Leaf pigmentation, its profiles and radical scavenging activity in selected Amaranthus tricolor leafy vegetables

The selected A. tricolor accessions contained abundant color attributes, betacyanin, carotenoids, betalains, betaxanthins, and antioxidants potentiality that varied in terms of genotypes. For the first time, we identified 4 betacyanins, and 5 carotenoid compounds in A. tricolor genotypes. The genotype VA14 and VA16 had abundant color attributes, betacyanin such as amaranthine, iso-amaranthine, betanin, iso-betanin, and antioxidants potentiality. These two genotypes having an excellent source of color attributes, betacyanins, betalains, betaxanthins, and antioxidants potentiality could be used as potent antioxidant varieties. The genotype VA11 and VA16 had abundant carotenoid components, such as zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls, and beta-carotene. The genotype VA11 and VA16 had abundant carotenoid components that could be used as carotenoid enrich varieties. It revealed from the correlation study that pigment profiles of A. tricolor genotypes exhibited high quenching capacity of radicals. These accessions have high antioxidant potentials and great opportunity to make drinks, preservatives, and colorant of food products to feed the community deficient in antioxidants. The identified components of betacyanins and carotenoids in A. tricolor require comprehensive pharmacological study. The baseline data on color attributes, betacyanins profile, carotenoids profile, betaxanthins, betalains and antioxidant potentiality obtained in the present study could contribute to pharmacologists for evaluating these components scientifically in A. tricolor.


Results and discussion
Color attributes. Figure 1 shows the leaf color attributes of four selected A. tricolor. The pronounced variability was noted with regard to b*, lightness (L*), a*, and chroma of the four accessions studied. The range of chroma, a*, b*, and lightness (L*) were 8.15 to 19.73, 7.35 to 17.68, 3.51 to 8.75, and 23.78.34 to 40.35, respectively. The maximum values of lightness was noticed in VA6 (40.35), while the minimum values of lightness was reported in VA14 (23.78) followed by VA16 (26.84). Similarly, the maximum a* (17.68), chroma (19.73), and b* (8.75) value were noted in VA14 followed by VA16, while the lowest a* (7.35), b* (3.51), and chroma (8.15) were found in VA6. We found corroborative findings with the findings of Colonna et al. 30 and Sarker and Oba 31 . Color attributes are essential parameters that have contributed significantly to the choice of consumers, their acceptability and preferability. These are main attributes for testing the potentiality of antioxidants in vegetables 30 . Deep red accessions VA14 and VA16 had high values of yellowness and redness representing the occurrence of plentiful pigments including betaxanthin, anthocyanins, betacyanin, and betalain. Conversely, the accession VA6 exerted low values of yellowness and redness representing the occurrence of low pigments including betaxanthin, anthocyanins, betacyanin, and betalain. The four bright red color accessions have ample betacyanin pigments showing better stability at pH 5-7 and lower temperatures (<14 °C) 32 . These accessions have great opportunity to make drinks, preservatives and colorant of food products.
Betacyanin components. The betacyanin is responsible for the red-violet and maroon color in the appearance of A. tricolor genotypes. Table 1 shows main fragment ions (MS 2 ), molecular ion, retention time, λmax, and detected betacyanin components. The liquid chromatograph separated values of betacyanin components from four genotypes (VA16, VA14, VA11, and VA6) were compared with standard betanin (Betanin from red beet were used as standards) using the respective peaks of the compounds. Figure 2 represents the individual betacyanin components and their relative proportions in the extracts. The HPLC determined betacyanin exhibited 2 principal peaks (peak1 and peak2) at 538 nm (Table 1). Peak1 and peak2 were eluted before the peaks of betanin and iso-betanin standards (peak3 and peak4), representing a greater intensity of glycosylation 33 . The betacyanin components amaranthine and iso-amaranthine were abundant pigments in Amaranthus 34 . Cai et al. 35 also reported that amaranthine as well as iso-amaranthine (C15 epimer of amaranthine) were widely distributed and available www.nature.com/scientificreports www.nature.com/scientificreports/  (Table 1). Based on the UV-Vis and MS data presented in Table 1 and those obtained by others, peak1 and peak2 were tentatively identified as amaranthine and iso-amaranthine (C15 epimer), respectively 32,33 .
