Studies on Phytochemistry and Antioxidant Capacity of Nine Hibiscus sabdariffa Accessions
Click on author name to view affiliation information
Respective chromatograms of a H. sabdariffa calyx extract. (A) Chromatogram at 520 nm. (B) Chromatogram at 325 nm. (C) Chromatogram at 372 nm. Identities of labeled peaks are as follows: delphinidin 3-sambubioside (1), delphinidin 3-glucoside (2), cyanidin 3-sambubioside (3), cyanidin 3-glucoside (4), chlorogenic acid (5), and quercetin (6).
Illustrations of calyx and leaf of nine H. sabdariffa accessions grouped according to their calyx color as red (A), light green and red (B), and light green (C).
Principal component analysis showing score (A) and loading plots (B) for principal component (PC) 1 × PC2. The score plot shows the accession number of three calyx color groups of nine H. sabdariffa accessions: red calyx is indicated with circles, light green and red is indicated with triangles, and light green calyx is indicated with rectangles. Variables in the loading plot represent the phytochemical concentration of cyanidin 3-glucoside (Cy3G), cyanidin 3-sambubioside (Cy3S), delphinidin 3-glucoside (D3G), delphinidin 3-sambubioside (D3S), total anthocyanin (TA), chlorogenic acid (CGA), quercetin (QUE), total phenols (Phenols), total flavonoids (Flavonoids), antioxidant capacity [2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)], and mineral content (N, P, K, Ca, Mg, S, Fe, Zn, Mn, and Cu) of the calyx.
Click on author name to view affiliation information
Hibiscus sabdariffa is considered a potential horticultural and pharmaceutical crop, yet there is limited information on beneficial traits of cultivars and germlines. We examined nine H. sabdariffa accessions for their horticultural traits, phytochemistry, and antioxidant capacity. Calyx, the main part of horticultural interest, was examined for its color, fresh weight, length, and width. Both calyx and leaf tissues were studied for phytochemical and antioxidant properties. Anthocyanins, chlorogenic acid, and quercetin were analyzed using high-performance liquid chromatography. Total phenolic content and total flavonoid content were determined by the Folin–Ciocalteu assay and the aluminum chloride assay, respectively. Antioxidant activity was examined by free radical scavenging assays, using 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radicals. The study showed that three accessions (PI 265319, PI 275414, and PI 500706) had red calyces (“red”), and four accessions (PI 256039, PI 275413, PI 291128, and PI 500713) had light green calyces tinged with red (“light green–red”). Two accessions with light green calyces (“light green”) were PI 273389 and PI 500724. The average weight of calyces was the greatest with PI 291128 (13.2 g) followed by PI 275413 (11.4 g) and PI 275414 (10.9 g). Calyx length ranged from 18.7 to 60.4 mm, and the width ranged from 17.8 to 34.8 mm. Among four anthocyanins in calyx, delphinidin 3-sambubioside and cyanidin 3-sambubioside were major anthocyanins with less amounts of delphinidin 3-glucoside and cyanidin 3-glucoside. The red accession PI 275414 had the highest concentration of the total anthocyanins (32.7 mg·g−1 dry weight). Chlorogenic acid concentrations were higher in calyces (13.1 to 25.4 mg·g−1) than in leaves (0.7 to 2.0 mg·g−1), while quercetin contents were greater in leaves (3.5 to 13.2 mg/100 g) than in calyces (1.3 to 3.5 mg/100 g). Similarly, total phenolic content was higher in leaves [18.6 to 22.5 mg gallic acid equivalent/g] than in calyces (10.4 to 17.3 mg gallic acid equivalent/g). For antioxidant capacity, DPPH was higher in leaves [16.6 to 22.0 mg vitamin C equivalent antioxidant capacity (VCEAC)/g] than calyces (6.6 to 7.3 mg VCEAC/g). In contrast, ABTS antioxidant capacity was greater in calyces (8.6 to 15.7 mg VCEAC/g) than leaves (3.6 to 3.9 mg VCEAC/g). Calyx had higher contents of K and Zn, while leaf tissues had more Ca, N, Mg, S, B, Mn, and Cu. This study indicated that there is a great variation in phytochemical contents and antioxidant properties between calyces and leaves and among H. sabdariffa accessions. Because red calyx is desirable for producing herbal tea or coloring, the most interesting accession in this study was PI 275414, with the largest red fresh calyx.
