Characterization of Polyphenolic Content in the Aquatic Plants Ruppia cirrhosa and Ruppia maritima —A Source of Nutritional Natural Products

Herein, the polyphenolic content in extracts of Ruppia cirrhosa (Petagna) Grande and Ruppia maritima L.was fully characterized for the first time. High amounts of the main compound chicoric acid (CA) (≤30.2 ± 4.3 mg/g) were found in both Ruppia species. In addition, eight flavonoids, namely the 3-O-glucopyranosides and 3-O-galactopyranosides, as well as malonylated 3-O-glycosides of quercetin and isorhamnetin, were isolated and identified. The antioxidant activity of Ruppia cirrhosa extracts and isolated compounds was investigated spectrophotometrically by a 1,1-diphenyl-2-picrylhydrazyl (DPPH·) radical scavenging assay. IC50 values were 31.8–175.7 μg/mL for Ruppia cirrhosa extracts and 12.1–88.4 μg/mL for isolated flavonoids. Both individual and total phenolic and flavonoid content were quantified in crude extracts using analytical HPLC. The relative high amount of total flavonoids ranged from 5.9 to 14.7 mg/g in both species, with concentrations of individual flavonoids ranging from 0.4 to 2.9 mg/g dry weight. The content of chicoric acid was twofold more in Ruppia maritima than in Ruppia cirrhosa. Seasonal variation of the quantitative content in Ruppia cirrhosa was examined. Total flavonoid content ranged from 8.4 mg/g in October to 14.7 mg/g in August, whereas the highest concentration of chicoric acid was observed in March (29.2 mg/g).


Introduction
The marine environment is a potential source for a wide variety of nutritional natural products. Seaweeds are used as human food or as raw materials for the production of compounds of nutritional interest [1]. On the other hand, marine angiosperms, such as seagrasses, are known for their content of secondary metabolites [2,3]; however, these are very little exploited to find commercially valuable natural products. A few seagrass species, especially of the genus Zostera, Halophila, Posidonia, Thalassia and Syringodium, have been investigated for their content of phenolics and flavonoids [3][4][5][6][7][8][9][10][11][12][13].
The widgeon grass family (Ruppiaceae) is a submersed aquatic angiosperm widely distributed in temperate and tropical regions all over the world. Ruppia species usually occur in brackish or saline waters, but can also be found in diluted fresh water or fresh water with high salinity, and only rarely under marine conditions [14][15][16]. In Norwegian coastal waters, two Ruppia species have been found, namely Ruppia maritima L. and Ruppia cirrhosa (Petagna) Grande, the latter occasionally synonymized under R. spiralis L. ex Dumort. Both species can be found in single populations with no other vascular plants present, and they are hardly ever found together. R. maritima can sometimes be found in proximity of Zostera noltii populations, while R. cirrhosa can be found with or close to Zostera marina L. populations.
The number of studies investigating secondary metabolites in Ruppia species are limited, and a full analysis of polyphenolic content is lacking [7,10,17]. In 1973 Boutard et al. [7] analyzed and identified two flavonoids in R. maritima based on chrysoeriol and possibly luteolin. Harborne and Williams reported in 1976 an unidentified glycosylflavone, as well as three caffeoyl conjugates in R. maritima, whereas no phenolic derivatives were found in R. cirrhosa [10]. Haynes and Roberts indicated later the presence of flavonols in one Ruppia species [17], yet these results remain unpublished, and no accurate identification of the flavonols has been concluded. The previous identification work is based on TLC retention times and electrophoretic surveys [7,10].
The aim of this work was to characterize the phenolic content of R. cirrhosa and R. maritima collected from Norwegian coastal waters with the aims of finding a new source of nutritional natural products. To our knowledge, this is the first report on complete structural characterization of both flavonoids and one phenolic acid in these two species and our quantitative studies revealed high amounts of the potent chicoric acid (CA) [18].
Molecules 2018, 23, 16 2 of 14 proximity of Zostera noltii populations, while R. cirrhosa can be found with or close to Zostera marina L. populations. The number of studies investigating secondary metabolites in Ruppia species are limited, and a full analysis of polyphenolic content is lacking [7,10,17]. In 1973 Boutard et al. [7] analyzed and identified two flavonoids in R. maritima based on chrysoeriol and possibly luteolin. Harborne and Williams reported in 1976 an unidentified glycosylflavone, as well as three caffeoyl conjugates in R. maritima, whereas no phenolic derivatives were found in R. cirrhosa [10]. Haynes and Roberts indicated later the presence of flavonols in one Ruppia species [17], yet these results remain unpublished, and no accurate identification of the flavonols has been concluded. The previous identification work is based on TLC retention times and electrophoretic surveys [7,10].
The aim of this work was to characterize the phenolic content of R. cirrhosa and R. maritima collected from Norwegian coastal waters with the aims of finding a new source of nutritional natural products. To our knowledge, this is the first report on complete structural characterization of both flavonoids and one phenolic acid in these two species and our quantitative studies revealed high amounts of the potent chicoric acid (CA) [18].
Molecules 2018, 23, 16 3 of 14 glucopyranoside and isorhamnetin 3-O-β-D-glucopyranoside have previously been identified in the seagrass C. nodosa [29]. As far as we know, this is the first report of 3-O-galactopyranosides and malonylated glycosides of quercetin and isorhamnetin in aquatic plants.

