Effect of Harvest Age on Total Phenolic, Total Anthocyanin Content, Bioactive Antioxidant Capacity and Antiproliferation of Black and White Glutinous Rice Sprouts

Black (cv. BGR) and white (cv. RD6) glutinous rice sprouts from fertilizerand pesticide-free farm in Khon Kaen province, Thailand were investigated for antioxidation and antiproliferative activity. Three different ages of rice sprouts were collected and prepared as the extract. BGR exerted higher antioxidant capacity than RD6 based on total phenolic (TPC) and total anthocyanin contents (TAC), DPPH, and FRAP assays. BGR at 10–15 days contained the highest TPC (29.72 ± 1.42 mg gallic acid equivalent/g extract) and reducing power (2.22 ± 0.014 mmole FeSO4/g extract). BGR at 20–25 days contained the highest TAC (0.86 ± 0.096 equivalence of cyanidin-3-glucoside/g extract) and DPPH radical scavenging activity (IC50 = 231.09 ± 12.99 μg/mL). Antiproliferative activity of the extracts was evaluated in the human T-lymphocyte (Jurkat), hepatocellular carcinoma (HepG2), colorectal carcinoma (HCT116), melanoma (SK-MEL-2) and noncancerous cells (Vero) by neutral red assay. BGR showed the most selective antiproliferation against Jurkat cells, by inducing apoptosis, and caspase 3/7 activity. BGR at 200 μg/mL from all ages significantly decreased ROS using DCFH-DA and increased endogenous glutathione levels in Jurkat cells compared to the control (p < 0.05). The higher antiproliferation of BGR than RD6 was via its antioxidation capacity and attributed to its higher phenolic and anthocyanin contents. BGR sprout is a potential source of biologically active substances good for wellness and health benefits.


Introduction
Cancer is an uncontrollable cell growth developed into abnormal cells that can invade normal body tissue. The major influencing factors include genetic and lifestyle that affect and trigger cancer formation in the body. Cancer formation is the multiple steps involving the transformation of the normal cells into cancer cells (initiated cells), which undergo tumor promotion to become preneoplastic During germination, hydrolytic enzymes in rice grain were activated to decompose macromolecules such as starch and protein for synthesizing new cell constituents, and to alter the micromolecular constituents. Rice grain has been reported to accumulate a significant amount of phenolic and flavonoid compounds such as chalcone, flavones, condensed tannins, ascorbic acid, tocopherol, and anthocyanin, which are antioxidants [23]. Our previous study reported the presence of chlorophylls, flavonoids, phenolics, and anthocyanins in the BGR and RD6 at the same harvest ages [24]. Rice sprouts could, therefore, play a profitable role for health. However, few studies on their biological activity as functional foods have been investigated. In addition, to be used as health benefiting foods, there was an insufficient study on the rice sprouts on their antioxidant as well as antiproliferative capability against cancer cells.
RD6 and BGR are two different glutinous cultivars that are popularly consumed in Northeastern Thailand. RD6 is the white grain, while the black glutinous rice (BGR) is the black grain. The rice sprout can be edible as it is characterized as soft and chewable. Our previous study indicated that the two cultivars consist of several phenolic compounds (i.e., protocatechuic acid, vanillic acid, and rutin) [24], which have been proposed to possess antioxidant properties and could contribute to the anticancer activity. Therefore, our objectives of the study were to determine phenolic compounds, as well as the antioxidant activities of RD6 and BGR, both in vitro and in cell models, that could attribute to the anticancer activity. In this study, rice sprouts of these two rice cultivars (RD6 and BGR) were then investigated for their phenolic compounds (total phenolic and total anthocyanin content) that potentially contributed to antioxidant activities (detected by DPPH and FRAP assay), and antiproliferative activity (using neutral red assay) against several cancer cell lines. The most sensitive cell line was further selected for the determination of intracellular oxidative stress (intracellular ROS and GSH) as well as the apoptotic induction (annexin V staining) and caspase 3/7 activity. Our obtained information could be useful for the future development of health food products and/or pharmaceuticals.