Hibiscus sabdariffa L., commonly known as “roselle” or “sorrel,” is an erect annual herb of the Malvaceae family (Ross 2003). It is widely cultivated in many countries, including India, Malaysia, Africa, and Central America for its important horticultural values (Morton 1987). The red calyces are economically important to produce herbal teas, jellies, and other desserts (Duke 1983). It is also a natural red food colorant due to the presence of anthocyanins (Du and Francis 1973; Khoo et al. 2017). The young leaves and tender stems can be eaten as leafy green vegetables (Zhen et al. 2016). The seeds can be ground into a meal or roasted as a coffee substitute (Morton 1987). In addition, bioactive compounds in H. sabdariffa function as strong antioxidants and have considerable nutraceutical value (Juliani et al. 2009; Kasote et al. 2015). Several review articles on the chemical constituents and pharmacological effects of H. sabdariffa have reported the presence of bioactive compounds such as vitamin C, anthocyanins, flavonoids, and organic acids that are responsible for their therapeutic effects (Ali et al. 2005; Da-Costa-Rocha et al. 2014; Owoade et al. 2019). In H. sabdariffa calyces, four major anthocyanins have been identified: cyanidin 3-sambubioside, cyanidin 3-glucoside, delphinidin 3-sambubioside, and delphinidin 3-glucoside (Borrás-Linares et al. 2015; Da-Costa-Rocha et al. 2014; Rodríguez‐Medina et al. 2009). Hibiscus extracts also contain other flavonoids, such as quercetin and its glycosides (Desmiaty and Alatas 2008), as well as phenolic acids, such as ascorbic acid and protocatechuic acid (Da-Costa-Rocha et al. 2014). Regarding their medicinal properties, infusions prepared from the calyces and leaves were used in traditional medicine for cancer prevention, diuretic effects, and hypotensive properties (Da-Costa-Rocha et al. 2014; Morton 1987). Leaf extracts exhibited anti-cancer properties both in vitro and in vivo due to their ability to induce apoptosis of human prostate cancer cells (Lin et al. 2012). Clinical studies have also shown that consumption of H. sabdariffa calyx extracts can aid in the treatment of hypertension (Herrera-Arellano et al. 2007; Mozaffari-Khosravi et al. 2009).
The growing interest in functional foods has become the basis for the selection of varieties with high phenolic contents and antioxidant capabilities. Although the phytochemical and pharmacological properties of H. sabdariffa have been reported in several review articles (Ali et al. 2005; Barhé and Tchouya 2016; Da-Costa-Rocha et al. 2014; Owoade et al. 2019), there are few that specify the crop site and compare different accessions with respect to their plant parts. While H. sabdariffa grows well in tropical and subtropical regions (Morton 1987), the mineral composition and chemical and nutraceutical qualities of the calyx are influenced by cultivars (Hinojosa-Gómez et al. 2018). However, there is no record of roselle production in Guam, a tropical island with calcareous soil. A preliminary study demonstrated that H. sabdariffa can be grown in Guam producing calyces (Clemente and Marutani 2019). The current study was conducted to examine the horticultural characteristics of roselle, especially size and color of calyx, phytochemical properties, and antioxidant capacity of calyx and leaf tissues, as potential horticultural and pharmaceutical crops in Guam.
Seeds of nine accessions of H. sabdariffa (PI 256039, PI 265319, PI 273389, PI 275413, PI 275414, PI 291128, PI 500706, PI 500713, and PI 500724) were obtained from the Plant Genetic Resources Conservation Unit of the US Department of Agriculture Agricultural Research Service in Griffin, GA, USA. The seeds were sown in 72-cell Styrofoam trays with a commercial potting medium (Sunshine Mix #4; Sun Gro Horticulture, Bellevue, WA, USA), and 3-week-old seedlings were transplanted into the field at the Guam Agricultural Experiment Station Yigo farm (lat. 13°31′46″N, long. 144°52′20″E, elevation 133 m) on 20 Sep 2021. Soil was classified as Guam cobbly clay loam (clayey, gibbsitic, nonacid, isohyperthemic, Lithic Ustorthents). The field consisted of two 15-m-long rows with 3 m between rows. Five plants for each accession were planted 0.6 m apart within each row as a group.
Flowering and fruit formation occurred from Oct 2021 to Jan 2022. Fifty samples of mature fruit (calyx and capsule) were collected from each accession from 10 and 27 Dec 2021, and the fresh weight of fruit with capsule was recorded. Fruit length was measured from the base to the tip of the calyx, and the width was determined by measuring the widest portion of the calyx. Qualitative traits, such as calyx color, leaf shape, and leaf color, were also recorded.
For phytochemical, mineral, and antioxidant analysis, a total of 100 calyces were collected from each accession when the calyces were mature, and samples were prepared by removing the seed capsules and the epicalyx, leaving only the fleshy calyx parts. Leaf samples of all accessions were collected on 17 Dec 2021, and 100 g of leaf tissues were used for the analyses by removing petioles from each blade. Both calyx and leaf samples divided into two groups of 50 calyces and 50 g of leaf tissues were freeze-dried (Buchi Lyovapor L-200; Buchi Corporation, New Castle, DE, USA) at −58 °C (4.00 mbar), and ground into a fine powder. Ground samples were stored in an airtight container at −16 °C for further study.