Considering betacyanin components, amaranthine was the most abundant betacyanin followed by iso-amaranthine in A. tricolor genotypes (Fig. 2). Our obtained amaranthine, iso-amaranthine, betanin and iso-betanin content in A. tricolor genotype VA14 and VA16 were greater than the results of Stinzing et al. 33 in A. spinosus stem. Amaranthine, iso-amaranthine, betanin, iso-betanin, and total betacyanin ranged from 7.82 to 22.84, 5.32 to 12.83, 3.37 to 6.84, 2.21 to 5.32, and 19.60 to 46.52 mg 100 g −1 , respectively (Fig. 2). The highest amaranthine (22.84 mg 100 g −1 ), betanin (6.84 mg 100 g −1 ), iso-betanin (5.32 mg 100 g −1 ) and total betacyanin (46.52 mg 100 g −1 ) were obtained from the genotype VA14 followed by the genotype VA16. Cai et al. 19 also reported that amaranthine was the major compounds in betacyanin components in the plants of Amaranthaceae which corroborated with the present findings. Whereas, the genotype VA16 showed the highest iso-amaranthine (12.83 mg 100 g −1 ) followed by VA14. In contrast, the genotype VA11 showed the lowest amaranthine (7.82 mg 100 g −1 ), iso-amaranthine (5.32 mg 100 g −1 ), iso-betanin (2.21 mg 100 g −1 ) and total betacyanin (19.60 mg 100 g −1 ) and the genotype VA6 exhibited the lowest betanin (3.37 mg 100 g −1 ). Within leafy vegetables, amaranth is a unique and excellent source of betacyanin components including amaranthine, iso-amaranthine, betanin, iso-betanin. These betacyanin components have high scavenging activity of free radicals and act as an excellent antioxidants 19 . Betacyanin components could be used as a food colorant in low-acid foods as these components have higher pH stability compared to anthocyanins 20 . Amaranthine, a major pigment of betacyanin components in amaranth had very strong quenching capacity of radicals and also act as an excellent antioxidants. It is an alternate source of betanins (red beets) for colorant of foods and potential antioxidants 19 . Hence, the genotypes VA14 and VA16 containing high betacyanin profiles including amaranthine, iso-amaranthine, betanin, iso-betanin could be used for extracting colorful juice as drinks, preservatives, and colorant of food products.  Table 1. Retention time (Rt), wavelengths of maximum absorption in the visible region (λ max ), mass spectral data and tentative identification of betacyanin components in four selected A. tricolor leafy vegetables. * Stintzing et al. 33 . www.nature.com/scientificreports www.nature.com/scientificreports/ Carotenoid profiles. Table 2 represents the data on the molecular ion, main fragment ions in MS 2 , λmax, retention time, and identified carotenoid compounds. The LC separated carotenoid values from four genotypes (VA16, VA14, VA11, and VA6) were compared with standard carotenoid compounds masses through the corresponding peaks of the compounds. A total of five carotenoid compounds were identified in A. tricolor. Among them, four compounds were xanthophylls (Lutein, violaxanthin, neoxanthin, and zeaxanthin) and one compound was pro-vitamin A (beta-carotene). Figure 3 represents the identified carotenoid profiles along with total xanthophylls and total carotenoids of four selected A. tricolor leaves, respectively.
Percentage of zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls, beta-carotene to total carotenoids ranged from 4.98 to 6 Table 2. Retention time, wavelengths of maximum absorption in the visible region (λ max ), mass spectral data and tentative identification of carotenoid profiles in four selected A. tricolor leafy vegetables. www.nature.com/scientificreports www.nature.com/scientificreports/ 100 g −1 FW, respectively (Fig. 4). The genotype VA6 demonstrated the highest percentage of zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls to total carotenoids followed by VA16 and VA11 albeit this genotype exhibited the lowest beta-carotene. In contrast, VA14 showed the lowest percentage of zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls albeit this genotype exhibited the highest beta-carotene. It indicated from the results that the percentage of total xanthophylls to carotenoids along with its components had the opposite relationship with the percentage of beta-carotene content to carotenoids of A. tricolor genotypes. (Fig. 4).
