Phytochemistry 116 (2015) 230–238 Contents lists available at ScienceDirect Phytochemistry journal homepage: www.elsevier.com/locate/phytochem Medicago spp. as potential sources of bioactive isoflavones: Characterization according to phylogenetic and phenologic factors ⇑ João C.M. Barreira a,b, , Tatiana Visnevschi-Necrasov a,c, Eugénia Nunes c, Sara C. Cunha a, Graça Pereira d, M. Beatriz P.P. Oliveira a a REQUIMTE, Departamento de Ciências Químicas, Faculdade de Farmácia, Universidade do Porto, Rua Jorge Viterbo Ferreira, n° 228, 4050-313 Porto, Portugal b CIMO-ESA, Instituto Politécnico de Bragança, Campus de Santa Apolónia, Apartado 1172, 5301-855 Bragança, Portugal c CIBIO-ICETA, Faculdade de Ciências, Universidade do Porto, R. Padre Armando Quintas 4485-661, Vairão, Portugal d INRB/IP – INIA – Instituto Nacional de Recursos Biológicos, Caia E São Pedro Estrada Gil Vaz, 7350-228 Elvas, Portugal article info abstract Article history: A high variety of plant species are often proposed as potential natural sources of specific bioactive com- Received 20 December 2014 ponents, with emphasis in phenolic compounds. However, the ability to produce a determined phyto- Received in revised form 20 April 2015 chemical might be variable, even among species with close phylogeny. Furthermore, the metabolic Available online 13 May 2015 dynamics vary greatly according to phenologic factors. Herein, it was verified whether isoflavone produc- tion in Medicago spp. is more associated with phylogenetic or phenologic determinants, to define the Keywords: optimal productive conditions. Isoflavone profiles were characterized in field-grown Medicago species Isoflavones in three phenologic stages. Isoflavones were extracted by matrix solid-phase dispersion method and ana- Medicago lyzed using high-performance liquid chromatography coupled with a diode-array detector. The obtained Phylogeny Phenology data were evaluated by a generalized linear model (GLM) and linear discriminant analysis (LDA). MSPD extraction Formononetin, genistein and irilone were the most abundant isoflavones, reaching values higher than HPLC-DAD those present in acknowledged plant sources like soy or red clover. Outputs from GLM and LDA indicate GLM that the phylogenetic factors are the most defining criteria. This study promotes Medicago spp. as poten- LDA tial isoflavone sources, particularly because the effects of these compounds are highly dependent on their type and concentration, with potential application as foodstuff, feedstuff, or in the nutraceutical and pharmaceutical industry. Ó 2015 Elsevier Ltd. All rights reserved. 1. Introduction are important sources of phytochemicals, including carotenoids, saponins or phytoestrogens (Yildiz, 2005), which are known to The genus Medicago is part of the botanical family of act as antimicrobial agents, phytoanticipins, phytoalexins, Leguminosae and includes about 56 different species mainly dis- structural barriers, modulators of pathogenicity, plant defense tributed in Mediterranean climatic conditions areas (Farag et al., genes activators, or fungitoxic agents (Von Baer et al., 2006). 2007). Besides alfalfa (Medicago sativa), which is the main Isoflavones are synthesized, accumulated and constitutively Medicago species grown throughout the world, inclusively as expressed as phytoalexins in response to pathogen attacks source of phytochemicals (Nunes et al., 2008; Silva et al., 2013), (Dakora and Phillips, 1996), contributing to the global strategies not much attention has been given to other Medicago species. of plant defense mechanisms (Aloui et al., 2012) and modulation Medicago truncatula, which is often chosen as a model for genomic of the interaction of Fabaceae species with nitrogen-fixing bacteria studies of Fabaceae due to its small diploid genome (5  108 bp), in rhizobium-legume symbiosis (Antunes et al., 2006). Isoflavones self-fertilization, easy genetic transformation and prolific nature, is are also known as having a wide range of beneficial biological the sole exception (Farag et al., 2007; Huhman and Sumner, 2002; activities in the human body (Mortensen et al., 2009), but their Schliemann et al., 2008; Stochmal et al., 2009). Medicago species overconsumption have been suggested as potentially causing adverse effects (Setchell and Cassidy, 1999). Hence, the intake of isoflavones has been limited by International Organisation (such ⇑ Corresponding author at: CIMO-ESA, Instituto Politécnico de Bragança, Campus as Food Safety Commission of Japanese Government or The de Santa Apolónia, Apartado 1172, 5301-855 Bragança, Portugal. Tel.: +351 Nutrient Data Laboratory of the Agricultural Research Service of 273303909; fax: +351 273325405. the United States Department of Agriculture) to very restricted E-mail address: [email protected] (J.C.M. Barreira). http://dx.doi.org/10.1016/j.phytochem.2015.04.011 0031-9422/Ó 2015 Elsevier Ltd. All rights reserved. J.C.M. Barreira et al. / Phytochemistry 116 (2015) 230–238 231 