The extraction method of Escobar-Ortiz et al. (2021) with modification was used. Powdered sample (1 g) was placed into a 15-mL centrifuge tube with 8 mL of acidified ethanol solvent, resulting in a solid-to-solvent ratio of 1:8 (w/v). The solvent was prepared by mixing 95% ethanol:water:acetic acid with the ratio of 80:18:2 (v/v/v) (2% acetic acid solution). The aqueous sample mixture was then stirred by an orbital shaker (Hotech Instrument Corp., Taiwan, China) at 50 rpm for 15 min. The sample mixture was filtered using a Whatman paper, grade 4, and filtered again through a 0.22-µm syringe filter. The filtrate was stored at −16 °C until ready for analysis. For each accession, four extracts were prepared for the calyx and leaf samples.
The high-performance liquid chromatography (HPLC) analyses were conducted using a modified method of Rodríguez-Medina et al. (2009). Analyses were performed on a model LC-2030 Plus Prominence i-Series HPLC (Shimadzu Corporation, Kyoto, Japan) equipped with a photodiode array detector. Reversed-phase HPLC experiments were carried out on a Shim-pack GIS C18 column (4.6 mm × 150 mm, 5-μm particle size) at 25 °C. Injection volume of standard and sample solution was 10 µL. A low-pressure gradient was used with mobile phase A: water/formic acid 90:10 (v/v) and mobile phase B: acetonitrile. The total runtime was 37 min, and the gradient elution program was as follows: 0 min, 0% (B); 13 min, 20% (B); 20 min, 30% (B); 25 to 28 min, 80% (B); and 30 to 35 min; 0% (B) with flow rate of 0.5 mL·min−1. Detection was set at 520 nm for the four anthocyanins [cyanidin 3-sambubioside (Cy3S), cyanidin 3-glucoside (Cy3G), delphinidin 3-sambubioside (D3S), and delphinidin 3-glucoside (D3G)], 325 nm for chlorogenic acid (CGA), and 372 nm for quercetin (Fig. 1). The analytical standards of the compounds (Sigma-Aldrich Inc., St. Louis, MO, USA) were separated using HPLC, and a five-point standard curve (5, 10, 25, 50, and 100 ppm) was recorded for each standard compound. All standard curves demonstrated good linearity with coefficient of determination, R2 > 0.99. Sample extracts were separated, and compound identification was determined based on comparison of the retention times of the standards. The total anthocyanin was calculated as the sum of the four anthocyanins: Cy3S, Cy3G, D3S, and D3G.
Citation: HortScience 60, 6; 10.21273/HORTSCI18483-25
Total phenolic content (TPC) was determined using the Folin–Ciocalteu assay, following the protocol of Singleton and Rossi (1965) with modifications. A sample extract of 50 μL was diluted with 250 μL of distilled water. For each calibration solution, sample, or blank, 300 μL of the solution was pipetted into a 15-mL centrifuge tube. An aliquot of 1950 μL of distilled water and 1125 μL of the Folin–Ciocalteu reagent were mixed into the solution, followed by an incubation period of 5 min at room temperature. Then, 5625 μL of 20% sodium carbonate was added and the solution was allowed to react for 45 min protected from light. Samples were measured at a wavelength of 765 nm using an ultraviolet-visible spectrophotometer (model S-2150; Cole-Parmer, Vernon Hills, IL, USA) against a blank solution. The gallic acid equivalent (GAE), expressed in milligrams/liter, was determined from the standard curve (calyces: y = 0.004x + 0.0325, R2 = 0.999; leaves y = 0.0042x + 0.0349, R2 = 0.997) from five concentrations of gallic acid standard (50, 100, 150, 250, and 500 ppm). The TPC was calculated as follows: [1] where C is the concentration of gallic acid equivalent (mg·L−1), V is the volume of extract (mL), W is the weight of the sample (g), and D is the dilution factor. The TPC results are expressed as milligrams of GAE per gram of dried sample.
The total flavonoid content (TFC) was determined by the aluminum chloride assay according to Nerdy et al. (2022). To create the test solution, the extracts were diluted by dissolving 250 μL of the crude extract to a volume of 1 mL with distilled water. For each 1 mL of standard solution, sample, or blank, the following solutions were added: 1.5 mL of methanol, 100 µL of 10% aluminum chloride solution (w/v), and 100 µL of 1 M sodium acetate. The solutions were then diluted to a volume of 10 mL using distilled water and mixed thoroughly. Four replications were prepared for each accession. The samples were allowed to stand for 15 min, and absorbance was measured at a wavelength of 432 nm using an ultraviolet-visible spectrophotometer (model S-2150; Cole-Parmer). TFC was determined from the standard curve (calyces: y = 0.0094x – 0.0147, R2 = 0.997; leaves: y = 0.0092x + 0.0515, R2 = 0.999) from five concentrations of quercetin standard (10, 20, 30, 40, and 50 ppm). The TFC was calculated as follows: [2] where C is the concentration of quercetin equivalent (mg·L−1), V is the volume of solvent used (mL), and W is the dry weight of the sample (g). The TFC is expressed as milligrams of quercetin equivalent (QE) per gram of dried sample.
