Seeds were obtained from a local pepper grower in Şanlıurfa, Turkey. The seeds were dried on trays refrained from sunlight at room temperature for 7 days and stored at 4 ºC until used in the experiments. The chemicals and standards were obtained either from Sigma or Merck (Darmstadt, Germany) unless otherwise stated.

PSO extraction from İsot pepper seeds was carried out using a laboratory scale cold pressing machine (12 kg seed/h capacity, single head, 750 W power). Centrifugation (1420 g, 10 °C) and filtration were applied to obtain pure oil. The resulting supernatant was collected and stored in amber-colored vials at -20 ºC after flushing with nitrogen until analysis. 10 g of oil were mixed with 2 mL of methanol (X2) in a vial to extract PSO phenolics. After vortexing, the mixture was centrifuged at 1420 g for 30 min and then the resulting supernatant was used for total phenolic content, total flavonoid content, antioxidant activity, antimicrobial activity, enzyme inhibition activity, NMR, and LC-ESI-MS/MS analyses.

Physicochemical analyses (peroxide value, iodine number, free fatty acid (FFA), color (L*, a* and b*), saponification, refractive index, acid value) (AOAC, 1984), total phenolic content (Chen and Young-Hwa, 2013), total flavonoid content (Zhishen et al., 1999), four different antioxidant activity methods: DPPH and ABTS (Çam et al., 2009), CUPRAC (Apak et al., 2007), FRAP (Benzie and Strain, 1996) assays and enzyme (α-glucosidase and α-amylase) (McDougall et al., 2005) inhibition activity were determined accordingly.

PSO antimicrobial activity was tested against two reference bacterial strains, one yeast and one mold including Staphylococcus aureus subsp. aureus (ATCC 6538P), Escherichia coli (ATCC 8739), Aspergillus brasiliensis (ATCC 16404) and Candida albicans (ATCC 29212). The agar-well diffusion method was employed for the evaluation of PSO antimicrobial activity. 9-mm holes were punched into the solidified agar and these holes were filled with 120 μl of PSO extract at 200 mg/mL concentration. The microbial count was set at 0.5 Mcfarland standard equivalent to give a concentration of 1*10⁷ bacteria, 1*10⁵ yeast, 1*10⁵ mold per milliliter. The bacteria, yeast and mold cultures were grown in the plates and incubated at 37 °C for 24 h, 25 °C for 24 h, and 25 °C for 5 days, respectively. For the determination of inhibitory concentration (MIC) and minimum bactericidal concentration (MBC), the PSO extracts were diluted in the range of 6.25 mg/mL to 200 mg/mL.

The fatty acid composition of PSO was analyzed by a gas chromatography (GC-FID, Shimadzu, GCMSQP2010, JAPAN) with a fused-silica capillary column (30 m 0.25 mm) coated with 0.25 mm dimethyl polysiloxane (RTX®-5MS, Restek). The column temperature was programmed from 40 °C (held for 2 min) to 240 °C at 3 °C/min increases. Injector, interface and detector temperatures were 250, 250 and 220 °C, respectively. Helium was used as carrier gas (1.0 mL/ min.) at a split ratio of 1:20.

PSO extracts were filtered through a 0.45 μm pore size membrane filter and then injected into Nexera Shimadzu UHPLC LC–ESI-MS/MS (Shimadzu, Japan) equipped with a binary gradient pump, an autosampler (SIL-20AC), a degasser (DGU-20A3R) and a column thermostat (CTO-10ASVP). The column was an Intersil ODS-4 analytical column with 3 mm i.d. x 100 mm and 2μm particle size. Experiments to elute phenolic compounds were conducted based on the following conditions: 0.3 mL min-1 flow rate, 40 °C column temperature and 2 μL injection volume, using a mixture of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in methanol) as mobile phase; gradient conditions: from 0 to 4 min 5% B, from 4 to 7 min 95% B, from 7 to 7.01 min 95% B and at 7.01 min back to the initial conditions of 5% B. The concentrations of phenolic compounds are expressed in μg/kg units.