Plant Extraction
In this study, the rice sprouts with two different cultivars-black glutinous rice (BGR) and white glutinous rice-(RD6) at different harvest ages-young age (5-7 days), middle age (10-15 days) and old age (20-25 days)-were collected from fertilizer-and pesticide-free farm in Khon Kaen province, Thailand. The fresh rice sprouts were chopped into small pieces and extracted with methanol containing 1% hydrochloric acids (20 g per 1000 mL) and then sonicated for 30 min (Crest Ultrasonics, Bangkok, Thailand). The extract was filtered through filter paper. Then, the solvents were removed by a rotary evaporator (IKA ® RV 10 Rotary evaporator, Staufen, Germany) at room temperature and moisture was removed by a freeze drier (Eyela FDU-1200 Freeze dryer, Tokyo, Japan) to obtain the dry residue. Finally, the crude extract was stored under 4 • C in an air-tight container before use.

Total Phenolic Content (TPC) by Folin-Ciocalteu's Reagent Method
Total phenolic content (TPC) was determined by Folin-Ciocalteu's reagent method [25,26]. The concentration of the stock solution of the rice sprout extracts was prepared at 10 mg/mL. Briefly, 15 µL of the extract was mixed with 120 µL of the prepared Folin-Ciocalteu's reagents and kept in the dark at room temperature for 5 min. Later, 60 g/L sodium carbonate buffer pH 7.5, 120 µL was added into the mixture and kept for another 90 min under the same condition. The absorbance of the blue solution of molybdenum (V) in Folin-Ciocalteu's reagents was measured at 725 nm with the microplate spectrophotometer (EnSight™, Waltham, MA, USA). Gallic acid was used as a standard antioxidant and as a positive control. The experiments were repeated in four replicates. The standard curve was created from the plot between the absorbances of the blue solutions against the gallic concentrations (final concentrations of 2, 10, 20 and 40 µg/mL). TPC was calculated from a standard curve of gallic acid (y = 0.643x + 0.0267, R 2 = 0.9997) and expressed as milligrams of gallic acid equivalent per gram of rice sprouts dried extract (mg GAE/g extract).

Total Anthocyanin Content (TAC)
The anthocyanin content was measured according to the pH differential method [27][28][29] to determine the colored oxonium formed at pH 1.0 and colorless hemiketal format at pH 4.5. Briefly, the mixture of rice sprout extracts solution in methanol (250 µL) was mixed with either sodium acetate buffer pH 4.5 or potassium chloride buffer pH 1.0 (750 µL) in each well. The extract in methanol was diluted 10 times with two buffers, shaking for 15 min under the dark condition and then centrifuged at 2500× g for 10 min at room temperature. The absorbance of the pigment concentration in each well was measured at 510 and 700 nm at room temperature by using the microplate spectrophotometer (EnSight™). The experiments were performed in triplicates. TAC was calculated following Equation (1). Data were presented as milligrams of cyanidin 3-glucoside equivalents per gram of extract (mg C 3 GE/g E).

