Radish peel (Daikon/White Radish, Raphanus sativus var. Longipinnatus) waste was collected from MK restaurant in Nakhon Pathom province, Thailand, whereas Thai indigenous mango peel (Bao mango, Mangifera indica L. var.) waste was received from Promsap (Thailand) Company Limited, Hat Yai, Songkla, Thailand. Only peels from either mango and radish were collected and washed to remove any dust or foreign particles.

The peel was dried overnight in a hot air oven at 50°C until the moisture content was reduced to less than 10%. Afterward, each dried sample was ground to powder with the blender and sieved with a stainless-steel sieve of mesh size 35. Each sample was extracted with 80% ethanol solvent (5 g/250 mL) at 40°C for 2 h using a temperature-controlled hot plate magnetic stirrer (IKA C-MAG HS7, Don Mueng, Bangkok, Thailand). Then, the mixture was filtered using Whatman filter paper no. 1, and the filtrate was concentrated using an Eyela rotary evaporator (Eyela USA, NY, United States) at 40°C. The concentrated extract was collected and lyophilized.

The extraction yield was calculated by dividing the weight of the extract powder by the weight of the initial sample and multiplying by 100.

Initially, properly diluted extract (200 μL) was added with 10% Folin–Ciocalteu reagent (1 mL) and stood for 3 min at room temperature. Thereafter, 1 mL of a 2% sodium carbonate solution was added to the mixture and left to stand for 30 min under dark conditions. The absorbance of the reaction mixture was read at 760 nm using a spectrophotometer (Perkin Elmer LS 55, United States). For the blank, deionized water was used instead of the Folin–Ciocalteu reagent in the sample blank, and the extract was excluded and ethanol was used instead in the control. The concentration of total phenolic compound in the extracts was calculated from the standard curve of quercetin with a range of 1–10 μg/mL and expressed as μg QE/ g of extract powder.

Briefly, diluted extract (1 mL) was added to 1 mL of DPPH (0.15 mM) prepared in 95% ethanol and incubated for 30 min at room temperature in dark conditions (Nirmal and Benjakul, 2011). The absorbance of the mixture was read at 517 nm using the UV spectrophotometer. The sample blank was prepared using ethanol instead of a DPPH solution, while the control consisted of a DPPH solution and ethanol. The percentage inhibition of DPPH radical was calculated using the following equation and results were expressed as the IC₅₀ value (μg/ mL):

Percentage inhibition of DPPH radical = 1 : A sample / A control × 100

where A_sample and A_control are the absorbance of sample and control, respectively.

A diluted extract in phosphate buffer (500 μL, 0.2 M, pH 6.6) was mixed with 1% potassium ferric cyanide (500 μL), and the mixtures were incubated at 50°C for 20 min (Nirmal and Panichayupakaranant, 2015). Thereafter, 500 μL of 10% trichloroacetic acid was added and cooled at room temperature. The mixtures were centrifuged at 1000 × g for 10 min and the supernatant (500 μL) was added to the deionized water (500 μL) and 0.1% ferric chloride (100 μL) mixture. The preparation of each sample blank and the control were managed similarly. The phosphate buffer was used instead of potassium ferric cyanide in the sample blank and instead of extract in the control. The absorbance of the reaction was read at 700 nm. A higher absorbance of the sample indicated a higher reducing capacity.

The assay was conducted as mentioned by Nguyen et al. (2023). Briefly, 1 mL of each sample (0.25–4 mg/mL) was added to the 25 μL of 2 mM FeCl₂ · 4H₂O. Afterward, 5 mM ferrozine (100 μL) was mixed with the mixture and vortexed. After incubation at room temperature for 10 min, the absorbance of the sample was recorded at 562 nm. EDTA was used as the positive control in this assay. The percentage of the chelating ability of the sample was determined using the following equation and the results were expressed as the IC₅₀ value (mg/mL).

Chelating ability (%) = A 0 − A s A o × 100

where A_o = Abs of control sample (95% Ethanol + FeCl₂ · 4H₂O + Ferrozine)

A_s = Abs of a sample

The fatty acid emulsion was prepared using linoleic acid (0.28 g), Tween-40 (0.28 g), and 50 mL of phosphate buffer (0.2 M, pH 7.0) as mentioned by Jayaprakasha et al. (2001). Each diluted sample in 75% ethanol (0.5 mL) was mixed with emulsion (2.5 mL) and phosphate buffer (2.5 mL) and incubated at 37°C for 120 h. The peroxidation of fatty acid emulsion was examined using the thiocyanate method. Briefly, 0.1 mL of the emulsion mixture was taken every 24 h to mix with 5.0 mL of ethanol (75%), 0.1 mL of ammonium thiocyanate (30%), and 0.1 mL of FeCl₂ (20 mM in 3.5% HCl). The absorbance of the reaction mixture was read at 500 nm after standing at room temperature for 3 min. The control was prepared using 75% ethanol instead of the sample in the same manner.

