The cultivated peanut Arachis hypogaea L. is an allotetraploid   (AABB)    that    originated    from  a  single crossing event between the diploid wild species Arachis duranensis  and  Arachis  ipaënsis  (Kochert  et al., 1996;  Bertioli et al., 2016). These two species belong to section Arachis, which also comprises the cultivated and its most close relatives (Krapovickas and Gregory, 1994). All the other species of section Arachis are diploids with the single exception of Arachis monticola, which is also a tetraploid (Smartt et al., 1978; Fernández and Krapovickas, 1994; Peñaloza and Valls, 2005; Stalker, 2017).

Section Arachis species constitute the secondary gene pool of the cultivated peanut (Stalker and Moss, 1987) and because of that, many accessions of those species have been characterized and evaluated for several agronomic traits, including resistance to biotic (Stalker, 1984; Pande and Rao, 2001; Michelotto et al., 2015) and abiotic stresses (Nautiyal et al., 2008; Leal-Bertioli et al., 2012).

The reproductive barrier between cultivated and wild Arachis due to ploidy level difference has been overcomed using interspecific synthetic allotetraploids. Sterile diploid hybrids obtained by crossing A and B genome Arachis species have been turned fertile after their tetraplodization with colchicine, which allowed their crossing with peanut and the introgression of alleles from the wild species into the cultivated (Simpson , 1991).

Peanuts are among the few plant species that produce resveratrol (Lanz et al., 1990; Sobolev and Cole, 1999; Arora and Japlan, 2018). This phenolic compound is a potent antioxidant (Frankel et al., 1993) whose healing and preventive potential for many human diseases were described in some recent reviews (Colica et al., 2018; Galiniak et al., 2019). Resveratrol is also a phytoalexin that has been associated with resistance to major peanut diseases (Sobolev et al., 2007). Moreover, ten species of section Arachis also synthesizes resveratrol and three of them had levels higher than those found in cultivar Caiapó of A. hypogaea (Lopes et al., 2013).

The effect of polyploidization and hybrization on different traits in Arachis interspecific synthetic allotetraploids have been studied (Burow et al., 2001; Fávero et al., 2009, 2015; Leal-Bertioli et al., 2012, 2017; Michelotto et al., 2015, 2016, 2017). The characterization of resveratrol content in Arachis allotetraploids could add new value to these genotypes, which have been developed to be used in peanut pre-breeding programs (Bertioli et al, 2011). In this context, the objective of the present study was to evaluate the impact of polyploidization and hybridization in the resveratrol content analyzing synthetic allotetraploids, their respective diploid wild parentals and hybrids between two synthetic allotetraploids and three A. hypogaea cultivars.

Plant material

Seventeen Arachis genotypes were analyzed for resveratrol content being six wild diploid species, three tetraploid peanut (A. hypogaea) cultivars   and    eight    tetraploid    hybrids    that    comprised   five interspecific synthetic allotetraploids, and three hybrids resulting from crosses between peanut and synthetic allotetraploids (Table 1). The Arachis wild species analyzed harbor different types of genome: A (villosa, stenosperma, A. duranensis), B (A. ipaënsis and Arachis gregoryi) and K (batizocoi). The cultivars of peanut analyzed were ‘IAC Caiapó’, ‘Runner IAC 886’ and ‘IAC 505’. All synthetic allotetraploids and hybrids analyzed were developed by Santos (2013). The plants were grown in greenhouses at Embrapa Genetic Resources and Biotechnology, Brasília, Brazil.

Induction of resveratrol synthesis using UV

The experiments were performed using detached leaves collected from six-month-old Arachis plants in greenhouse conditions, as previously described (Lopes et al., 2013). In short, detached leaves were exposed to an ultraviolet light for 2 h 30 min and maintained in the dark for additional 15 h at room temperature. UV-treated and non-treated control leaves of each genotype were divided into three aliquots of 1 g and stored at -80°C.