In this study, we found remarkable pigments profile such as betacyanins, betalains, betaxanthins, and carotenoid profiles such as lutein, violaxanthin, neoxanthin, zeaxanthin, total xanthophylls, beta-carotene in A. tricolor genotypes. The results of total carotenoids of our study corroborated with the results of Khanam and Oba 37 where they observed higher carotenoids in the red amaranth genotype compared to green amaranth. The genotype VA11 and VA16 contained higher lutein, violaxanthin, neoxanthin, zeaxanthin, total xanthophylls, beta-carotene, and total carotenoids compared to the genotype VA14 and VA6. Hence, the carotenoid profiles of A. tricolor genotype could be an important parameter for consumers, playing a crucial role in detoxification of ROS in the human body and preventing antiaging and many degenerative human diseases 17,29 . Our result showed that the A. tricolor genotype is an excellent source of lutein, violaxanthin, neoxanthin, zeaxanthin, total xanthophylls, beta-carotene, and total carotenoids among leafy vegetables that has important free radical-scavenging activity 19 . A. tricolor genotypes VA11 and VA16 had high carotenoid profiles, such as zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls, beta-carotene, and total carotenoids content. The genotypes VA11 and VA16 might be utilized as high-yielding varieties. These genotypes containing abundant carotenoid profiles could be used for extracting colorful juice as drinks, preservatives, and colorant of food products. It revealed from the current investigation that VA11 and VA16 have abundant carotenoids that explore new insight for comprehensive pharmacological study.
A. tricolor accession VA14 demonstrated the maximum antioxidant capacity (ABTS + ) followed by VA16. On the contrary, antioxidant capacity (ABTS + ) was the minimum in VA6. Our results were corroborated with the findings of Khanam and Oba 37 . They reported greater betaxanthins, betalains, and antioxidant capacity in the red color amaranth compared to green color amaranth. The A. tricolor genotype VA14 and VA16 contained higher betaxanthins, betalains, and antioxidant capacity compared to the genotype VA6 and VA11. Hence, these antioxidant constituents of A. tricolor genotype could be an important parameter for consumers, playing a crucial role www.nature.com/scientificreports www.nature.com/scientificreports/ in quenching of ROS, curing many degenerative diseases and preventing anti-aging and 17,29 . Our result revealed that A. tricolor contained ample betaxanthins, betalains, and antioxidant potentials among leafy vegetables that have important free radical-scavenging activity 19 .
In this study, we found remarkable color attributes, betacyanin profiles such as amaranthine, iso-amaranthine, betanin, iso-betanin, carotenoid profiles, such as zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls, beta-carotene, total carotenoids, betaxanthins, betalains, and antioxidant capacity in A. tricolor genotypes. Our results were corroborated with the findings of Khanam and Oba 37 . They reported greater antioxidant capacity, betacyanins, betaxanthins, betalains, total carotenoids in the red color amaranth compared to green color amaranth. Betacyanin, total carotenoids, antioxidant capacity (ABTS + ) and antioxidant capacity (DPPH) of current investigation were corroborated with the findings of A. tricolor by Khanam et al. 38 . We found 2 to 3 time's higher β-carotene in red color amaranth compared β-carotene of Raju et al. 36 in A. gangeticus. The β-carotene of the leaves of red color amaranths were 2 to 3 times and green color amaranths were twice than the β-carotene of A. caudatus leaves 29 . Li et al. 39 in A. hypochondriacus leaves noticed the highest quenching capacity of radicals (ORAC and FRAP methods) in comparison to the leaves of A. caudatus. They also noted the maximum quenching capacity of radicals (FRAP) in leaves in comparison to seed, stalks, sprouts, and flowers. The genotypes VA14 and VA16 had high color attributes, betacyanin profiles, betaxanthins, betalains, and antioxidant capacity. The genotypes VA14 and VA16 containing high antioxidant profiles and high quenching capacity of radicals might be utilized as high-yielding varieties. It revealed from the current investigation that quenching capacity of radicals have high possibility to feed the community deficient in antioxidants. These genotypes also could be used for extracting colorful juice as drinks, preservatives, and colorant of food products.
