Phenolic constituents and antioxidant properties of five wild species of Physalis (Solanaceae)

Background Fruits of wild species of the genus Physalis are consumed as food and calyces and leaves are used in traditional medicine. The phenolic composition of the species of this genus have been scarcely studied. To contribute to a better knowledge for the use of all the potential of these wild species of plants, leaves, fruits, and calyces of five wild species of the genus were analyzed for their phenolic composition and antioxidant properties. Results Important tissue- and species-dependent variations were found. Calyces of Physalis subulata showed the highest contents of phenolics (176.58 mg of gallic acid equivalents/g dry tissue), flavonoids (39.63 mg/g dry tissue), and phenolic acids (50.57 mg of quercitrin equivalents/g dry tissue), and its leaves displayed the highest total antioxidant capacity (3.59 mg of ascorbic acid equivalents/mL) and one of the highest reduction powers (0.54 µg of ascorbic acid equivalents/mL). A high performance liquid chromatography with photodiode array detection analysis revealed a total of 28 phenolic compounds in foliar tissues (mainly kaempferol-3-O-glycosides), 16 in fruits (mainly phenolic acids), and 16 in calyces (mainly kaempferol-3-O-glycosides); the profiles of these compounds in the three types of tissue were species-specific. Conclusions The studied species of Physalis are important sources of phenolics with relevant antioxidant activity. The current results indicate that phenolic profiles are valuable specific chemical markers and can be relevant as food tracing and authenticity indicators for plant-based preparations involving species of Physalis.


Background
Plants are a promising source of beneficial compounds for human health and represent a starting point for the discovery and development of drugs for the treatment of human diseases (Kouloura et al. 2014). There is an impressive plant diversity worldwide, in which the chemical variation is represented by a great deal of bioactive compounds. The interest in wild plants as a source of natural antioxidants is valid as a healthier and cheaper alternative to the synthetic ones, which have been perceived as toxic and carcinogenic (Krishnaiah et al. 2011).
The family Solanaceae is constituted by around 2300 species (D' Arcy 1991), which are significant sources of phytochemicals (Eick 2008). This family includes important cultivated species, such as chili (Capsicum annuum L.), tomato (Lycopersicon esculentum Mill.), and potato (Solanum tuberosum L.). The family also includes numerous wild species, like most belonging to the genus Physalis, which is the fifth largest genus of the family, including 70-90 species (Whitson and Manos 2005). Mexico, with about 50 species of Physalis growing in its territory, is considered the center of origin, diversity (D' Arcy 1991), and domestication of this genus (Santiaguillo et al. 1994). The fruits of many species of Physalis have been important elements in the culinary traditions of the people of Mesoamerica, consumed fresh or cooked, and calyces and leaves have been used in folk medicine since pre-Columbian times. Calyces are also consumed as seasoning and leaves as food (Hernández and Yáñez 2009). However, at present, only a few species are cultivated, P. ixocarpa, P. peruviana, and P. alkekengi are among them.
Phenolic compounds are synthesized and accumulated in practically all plant tissues. Many of these compounds, mainly flavonoids and phenolic acids, have biological properties with medical implications (Zhang and Cui 2005). The antioxidant capacity is one of the most important of these properties, because oxidation damages nucleic acids, proteins, lipids, and other macromolecules, producing cardiovascular disorders, neurodegenerative diseases, and cancer (Ames et al. 1993). Flavonoids and phenolic acids have relevant antioxidant properties (Barriada-Bernal et al. 2014). The antioxidant capacity of plant phenolic extracts depends on their concentration (Dobre et al. 2010), but also on their accumulation profiles inside plant tissues (Barriada-Bernal et al. 2014). The concentration is affected by environmental conditions, age, and phenological stage (Almaraz-Abarca et al. 2013), while the qualitative phenolic profiles are more stable and vary among different groups of plants with a species-specific tendency (Emerenciano et al. 2001).
Few species of Physalis have been analyzed for their phenolic composition and antioxidant properties, among them are P. ixocarpa (González-Mendoza et al. 2011), P. peruviana (Rockenbach et al. 2008;Wu et al. 2009), P. angulata (Ismail and Alam 2001), and P. alkekengi var. franchetii (Diaz et al. 2012). The aim of the present study was to evaluate the abundance and diversity of phenolics and the antioxidant properties of leaves, edible fruits, and calyces of five wild species of Physalis [P. angulata L., P. patula Mill., P. hederifolia A. Gray var. hederifolia, P. solanacea (Schltdl.) Axelius, and P. subulata Rydb.] from Durango, Mexico to determine their potential as sources of natural antioxidants. The significance of the phenolic profiles as specific chemomarkers and their potential as tool for food tracing and authenticity was also evaluated.