DPPH Radical Scavenging of Ruppia Polyphenols
DPPH is a stable free radical with a maximum absorbance at 517 nm (deep purple colour). When reacting with a radical scavenger it donates a hydrogen and acquires a colorless reduced form. The loss of purple colour correlates with scavenging activity of the compound, and IC50 values are commonly used to determine the compounds ability to scavenge radicals. The IC50 values of R. cirrhosa extracts and isolated compounds are shown in Table 1. Due to insufficient amounts of sample material, DPPH· scavenging activity of R. maritima was not tested. The R. cirrhosa extract exhibited an IC50 value of 152.9-175.7 μg/mL, which is considered low to moderate radical scavenging activity [34]. These results are comparable to antioxidant activities of crude extracts of the seagrasses Halodule ovalis (IC50 130 μg/mL) [35], Syringodium isoetifolium (IC50 96.34 μg/mL), Enhalus acoroides (IC50 115.79 μg/mL), Cymodocea rotundata (IC50 123.72 μg/mL) and Thalassia hemprichii (IC50 214.68 μg/mL) [36]. However, after partition with ethyl acetate, the aqueous phase of R. cirrhosa exhibited very strong radical scavenging activity, with an IC50 value of 31.8 ± 3.2 μg/mL. To our knowledge, this is the first reported results on DPPH· scavenging activity of R. cirrhosa extracts.