DPPH Radical Scavenging Assay
The DPPH assay was based on hydrogen atom transfer (HAT) reactions [30,31]. 2,2-Diphenyl-1picrylhydrazyl or DPPH will generate a stable free radical with an unpaired electron that is delocalized over the entire molecule. Then, the changing of the color from yellow to purple was detected by the UV spectrophotometer. Briefly, various concentrations of the rice sprout extracts dissolved in methanol (between 5 to 1000 µg/mL) and the DPPH reagents were added and mixed in the ratio of 1:1 in the 96 well plates for 30 min in the dark at room temperature. The loss of DPPH radical's absorbance at 517 nm was measured by using a microplate spectrophotometer (EnSight ™). Gallic acid, a standard antioxidant was used as a positive control (y = 21.216x + 2.5088, R 2 = 0.9822). The linear curve was obtained from the plot between gallic acid concentration and DPPH radical scavenging power. The experiments were repeated in five replicates. The DPPH radical scavenging capacity is represented as the percentage of DPPH radical inhibition at 50% (IC 50 ). The percentage of inhibition or %DPPH scavenging effect was calculated following Equation (2). The FRAP assay describes ferric reducing antioxidant power, the ability of an antioxidant to reduce Fe 3+ -TPTZ complex to Fe 2+ -TPTZ [27,32]. Various concentrations of each test compound (e.g., rice sprout extracts or standard antioxidant gallic acid was mixed with 1 mL DMSO. The FRAP reagent comprised of 300 mM acetate buffer (pH 3.6), 10 mM 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ) solution and 20 mM FeCl 3 ·6H 2 O in a 10:1:1 ratio. The reaction mixture was pipetted into each well of a 96-well plate and incubated for 30 min in the dark at room temperature. The absorbance of the colored product (ferrous tripyridyltriazine complex) was measured at 593 nm with the microplate spectrophotometer (EnSight TM ) against a blank with deionized water. The experiments were repeated in five replicates. The FRAP value was calculated from the standard curve of FeSO 4 (y = 0.5065 − 0.3073, R 2 = 0.9782) and expressed in mmol Fe 2+ per gram of extract (mmole Fe 2+ /g extract). Gallic acid was used as a positive control (y = 0.3821x + 0.0653, R 2 = 0.998).

Cytotoxicity Assay by Neutral Red Assay
Neutral red (NR) can diffuse and accumulates intracellularly in the lysosomes of the living cells [32,33]. Cancer cell viability was, therefore, measured using neutral red (NR) assay. Briefly, cells at a density of 5 × 10 5 cells/mL were seeded with various concentrations of rice sprout extracts for 24 and 48 h. NR solution was added to each well (with a final concentration of 50 µg/mL) and incubated at 37 • C for 2 h. Then, cells were washed twice with 0.01 M PBS and the supernatant was then discarded. Cells were lyzed by 0.33% HCl in 100 µL isopropanol. The absorbance of NR dye was detected by the microplate spectrophotometer (EnSight™) at 537 nm with a 650 nm as a reference wavelength. The percentage of cell viability was determined between the absorbance of the treated cells at each concentration of the rice sprout extracts against the absorbance of treated cells with DMSO at the concentration used in the extract concentration. The experiments were repeated in five replicates. The results were expressed as the concentration of the test compound that caused a 50% loss of the cell viability (IC 50 ) when compared to the untreated cells or control. The selectivity index (SI) was also calculated from the ratio of the IC 50 of normal cells versus the IC 50 of cancer cells [34]. DCFH-DA to 2',7'-dichlorofluorescein (DCF), and its fluorescent intensity can be measured by flow cytometry [32]. Briefly, DCFH-DA was added to the extract-treated Jurkat cells for 12 h. H 2 O 2 was used as a positive control. After the treatment of 12 h of the positive control and samples, all the cells were collected, washed in 0.01 M PBS, and centrifuged at 700× g for 5 min. The supernatant was discarded, and cells were re-suspended in 0.01 M PBS before staining the cells in the dark with 10 µM DCFH-DA in serum-free medium for 30 min. Once the incubation period was completed, cells were analyzed by flow cytometry (BD FACSCanto II, BD Biosciences, San Jose, CA, USA) within 1 h of staining period. Intracellular ROS was expressed as DCF fluorescence intensity and calculated by BD FACSDiva software. The data were calculated in comparison to the untreated cells and represented as a ratio compared with the untreated cells.