Plant material
Leaves, fruits, and calyces of mature individuals (blooming and bearing fruits) of six natural populations of Physalis were collected in different locations of Durango, Mexico (Table 1). Tissues of four individuals per population were randomly sampled, combined, and three pools of samples were formed and separately analyzed. Voucher specimens were deposited at the Herbarium CIIDIR. Samples were individually dried in a ventilated oven at 40 °C, and then ground in a domestic blender. The dry, ground tissues were kept in paper bags at room temperature, in a desiccator with silica, in darkness until analysis.

Preparation of extracts
Phenolic compounds were extracted from dry ground tissues (1 g) by maceration in 20 mL of 80 % methanol (v/v), for 24 h, by shaking at 100 rpm, in darkness, at room temperature. The extracts were centrifuged (4800g) for 5 min at room temperature and the supernatants formed the total extracts. Different aliquots were taken to be used in the determination of phenolic content, antioxidant potential, and HPLC-DAD analysis.

Determination of total phenolics
The concentrations of total phenolics of each sample were determined using the Folin-Ciocalteu reagent, according to Nurmi et al. (1996). The phenolic contents were calculated by using a calibration curve of gallic acid (A 760 = 0.003[gallic acid] − 0.0241, r = 0.9985) constructed with five concentrations of this compound (0-92 μg/mL). Phenolic contents were expressed as milligrams of gallic acid equivalents per gram of dry tissue (mg GAE/g dt).

Free radical scavenging activity
Determination of the free radical scavenging activity was carried out following the DPPH* method described by Barriada-Bernal et al. (2014). Decreases in absorbance (at 523 nm) of an initial DPPH* solution (62 µg/mL methanol) against increasing flavonoid concentrations of samples were plotted to determine, by linear regression, the efficient concentration at 50 %, defined as the amount of antioxidant needed to decrease by 50 % the initial DPPH* concentration (EC 50 ). The following curve made with DPPH* between 2 and 60 µg/mL was used to estimate the DPPH* concentration (µg/mL) in the reaction medium: A 523nm = 0.0019 + 0.0309 [DPPH*], correlation coefficient r = 0.9996. Antiradical activities were expressed in terms of EC 50 in microgram per milliliter (µg/mL). Quercetin and epicatechin were analyzed in the same manner as references.

Total antioxidant capacity (TAC)
Total antioxidant capacity was evaluated according to Prieto et al. (1999). In this method, the reduction of Mo (VI) to Mo (V) by an antioxidant, forming a green phosphate/Mo (V) complex at an acidic pH is measured. TAC values were expressed as milligrams of ascorbic acid equivalents/mL (AAE/mL), which were calculated using the following curve: A 695 = −0.2365 + 4.2133 [ascorbic acid], correlation coefficient r = 0.9987, constructed with six concentrations of ascorbic acid between 0.1 and 1 mg/ mL. Quercetin (0.1 mg/mL) and epicatechin (0.1 mg/mL) were analyzed in the same manner as references.

Iron reducing power
The iron reducing power method is based on the measurement of the formation of Fe 2+ from Fe 3+ in the presence of antioxidants; Fe 2+ was estimated according to Siddhuraju and Becker (2003). Results were expressed in terms of IC 50 in micrograms of ascorbic acid equivalents per milliliter (µg AAE/mL), calculated from the following curve of ascorbic acid: A 700 = 25.372 [ascorbic acid] − 0.0242, correlation coefficient r = 0.9956, constructed with five concentrations of ascorbic acid between 0.03 and 0.1 mg/mL. The highest absorbance values indicated the greatest reducing capacity. Quercetin and epicatechin were analyzed in the same manner as references.