The correlation coefficient study. Correlation of betacyanins, betaxanthins, betalains, and antioxidant capacity of A. tricolor leafy vegetables are shown in Table 3. Total betacyanins and betaxanthins, betalains had highly significant positive associations among themselves, with total carotenoids, antioxidant capacity (DPPH and ABTS + ). It revealed that total betacyanins betaxanthins, and betalains, exhibited strong antioxidant activity. Total xanthophylls, total carotenoids, and β-carotene had positive and significant interrelationships with β-carotene, antioxidant capacity (ABTS + and DPPH), and total carotenoids which revealed high quenching   www.nature.com/scientificreports www.nature.com/scientificreports/ capacity of radicals of major carotenoids. The findings of current investigation were corroborated with the results of our earlier study of drought and salt-stressed A. tricolor 24 .
In conclusions, we identified betacyanin profiles, such as amaranthine, iso-amaranthine, betanin, iso-betanin, carotenoid profiles, such as zeaxanthin, lutein, violaxanthin, neoxanthin, total xanthophylls, beta-carotene, and total carotenoids, betaxanthins, betalains, and antioxidant potentiality (DPPH and ABTS + ) in A. tricolor genotypes. A. tricolor genotype VA14 and VA16 had abundant betacyanin components, betaxanthins, betalains and antioxidant potentiality might be utilized as high-yielding varieties. VA11 and VA16 had abundant carotenoids components might be utilized as high-yielding varieties. It revealed from the correlation study that all phytopigments of A. tricolor exhibited good quenching capacity of radicals. It also revealed from the current investigation that these accessions had ample antioxidants to quench ROS and had high possibility to study pharmacologically in details. These genotypes also could be used for extracting colorful juice as drinks, preservatives, and colorant of food products. The data on color attributes, betacyanins, carotenoids, betaxanthins, betalains and antioxidant potentiality obtained in the present study could contribute to the scientists for scientific evaluation of these active compounds in A. tricolor.

Methods
Experimental materials. We selected four high yields and antioxidant potential genotypes from 43 genotypes. It is the first report on color attributes, betacyanin and carotenoid profiles, and antioxidant potentials in A. tricolor.
Design and layout. We executed the experiment in three replicates following a completely randomized block design (RCBD) at Bangabandhu Sheikh Mujibur Rahman Agricultural University. Each genotype was grown in 1 m 2 experimental plot following 20 cm and 5 cm distance between rows and plants, respectively.
Determination of color attributes. We measured the color attributes C*, L*, b*, and a* using a TES-135A Plus color meter (Taiwan) in fifteen replicates. Positive b value (+b*) represents yellowness, while negative b value (−b*) indicates blueness. Positive a value (+a*) indicates the magnitude of redness, while negative a value (−a*) indicates magnitude of greenness. L* represents lightness and C* value represents magnitude of leaf color designated as chroma. The chroma value was estimated following the formula, Chroma C* = (a 2 + b 2 ) 1/2 .

Extraction of samples for LC-MS and HPLC analysis.
Ten ml methanol (80%) containing acetic acid (1%) was added in leaf sample (1 g). The mixture was thoroughly homogenized and transferred to a 50 ml tightly capped test tube. The test tubes were placed in a Scientific Industries Inc. shaker (USA) for fifteen h at 400 rpm.
A MILLEX ® -HV syringe filter, USA (0.45 µm) was utilized to filter the mixture. It was centrifuged for fifteen min at 10,000 × g. Betacyanin components were detected from this filtrate. Betacyanin analysis in the samples could interfere through precipitation of methanol with the proteins and other insoluble substances in the samples. A Polymeric Reversed-Phase cartridges (33 µ) (Strata TM -X, USA) was utilized to purify betacyanin. The samples were extracted in triplicates.