Determination of the Total Endogenous Glutathione Level
The total endogenous glutathione (GSH) level was analyzed colorimetrically using a glutathione assay kit (Abnova, Taipei, Taiwan). Glutathione (Sisco Research Laboratories Pvt. Ltd., Mumbai, India) with the final concentrations of 125, 250, 500, 750, 1000, 1500 ng/mL was used as a standard in this assay to generate the standard curve to determine the intracellular GSH level (y = 0.0019x + 0.3842, R 2 = 0.9942). The Jurkat cells were treated with BGR for 12 h. Afterward, cells were lysed with glutathione buffer and incubated on ice for 10 min. Then, the cells were centrifuged at 8000× g for 10 min and the supernatant was collected for the glutathione assay. The reaction mixture of NADPH generation mixture (lyophilized), glutathione reductase, and glutathione reaction buffer was prepared and added to each well and incubated at room temperature. Sample or standard was added following by the substrate (5,5'-dithio-bis-[2-nitrobenzoic acid], DTNB). The total amount of glutathione was detected by the microplate spectrophotometer at a wavelength of 415 nm. The percentage of glutathione levels of treated cells was calculated in comparison to the untreated cells.

Apoptotic Cell Death Mode by Flow Cytometry
When cells undergo apoptosis, the membrane phospholipid phosphatidylserine (PS) translocates to the outer membranes and binds with annexin V [32]. Briefly, Jurkat cells at 1 × 10 6 cells/mL were seeded in the plates. After being treated with BGR extract for 24 h, Jurkat cells were washed with 0.01 M PBS centrifuged at 700× g for 5 min. Then, cells were resuspended in annexin V binding buffer and stained with 5 µL annexin V-FITC for 15 min in the dark at room temperature. Docetaxel (100 µM), an anticancer drug, was used as a standard apoptosis inducer. Once the incubation period was completed, cells were analyzed by flow cytometry (BD FACSCanto II, BD Biosciences, San Jose, CA, USA) within 1 h after staining. The cells binding with annexin V were classified as apoptotic cells, and the results were displayed as a percentage to the total cell population.

Caspase 3/7 Activity Assay
Caspase 3/7 assay was done according to manufacturer's instruction in white polystyrene 96 well plates (Promega). The proluminescent caspase-3/7 substrate is cleaved to release aminoluciferin, which will be catalyzed by luciferase to produce the luminescence light. Briefly, 3 × 10 4 cells/mL of Jurkat cells were seeded in the plates and were treated with BGR extract for 24 h. Docetaxel (10 µM) was used as a positive control. After completing the treatment, caspase Glo ® -3/7 reagents were added into the cells and equilibrated for 1 h at room temperature. Luminescence mode was measured by using a microplate spectrophotometer. The percentage of caspase 3/7 activity was calculated in comparison to the untreated cells.

Statistical Analysis
Experimental results are presented as mean ± standard deviation (SD). Statistical analyses between groups were performed by using one-way ANOVA followed by Tukey multiple comparison tests. A significant difference was set at p < 0.05. Regression and partial correlation analysis were performed using SPSS 25 (SPSS Inc, Chicago, IL, USA).

Determination of Total Anthocyanin Content
In general, BGR contained significantly higher total anthocyanin content (TAC) than RD6 (Figure 1b). TAC in BGR was positively correlated to the increasing age. BGR at 20-25 days exhibited the highest TAC content of 0.86 ± 0.096 mg C3GE/g extract followed by BGR at 10-15 days according to the pH differential methods. Results expressed as mean ± SD with four replicates for TPC and three replicates for TAC. The letters indicate statistically significant (p < 0.05) among samples.

Determination of Ferric Reducing Antioxidant Power
FRAP values indicated that the BGR possessed the greater ferric reducing power than RD6. The reducing power of BGR at each age increases in a concentration-dependent manner. BGR at 10-15 days and 20-25 days showed higher reducing ability (2.22 ± 0.014 and 1.58 ± 0.0302 mmoles of FRAP power per weight in gram unit of extracts at 1000 µg/mL, respectively) (Table 3) than other harvest ages within the same and different cultivars. Table 3. Ferric reducing power of rice sprout extracts. Data are expressed as mean ± SD of absorbance ferric-to-ferrous reduction capacity of BGR and RD6 from five replicates (p < 0.05).