Data analysis
All the assays were carried out for three independent pools of each sample. Data were subjected to an analysis of variance (p ≤ 0.05), and means were separated by Tukey test. Correlations between different parameters were carried out with Pearson test, by using the SPSS statistics 17.0 software. A principal component analysis (PCA), considering all the quantitative parameters, was carried out by using Past 3.0; the contribution of each parameter for the differentiation of samples was evaluated. For the different tissues (leaves, fruits, and calyces), the phenolic profiles were made up of all compounds present in the respective HPLC-DAD chromatogram. Each compound was treated as a single chemical character. A binary matrix coded by 1 (presence) or 0 (absence) formed by all individual samples vs. all resolved compounds for each type of tissue was subjected to a cluster analysis (paired group algorithm and Jaccard similarity measure) by using Past 3.0.

Phenolic, flavonoid and phenolic acid contents
The phenolic contents are displayed in Table 2. Significant tissue and species-dependent variations were found. Among the foliar tissues, the leaves of P. patula accumulated the highest level of phenolics (129 mg/g dt) and among the fruits, those of P. hederifolia var. hederifolia were the richest ones in total phenolics (86.51 mg/g dt).
The calyx phenolic content of P. subulata was the highest found in the current study (176.58 mg/g dt). Table 2 shows the flavonoid contents found in the different tissues of the analyzed species of Physalis. Significant tissue and species-dependent variations were found. The foliar flavonoid contents explained between 7.55 and 56.86 % of total phenolics, the highest level was found for P. angulata (23.036 mg/g dt). Flavonoid contents in fruits represented between 0.0 and 16.91 % of total phenolics, the fruits of P. subulata showed the highest level (5.462 mg/g dt). Flavonoid levels in calyces (representing between 3.08 and 30.43 % of total phenolics) ranged from 3.67 mg/g dt in P. solanacea (sample 412) to 39.63 mg/g dt in P. subulata.
Species-and tissue-dependent variations were also found in the phenolic acid contents (Table 2). In leaves, phenolic acids corresponded to between 2.11 and 16.55 % of total phenolics and P. angulata leaves showed the highest level (6.7 mg/g dry tissue). In fruits, phenolic acids represented between 0.0 and 21.25 % of total phenolics; the fruits of Physalis angulata were the richest in these type of phenolics (7.84 mg/g dt). Excepting the calyces of P. hederifolia var. hederifolia, which practically did not accumulate phenolic acids, these compounds explained from between 2.86 and 28.63 % of total phenolics. The content of phenolic acids in the calyces of P. subulata (50.57 mg/g dt) was outstanding.
The HPLC-DAD analysis revealed a total of 28 phenolic compounds in the leaves of the five species of Physalis analyzed (the respective chromatograms are shown in Fig. 1). Twenty of those compounds were flavonols (among them, 14 were kaempferol-3-O-glycosides and 5 were quercetin-3-O-glycosides) and 8 were phenolic acids. The respective retention times and λ max are displayed in Table 3. The HPLC chromatograms of fruits (Fig. 2) revealed a total of 16 compounds: 11 phenolic acids, 3 flavonols (2 quercetin-3-O-glycosides and 1 kaempferol-3-O-glycoside), and 2 dihydroflavonols (retention times and λ max are displayed in Table 3). A total of 16 phenolic compounds were unveiled by the chromatograms of calyces (Fig. 3), among which 12 were flavonols (10 kaempferol-3-O-glycosides and 2 quercetin-3-O-glycosides) and 4 were phenolic acids ( Table 3). The concentration of each phenolic compound found in the different analyzed species is shown in Table 3.

Cluster analysis
The results of the cluster analysis based on the matrices constructed with each foliar, fruit, and calyx phenolic profile of each sample of Physalis are displayed in Fig. 4. All species were discriminated one from each other at levels of similarity from around 0.06 (Jaccard similarity measure, according to the foliar phenolic profiles) to around 0.42 (according to the calyx phenolic profiles).

Antioxidant assays
The results of the free radical scavenging activity, total antioxidant capacity, and iron reducing power are shown in Table 2. Specific and tissue dependent variations were found in the three antioxidant assays.