Resveratrol extraction and sample preparation

The resveratrol extraction protocol was based on Potrebko and Resurreccion (2009). Prior to high performance liquid chromatograph (HPLC) injection, the dried residue was reconstituted in 6.8 ml of 15% (v/v) ethanol. The samples were vortexed for 1 min and left in ultrasonic bath for four minutes. The procedure was repeated twice to ensure the complete recovery of the extract. The samples were then transferred to 2.0 ml tubes and centrifuged for 15 min at 25°C at 13,400 rpm. The supernatant of the centrifuged material was conditioned in a 2-ml tube and then used for injection in a HPLC (CLAE, Varian®) with ternary pump, automatic dial and coupled photodiode array detector (PDA Varian® PS- 240 / PS-410 / PS-335 / Galaxie Software 1.9).

HPLC analysis

The column used in HPLC was Zorbax XDB Agilent (250 x 4.6 mm, 5 μm), without guard column. A gradient of acetonitrile and a 0.02% aqueous phosphoric acid (J. T. Baker) were used as mobile phase. The conditions were: acetonitrile for 0 min at 13%; 6 to 9 min at 15%; 17 min at 17%; 28 to 33 min at 28%; 40 min at 50%; 45 min at 60%; 46 to 48 min at 80%; 49 to 54 min at 13%; flow rate of 1.0 ml/min. The UV absorption was monitored at 308 nm, 280 nm and also at the maximum absorption wavelength of each eluent (PDA). The injection volume of each sample was 10 μl.

The peak of resveratrol was identified by comparison with the retention time of the commercial standard solutions of resveratrol (> 99%, 230-240 μg/ml, Sigma-Aldrich) and phenolphthalein (> 98%, 2927-2835 μg/ml, Sigma-Aldrich) that were injected daily for area verification. Additional procedures for resveratrol identification were analysis of the spectrum provided by the diode array detector, and the quantification by co-elution with the resveratrol pattern and further comparison of the chromatograms of the induced and control samples. The final concentration of resveratrol per gram of leaf was calculated according to Potrebko and Resurreccion (2009).

Data analysis

The means of the resveratrol production were compared using the Scott and Knott test at 5% probability, considering the groups of plants over time (3 blocks) as covariate, aiming at filtering the variability  observed  due  to  these   repetitions.  The  analysis  was developed in the statistical language program R, free for download at the site http://www.r-project.org/.

All genotypes analyzed were able to produce resveratrol in response to UV induction (Table 1). Traces of resveratrol (below 0.1 μg) were detected in the samples not exposed to UV (data not shown).

The resveratrol content varied greatly among the six wild diploid species analyzed going from 42.53±12.5 μg/g in A. stenosperma to 415.8±118.0 μg/g in A. duranensis (Table 1). Lopes et al. (2013) detected 370.0 μg/g of resveratrol in UV-treated plants of A. gregoryi (accession V6389) which was very similar to the value found in the present study (390.3±30.9 μg/g). Conversely, for A. batizocoi (accession K9484) and A. ipaënsis (accession K30076), Lopes et al. (2013) detected higher contents (524.5 and 314.0 μg/g, respectively) than those found here (162.1±147.0 μg/g and 84.5±46.4 μg/g, respectively). Also, Carvalho et al. (2017) detected resveratrol in UV-treated leaves of A. duranensis (accession V14167) in concentration (371.97 μg/g) similar to that observed here (415.8±118.0 μg/g), whereas in A. stenosperma, (accession V10309) resveratrol concentration was at least 13-times higher (512.6 μg/g) than in our study (42.49±12.5 μg/g). The differences in the resveratrol content of a same accession observed in these studies may be due to different factors, such as the intrinsic nature of resveratrol as a secondary metabolite, whose production is prone to changes according to the environment temperature (Wang and Zheng, 2001), plant age (Chung et al., 2001), water availability in the soil (Esteban et al., 2001), and cultivation season (Chen et al., 2002). Genetic, ontogenic, morphogenetic, and environmental factors that could cause variation on plant secondary metabolite content in different species were reviewed by Yang et al. (2018).