Antioxidant capacities were evaluated according to two radical-scavenging assays: the 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay and the 2,2’-azino-di-[3-ethylbenzythiazoline-6-sulfonic acid] (ABTS) assay. In each assay, the standard curve was constructed using six concentrations of ascorbic acid (5, 10, 50, 100, 150, and 250 ppm). The radical scavenging activity was calculated as follows: [3]
The results were determined from the standard curve and were expressed as milligrams of vitamin C equivalent antioxidant capacity (VCEAC) per gram of dried sample as described by Kim et al. (2002).
The DPPH antioxidant capacity was determined according to the method described by Brand-Williams et al. (1995). A 0.1 mM DPPH solution was prepared by dissolving 39.4 mg of DPPH in 1 L of methanol. For each standard, blank solution, sample extract, and an aliquot of 150 µL of the solution was added into test tubes containing 3.85 mL of the DPPH solution. The blank solution was prepared by adding 150 µL of methanol. The resulting solutions were mixed and stored in the dark at room temperature for 1 h. The samples were measured at a wavelength of 517 nm using an ultraviolet-visible spectrophotometer (model S-2150; Cole-Parmer) against the blank solution, and the results were determined from the standard curve (calyces: y = 0.3716x + 0.8436, R2 = 0.984; leaves: y = 0.374x + 0.3812, R2 = 0.983).
The ABTS antioxidant capacity was determined according to the method by Re et al. (1999). A 7 mM ABTS solution was prepared by dissolving 38.4 mg of ABTS in a 10-mL volumetric flask with distilled water. A 2.45 mM potassium persulfate solution was prepared by dissolving 6.6 mg of potassium persulfate in a 10-mL volumetric flask with distilled water. To form the stable radical cation, equal parts of the ABTS solution and the potassium persulfate solution were mixed and stored in the dark at room temperature for 16 h. The ABTS solution was then diluted with ethanol to an absorbance of 0.70 ± 0.02 at a wavelength of 734 nm. The assay was conducted by mixing 2.970 mL of the diluted ABTS solution with 30 μL of the sample extract. The resulting solutions were mixed and stored in the dark at room temperature for 1 h. The samples were measured at 734 nm using an ultraviolet-visible spectrophotometer (model S-2150; Cole-Parmer) against a blank solution, and the results were determined from the standard curve (calyces: y = 0.1847x – 0.2853, R2 = 0.999; leaves: y = 0.1991x – 1.892, R2 = 0.992).
Dry ground calyx and leaf samples were sent to a commercial testing laboratory (A & L Great Lakes Laboratories, Fort Wayne, IN, USA) for mineral analysis. There were two samples of calyx and leaves for each accession. A microwave-assisted acid digestion procedure established by the US Environmental Protection Agency test method SW-846 3051A (US Environmental Protection Agency 2007) was used to process the samples. For each sample, 2 mL of nitric acid was added to 0.2 g of sample. The samples were then microwaved in two steps. In the first step, the microwave was programmed to ramp up to 90 °C, and that temperature was held for 90 s. The samples were then cooled to below 50 °C, and 1 mL of peroxide was added. The samples were then returned to the microwave for the second time, and the temperature was ramped up to 105 °C and held for 10 min. The samples were then cooled and brought to a final volume of 25 mL, capped, and thoroughly mixed. The resulting solution was then used for analysis.
Mineral analysis for P, K, Ca, Mg, S, Na, Fe, Mn, Zn, and Cu was conducted using an inductively coupled argon plasma according to Association of Official Analytical Chemists method 922.02 (Association of Official Analytical Chemists 2000a). Total N content was determined using the Dumas method with the Elementar Rapid-N cube (Elementar Americas Inc., Mt. Laurel, NJ, USA) according to Association of Official Analytical Chemists method 985.01 (Association of Official Analytical Chemists 2000b).
Quantitative results were analyzed using simple statistics and analysis of variance, and Tukey’s honestly significant difference was used to compare means between accessions if applicable. t tests were used to compare mineral content between calyces and leaves. The means were considered significantly different at P < 0.05 using R statistical software (R version 4.0.3; R Core Team 2022). In addition, the results were analyzed using principal component analysis (PCA) by JMP statistical software (version 17; SAS Institute Inc., Cary, NC, USA) to observe the correlations or relationships among the variables to determine the characteristics of roselle accessions.