Betacyanin analysis through HPLC. The method previously described by Stintzing et al. 33 was followed to determine betacyanin components in A. tricolor leaf sample using HPLC. The high performance liquid chromatograph Shimadzu SCL10Avp, Kyoto, Japan was furnished with a degasser (DGU-14A), an LC-10Avp binary pumps, and a Shimadzu SPD-10Avp detector (UV-vis). A STR ODS-II, 150 × 4.6 mm I.D., CTO-10AC column (Shinwa Chemical Industries, Ltd., Kyoto, Japan) was utilized to detected the betacyanin components. The binary mobile phase was pumped with solvent A (acetic acid 6% v/v) in water and solvent B (acetonitrile) at the flow rate of 1 ml/min for seventy min. High performance liquid chromatograph was run using a gradient program containing acetonitrile 0-15% for forty five min, 15-30% for fifteen min, 30-50% for five min and 50-100% for five min. Exactly 35 °C temperature was continued in the column with 10 μl injection volume. The betacyanin was simultaneous monitored by setting the detector at 538 nm. The betacyanin components were detected through comparing UV-vis spectra and retention time with their respective standard compounds. We confirmed the betacyanin components through the method of mass spectrometry. The betacyanin components were detected in triplicates. The betacyanin components were estimated as mg 100 g −1 of fresh weight. The LC-MS analysis www.nature.com/scientificreports www.nature.com/scientificreports/ was done using a mass spectrometer (Tokyo, Japan) equipped with HPLC (Agilent 1100 Series) and a detector coupled with on-line ElectroSpray Ionization (ESI) source. The LC-MS analysis was done in negative ion mode with the needle voltage at −2000 V and the range of the column elutes was m/z 0-1000. Extract constituents were identified by LC-MS-ESI analysis.
Quantification of betacyanin components. Calibration curves of the respective standards were used to quantify individual phenolic compounds. The betanin was dissolved in methanol (80%) to prepare the stock solutions of 10, 20, 40, 60, 80, and 100 mg/ml. A standard curves were made from these solutions to quantify the individual betacyanin components with external standards. The retention times, co-chromatography, and UV spectral characteristics of samples spiked with available standards (commercially) to identify and match the betacyanin components. Betanin standard was used to prepare standard curves based on the equimolecular conversion for estimating amaranthine and iso-amaranthine in the different samples.
Sample preparation for extraction of carotenoids. The fresh leaves were thoroughly washed and dried using blotting papers. All precautions were taken to protect photo-isomerization and photo-oxidation and prevent any losses of carotenoids. Sampling was performed at 20 °C with subdued lighting. We ground the dried leaves using a mechanical blender. The powdered leaves were kept in a bag (self-sealing) containing aluminum foil inside the bag at below −20 °C.

Extraction of carotenoids.
We extracted carotenoids following the procedure of Lakshminarayana et al. 41 .
We extracted carotenoids with acetone (ice-cold) until the powdered leaves became colorless. To prevent oxidation of carotenoid rapid extraction in acetone (ice-cold) was performed. Petroleum ether (100 ml) and 25% aqueous sodium chloride (w/v) (100 ml) were added to leaf extract (50 ml) in a funnel. After mixing, we separated the upper layer. We repeated it for three times (final volume: 250 ml). We dried the extract over 20 g anhydrous sodium sulphate and filtered through a filter paper (Whatman No.1). A rotary evaporator was used to evaporate the filtrate at 35 °C to dryness. We re-dissolved the dried filtrate in hexane (known volume) and an aliquot (100 µl) for analyzing it using high performance liquid chromatograph. We carried out homogenization, extraction, and sample handling under dim yellow light at 4 °C to reduce photo-oxidation and photo-isomerization of carotenoids.