Concerning the three peanut cultivars evaluated, ‘IAC Caiapó’ presented the highest resveratrol concentration (579.5 ± 85.8^b μg/g), followed by ‘Runner IAC 886’ (275.2 ± 88.0^e μg/g) and ‘IAC 505’ (212.7 ± 88.0^e μg/g) that showed similar concentrations to each other. Over the years, many studies have shown a variable resveratrol content among A. hypogaea varieties/cultivars, with differences due to the plant organ studied, crop location, annual season and pathogens infestation levels. Sanders et al. (2000) found differences among resveratrol content (from 0.03 to 0.147 µg/g) in seeds without coat of three peanut market types (Virginia, Runner, and Spanish) produced in different areas and without any specific induction of resveratrol, as UV used in this study. Significant variations in resveratrol content (from 0.125 to 1.626 μg/g) was also found when seeds of 20 germplasm accessions of A. hypogaea harvested from the same field were analyzed using HPLC (Wang and Pittman, 2009). Variation on resveratrol content was also found in roots of three peanut  cultivars grown in 2000 fall and 2001 spring being the content of fall crops much higher than those of spring (Chen et al., 2002). Peanut cultivar ‘IAC Caiapó’ higher resveratrol content (ranging from 300 to 600 μg/kg) when compared to the cultivar ‘IAC 886’ (Zorzete et al., 2011). This last cultivar is less susceptible to thrips Enneothrips flavens infection (Moraes et al., 2005). Considering that resveratrol is a phytoalexin, peanut cultivars with higher concentrations of this metabolite are likely more resistant against pathogen attack. Resveratrol content in peanut seeds was negatively correlated with aflatoxin production and in vitro trials demonstrated that resveratrol could inhibit aflatoxin production (HouMiao et al., 2012). An association between total phytoalexin production and genotype resistance to major peanut diseases was observed being trans-resveratrol was one of the main compounds found in stress-resistant genotypes (Sobolev et al., 2007).

The grouping of the genotypes according to their ploidy level (Table 2), helped to observe that the tetraploids genotypes (peanut cultivars and hybrids) showed significantly higher resveratrol contents than the wild diploid species. The effect of polyploidization has been studied in Arachis comparing synthetic allotetraploids, their corresponding diploid parental species and peanut cultivars. The comparison among A. duranensis (V14167), A. ipaënsis (KG30076), a synthetic allotetraploid (A. duranensis V14167 × A. ipaënsis K 30076)^4× and A. hypogaea subsp. hypogaea var. hypogaea ‘Runner IAC 886’ showed some diploid traits such as chlorophyll meter readings are maintained through hybridization and polyploidization and most characters are substantially modified (Leal-Bertioli et al., 2012). An increase in resistance to the foliar diseases rust (Puccinia arachidis) and late leaf spot (Cercosporidium personatum) was observed in the synthetic allotetraploids compared to their diploid parental species (Kumari et al., 2014). More recently, it was also demonstrated that Arachis allotetraploids have some general phenotypic trends that are common, regardless of the combination of their wild parental diploid suggesting that nucleotypic effect is more important than new allelic combination (Leal-Bertioli et al., 2017). The effect of polyploidization on the increase of bioactive compounds has also been studied in some other species. The concentrations of some phytoconstituents, such as emodin, physcion, piceatannol, resveratrol and rutin were determined by LC–MS in three species of Rumex and a positive correlation could be detected with the increasing ploidy status in different chromosomal races (Jeelania et al., 2017).

This study data suggest a positive effect of polyploidy and hybridization in resveratrol content in Arachis hybrids. Resveratrol can be synthesized by a few species and the major dietary natural sources include grapes, wine, peanuts, and soybeans (Burns et al., 2002). Our data opens the possibility to create and provide new sources of natural resveratrol by the used of interspecific synthetic Arachis hybrids analyzed, mainly the BCs genotypes, which displayed higher resveratrol contents than wild and the cultivated species.

The authors thank Embrapa (Empresa Brasileira de Pesquisa Agropecuária) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for the financial support (Embrapa grant number 03.09.01.014.00.00 and CNPq grant number 482562/2011-0) along with CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for Paula Vasconcelos´ scholarship.

The authors have not declared any conflict of interests.