Minimal inhibitory concentration (MIC) was determined using the broth microdilution assay as described by Htoo et al. (2019). Bacterial cultures Gram-positive (Staphylococcus aureus-ATCC 29213 and Bacillus subtills-PY 79), and Gram-negative (Escherichia coli-ATCC 25922, and Salmonella typhimurium-DMST 562) were grown on trypticase soy agar (TSA) plates. The freshly grown colonies were transferred to tryptic soy broth (TSB) until the absorbance of 0,2 was achieved at 600 nm. A 96-well plate was prepared with mango peel extract (MPE) or radish peel extract (RPE) with serial dilution. Subsequently, bacterial culture was diluted to 1:100 into each 96-well plate to obtain the final bacterial concentration of 10⁵ CFU/mL. The plates were incubated at 30°C for 24 h and were visually observed for bacterial growth. The MIC of the sample was noted where there was no bacterial growth.

The medium from each MIC well that showed no apparent growth was removed and subculture onto Mueller-Hinton agar (MHA) plates. To determine the minimal bactericidal concentration (MBC), the MHA plates were incubated at 30°C for 24 h to observe the growth of bacteria.

The sample with higher antioxidant and antimicrobial activities was subjected to high-performance liquid chromatography-electrospray ionization tandem mass spectrometry (HPLC-ESI-MS/MS) to examine the phytochemicals in the selected extract. The procedures mentioned by Sirichai et al. (2022) and Chupeerach et al. (2022) were followed. The standard phenolic compounds include 3,4-dihydroxybenzoic acid, genistein, caffeic acid, hesperidin, (−)-epigallocatechin gallate, 4-hydroxybenzoic acid, apigenin, kaempferol, cinnamic acid, chlorogenic acid, naringenin, p-coumaric acid, ferulic acid, mangiferin, sinapic acid, syringic acid, quercetin, myricetin, vanillic acid, gallic acid, rutin, and galangin and rosmarinic acid (Sigma-Aldrich, St. Louis, MO, United States) were used as the standards.

All experiments were performed in triplicates and the results are expressed as mean ± standard deviation. One-way analysis of variance (ANOVA) followed by Duncan’s tests was used to evaluate significant differences. All the statistical analysis was conducted using SPSS, and significant differences between treatments were considered significant at p < 0.05.

The percentage extraction yield of radish peel (39.12%) was higher compared to the mango peel extract powder (18.41%) (p < 0.05). Safdar et al. (2017) reported that mango peel extracted using 80% methanol achieved higher extraction yield, but higher TPC was noted in 80% ethanol solvent. On the other hand, Janjua and Shahid (2013) reported that radish peel (R. sativus) extracted using 80% ethanol achieved a higher extraction yield (5.6%) compared to the petroleum ether (5%), and methanol (4.2%) extract. The extraction yield of radish peel in this study was higher compared to the report of Janjua and Shahid (2013). It was noted that Janjua and Shahid (2013) extracted radish peel samples at room temperature, whereas this study used 40°C temperature for the extraction. Hence, the extraction yield of any plant material is affected by the type of solvent, temperature, time of incubation, and most importantly plant material structural integrity and presence of phytochemicals (Nirmal et al., 2023b).

TPC in the mango peel powder was significantly (p < 0.05) higher than that of the radish peel powder. The TPC in the MPE sample was 280.5 ± 7.95 μg quercetin/ g of extract powder, whereas 25.92 ± 0.32 μg quercetin/ g of extract powder was found in the RPE sample (Table 1). Earlier, Noor et al. (2015) reported that the Mangifera indica L. var. Perlis Sunshine mango peel contained a TPC value of 21.7 mg/g of peels. The mango peel is known to have a higher TPC and free radical scavenging activity compared to the mango pulp (Abbasi et al., 2015; Agatonovic-Kustrin et al., 2018). On the other hand, Noman et al. (2021) reported that the TPC content in white radish leaves (125.3 GAE/g) was higher than the root (95.8 GAE/g). Also, extraction methods have a significant effect on the TPC of fruits and vegetable wastes, as Martínez-Ramos et al. (2020) documented that the optimal extraction is dependent on the solvent employed. Ethanol-acetone blend yielded the extract exhibiting the highest activity compared to the extract obtained from conventional extraction of mango (Mangifera indica L.) peel. Therefore, MPE had higher TPC content than RPE.

