANOVA Significance ANOVA Significance Origin ** Origin ** Source ** Source NS Origin × Source * Origin × Source NS Samples ** Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant

3-caffeoylquinic acid Kaemp-3-fer-diglu-7-diglu 0.12 Local samples 0.2 Local samples Field experiment Field experiment 0.09 A AB B A B B total phenolics ) total total phenolics ) -1 0.06 -1 0.1

A A 0.03 A A A A B A A B A A 0.00

Fraction Fraction ( mg mg 0.0 Fraction ( mg mg Italy Portugal Turke y Ita ly Portugal Turkey

ANOVA Significance ANOVA Significance Origin * Origin ** Source ** Source NS Origin × Source * Origin × Source NS Samples ** Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Relative content of relevant phenolics – correlated to DF2

3-p-coumaroyl quinic acid 4-p-coumaroyl quinic acid Quer-3-diglu-7-glu 0.03 0.06 0.09 Local samples Local samples Local samples Field experiment Field experiment Field experiment A B B A B B A AB B 0.02 0.04 0.06 total phenolics ) phenolics total total phenolics ) phenolics total total phenolics ) phenolics total -1 -1 -1

A A 0.01 0.02 0.03 A A A A A A A A A A A A A A A A 0.00 0.00 0.00 Fraction ( mg mg mg mg ( Fraction Fraction ( mg mg mg mg ( Fraction Fraction ( mg mg mg mg ( Fraction Italy Portugal Turkey Italy Portugal Turkey Italy Portugal Turkey

ANOVA Significance ANOVA Significance ANOVA Significance Origin ** Origin ** Origin ** Source NS Source NS Source NS Origin × Source NS Origin × Source NS Origin × Source NS Samples ** Samples ** Samples * ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant

Quer-3-caf-diglu-7-glu+kaemp-diglu Quer-3-sin-diglu-7-glu Kaemp-3-fer-triglu-7-diglu Local samples 0.08 0.12 Field experiment Local samples Local samples 0.05 Field experiment Field experiment B A A 0.06 A B B 0.09 A B C

A A total phenolics ) phenolics total total phenolics ) phenolics total total phenolicstotal )

A A -1 -1 -1 A 0.06 A A A 0.04 A A A 0.03 A A A A B 0.02 0.03 A A

0.00 0.00 Fraction ( Fraction mg mg 0.00 Fraction ( mg mg mg mg ( Fraction Fraction ( mg mg mg mg ( Fraction Italy Portugal Turkey Italy Portugal Turkey Italy Portugal Turkey

ANOVA Significance ANOVA Significance ANOVA Significance Origin ** Origin ** Origin ** Origin ** Source NS Source NS Source NS Source NS Origin × Source NS Origin × Source ** Origin × Source NS Samples ** Samples ** Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Results Relative content of relevant phenolics – correlated to DF3

Sin-glu Local samples Quer-3-fer-diglu-7-glu + Kaemp-3-fer-diglu-7-glu Local samples Field experiment quer-3-diglu-7-sin-diglu 0.2 Field experiment 0.05 0.10 Local samples C B A A A A Field experiment a b ab A A A A

A phenolicstotal ) total phenolics ) phenolics total total phenolics ) phenolics total A -1 A -1 B -1 0.03 0.05 0.1 A A B A B B A B B B

0.0 0.00 0.00 Fraction mg( mg Fraction ( mg mg mg mg ( Fraction Italy Portugal Turkey mg mg ( Fraction Italy Portugal Turkey Italy Portugal Turkey ANOVA Significance ANOVA Significance ANOVA Significance Origin ** Origin NS Origin * Source ** Source ** Source ** Origin × Source NS Origin × Source ** Origin × Source ** Samples ** Samples ** Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05;NS, not significant

Quer-3-disin-triglu-7-glu Local samples Kaemp-3-disin-triglu-7-diglu Trisin-diglu Local samples 0.10 Field experiment 0.08 0.03 A A A Field experiment Local samples A C B Field experiment A B B 0.02 A A total phenolics ) phenolics total total phenolics ) phenolics total total phenolics ) phenolics total -1 -1 0.05 A -1 0.04 B A A A B 0.01 A B B B A A A A A A 0.00 0.00 0.00 Fraction ( mg mg mg mg ( Fraction Fraction ( mg mg mg mg ( Fraction Fraction ( mg mg mg mg ( Fraction Italy Portugal Turkey Italy Portugal Turkey Italy Portugal Turkey ANOVA Significance ANOVA Significance ANOVA Significance Origin NS Origin ** Origin ** Source ** Source ** Source ** Origin × Source ** Origin × Source ** Origin × Source ** Samples ** Samples ** Samples ** ** p < 0.01; * p < 0.05; NS, not significant ** p < 0.01; * p < 0.05;NS, not significant ** p < 0.01; * p < 0.05; NS, not significant 9th International Food data Conference, Norwich, UK, September 14-16, 2011

Conclusions

 First time exploration of populations from different countries in a common environment  Glucosinolate content in the range of other experiences  High variability due to environmental factors  Portuguese populations show a more different profile (sinigrin)

 Scarce clear references available for phenolics  Very high content in the populations examined  Some phenolics are more effective to discriminate growing conditions

 Kales confirmed to be very interesting food crops and potential easy to grow raw materials for food extracts or preparations.