Great variation was observed among nine accessions in their calyx color and size (Fig. 2). Calyx color was classified as “red” (red pigmented), “light green–red” (mostly green with some red pigment), or “light green” (very little to no red pigment). Three accessions (PI 265319, PI 275414, and PI 500706) produced red calyces, four accessions (PI 256039, PI 500713, PI 275413, and PI 291128) had light green–red calyces, and two accessions (PI 273389 and PI 500724) produced light green calyces (Fig. 2; Table 1). The nine accessions varied in their weight, length, and width of fruit (P < 0.05). The average weight of fruit was greatest with PI 291128 (13.2 ± 2.4 g) followed by PI 275413 (11.4 ± 2.3 g) and PI 275414 (10.9 ± 1.8 g), and PI 500724 had the lowest average fruit weight (3.2 ± 0.5 g). The ranges of fruit length and width were from 18.7 to 60.4 mm and from 17.8 to 34.8 mm, respectively. It was noted that two accessions (PI 256039 and PI 500724) produced distinct trichomes on the surface of the calyx.
Citation: HortScience 60, 6; 10.21273/HORTSCI18483-25
Leaf shape and pigmentation also varied among accessions. Five accessions (PI 273389, PI 275413, PI 275414, PI 500706, and PI 500713) produced palmate and deeply divided leaves (Fig. 2). Three accessions (PI 265319, PI 256039, and PI 291128) produced palmate and fully divided leaves, and one accession (PI 500724) had an ovate leaf. Red pigmentation was present in the petiole and leaf veins of seven accessions, which also had red or light green–red calyces. Two accessions with light green calyces (PI 273389 and PI 500724) lacked red pigment in their leaves.
Tables 2 and 3 show the concentration of four anthocyanins: D3S, D3G, Cy3S, and Cy3G, and the total amount of anthocyanins detected in the calyces and leaves in H. sabdariffa accessions, respectively. The concentrations of anthocyanins isolated from the calyx and leaf tissues were significantly different among accessions (P < 0.05). In the calyces, the total content of anthocyanins was the highest in the red calyces of PI 275414 (32.7 mg·g−1), followed by the calyces of PI 500706 (25.6 mg·g−1) and PI 265319 (23.3 mg·g−1). The three accessions of the red group contained D3S with 69% to 85% of the total anthocyanins, followed by Cy3S with 13% to 28%, and very low amounts of D3G and Cy3G were detected with concentrations of less than 3% and 1%, respectively. PI 275414 had the highest concentration of total anthocyanins (32.67 mg·g−1), consisting of D3S (27.79 mg·g−1), Cy3S (4.41 mg·g−1), D3G (0.43 mg·g−1), and Cy3G (0.03 mg·g−1) (Table 2). Four accessions in the light green–red group, PI 256039, PI 275413, PI 291128, and PI 500713, contained less than 1.00 mg·g−1 of all four anthocyanins except D3S in PI 256039, with 3.01 mg·g−1. The total anthocyanins in the light green–red group ranged from 1.05 to 3.88 mg·g−1. Two accessions from the light green group had less than 0.1 mg·g−1 of total anthocyanins. Anthocyanins were also detected in leaf tissues of all H. sabdariffa accessions except PI 500724 (Table 3). The highest amount of the total anthocyanins in leaves was from PI 291128 (1.06 mg·g−1).
Table 4 shows the concentration of chlorogenic acid and quercetin from calyx and leaf of nine accessions. The amount of chlorogenic acid was greater in calyx ranging from 13.07 to 26.13 mg·g−1 compared with leaf with the range of 0.66 to 1.97 mg·g−1. On the other hand, quercetin concentrations were higher in leaf with the range of 3.53 to 13.23 mg/100 g than in calyx with 1.29 to 3.53 mg/100 g. Variations in the content of chlorogenic acid and quercetin among accessions were found for both calyces and leaves.
The concentrations of the total phenols (TPC) and the total flavonoids (TFC) for calyx and leaf samples are shown in Tables 5 and 6, respectively. A variation was seen in both phytochemicals among accessions (P < 0.05). The TPC from calyces ranged from 10.43 to 17.28 mg GAE/g, while in leaves, the TPC ranged from 18.59 to 22.49 mg GAE/g. The highest TPC was found in the light green calyx from PI 273389 (17.28 mg GAE/g). Within the leaf samples, four accessions contained more than 20 mg GAE/g, including PI 275414 and PI 500706 (both in the red group), PI 500713 (light green–red group), and PI 273389 (light green group).