Table 3 displays the antibacterial activities of MPE and RPE against Gram-positive and Gram-negative bacteria. The MIC and MBC values of MPE showed a stronger inhibitory capacity than those of RPE toward all tested bacteria (p < 0.05). In general, Gram-positive bacteria showed higher sensitivity to natural compounds than Gram-negative bacteria due to their different cell wall compositions (Nirmal et al., 2023a). Loan et al. (2021) examined that a crude extract of mango peel waste inhibited E. coli growth with a zone of inhibition ranging from 7.7 ± 0.3 to 15.6 ± 0.5 (mm), and a dose-dependent inhibition manner was observed. Previous studies stated that mangiferin possesses excellent antibacterial activity against a wide range of Gram-positive and Gram-negative bacteria (Thambi et al., 2016; Kanatt and Chawla, 2018; Loan et al., 2021). Stoilova et al. (2005) also documented that mangiferin demonstrated antibacterial activity against Staphylococcus, Bacillus, Escherichia coli, and certain Salmonella spp. Also, Kanatt and Chawla (2018) found that mango peel extracted with 70% ethanol showed the maximum zone of inhibition against Gram-positive and Gram-negative bacteria. Radish peel, on the other hand, had a lower potential to inhibit bacterial growth. As illustrated in Table 3, the MBC of RPE to inhibit different bacteria was 5 mg/mL or higher, which was 20-fold higher than mango peel (0.25 mg/mL). It was reported that the ethanolic extracts of R. raphanistrum subsp. sativus powder possesses antimicrobial activity against specific foodborne pathogens, such as Bacillus cereus ATCC 10876, Salmonella enteritidis 110, and Staphylococcus aureus ATCC 6538. However, no inhibitory activity was observed against Escherichia coli 23,716 (Lim et al., 2019) (see Figure 1).

The polyphenolic compounds in MPE, which showed higher antioxidant and antimicrobial potentials, were determined using LC-ESI-MS/MS. Three major and three minor compounds were identified in the mango peel extract as shown in Table 4. The major phenolic groups were xanthonoids, gallotannins, and flavonoids. Mangiferin, a xanthonoids, was the main phenolic, which was 82,806.97 μg/ g of mango peel extract powder, followed by 3,4-Dihydroxybenzoic acid, and gallic acid. Figure 2 represents the LC-ESI-MS/MS chromatogram of mango peel extract indicating three major peaks. Also, trace amounts of naringenin, catechin, and epicatechin gallate were identified. The identification and quantification of the phenolic profile of different Brazilian mango cultivars byproducts (barks, kernels, peel, and old and young leaves) was conducted by Barreto et al. (2008). The authors indicated that mangiferin, penta-O-galloyl-glucoside, gallic acid, and methyl gallate were the four major compounds identified in all samples. Among them, mangiferin was reported in higher concentration in all byproducts and showed higher antioxidant activities (DPPH radical scavenging, FRAP, and ORAC). Marcillo-Parra et al. (2021) investigated the presence of a phytochemical profile in MPE and reported that mangiferin ranged from 5.72 mg to 314 mg/100 g, gallic acid from 28.6 to 43.1 mg/100 g, and epicatechin from 9.01 to 9.24 mg/100 g of the dry weight of MPE. The variety, the region where the mango is grown, and the methods of extraction significantly affect the phenolic content of mango and its peel extracts (Coelho et al., 2019; Marcillo-Parra et al., 2021). The compounds identified in the present study have been recognized for their vital biological and health-beneficial properties, such as antioxidant, anti-inflammatory, anticancer capacity, antidiabetic, and antimicrobial properties (Ordoñez-Torres et al., 2021; Tacias-Pascacio et al., 2022). Therefore, the byproduct of mango, i.e., peel, has a wide range of applications in the food, pharmaceutical, and nutraceutical industries.

Considering that several antioxidant mechanisms contribute to the total antioxidant activity, it is unclear whether the TPC results are responsible for the actual antioxidant capacity. Previous studies revealed a significant relationship between the total phenolic components of peel extracts and their antioxidant activities. Table 5 displayed correlation coefficient matrices between TPC in the mango peel extract and their antioxidant activities. In the mango peel samples, significant coefficients of determination were calculated between mangiferin-TPC, mangiferin-DPPH, and mangiferin-lipid peroxidation. These findings supported earlier research by Abbasi et al. (2015) which showed that phenolic compounds from mango peel play a key role in antioxidant capacity. According to our research, mango peel is one of the promising sources of phenolic compounds, which are a key source of naturally occurring antioxidants.