Determination of Cell Viability in Various Cancer and Noncancer Cell Lines
The cell viability assay of rice sprout extracts at 24 and 48 h was performed using a neutral red assay in various cell lines including HepG2, HCT116, SK-MEL-2, and Jurkat, which have never been previously reported. Cisplatin, a standard anticancer drug, was used as a positive control [35]. Results showed that at 24 h cisplatin showed strong to low antiproliferation on HCTT116, SK-MEL-2, Jurkat, HepG2, and Vero, respectively (Table 4). The cytotoxicity percentage were presented as mean ± SD (n = 5). Inactive is indicated as > 50% cytotoxicity when treated with the samples at the maximum concentration (1000 µg/mL). Different lower-case letters indicate a significant difference of test samples in a row (different cell lines, same treatment) and different capital letters indicate a significant difference in column (same cell line, different treatments) (p < 0.05).
At 24 h, Jurkat was the most sensitive cell line responding to BGR at all ages and RD6 at 10-15 days as indicated by the lowest IC 50 value. BGR at age 5-7, 10-15 and 20-25 days decreased cell viability of Jurkat with the respective IC 50 values of 206.2 ± 15.1, 261.9 ± 6.6, 438.6 ± 11.6 µg/mL when compared to control (p < 0.0001). RD6 at 10-15 days reduced the cell viability of Jurkat cells (IC 50 value of 310.2 ± 28.0 µg/mL) while the other ages of RD6 were inactive to Jurkat cells. HCT116 was the second cell line upon which BGR exerted a low reduction of cell viability. BGR and RD6 extracts were inactive in HepG2, SK-MEL-2, and Vero cells at 24 h (Table 4).
At 48 h, Jurkat was the most sensitive cell line responding to BGR following by HCT116 and HepG2. BGR showed greater inhibition of cell viability at 48 h than those of 24 h. BGR at 5-7, 10-15 and 20-25 days inhibited cell viability of Jurkat cells with the respective IC 50 values of 67.62 ± 6.8, 166.7 ± 13.2, 136.4 ± 4.9 µg/mL compared to the control (p < 0.05). HCT116 and HepG2 displayed low sensitivity to BGR (p < 0.05). Both BGR and RD6 were inactive in SK-MEL-2 and Vero at 48 h (Table 4). Notably, BGR and RD6 were inactive to Vero cells at both 24 and 48 h.
The selectivity index (SI) indicates selective inhibition of cancer cell viability of the tested compound in comparison to the noncancerous cells [34,36]. The SI values of BGR extracts on Jurkat cells were increased in a time-dependent manner. Moreover, HepG2 cells become more sensitive to some BGR extract with increasing SI values at 48 h when compared to 24 h. RD6 at only 10-15 days highly selective inhibited cell viability of Jurkat cells at 24 h (SI = 3.2) ( Table 4) and this extract did not show higher SI value at the longer time point (48 h) in Jurkat cells but gain slight selectivity (SI = 1.5) in SK-MEL-2 cells at 48 h.

Determination of Intracellular Hydrogen Peroxide Level in Jurkat Cells 1
Antioxidant activity of BGR in the Jurkat cells was determined whether it was contributed to its 2 anticancer activity. The intracellular ROS level was presented as a bar graph in Figure 2a.

Determination of Endogenous Glutathione Level in Jurkat Cells 19
The BGR at 100 and 200 μg/mL significantly increased the endogenous glutathione level (p < 20 0.05) compared to the untreated group (100.00 ± 5.72%) (Figure 2b) Apoptosis is the programmed cell death that is a hallmark of the anticancer action of the tested 28 compound [32,37]. BGR at two different concentrations according to the cell viability assay (0.5 × IC50 29

Caspase 3/7 Activity Assay
To confirm the apoptosis-inducing effect of BGR, the caspase 3/7 activity was determined as the enzymes are activated during apoptosis. BGR at both 100 and 200 µg/mL at 5-7, 10-15 and 20-15 days induced apoptosis at 24 h that indicated by the significantly increased activity of caspase 3/7 in Jurkat cells in comparison with the untreated cells or the control (100.00 ± 9.26%) (p < 0.05) (Figure 3b). The standard anticancer drug docetaxel at 10 µM was used in this study as it is clinically used in leukemia [38][39][40]. Docetaxel strongly increased the caspase 3/7 activity at 281.56 ± 7.26% compared to the control. A higher caspase 3/7 activity was found when Jurkat cells were treated with 200 µg/mL BGR than those with 100 µg/mL.