The TFC ranged from 56.74 to 92.79 mg QE/100 g for calyces (Table 5) and from 66.48 to 156.20 mg QE/100 g for leaves (Table 6). There were significant differences in TFC among the nine accessions for both calyx and leaf samples (P < 0.05). The highest TFC in calyx was from PI 500724 (light green), and the minimum content was found in PI 265319 (red). Leaves of PI 500713 (light green–red) had the highest concentration of TFC (156.20 mg QE/g).
The results of antioxidant capacities determined by DPPH and ABTS assays for calyx and leaf samples are shown in Tables 5 and 6, respectively. Antioxidant capacity by DPPH assay ranged from 6.55 to 7.27 mg VCEAC/g for calyces, while leaf tissues had a range of 16.55 to 21.95 mg VCEAC/g. The ABTS assay found that the range of antioxidant capacities of calyces was from 8.09 to 14.93 mg VCEAC/g, while leaf tissues ranged from 3.55 to 3.91 mg VCEAC/g. There were significant differences in both antioxidant capacities determined by both DPPH and ABTS assay for both calyces and leaves among the accessions (P < 0.05) (Tables 5 and 6).
Mineral contents in the calyces and leaves of H. sabdariffa are shown in Table 7 (for N, P, K, Ca, Mg, and S) and Table 8 (for Fe, Mn, Zn, Cu, and B). The concentrations of all elements except P were affected by plant part (P < 0.05), with calyces containing higher mean concentration for K (1.5%) and Zn (17.9 ppm) than leaf tissues having K (0.44%) and Zn (15.4 ppm). Leaf tissues contained higher contents of Ca (3.7%), N (2.1%), Mg (0.4%), S (0.2%), B (67.7 ppm), Mn (62.2 ppm), and Cu (3.2 ppm) than calyces with Ca (1.2%), N (1.0%), Mg (0.3%), S (0.14%), B (47.1 ppm), Mn (35.9 ppm), and Cu (2.6 ppm).
To evaluate the differences in phytochemical properties of calyx from nine H. sabdariffa accessions, PCA was performed with concentrations of anthocyanins, chlorogenic acid, quercetin, total phenols, total flavonoids, antioxidant capacities (DPPH and ABTS), and minerals (N, P, K, Ca, Mg, S, Fe, Zn, Mn, and Cu). Figure 3 shows the scores and loading plots obtained by PCA. The first two principal components (PCs) accounted for about 57.2% (PC1: 36.9%, PC2: 20.3%) of the total variance. The score plot shows dispersion of three red calyx accessions (PI 275414, PI 265319, and PI 500706) with positive values of PC1 from six accessions of the light green–red and light green groups having negative PC1 values (Fig. 3A). Observation of the score plots and loading plots together suggests that the main anthocyanins responsible for separation of three red calyx accessions from the light green–red and light green groups were D3S, Cy3S, and D3G. The loading plot shows that D3S, Cy3S, and D3G pointed the same positive direction of PC1 and were closely related to each other (Fig. 3B). The factors that contribute to PC1 (positive side) also included Cu, S, Mg, N, P, and Zn, and the loading plot indicated that N and P were very closely related to each other.
Citation: HortScience 60, 6; 10.21273/HORTSCI18483-25
Nine H. sabdariffa accessions grown in Guam showed that calyces had great variations in color, size, and presence of trichomes. Variations were also found in the shape and color of leaves. The accessions were divided into three groups according to calyx color: red (PI 265319, PI 275414, and PI 500706), light green–red (PI 256039, PI 500713, PI 275413, and PI 291128), and light green (PI 273389 and PI 500724). Red calyx is the preferred color for horticultural value-added products reported by several researchers (Borrás-Linares et al. 2015; Juliani et al. 2009; Khoo et al. 2017). Among the red group, PI 275414 originating from Poland, had a fresh fruit weight including calyx and seed capsule as 10.9 g, whereas the weights of the two other red accessions, PI 265319 and PI 500706, were 4.1 and 5.2 g, respectively. Earlier, Haryati et al. (2018) reported that fresh weight of calyx was 11.9 g at 33 d after anthesis of a cultivar grown in Indonesia. The fruit weight of PI 275414 is compatible with dark red cultivar Terengganu from Malaysia at about 10.2 g (Javadzadeh and Saljooghianpour 2017). The color of red calyx of PI 275414 was determined for three coordinates of CIELAB (L*, a*, b*) using a CR-400 chromameter (Konica Minolta, Inc., Tokyo, Japan) and obtained readings of L* = 29.90 ± 4.19; a* = 31.70 ± 5.55; b* = 15.14 ± 3.08 (n = 100) (Delfin MM and Marutani M, unpublished data). These quantitative measurements of color will strengthen in analytical interpretation to compare characteristics of accessions in our future studies.