DPPH Radical Scavenging of Ruppia Polyphenols
DPPH is a stable free radical with a maximum absorbance at 517 nm (deep purple colour). When reacting with a radical scavenger it donates a hydrogen and acquires a colorless reduced form. The loss of purple colour correlates with scavenging activity of the compound, and IC 50 values are commonly used to determine the compounds ability to scavenge radicals. The IC 50 values of R. cirrhosa extracts and isolated compounds are shown in Table 1. Due to insufficient amounts of sample material, DPPH· scavenging activity of R. maritima was not tested. The R. cirrhosa extract exhibited an IC 50 value of 152.9-175.7 µg/mL, which is considered low to moderate radical scavenging activity [34]. These results are comparable to antioxidant activities of crude extracts of the seagrasses Halodule ovalis (IC 50 130 µg/mL) [35], Syringodium isoetifolium (IC 50 96.34 µg/mL), Enhalus acoroides (IC 50 115.79 µg/mL), Cymodocea rotundata (IC 50 123.72 µg/mL) and Thalassia hemprichii (IC 50 214.68 µg/mL) [36]. However, after partition with ethyl acetate, the aqueous phase of R. cirrhosa exhibited very strong radical scavenging activity, with an IC 50 value of 31.8 ± 3.2 µg/mL. To our knowledge, this is the first reported results on DPPH· scavenging activity of R. cirrhosa extracts. The extract from the plant material collected in October had a slightly lower scavenging activity than the R. cirrhosa extract from August. This may be related to the lower phenolic content found (Table 4). In addition, the percent scavenging of four crude extracts of R. cirrhosa with known concentrations of both flavonoids and chicoric acid was examined (Figure 3), revealing a correlation between antioxidant scavenging and concentration of total flavonoids and chicoric acid.  The extract from the plant material collected in October had a slightly lower scavenging activity than the R. cirrhosa extract from August. This may be related to the lower phenolic content found (Table 4). In addition, the percent scavenging of four crude extracts of R. cirrhosa with known concentrations of both flavonoids and chicoric acid was examined (Figure 3), revealing a correlation between antioxidant scavenging and concentration of total flavonoids and chicoric acid.  (4) showed very strong antioxidant activity, with an IC50 value of 12.1 ± 3.3 μg/mL. The measured value is similar to the IC50 values obtained for the reference standards quercetin (5.5 ± 0.3 μg/mL), quercetin 3-O-β-D-glucopyranoside (11.0 ± 1.0 μg/mL) and rutin (13.9 ± 0.7 μg/mL), once molar mass is accounted for. Flavonoids with an isorhamnetin aglycone (compounds 5-8) showed lower antioxidant activity than the quercetin-based flavonoids (3 and 4), explained by the number of free hydroxyl groups on the aglycone B-ring [37]. Interestingly, the malonylated isorhamnetin O-glycosides 7 and 8 showed much higher antioxidant activity than the corresponding isorhamnetin O-glycosides 5 and 6, with IC50 values of 51.7 ± 6.8 μg/mL and 88.4 ± 7.0 μg/mL, respectively.
DPPH· scavenging with chicoric acid (CA), isolated from R. cirrhosa, resulted in a higher IC 50 value (23.0 ± 3.2 µg/mL) than the one seen for the mixture of quercetin (4). Compared to the isolated isorhamnetin-based flavonoids (5 & 6 and 7 & 8) however, CA showed stronger scavenging and lower IC 50 value. The chicoric acid (CA) isolated in this study had a higher IC 50 value (23.0 ± 3.2 µg/mL) ( Table 1) than the one measured for the reference compound (9.7 ± 1.7 µg/mL) ( Table 2). Since DPPH is a highly concentration sensitive method, variations in IC 50 values for the same compound is often seen [38][39][40][41][42][43][44][45]. No significant impurities were observed for the isolated sample of CA in the present study using HPLC and NMR for purity determination. However, water content, especially if the compound is hygroscopic, and inorganic salt content will normally not be determined by these methods [46]. Nonetheless, both the isolated CA and reference compound showed very strong antioxidant activity.

Quantitative Analysis of Polyphenolic Content in Ruppia
The quantitative content of individual flavonoids 1-8 and chicoric acid was characterized in three R. cirrhosa and two R. maritima populations, collected from different localities at the east and west coast of Norway (A-E) ( Table 3). As seen in Figure 4a, the flavonoid content was significantly higher in R. cirrhosa from the Bergen location (A) compared to the other R. cirrhosa populations from the west coast of Norway (B and C).  No significant differences in the total flavonoid or phenolic content of the two R. maritima populations from the east coast were observed (D and E). However, significant differences in the distribution of the individual flavonoids were seen. The R. maritima samples from Tønsberg (D) showed a higher content of the quercetin O-glycosides 1 and 2, whereas R. maritima samples from the Råde (E) location contained higher amounts of the malonylated isorhamnetin O-glycosides (7) and (8).
The concentrations of chicoric acid (CA) were significantly higher in R. maritima (30.2 and 27.9 mg/g) than in R. cirrhosa (11.1-12.7 mg/g). It seems natural to conclude that R. maritima generally have a higher production of CA although, although it should be taken into consideration that the R. maritima samples were collected from a different part of Norway. Differences in chicoric acid accumulation may be a function of nutritional and/or environmental stress, but there is a need for more research on how chicoric acid accumulation in plants is regulated [18]. In leaves of Cymodocea nodosa and Syringodium filiforme, the amounts of chicoric acid have been reported to range from 8.13-27.4 mg/g and 0.94-5.26 mg/g, respectively [12,29]. Chicoric acid has also been found in Posidionia oceania from the Mediterranean Sea, however, the quantitative content varied greatly. The maximum content of chicoric acid was 0.1386 mg/g in young leaves of P. oceanica collected in the Aegean sea outside Turkey, whereas both detrital and fresh leaves of P. oceanica from four different localities in the western part of the Mediterranean sea were found to contain up to 12.78 mg/g chicoric acid [31,32]. The high level of CA (≤30.2 ± 4.3 mg) found in this study is comparable to the content of CA in the known source Echinacea purpura [52][53][54], proposing Ruppia to be a new and valuable source of chicoric acid (CA). Chicoric acid is high value-added on the nutraceutical market, due to its possible health benefits and its relative rare occurrence in the plant kingdom [12,18].
Fluctuations in natural product concentrations should be taken into consideration before scheduling harvest dates or planning herbal product manufacturing [18]. In order to get an impression of the seasonal fluctuations of phenolics in Ruppia, the total flavonoid and CA content in R. cirrhosa collected from the Bergen location (A) in October, March and August were analyzed (Table 4, Figure 4b).
During the winter season (December-February) the biomass on the examined locality was scarce. The concentration of flavonoids in R. cirrhosa was significantly higher in August (14.7 ± 1.9 mg) compared to October (8.4 ± 1.1 mg) and March (11.1± 2.4 mg). The concentration of CA in R. cirrhosa measured in March (29.2 ± 6.3 mg) was over twice the amounts found in August (12.7 ± 2.5) and October (10.6 ± 2.5). The observed seasonal variation of flavonoids and phenolic acid indicates a similar pattern as we have previously seen in Zostera marina [51], with higher concentrations in spring and summer. These trends are associated with environmental stress factors, mainly UV radiation-as seen for terrestrial plants [55,56]. It is also likely that because the young leaves are still growing, they are consequently more vulnerable for microbial/fungal and herbivory attacks, which will result in an increased production of phenolics [57]. Yet, to achieve more accurate and reliable data on the seasonal variation in relation to environmental factors, a more comprehensive study of the content of both flavonoids and chicoric acid in R. maritima and R. cirrhosa is needed.