Discussion
Sprouts of black (cv. BGR) and white (cv. RD6) glutinous rice with the different harvest ages (5-7, 10-15, and 20-25 days or young, middle and old age) showed different degree of antioxidant capacity. The in vitro antioxidant capacity of BGR is higher than RD6 based on TPC, TAC, DPPH, and FRAP assays. The hydroxyl group of phenolic compounds and anthocyanin has been reportedly responsible for antioxidant activity [41][42][43][44]. DPPH yields a stable free radical of unpaired electron delocalized over the entire molecule. The scavenging activity of BGR and RD6 resulted in a decrease in DPPH radical absorption proportional to the concentration of radicals that were scavenged. The lowest IC 50 from DPPH assay of middle and old age of BGR showed the greatest scavenging activity among harvest ages and cultivars (p < 0.05). Moreover, the single-electron transfer mechanism based on the ferric ion reducing capability of BGR and RD6 was evident. Interestingly, BGR at middle and old age exhibited greater ferric reducing antioxidant power (FRAP) than RD6.
A previous study reported on the antioxidant activity of the rice grass juices of various rice cultivars based on DPPH and FRAP assays [27]. The colored grass juice can prevent oxidative damage of supercoiled pBR322 DNA in a dose-dependent DNA protective effect. Their results showed that the colored rice (cv. Kum) possessed higher TPC than the white rice cultivar (cv. RD6). Moreover, TAC was also high in colored rice grass juices. The colored grass juice of Kum Doisaket (cv. C-KDS) contained TAC higher than the other colored rice. These results are in agreement with our study regarding the higher TPC, DPPH scavenging activity, and ferric reducing antioxidant power of colored rice (BGR) than that of white-colored rice (RD6 The excessive production of ROS in normal cells leads to the initiation of DNA damage and carcinogenesis [1,5,46,47]. Many plants have been reported to possess the antioxidant properties to decrease the ROS level in cancer cells [48][49][50]. Low ROS level plays an important part in cellular signaling, cell proliferation, and apoptosis [1,51]. Moreover, cells have mechanisms to repair damaged DNA. However, if the damaged DNA cannot be repaired, cells will undergo apoptotic cell death. DNA damage-induced apoptosis is a cellular defense mechanism to eliminate genetically damaged cells [52,53]. Apoptosis induction is, therefore, an endpoint for screening anticancer action [54,55].
Our study found that BGR induced apoptosis in Jurkat cells at 24 h. During apoptosis, DNA is condensed and fragmented, and there is an externalization of phosphatidylserine to the outer side of the cell membrane of apoptotic cells. Later, phagocyte molecules involve in recognition and remove unrepairable apoptotic cells without triggering inflammation [55]. The flipping of phosphatidylserine can specifically bind with annexin V that can be detected by flow cytometry to indicate the early stage of apoptosis [55]. The cytotoxic effect of BGR occurred via apoptosis as the Jurkat treated cells were stained with annexin V. Moreover, caspase 3/7, that is the downstream effector caspases in the caspase cascade pathway [56], was increased in the BGR-treated Jurkat cells. Our results indicated that BGR induced an early stage of apoptosis detected by cells stained with annexin v measuring by flow cytometry and induced Jurkat cells to die via apoptotic death mode as observed from the increasing of caspase 3/7 activity. A previous study also reported that the fermented mixture of brown rice and rice bran induced apoptosis in human acute lymphoblastic leukemia cells [57], which is in agreement with our study. Several studies reported the contribution of plant phenolic content to antioxidant, cytotoxicity, and apoptosis in cancer cell lines [58,59] as well as anthocyanin, one of the flavonoid groups, for its cancer prevention activity [60,61]. Hence, the phenolic content and anthocyanin content of BGR may contribute to its antioxidant, cytotoxicity, and apoptosis in the Jurkat cells.
It should be noted that rice sprout is a young green leaf where the second metabolites or active compounds are synthesized after photosynthesis. The difference in the types and contents of compounds in the same plant found between studies could be affected by several factors such as different harvest time, age, parts, growing condition, and the type of solvent used for the extraction. The previous study extracted the rice sprout (20 g) with 1% hydrochloric acid in methanol (100 mL) twice each by sonication for 30 min [24]. Our study extracted the rice sprout (20 g) with 1% hydrochloric acid in methanol 1000 mL and sonicated for 30 min.