In phytochemical analyses, the most abundant anthocyanin detected was D3S followed by Cy3S in red calyces, which coincides with the reports by Beye et al. (2017), Juliani et al. (2009), and Kouakou et al. (2015). In this study, the total anthocyanin contents in red calyces ranged from 23.30 to 32.67 mg·g−1, which were between the range of 6.88 to 15.29 mg·g−1 reported by Beye et al. (2017) and the value of 42.99 ± 0.70 m·g−1 reported by Kouakou et al. (2015). Low concentrations of anthocyanins were detected from both the calyces and leaves of light green–red and light green accessions. No anthocyanins were detected from the leaves of light-green PI 500724 (Tables 2 and 3). The role of anthocyanins in pigmentation was also shown by Christian and Jackson (2009); the red color of the calyces intensified when the anthocyanin content increased. Lyu et al. (2020) also verified that anthocyanin content in Hibiscus acetosella was positively associated with the expression of red pigmentation of leaves. In this study, we also found that the leaves of some accessions also contained red-pigmented veins.
The amounts of CGA found in H. sabdariffa leaves (0.66 to 1.97 mg·g−1) in this study were similar to the reports (0.49 to 1.49 mg·g−1) by Zhen et al. (2016). The CGA was the one of major phenolics in plant tissues with the concentration range of 13 to 26 mg·g−1 in calyces of accessions (Table 4).
The TPC and the TFC were examined as indicators of phytochemical compounds contributing to H. sabdariffa as a functional food. The highest TPC in H. sabdariffa calyces was observed in light green PI 273389 (17.28 ± 0.35 mg GAE/g), followed by red PI 500706 (15.50 ± 0.29 mg GAE/g). Because the light green accession PI 273389 lacked red pigmentation, the high TPC value suggests that calyces contain other compounds besides anthocyanins that are contributing to the TPC. TFC concentrations of calyces detected in this study ranging from 56.74 to 92.79 mg QE/100 g were much lower than those reported by Nerdy et al. (2022) and Borrás-Linares et al. (2015), possibly due to differences in extraction procedures or growing environment of the plant.
We examined antioxidant properties and DPPH and ABTS radical scavenging activity. The DPPH antioxidant capacity in H. sabdariffa accessions was greater in the leaves than in the calyces (Tables 5 and 6). Mohd-Esa et al. (2010) also reported that leaf extracts exhibited a greater DPPH radical scavenging activity compared with the calyces. In our study, ABTS antioxidant capacity was greater in the calyces than leaves. In comparison between the ABTS and the DPPH assay, Floegel et al. (2011) stated that highly pigmented foods have a greater ABTS antioxidant capacity than the DPPH antioxidant capacity. Our study agreed with their reports that the red calyces of PI 275414, PI 265319, and PI 500706 had higher ABTS antioxidant capacity than DPPH antioxidant capacity because water-soluble anthocyanins affected more the ABTS assay results. Additionally, this study showed that the TPC was positively correlated to DPPH antioxidant capacity for calyces, which aligned with the report by Mohd-Esa et al. (2010). The solubility of flavonoids should also be considered because the ABTS assay is suitable for both hydrophilic and lipophilic antioxidants, whereas the DPPH assay is best suited for lipophilic antioxidants (Magalhães et al. 2008).
Among the six compounds examined in the study, anthocyanins were the highest in concentration in hibiscus calyces. Water-soluble anthocyanins can be easily extracted, and the water solubility of anthocyanins facilitates rapid absorption into the body immediately following consumption (McGhie and Walton 2007). When consumed as an infusion, reports have demonstrated its medicinal properties in cancer prevention and for treating hypertension (Da-Costa-Rocha et al. 2014).
For mineral analysis, our study showed that calyx contained the mean concentration of macroelements, N (1.0%), P (0.26%), K (1.46%), Ca (1.2%), Mg (0.3%), and S (0.14%), and microelements, Fe (58.7 ppm), Mn (35.9 ppm), Zn (17.9 ppm), Cu (2.6 ppm), and B (47.1 ppm). In earlier studies in Africa, Salami and Afolayan (2021) indicated that calyces contained compatible amounts of P (0.37% for green calyx and 0.13% for red) and Ca (1.1% for green and 1.7% for red) with our findings. Other elements including K, Mg, Fe, Mn, Zn, and Cu had much higher contents in their calyces than our reports; especially microelements of Mn (64.5 ppm for green and 294 ppm for red), Zn (39.5 ppm for green and 40 ppm for red), Cu (8.5 ppm for green and 5.0 ppm for red), and Fe (179 ppm for green and 212 ppm for red) were almost more than double the amount of our measurements. The high Zn and Fe contents of calyces were also reported by Babalola et al. (2001). Green, red, and dark red calyces contained 58, 65, and 63 ppm of Zn and 328, 378, and 346 ppm of Fe, respectively. This suggests that African roselle varieties have high mineral nutrient values and that they are encouraged highly for consumption as vegetable crops in the region.