NMR experiments were conducted on an Agilent Instrument NMR Spectrometer operating at 400 MHz. The PSO was dissolved in chloroform and placed into 5 mm NMR tubes, and injected into the NMR spectrometer to obtain ¹H NMR spectra. The acquisition parameters were: spectral width, 6410.3 Hz; spin: 20; number of scans 32; relaxion delay 5s; pulse angle 90°; observe pulse 8.50 μs for ¹H NMR spectra.

For the thermal parameters, a Shimadzu DSC-60 differential scanning calorimeter (DSC) (Shimadzu, Kyoto, Japan) was used. The DSC instrument was calibrated with indium (heat of fusion 28.45 J/g). Temperature calibration was carried out using hexane (m.p 93.5 °C), water (m.p. 0.0 °C) and indium (m.p. 156.6 °C). Samples were weighed into aluminum pans at 10–15 mg and covers were hermetically sealed into place. An empty pan was used for reference. Samples were heated to 60 °C under nitrogen flow (0.5 mL/min), held for 10 min at constant temperature and cooled to -60 °C, and then heated again to 60 °C. The scanning rate was 2 °C/min. The machine equipment program (TA-60WS) was used to plot and analyze the thermal data (Calligaris et al., 2004).

The PSO extracted from three different seeds of the same origin was used for each analysis and all analyses within each PSO were carried out at least in duplicate. Origin Pro 8 (Origin Lab Inc.) and Lab Solutions software (Shimadzu) was used to obtain DSC graphs. The differences among data were compared with independent samples according to the t-test and One-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test at the confidence level of 95% (p ≤ 0.05) using SPSS 22.0.0 statistical package for Windows (SPSS Inc., Chicago, IL, USA).

The physicochemical characteristics of İsot seed oil are given in Table 1. All three color parameters for PSO were higher compared to capia pepper seed oil (Yılmaz et al., 2015) and its color was a deeper reddish and yellowish than that extracted from seed of capia. There is no internationally accepted PSO standard defined in the codex alimentarius. Therefore, we compared the obtained data regarding PSO physicochemical properties with the virgin olive oil standards published in codex alimentarius. The FFA content of PSO (0.22%) was below 0.8%, which was required for an olive oil to be included in the extra virgin category according to the International Olive Oil Standards. The FFA of pepper seed oil for California Wonder (C. annuum) and Aji Colorado were 0.71 and 4.74% (Jarret et al., 2013), respectively, both higher compared to what we measured for İsot PSO. The peroxide value of PSO was (6.35 meq O₂/kg), below 20 meq O₂/kg, which meets the criteria for high quality olive oils. Saponification value was slightly lower and Iodide value was slightly higher for PSO when compared to standards, as determined by 181.45 and 94.26, respectively (Table 1).