A previous study of BGR and RD6 at the same harvest ages reported the existence of important phytochemicals, including total chlorophyll content (TCC), total flavonoid content (TFC), total phenolic content (TPC), total anthocyanin content (TAC), and the proximate composition (i.e., ash, moisture, fat, protein, carbohydrate, energy, and dietary fiber content) [24]. Moreover, protocatechuic acid, vanillic acid, and rutin were quantified in a different amount in BGR and RD6 [24]. The BGR and RD6 rice sprout used in the previous and the present study were at the same harvest age and grown under the same condition but extracted with a slightly different method. Moreover, the extract from BGR and RD6 could comprise of other phytoconstituents besides these identified compounds. The content of protocatechuic acid, vanillic acid, and rutin was not re-analyzed in the present study. Instead, the marker compounds-TPC and TAC-were quantified in the extract, as they reportedly attribute to antioxidation, antiproliferation, and apoptosis induction effects in other plants of other studies.
In the previous study [24], BGR at 10-15 days contained the highest TCC (40.8 ± 0.8 mg/g extract), and BGR at 5-7 days contained the highest TAC (22.13 ± 0.002 mg cyanidin 3-glucoside (C 3 GE)/g extract), whereas RD6 at 5-7 days comprised the highest TFC and TPC (34.2 ± 3.4 mg quercetin equivalent (QE)/g extract, and 44.8 ± 1.6 mg gallic acid equivalent (GAE)/g extract, respectively). Moreover, the proximate compositions with 100 g fresh weight of the rice sprouts from both BGR and RD6 had a total ash 2.86 to 5.78 g, moisture 64.72 to 79.11 g, fat 0.32 to 1.2 g, protein 3.38 to 5.99 g, carbohydrate 12.22 to 24.54 g, and energy 73.44 to 124.0 Kcal. The dietary fiber of BGR at 10-15 days was 1.4 times higher than that of RD6 (16.92, and 11.98 and g/100 g fresh weight, respectively). The higher TPC was observed in the BGR group at 10-15 days in the previous study (31.9 ± 1.1 mg GAE/g extract) than our result of BGR at the same age (29.72 ± 1.42 mg GAE/g extract). In the previous study, the highest TAC was found in BGR at 5-7 days (22.13 ± 0.002 mg C 3 GE/g extract). The highest TAC in our present study was found in BGR at 20-25 days only at 0.86 ± 0.096 mg C 3 GE/g extract. The TAC in the present study was less than the presvious reported result, which might be due to the different extraction methods.
Interestingly, the total bioactive compounds (TAC and TPC) were well correlated to the increasing of BGR ages (r > 0.6 with p < 0.05), not RD6 (r < 0.6 with p > 0.05) (Supplementary Materials, Figure S1 and Figure S2). In contrast, there is no correlation between BGR ages and FRAP (r = 0.314 with p > 0.05) (Supplementary Materials, Figure S1), whereas RD6 age was correlated to only DPPH (r = 0.606 with p < 0.05) (Supplementary Materials, Figure S2). TAC displayed no correlation with both DPPH and FRAP but TPC from both BGR and RD6 demonstrated well correlation to FRAP (r = 0.891 and = 0986, respectively with p < 0.05). The result implied that TPC was the major bioactive constituents that contribute to the direct antioxidative activity of BGR and RD6 in vitro. Moreover, BGR displayed the time-dependent cytotoxic activity to the sensitive cell lines (Jurkat and HCT116), whereas RD6 possessed less potency to the treated cell lines, except SK-MEL-2, with no time-dependent cytotoxicity. The correlation data indicated that Jurkat cell viability increased as BGR ages increased from young to old age at 24h (r = 0.952 with p < 0.05) and at 48h (r = 0.586 with p < 0.05), while HCT116 cell viability decreased with ages (r = −0.951 with p < 0.05) and at 48h (r = −0.609 with p < 0.05) (Supplementary Materials, Figure S3).
Since Jurkat was the most sensitive cell line to BGR, the further correlation analysis between BGR ages and intracellular biological activities (ROS reduction, increase in GSH, caspase activity, and apoptosis induction) had been performed. BGR decreased intracellular ROS (r = -0.943 with p < 0.05) and increased intracellular GSH (r = 0.922 with p < 0.05), with increasing ages. The good correlation between the reduction of intracellular ROS and an increase in GSH was observed (r = -0.891 with p < 0.05). Caspase 3/7 activity (r = 0.782 with p < 0.005) and the percentage of apoptotic cells (r = 0.824 with p < 0.05) were increased with the increasing ages (Supplementary Materials, Figure S1). The antioxidant capacity in the cell-based assays (from DCFH-DA and GSH assays) was directly correlated with TPC (r > 0.6; p < 0.05), while cytotoxicity was inversely correlated with TPC (r < -0.6; p < 0.05). Results implied that TPC predominantly contributed to antioxidant activities in vitro and indirectly enhanced intracellular GSH, resulting in a lower intracellular ROS. In addition, TPC could trigger cancer cells to undergo apoptosis as indicated by the increase in caspase 3/7 activity.
In the present study, the phenolic and anthocyanin compounds play a pivotal role for a higher antioxidant capacity as well as apoptosis induction in the Jurkat cells via the activation of the caspase family and other key cellular components of the colored rice (BGR) than those of the white-colored rice (RD6). To compare, BGR exerted greater cytotoxic effect, antioxidant effects, and induce cell death via apoptosis than RD6. The colored rice BRG could be the preferable sprouts for developing as a functional food in the future.