HPLC analysis. We estimated the carotenoid profiles in A. tricolor leaf sample using high performance liquid chromatograph (Shimadzu SCL10Avp, Kyoto, Japan) 41 . A variable detector, binary pumps, and a degasser were equipped with the HPLC system. Briefly, the carotenoids were separated on a column (STR ODS-II, 150 × 4.6 mm I.D., Kyoto, Japan). The carotenoids were separated using methanol:acetonitrile:dichloromethane with a ratio of 20:60:20 (v/v/v) containing ammonium acetate (0.1%) as a mobile phase. For HPLC analysis, leaves (20 µl) were injected under the isocratic condition at a flow rate of 1 ml/min. The detector was set at 450 nm. The SPD-10AVP detector were equipped with Shimadzu model LC-10Avp series to confirm the peak of carotenoids in comparison with their retention time of standard chromatograms. While the characteristic spectrum record with a PDA detector was taken to confirm the λ max values of these compounds. We quantified the carotenoid profiles estimating the area of peak of corresponding standards.
Betaxanthins content measurement. The leaves of A. tricolor leafy vegetables were extracted in methyl alcohol (80%) containing ascorbate (50 mM) to measure betaxanthins following the method of Sarker and Oba 42 . A Hitachi, U-1800 spectrophotometer (Tokyo, Japan) was utilized to read the absorbance at 475 nm for betaxanthins. We calculated the betaxanthins as miligrams per gram of fresh weight equivalent to indicaxanthin. Antioxidant capacity assay. Thirty days old A. tricolor leaves were harvested. Antioxidant capacity assay, the leaves were dried in the air in a shade. Aqueous methanol (40 ml, 90%) was utilized to extract grounded dried leaves (1 g) from each cultivar in a capped bottle (100 ml). A Thomastant T-N22S (Japan) shaking water bath was utilized to extract leaf samples for 1 h. A MILLEX ® -HV syringe filter (0.45 µm) (USA) was used to filter the homogenized mixture. After centrifugation for fifteen min at 10,000 × g, the antioxidant capacity was estimated from the extract.
Diphenyl-picrylhydrazyl (DPPH) radical degradation method 43,44 was used to estimate the antioxidant activity. We added DPPH solution (250 µM, 1 ml) to extract (10 µl) in a test tube (in triplicate). Finally we added 4 ml distilled water the extract and kept in the dark (30 min). A spectrophotometer (Hitachi, Tokyo, Japan) was used to take the absorbance at 517 nm. The antioxidant activity was also estimated following ABTS + assay 45,46 . To prepare two stock solutions separately ABTS + solution of 7.4 mM and potassium persulfate of 2.6 mM were used. We mixed both solutions in equal proportion to prepare the working solution at room temperature. The working solution was allowed to react in the dark for 12 h. One hundred fifty μl extract was added to 2.85 ml of ABTS + solution and allowed to react in the dark for 2 h. For the preparation of the solution, one ml of ABTS + solution was mixed with sixty ml of methanol. A spectrophotometer (Hitachi, Tokyo, Japan) was utilized to take the absorbance at 734 nm. The inhibition of ABTS + (%) and DPPH (%) with control were utilized to calculate quenching capacity of radicals using the equation: − . . × Antioxidant activity (%) (A blank A sample/A blank) 100 www.nature.com/scientificreports www.nature.com/scientificreports/ Where, A. sample is the absorbance of the test compound and A. blank is the absorbance of the control reaction [10 µl methanol (DPPH), 150 μl methanol (ABTS + ) instead of extract]. The quenching capacity of radicals were calculated as μg g −1 of dry weight equivalent to Trolox. Statistical analysis. Statistix 8 software was used to analyze the data for analysis of variance (ANOVA) 47 .
Duncan's Multiple Range Test (DMRT) at a 1% level of probability was used to compare the means. The results were reported as the mean ± SD of three separate replicates.
Ethical statement. The lab and field experiments in this study were carried out as per guidelines and recommendations of "Biosafety Guidelines of Bangladesh" published by the Ministry of Environment and Forest, Government of the People's Republic of Bangladesh (2005).

Data availability
Data used in this manuscript will be available to the public.