Capsaicin, dihydrocapsaicin, nordihydrocapsaicin, homodihydrocapsaicin, and homocapsaicin are members of the capsaicinoid family. Capsaicin and dihydrocapsaicin represent 80-90% of the total capsaicinoids present in the capsicum genus (Figueiredo et al., 2013). The ratio of capsaicin/dihydrocapsaicin ranges from 1:1 to 2:1 (Hayman and Kam, 2008). The mean concentrations of capsaicin and dihydrocapsaicin in red chili pepper are 309.3 ppm and 238.2 ppm and 138.5 ppm and 146.4 ppm in green chili pepper (Othman et al., 2011). The capsaicin and dihydrocapsaicin contents of PSO were 762.92 ppm and 725.73 ppm (Table 2), a higher concentration compared to those of red chili pepper and lower than hot chilies. Capsaicin (8-methyl-N-vanillyl-6-nonenamide) is a unique alkoloid found primarily in pepper fruits with analgesic effects, which give them their hot taste and spicy flavor. Capsaicin exerts selective effects on the peripheral part of the sensory nervous system and relieves pain by depleting the neurotransmitter of painful impulses known as substance P from the sensory nerve terminals. The main source of capsaicinoids in nature is Capsicum fruits. The capsaicin content of peppers is closely linked to their commercial quality and pharmaceutical properties. Tocopherols are considered lipophilic antioxidants, which belong to the Vitamin E group of natural compounds. Tocopherols are found in distinct plant tissues such as fruits, flowers and seeds. Peppers contain tocopherols in amounts comparable to those in spinach and asparagus; while higher than tomatoes (Bosland and Votava, 2000). The PSO evaluated in this study included both α (62.40 ± 0.85 ppm) and δ tocopherols (643.23 ± 1.40 ppm) (Table 2). A group of carotenoids including violaxanthin, capsanthin, crytocapsin, zeaxanthin, β-apo-8’carotenal, cryptoxanthin, and β-carotene was detected in PSO. The unique keto carotenoids capsanthin and cryptocapsin provide the brilliant red color; while β-carotene, zeaxanthin, and violaxanthin impart the yellow-orange color. β-apo-8’carotenal is an aldehydic carotenoid which is widely distributed in peppers and has an orange-red color. Besides their use as natural colorants carotenoids promote human health due to their nutritional characteristics and preventive-protective roles against degenerative diseases (Arimboor et al., 2014). Statistically significant differences (p < 0.05) were found among the carotenoid fractions of PSO and zeaxanthin was the principal carotenoid (29.51 ± 2.00 mg/kg) in PSO. The phenolic and flavonoid contents of the PSO samples are shown in Table 2. Total phenolic content of PSO was higher than that of total flavonoids (p < 0.05). The concentration of total phenolic compounds was 201.84 ± 2.03 (mg GAE/kg) and the flavonoid content was 164.80 ± 1.56 (mg QE/kg). Total phenolic content of İsot seed oil was higher compared to that of capia peppers (Yılmaz et al., 2015). The capsaicin, tocopherol, phenolic and flavonoid contents together with the remarkable antioxidant activity of PSO reveal its rich nutraceutical value and importance for human nutrition.

The fatty acid content of PSO is shown in Table 3. The extracted PSO was mainly composed of unsaturated fatty acids (UFA), approximately 85.3%, including mono-unsaturated fatty acids (oleic) and the poly-unsaturated fatty acid (linoleic). The saturated fatty acids (SFA; palmitic and stearic acids) were also present in PSO but at lower concentrations (14.7%). These results are in agreement with previously published reports (Yılmaz et al., 2015). The ratio of UFA to SFA was 5.8, which was higher than previously reported (Chougui et al., 2013). Linoleic acid was the dominating fatty acid (73.77%) followed by palmitic acid (10.86%) and oleic acid (10.51%). The percentage of linoleic acid (C18:2) found in İsot pepper seed was higher than that found in Capsicum galapagoense, similar to Capsicum flexuosum and lower than Capsicum tovarii (Jarret et al., 2013). The observed differences in the fatty acid composition and percentages of PSO could be variety-specific or due to environmental factors and agronomic practices. The essential fatty acids were named as f-vitamins and draw significant attention due to the health-promoting potential of polyunsaturated fatty acids. It has been reported that unsaturated fatty acids play a role in preventing cardiovascular diseases, diabetes, cancer, Alzheimer’s disease and the alleviation of certain chronic defects (Shahidi and Ambigaipalan, 2018).