Correlation analysis
While the kinetic evaluation of all the antioxidant assays was highly related to flavonoid contents (0.9731 < r < 0.999), the correlation analysis revealed lower associations between the antioxidant properties and the phenolic, flavonoid, and phenolic acid contents in the samples (0.224 between calyx total flavonoids and calyx IC 50 <Pearson correlation value <0.856 between calyx total phenolics and calyx IC 50 ).

Principal component analysis
The results of a PCA, based on the quantitative determinations (phenolic, flavonoid, and phenolic acid contents and the estimation of antioxidant capacity by three methods) are showed in the Fig. 5. Three principal components accounted for 93.77 % of total variance, being the calyx phenolic content (PC1) the main one, taking 73.467 %, with the highest relative discriminating power (eigenvalue 4860.76). The calyx antiradical activity (PC2) explained the 12.425 % of total variance (eigenvalue 822.079), and the foliar phenolic content, the 7.88 % (eigenvalue 521.505).

Phenolic, flavonoid, and phenolic acid contents
Phenolic compounds contribute to the organoleptic and antioxidant properties of foods and they are among

Table 2 Phenolics, flavonoids, phenolic acid contents and antioxidant evaluations of five wild species of Physalis
The values represent the mean and standard deviation for three independent samples. Different letters in the same column for each type of tissue mean significant differences (p < 0.05)  (Tapas et al. 2008). Some, like phenolic acids, play important roles in fruit maturation, prevention of enzymatic browning, and food preservation (Robbins 2003). Phenolics also represent important tools for food tracing and to prove authenticity of foods (Zimmermann and Galensa 2007). The understanding of their abundance, diversity, and distribution in the plant kingdom and in the different plant tissues offers the opportunity of developing new drugs to improve human health, and to develop new varieties with enhanced accumulation of these compounds through breeding programs. The potential of different species to synthesize and accumulate flavonoids, and phenolics in general, is the result of the interaction of genetic and environmental factors (Veit et al. 1995) and different species of plants can synthesize and accumulate widely variable levels (Diaz et al. 2012). Aside from the differences in the flavonoid levels due to diverse environmental conditions of growing and to the specific variations, differences between tissues can be expected as the result of the differential regulation of the genic expression in well differentiated cells.

Species and reference compounds
The total phenolic concentrations observed in the leaves and fruits of the analyzed species of Physalis were comparable to those reported by Wang and Lin (2000) in the leaves (30.9-129.2 mg/g dry matter) and fruits (9.16-23.10 mg/g dry matter basis) of several cultivars of black raspberry, red raspberry, and strawberries, all considered as important source of phenolics. To our knowledge, this is the first report of phenolic contents in calyx for the species here analyzed of Physalis and reveals that these structures can accumulate relevant phenolic levels. The present results indicate that the species analyzed of Physalis are important sources of total phenolics, mainly accumulated in leaves and calyces.
Comparatively, the found foliar flavonoid contents were lower than those reported by Wu et al. (2009) for the leaves of P. peruviana (37.39-226.19 mg/g dry extract, depending on the type of solvent used for the extraction). Excepting P. angulata, in whose fruits no flavonoids were found, the fruit flavonoid contents found in the present study (Table 2) were higher than that reported for ripe fruits of Lycium barbarum (Fructus Lycii) (0.72 mg/g dry fruit), whose importance as source of antioxidants has been reported by Le et al. (2007). The flavonoid levels found in the calyces of P. subulata (Table 2) can be comparable to that reported for the aqueous extract of P. peruviana leaves (37.39 mg/g) by Wu et al. (2009). Flavonoids were mainly accumulated in leaves and calyces.
Studies on determinations of phenolic acids in species of Physalis have been mainly focused on fruits (Rockenbach et al. 2008); however, the current results reveal that leaves and calyces of the species of Physalis can accumulate these phenolics at similar or even higher levels than fruits (Table 2).