Conclusions
Our findings suggested that BGR extract could probably induce the apoptotic death of human acute lymphoblastic leukemia cells, mainly through the death receptor-mediated pathway and through antioxidative effects, proposing the possibility that BGR was a potential functional food and might be beneficial to patients with hematological cancer. The obtained information illustrated the easy-material-accessibility, economic value, environmental benefits from an edible rice sprout. Further study is required to confirm the clinical effectiveness and safe consumption.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-3417/10/20/7051/s1, Figure S1: Correlation coefficient (r) of BGR group at different harvest ages with ntioxidant capacity (TPC, TAC, DPPH, FRAP, DCHF-DA and GSH), and anticancer effect (cytotoxicity, apoptosis, and caspase activity). If the r value is close to +1 or -1, it will indicate a strong positively relationship or inverse relationship, respectively. If r is near to 0, it will indicate weak or no relationship. * Correlation is significant at p < 0.05, Figure S2: Correlation coefficient (r) of antioxidant capacity (TPC, TAC, DPPH, and FRAP) of RD6 at different harvest ages. If the value is close to +1 or -1, it will indicate a strong positively relationship or inverse relationship, respectively. If r is near to 0, it will indicate weak or no relationship. *Correlation is significant at p < 0.05, Figure S3: Correlation coefficient of cytotoxicity between Jurkat and HCT116 after BGR treatment (A) at 24 h and (B) at 48 h. If the value is close to +1 or -1, it will indicate a strong positively relationship or inverse relationship, respectively. If r is near to 0, it will indicate weak or no relationship. *Correlation is significant at p < 0.05.