NMR-Spectroscopy
One-dimensional 1 H and 13   Flowers of both species are hermaphroditic and small, in two-flowered, pedunculate spikes. Perianth is absent. Peduncles in R. cirrhosa are 8-15 cm long, sometimes longer, and spirally coiled when fruits are mature. Peduncles in R. maritima are shorter; 4-6 cm long, often somewhat recurved in fruit but never spirally coiled. R. cirrhosa is typically 30-50 cm long, whereas R. maritima often is 10-15 cm long, sometimes up to 30 cm long. R. maritima is found mostly in the hydrolittoral zone, sometimes down to the upper part of the sublittoral zone, growing at ±0.5 m deep, whereas R. cirrhosa occurs in the sublittoral zone and is permanently submerged at depths of 0.5-1.5 m. Both species are found on soft substrata, such as mud and silt. R. maritima is also found on fine sand.

Extraction, Purification and Identification
The collected plant material was washed thoroughly in fresh water and air-dried. The root was separated from the rest of the plant, and the material was cut in small pieces and stored at −20 • C, when not used. Air-dried leaves of R. cirrhosa were extracted with 50% aqueous methanol (HPLC) for 24 h at room temperature. The extraction was repeated 4 times. The combined extracts were filtered through glass wool, and the volume was further reduced using a rotavapor. The concentrated aqueous extract was partitioned against ethyl acetate three times. The content of both the ethyl acetate and water phase was examined on HPLC. About a third of the aqueous extract was applied to an Amberlite XAD-7 column (5 × 20 cm), and eluted with distilled water until no colour was observed, then methanol was applied. Collected fractions were analyzed on analytical HPLC and concentrated using a rotavapor. The semi-purified plant extract was submitted to preparative HPLC to obtain purified compounds. The physiochemical and spectral data of the flavonoids and chicoric acid were as follows:  C-6"). The structure was confirmed by comparison with literature data [19][20][21]. .6 (C-5 ), 60.9 (C-6 ). The structure was confirmed by comparison with literature data [19,21,22,58].  C-6 ). The structure was confirmed by comparison with literature data [24,25].  .5 (C-9 , C-9 ). The structure was confirmed by comparison with literature data [28].