Phenolic profiles
Quercetin derivatives, kaempferol derivatives, and phenolic acids were the major phenolic compounds  synthesized by the five species analyzed of Physalis, and were accumulated with a species-specific tendency. The leaves of P. angulata accumulated the highest number of phenolics (12 compounds) and those of P. solanacea the lowest (3 compounds) (Table 3). Rutin (compound 22) was found in all species (as one of the dominant compounds), except in P. hederifolia var. hederifolia, whose leaves were rich in  kaempferol-3-O-glycosides (9 among 11 compounds were kaempferol derivatives). Rutin has also been reported in aerial parts of Physalis orizabae (Maldonado et al. (2012). Myricetin-3-O-neohesperidoside (λ max : 266, 299sh, 356) was reported in methanol extracts of leaves of P. angulata by Ismail and Alam (2001); however, the current results revealed no myricetin derivatives for any of the analyzed species. In fruits, phenolic acids were the dominant compounds (Table 3), except for P. hederifolia var. hederifolia.
A similar dominance has been reported for the fruits of P. peruviana (Rockenbach et al. 2008). Relevant interspecific differences were observed in the phenolic composition, for example, for the fruits of P. hederifolia var. hederifolia the only phenolic compound found was one kaempferol-3-O-glycoside (F5) (Fig. 2), this contrasts with the composition found for the other four species and with the results reported for P. peruviana (Rockenbach et al. 2008). Rutin and myricetin have been found in the fruits of P. peruviana, rutin being present as a  Table 3 Fig. 3 HPLC chromatograms and UV spectra of major calyx phenolics of five species of Physalis. The HPLC chromatograms were registered at 260 nm and the UV spectra of the resolved compounds were obtained from 200 to 400 nm. The number of compounds corresponds to those of Table 3 predominant compound (Licodiedoff et al. 2013). Rutin (compound F7) was also present in the fruits of P. patula, P. solanacea, and P. subulata, and in the two first species it was one of the dominant compounds (Fig. 2). However, none of the spectral data found in the present study corresponded to those of myricetin reported by Campos and Markham (2007) (λ max : 252, 263sh, 298sh, 371) or by Mabry et al. (1970) (λ max : 254, 272sh, 301sh, 374). The fruits of P. subulata contained dihydroflavonoids (F8 and F9), which are less common phenolics in plant kingdom than flavonoids. The concentrations of F8 and F9 were similar to those of most phenolic acids (Table 3). Dihydroflavonoids can be important inhibitors of tumor development, antioxidant, and analgesic (Wong and Rabie 2009). The fruits of P. patula accumulated the highest number of phenolic compounds (seven).
In calyces of P. hederifolia var. hederifolia, as in its leaves and fruits, the dominant compounds were kaempferol-3-O-glycosides, and this was the only species in whose calyces, rutin (compound C11) was missing. Rutin has also been reported in the calyx of P. solanacea by Pérez-Castorena et al. (2013). The number of compounds accumulated in the calyces varied from 4 in P. solanacea and P. patula to 7 in P. subulata.
The concentrations of the individual phenolics were variable between species and between tissues. The highest level was found for compound C1, a phenolic acid accumulated in the calyces of P. subulata (Table 3). Comparatively, the individual phenolic acid concentrations of fruits and calyces were higher than the concentrations reported by Rockenbach et al. (2008) for the fruits of Physalis peruviana.

Cluster analysis
The results of the cluster analysis suggests that the phenolic profiles obtained by HPLC-DAD (Fig. 4) were species-specific, independent from location and time of collection, as indicated for the two samples of P. solanacea (Jaccard similarity value of 1), in such a way that each species can be typified by its phenolic profile. These phenol profiles can be important specific chemotaxonomic markers for Physalis and may contribute to solve taxonomic controversies around the delimitation of species mentioned by several authors (Whitson and Manos 2005). The species-specific feature of the profiles can be used also as a tool to typify, trace, and define the authenticity of fruits and preparations of these species. The chemical diversity revealed for the five wild analyzed species of Physalis can be a starting point to improve the quality of fruit of the cultivated species of the genus, and could represent relevant allelic forms. Our results corroborate previous reports about the specificity of phenolic profiles (Almaraz-Abarca et al. 2013).