values (Sakamoto et al., 2015). Nevertheless, the intake amounts of isoflavones are usually evaluated on the basis of aglycone forms because the isoflavone glycosides are converted into aglycones by intestinal flora in vivo and in vitro (Sakamoto et al., 2015). The three main classes of phytoestrogens are isoflavones, lig- nans and coumestans (Jacobs et al., 2009), but their biosynthesis varies greatly with environmental and genotypic factors (Hoeck et al., 2000). Soybean, for instance, present mainly isoflavone agly- cones (daidzein, glycitein and genistein) and glycoside, acetylgly- coside and malonylglycoside forms. In contrast to soybean, red clover contains biochanin A and formononetin (aglycones), and their glycosides and malonyl derivatives as the major components, while in Medicago species, the most abundant isoflavones are for- mononetin, irilone and genistein (Visnevschi-Necrasov et al., 2014a). The isoflavones profile also varied greatly with the pheno- logical stage, varying along the plant maturity (D’Agostina et al., 2008). In lupin plants, the content of isoflavones was reported as negligible in seeds when compared to the amounts detected in leaves and roots (Bednarek et al., 2001). In a different study with M. truncatula, constitutive root isoflavonoids, like formononetin F: late flowering (one or more nodes with 50% open flowers, no seed malonyl glycoside or medicarpin malonyl glycoside, accumulate increasingly with the development of mycorrhizal symbiosis (Aloui et al., 2012). Furthermore, isoflavones profile is highly affected by genotype  environment interactions (Morrison et al., 2010). Herein, the profiles in free and conjugated isoflavones were compared among open-field grown Medicago spp. and the changes along vegetative cycle were monitored by evaluating three differ- ent phenologic stages: vegetative elongation, late bud and late flower. With this approach, it was intended to evaluate the effect of the plant species and the phenologic stage (as well as the inter- action of both factors) in the potential yield of individual and total isoflavones. 2. Results and discussion 2.1. Phylogenetic and phenologic influence on isoflavone profiles In this study eleven isoflavones were quantified, eluting in the following order: (1) puerarin (7,40-dihydroxy-8-C-glucosyli- soflavone), (2) daidzin (daidzein-7-O-b-D-glucoside), (3) genistin (genistein-7-O-b-D-glucoside), (4) daidzein (40,7-dihydroxy- isoflavone), (5) glycitein (40,7-dihydroxy-6-methoxyisoflavone), (6) genistein (40,5,7-trihydroxyisoflavone), (7) pratensein (40-meth- oxy-30,5,7-trihydroxyisoflavone), (8) formononetin (7-hydroxy-40- methoxyisoflavone), (9) irilone (9-hydroxy-7-(4-hydroxyphenyl)-[1, species. 3]dioxolo[4,5-g]chromen-8-one), (10) prunetin (40,5-dihydroxy-7- methoxyisoflavone) and (11) biochanin A (5,7-dihydroxy-40- methoxyisoflavone). Besides the chromatographic data, the Medicago proposed identification was also supported by UV spectra as described previously (Rodrigues et al., 2014). Chromatographic Puerarin Daidzin1 ± 1 Genistin7 ± 10nd Daidzein nd22 ± Glycitein 32 ndnd Genistein 6nd ± nd 8 Pratensein 1nd ± 1 3 ± 326 Formononetin ± nd 37 4 ± 1nd Irilone 5 nd ± 2 19 ± 1 7 ± nd 7 2 ± 3 4 ± 9 1 ± 2 Prunetin nd 5 nd nd ± 169 4 ± 78 Biochanin 38 7 ± A ± 15 5 3 1 ± ± 63 4 1 ± Total 10 quantified isoflavones nd 15 nd ± 21 13 ± 104 9 ± 38 664 ± 115 118 ± 18 7 9 1 ± ± ± 4 13 1 19 nd ± 5 42 ± 11 4 4 nd ± ± 5 3 206 ± 23 103 1 ± ± 51 25 1 ± 2010 6 ± 490 390 21 ± 1546 ± 227 ± 104 16 3 1215 395 ± ± ± 51 4 298 47 440 ± ± 5 52 119 1158 ± ± 22 435 nd 2806 225 ± ± 192 64 45 73 ± ± 14 23 620 ± 219 11 141 96 ± ± ± 16 30 37 7 103 ± ± 11 16 824 8 ± 243 ± 1185 3 ± 353 40 ± 18 34 217 ± ± 26 113 37 nd ± 10 1432 ± 696 272 18 26 ± ± 28 38 ± 249 4 ± ± 18 25 18 2342 nd ± 538 nd 2502 16 ± 63 ± 572 ± 13 36 nd 1437 ± 1527 448 ± 279 nd nd 3166 1250 ± ± 231 382 25 ± 6 1889 ± 455 2021 1378 ± ± 535 378 parameters, namely limit of detection (LOD), limit of quantification (LOQ), linearity, recovery and repeatability were accepted as previ- = 36) <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 = 108) <0.001= 324) <0.001 <0.001 <0.001 0.009 <0.001 0.277 <0.001 0.234 <0.001 0.122 <0.001 <0.001 0.053 <0.001 <0.001 <0.001 0.005 <0.001 0.058 <0.001 <0.001 <0.001 <0.001 n ously assessed (Visnevschi-Necrasov et al., 2014a). n n The effects of plant species (PSp) and phenologic stage (PhS), as -Value ( well as the interaction of both factors, were assessed simultane- -Value ( -Value ( M. arabica M. doliata M. minima M. murex M. orbicularis M. polymorpha M. rigidula M. tornata M. truncatula p 1 – VE2 – LB3 – LFp p 10 ± 21 nd 2 9 ± ± 5 24 nd 26 1 ± ± 43 1 4 ± 8 15 39 ± ± 30 84 4 5 ± ± 11 5 7 ± 15 208 ± 172 15 11 ± ± 25 19 17 ± 18 245 269 ± ± 229 252 16 1104 22 ± ± ± 17 779 24 1208 242 1478 ± ± ± 903 317 751 16 ± 23 354 390 ± ± 536 473 42 ± 36 25 15 ± ± 41 34 13 22 ± ± 16 14 1682 ± 707 1904 2252 ± ± 790 596 ously by evaluating changes in isoflavones composition of Medicago spp.