Quantitative Determination
Leaves of R. cirrhosa and R. maritima were cut into small pieces and extracted with 50% aqueous methanol, the flavonoid content of the extract was characterized by analytical HPLC with DAD and HR-LCMS. Quantitative determination: 10-40 mg of dried plant material was weighed and extracted with 3-5 mL of 50% aqueous methanol for 2 hours at room temperature. Four replicate samples were made. Prior to injection, the solutions were filtered through a 0.45 µm Millipore membrane filter. HPLC calibration curves of quercetin 3-O-β-D-glucopyranoside (≥90% (HPLC), Sigma-Aldrich, Sigma-Aldric, St. Louis, MO, USA) and caffeic acid (≥98% (HPLC), Sigma-Aldrich) were used to determine the quantitative amounts of flavonoids and phenolic compounds, respectively. The results are presented as milligrams quercetin 3-O-β-D-glucopyranoside or caffeic acid equivalents ± one standard deviation (SD) per gram of dry weight (DW) plant material. Two sample t-test assuming unequal variances with a p-value of 0.05 was used to determine if the means of two different measurements were equal or not. Standard error bars were calculated using the STDEV. P function in excel, and represent one standard deviation (n = 4 or number of replicates).

Method Validation
The established HPLC method was validated for linearity, sensitivity, precision and accuracy, as previously described [51]. LOD and LOQ were calculated based on standard deviation of y-intercepts of the regression line (SD) and the slope (S), using the equations LOD = 3.3 × SD/S and LOQ = 10 × SD/S. Recovery study was performed in triplicate by adding known amounts of quercetin 3-O-β-D-glucopyranoside to crude extracts of R. cirrhosa. Data for calibration curves, test ranges, limit of detection (LOD) and limit of quantification (LOQ) for quercetin 3-O-β-D-glucopyranoside (90%, Sigma-Aldrich Sigma) and caffeic acid (Sigma-Aldrich) are presented in Table 5. The recovery was ranging from 93.3% to 94.8% for quercetin 3-O-β-D-glucopyranoside with a mean of 94.0 ± 2.0% (Table 5).

DPPH Radical Scavenging
The stable 1,1-diphenyl-2-picryl hydrazyl radical (DPPH·) was used for determination of free radical-scavenging activity of R. cirrhosa extracts and isolated mixtures of flavonoids (purity ≥ 75% (HPLC)). Different sample concentrations of the extracts were prepared, and 0.05 mL of each sample was added to a 2.95 mL methanolic solution of DPPH· (45 µg/mL). A UV-1800 UV spectrophotometer (Shimadzu Scientific Instruments, Columbia, MD, USA) was used for the antioxidant assays. The UV/Vis absorbance at 517 nm was measured every 30 s for 5 min. The experiment was repeated three times, and the results are presented as mean ± standard deviation (n = 3). Trolox (97%, Sigma-Aldrich), chicoric acid (≥95% (HPLC), Sigma-Aldrich), quercetin (≥95% (HPLC), Sigma-Aldrich), quercetin 3-O-β-D-glucopyranoside (≥90% (HPLC), Sigma-Aldrich) and rutin (≥95% (HPLC), Sigma-Aldrich) were used as standard controls. Percent radical-scavenging was calculated as 100 × (A start − A end )/(A start ), where A start is the absorbance before addition of the sample, and A end is the absorbance value after 5 min of reaction time. Percent scavenging IC 50 values were calculated from a linear regression plot of percent scavenging (%) against logarithmic concentration of the test compound [59]. IC 50 values denote the concentration of sample which is required to scavenge 50% of DPPH· free radicals.

Conclusions
In this study, the polyphenolic content of Ruppia cirrhosa and Ruppia marittima was characterized for the first time using NMR-spectroscopy, HRLC-MS and HPLC-UV. Both Ruppia species contained high amounts of chicoric acid (10.6-30.2 mg/g DW), followed by relatively high amounts of flavonoid glycosides (5.9-14.7 mg/g DW). The eight flavonoids identified were based on quercetin and isorhamnetin with 3-O-galactopyranosides or 3-O-glucopyranosides, four of these were malonylated. This is the first report of 3-O-galactopyranosides and malonylated flavonoids of quercetin and isorhamnetin isolated from aquatic plants. The seasonal variations of flavonoids and phenolics were examined by analyzing R. cirrhosa samples in October, March and August. Highest flavonoid content was found in August, whereas the highest concentration of chicoric acid was observed in March.
Extracts of R. cirrhosa showed low to moderate DPPH· antioxidant activity, however, partially purified extract and isolated compounds showed strong to very strong antioxidant activities, with IC 50 values ranging from 12.1 to 88.4 µg/mL.