Antioxidant properties and correlation analysis
All samples displayed important antioxidant properties. As scavengers of the free radical DPPH highlighted the fruits of P. hederifolia var. hederifolia (Table 2). All extracts showed higher scavenging activity than that reported by Chang et al. (2008) for the fruits of P. peruviana (EC 50 value around 300 µg/mL). All extracts showed higher power to reduce Mo (VI) to Mo (V) than quercetin and epicatechin (Table 2), both considered as powerful antioxidants (Tapas et al. 2008). Leaves of P. subulata displayed until 44-fold higher values of TAC than any of two standards. All the foliar extracts showed higher iron reduction potential than the standards quercetin and Fig. 4 Results of the cluster analysis based on the phenolic profiles of five species of Physalis. The dendrogram was generated with the Paired Group Algorithm and the Jaccard's Similarity Measure, for comparing foliar, fruit and calyx phenolic profiles for five wild species of Physalis. The reference number of samples corresponds to those of Table 1 epicatechin (Table 2). However, the highest value, which was reached by the extract of calyces of P. solanacea (426) was lower than the value reported by Diaz et al. (2012) for the fruits of P. alkekengi (IC 50 = 36.58 mg/g dry weight).
The variations found among the two samples of P. solanacea suggest that as the phenolic profiles were the same for both samples (Figs. 1, 2, 3), plant age caused variations in the concentrations of phenolics (Tables 2, 3), all which affected the antioxidant properties of tissues. These kind of effects have been reported for other plant species (Del Baño et al. 2003). In fruits, the accumulation of other antioxidants, like ascorbic acid, could affect the antioxidant properties. Synthesis of ascorbic acid has been reported in the fruits of P. peruviana (Rop et al. 2012) and may occur also in the species of Physalis here analyzed.
The low association between the antioxidant properties and the phenolic, flavonoid, and phenolic acid contents found in the present study agree with the reports of other authors (Morais et al. 2011;Barriada-Bernal et al. 2014), who suggested that the antioxidant potential of a given total extract depends not only on the amount of phenolic compounds present in it but also on the types of phenolics accumulated, their proportions, as well as the presence of compounds of different chemical nature.

Principal component analysis
The results of the PCA revealed a tendency of grouping based on the developmental stage (Fig. 5), group A included samples collected in August (P. hederifolia var. hederifolia and P. angulata), group B was formed by samples collected in September (P. patula and P. solanacea 412), and group C was formed by samples collected in October (P. solanacea 426 and P. subulata). These results reveal the effect of time of collection and the associated developmental stage, on the phenolic, flavonoid, and phenolic acids contents and the antioxidant properties of different tissues of five wild species of Physalis. Systematic studies concerning the seasonal variation of phenolics of the species of Physalis are needed, as it has been done for some species of Betula (Raal et al. 2015), and concerning the developmental variation, as it has been carried out for Rosmarinus officinalis (Del Baño et al. 2003). Those studies would contribute to identify the best collecting time and developmental stage to obtain the highest concentration of some phenolic compounds and the environmental conditions in which they are produced.

Conclusions
The wild analyzed species of Physalis synthesize important amounts of diverse phenolic compounds. The phenolic profiles of leaves, fruits, and calyces were species-specific; this reveals phenolic patterns as significant specific chemomarkers in the genus. The dominant phenolic compounds accumulated in leaves and calyces were 3-O-glycoside derivatives of kaempferol, and the dominant ones in fruits were phenolic acids; all these compounds are recognized for their important activities as antioxidants. Because of their phenolic compositions, P. angulata, P. patula, P. hederifolia var. hederifolia, P. solanacea, and P. subulata may be considered as relevant potential sources of natural antioxidants to be used in the food and pharmaceutical industries. The present phenolic fingerprinting contributes to generate a better knowledge for using all the potential of these wild species of plants. The chemical variability of wild species of Physalis offers the opportunity for domesticating and developing new cultivars with enhanced accumulation of phenolics having positive effects on human health. The results inform about the importance of consuming these plant species and support the use of calyces and leaves in traditional medicine. The not always high correlation between phenolic or flavonoid contents and antioxidant activities suggests the relevance of the phenolic profiles, besides the contents of phenolics, in determining these activities.