There is limited study in the literature on the phenolic composition of Capsicum annuum L. seed oils. However, the phenolic compositions of peppers have been studied (Baenas et al., 2019). Phenolic compounds belong to phenolic acid derivatives and flavonoids represent major secondary metabolites in peppers (Jayaprakasha et al., 2012). Quercetin, luteolin, myricetin, kaempferol, and apigenin are present in pepper fruits (Asnin and Park, 2015; Baenas et al., 2019; Jayaprakasha et al., 2012). In addition to these compounds, phenolic acids such as caffeic, p-coumaric, sinapic, and ferulic acid are found in peppers (Materska and Perucka, 2005). In this study, the phenolic compounds in PSO were determined using a LC-ESI-MS/MS method. Fourteen phenolic compounds were detected in PSO and resveratrol (6269.10μg/kg) was the most abundant one, followed by luteolin (733.81μg/kg) and 4-hydroxycinnamic acid (517.90μg/kg) (Table 4). Resveratrol has curing effects on stress and disease in rodent models (Baur and Sinclair, 2006); luteolin significantly inhibits neuronal cell apoptosis, elevates cell viability, and attenuates diabetes-associated disorders (Liu et al., 2017); and hydroxycinnamic acid and their derivatives are important components of medicinal herbs which are useful for treatment of chronic diseases (El-Seedi et al., 2018). The synergistic activity of these compounds and vitamin C and vitamin E are responsible for the potent antioxidant capacity of peppers which is associated to the protection of the human body against arteriosclerosis, osteoporosis, diabetes, cancer, and coronary diseases (Pandey and Rizvi, 2009).

Gas chromatography equipped with a flame ionization detector (GC-FID) is one of the best known methods for detecting the presence of fatty acids in the samples or the fatty acid composition of vegetable oils. NMR is currently being used along with mass spectrometry for detecting a wide range of fatty acids and phytochemicals including phenolic compounds. Compared to other analytical analyses, ¹H-NMR is relatively fast and different chemical structures are accurately and simultaneously determined (Savorani et al., 2013). The ¹H-NMR spectrum obtained in this study represented olefin protons at 5.28 ppm, bis-allylic protons at 2.71 ppm, allylic protons at 1.98 ppm and terminal methyl group protons at 0.83 ppm (Figure 1A). The unsaturated nature of PSO was also confirmed by NMR analysis. Linoleic acid was the most abundant fatty acid present in PSO and accounted for almost 74% of the total fatty acids found in PSO. Linoleic acid carbons, including the protons attached to olefinic carbons (–CH=CH–; C9,10,12,13), bis-allylic protons (=CH–CH2–CH=; C11), the allylic/α-C protons (=CH–CH2-alkyl; C: 2,8,14), and alkyl protons (–CH2/–CH3; C: 3,4,5,6,7,15,16,17,18) were clearly visible on the ¹H-NMR spectrum (Figure 1A). The homonuclear 2D – ¹H NMR COSY experiment contained apparent cross-peaks between the –CH= protons at 5.2 ppm and neighboring –CH2– groups at 2.01 ppm. The other cross-peak of the –CH= protons at 5.2 ppm was at 2.74 ppm and arose from the –CH2– groups between two double bonds (Figure 1B). This peak is generally found in the samples containing polyunsaturated fats. The DSC heating and cooling curves from -40 °C to 40 °C of PSO are given in Figure 2. The volume narrowing and exothermic effect resulting from crystallization started at -11.51 °C (Ton) and finished at -14.14 °C (Toff) with ΔH of 0.92 j/g. Ton, Toff and ΔH of the endothermic peak represented melting and were -15.84 °C, -17.38 °C and 0.64 j/g, respectively. One endothermic peak and 1 exothermic peak were detected. Previous thermal behavior evaluated in sunflower oil indicated the presence of three endothermic peaks and one exothermic peak (Calligaris et al., 2004). The differences in the number of the peaks could be due to the complexity of the Triacylglycerol (TAG) distribution (Tan and Che Man, 2002) because TAG could be present in different forms such as polymorphic and crystalline structures (Desmedt et al., 1990). In addition, TAG with high saturated fatty acids melt at higher temperatures compare to highly unsaturated ones (Calligaris et al., 2004). According to Myat et al. (2009), crystallization and melting temperatures directly depend on SFA/PUFA and SFA/MUFA ratios.