The results obtained by PCA demonstrate a clear dispersion of red calyx accessions from the light green–red and light green calyx groups due to three anthocyanins, D3S, Cy3S, and D3G, and six elements, Cu, S, Mg, N, P, and Zn (Fig. 3A and 3B). This study highlights the use of the H. sabdariffa calyces as a potential functional food source. Accessions with red fleshy calyces produced greater amounts of anthocyanins and exhibited higher antioxidant capacities.
The study characterized nine H. sabdariffa accessions for their morphological traits, phytochemical contents, and antioxidant capacity of calyx and leaf tissues in search of a new functional crop in Guam. Four anthocyanins were detected by HPLC, and the high contents of total anthocyanins were found from calyx of three accessions in the red group. The major anthocyanins were delphinidin 3-sambubioside followed by cyanidin 3-sambubioside, which were responsible for the red calyces. Because the red calyx was desirable to produce herbal tea or coloring, the most interesting accession in this study was PI 275414, which had the largest red fresh calyx. The study revealed that both calyces and leaves contained phytochemicals, minerals, and antioxidant properties that can be used for benefits to human health. Future studies will include the identification of methods for commercial production and creation of value-added products from H. sabdariffa.
Respective chromatograms of a H. sabdariffa calyx extract. (A) Chromatogram at 520 nm. (B) Chromatogram at 325 nm. (C) Chromatogram at 372 nm. Identities of labeled peaks are as follows: delphinidin 3-sambubioside (1), delphinidin 3-glucoside (2), cyanidin 3-sambubioside (3), cyanidin 3-glucoside (4), chlorogenic acid (5), and quercetin (6).
Illustrations of calyx and leaf of nine H. sabdariffa accessions grouped according to their calyx color as red (A), light green and red (B), and light green (C).
Principal component analysis showing score (A) and loading plots (B) for principal component (PC) 1 × PC2. The score plot shows the accession number of three calyx color groups of nine H. sabdariffa accessions: red calyx is indicated with circles, light green and red is indicated with triangles, and light green calyx is indicated with rectangles. Variables in the loading plot represent the phytochemical concentration of cyanidin 3-glucoside (Cy3G), cyanidin 3-sambubioside (Cy3S), delphinidin 3-glucoside (D3G), delphinidin 3-sambubioside (D3S), total anthocyanin (TA), chlorogenic acid (CGA), quercetin (QUE), total phenols (Phenols), total flavonoids (Flavonoids), antioxidant capacity [2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)], and mineral content (N, P, K, Ca, Mg, S, Fe, Zn, Mn, and Cu) of the calyx.
Contributor Notes
This work is supported by the Hatch/Multistate Plant Genetic Resources Conservation and Utilization (Accession No. 1918320) and the Resident Instruction Grants Program for Institutions of Higher Education in Insular Areas and Agriculture and Food Sciences Facilities and Equipment (2017-70004-27229; CRIS No. 1014067) from the US Department of Agriculture’s National Institute of Food and Agriculture. This paper is a portion of a thesis completed by M. M. Delfin. We thank Jian Yang and Maika Vuki for research suggestions. Our appreciation goes to Chieriel Desamito, Lara Mazloum, and Aubrie Uson for technical assistance. We also thank Sahena Ferdosh and Yin Yin Nwe for reviewing the article.
Respective chromatograms of a H. sabdariffa calyx extract. (A) Chromatogram at 520 nm. (B) Chromatogram at 325 nm. (C) Chromatogram at 372 nm. Identities of labeled peaks are as follows: delphinidin 3-sambubioside (1), delphinidin 3-glucoside (2), cyanidin 3-sambubioside (3), cyanidin 3-glucoside (4), chlorogenic acid (5), and quercetin (6).
Illustrations of calyx and leaf of nine H. sabdariffa accessions grouped according to their calyx color as red (A), light green and red (B), and light green (C).
Principal component analysis showing score (A) and loading plots (B) for principal component (PC) 1 × PC2. The score plot shows the accession number of three calyx color groups of nine H. sabdariffa accessions: red calyx is indicated with circles, light green and red is indicated with triangles, and light green calyx is indicated with rectangles. Variables in the loading plot represent the phytochemical concentration of cyanidin 3-glucoside (Cy3G), cyanidin 3-sambubioside (Cy3S), delphinidin 3-glucoside (D3G), delphinidin 3-sambubioside (D3S), total anthocyanin (TA), chlorogenic acid (CGA), quercetin (QUE), total phenols (Phenols), total flavonoids (Flavonoids), antioxidant capacity [2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)], and mineral content (N, P, K, Ca, Mg, S, Fe, Zn, Mn, and Cu) of the calyx.