The antimicrobial properties of PSO were assessed against Gram-negative, Gram-positive bacteria, yeast and fungi, including E. coli, S. aureus, C. albicans, A. brasiliensis. The Agar well diffusion assay was used to test the antimicrobial activity of PSO and the diameters of the inhibition zones obtained are shown in Table 5. The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of PSO on pathogenic microorganisms were also determined. The aqueous extract of PSO showed bactericidal activity only against S. aureus, which is a Gram-Positive bacterium. There is no study in the literature describing the antimicrobial activity of PSO. However, some studies report the antimicrobial activity of aqueous extracts of olive oil. The antimicrobial activity of aqueous extracts of olive oil was tested against Salmonella enteritidis, Listeria monocytogenes, S. aureus, E. coli O157:H7, Yersinia sp. 5057655, and S. sonnei JCP, and showed bactericidal activity against the food borne pathogens tested (Medina et al., 2007). The aqueous extract of virgin olive oil exerted strong antimicrobial activity. Neither S. aureus nor Yersinia sp. was detected after 5 min of contact with the aqueous extract of virgin olive oil, which represents as 5 log reduction in the count of pathogens (Medina et al., 2007). In another study, extra virgin olive oil (EVOO) showed strong bactericidal activity, although refined oils were found to be ineffective. The percent inhibition results of the phenolic compounds in EVOO were found at up to 13.13% for E. coli O157:H7, up to 27.68% for L. monocytogenes and up to 11.64% for S. enteritidis (Karaosmanoğlu, 2010). There are few studies regarding the antimicrobial activity of edible oils when compared to essential oils. The essential oils obtained from various plant tissues including leaves, flowers, stems, fruits and seeds displayed antimicrobial activity against various bacterial strains (Semeniuc et al., 2017). Much of the emphasis was devoted to the antimicrobial activity of essential oils rather than edible ones. It was generally shown that plant extracts are usually more potent inhibitors of Gram-positive bacteria than Gram-negative bacteria (Lin et al., 1999). Similar findings were also presented in this study. The aqueous extract of PSO exerted antimicrobial activity against a Gram-positive bacterium while no inhibitory activity was observed against Gram-negative E. coli. We were not able to detect any inhibitory activity against mold (A. brasiliensis) or yeast (C. albicans) cultures either. The inhibitory effects of PSO extracts on α-amylase and α-glucosidase enzymes are shown in Table 5. The extracts were found incapable on preventing α-amylase activity but exhibited inhibitory activity against α-glucosidase with an IC₅₀ value of 22.29±0.20 mg/mL. Acarbose was used as a positive control. The PSO extracts possessed the same inhibitory effect (p > 0.05) against α-glucosidase as determined for acarbose (20.28±0.91 mg/mL), indicating that it was capable inhibiting the active center (Rubilar et al., 2011). The enzyme activity of PSO could be explained by its bioactive compounds, including phenolic, carotenoid, capsaicinoid and tocopherol contents. Due to lack of studies on the enzymatic activities of oils, the results were compared to the results of other extracts, such as spinach (6.03 μg/mL) (Vyas, 2017). As expected, the plant extracts were more potent than the PSO aqueous extract in blocking the active center of the assayed enzymes. We tested the antioxidant activity of PSO by applying four different methods, including DPPH, ABTS, CUPRAC, and FRAP as a single method did not fully describe the antioxidant capacity of the samples (Table 5). DPPH-ABTS are radical scavenging and FRAP-CUPRAC are reducing power assays, and the results were expressed as μmol TEAC/g, as shown in Table 5. The radical scavenging ability of the assayed PSO was approximately 1.5 times higher in the DPPH assay (11.38±0.31 μmol TEAC/g) (p < 0.05) when compared to the ABTS assay (7.81 ± 0.01 μmol TEAC/g). The reducing power activities of PSO according to the CUPRAC and FRAP assays were also evaluated and varied significantly (p < 0.05), revealing 8.55 ± 0.30 and 3.89 ± 0.13 μmol TEAC/g, respectively.