Extraction of Active Compounds from Mixtures of Hemp (Cannabis sativa) with Plants of the Zingiberaceae Family

2.1. Antioxidant Activity and Total Phenolic Content

Boxplots of antioxidant activity and total phenolic content depending on the plant material, marked with separate colors, are shown in Figure 1A,B and Appendix A respectively, showing the distribution of data for extraction replicates. A five-digit summary is shown for each material, including the minimum and maximum values at the outer ends of the plot, the first and third quantiles at the two box edges, and the medians in between. Figure 1 was created with the following program sequence:

p1 < -ggplot (df, aes (x = Material, y = Antioxidants, fill = Material)) +

geom_boxplot ()+

labs (y = “Antioxidants [%]”,x = element_blank ())+

theme (legend.position = “none”)+

scale_fill_viridis (discrete = TRUE)

p2 < -ggplot (df, aes (x = Material, y = Total_Phenols, fill = Material)) +

geom_boxplot()+

labs (y = “Total phenolic content [mg GA/g extract]”, x = element_blank ())+

theme (legend.position = “none”)+

scale_fill_viridis (discrete = TRUE)

ggarrange (p1, p2,

labels = c (“A”, “B”),

nrow = 2)

Depending on the plant material, the antioxidant activity varied significantly, which was also confirmed with the Kruskal–Wallis statistical test (F(6) = 14.372; p < 0.026). Figure 1A shows the values of antioxidant activity depending on the plant material, while the corresponding data are also listed in the Table A1 in Appendix A. Ginger and turmeric had the highest median antioxidant activity of 57.69 (±17.98)% and 48.38 (±32.48)%, respectively, while hemp and cardamom had the lowest values of 6.64 (±2.13)% and 1.74 (±1.35)%, respectively. The antioxidant activity of the plant mixtures fell between these values for hemp and Zingiberaceae mixtures, except for cardamom, whose mixture with hemp had higher radical scavenging activity than other plant materials extracted separately. Hemp–cardamom mixtures had a median antioxidant activity of 9.23 (±3.12)%, but post hoc Dunn test did not confirm significant differences between the hemp–cardamom mixture and separate plant materials (p > 0.05). Nevertheless, ultrasonic-assisted extraction (UAE) showed higher values for antioxidant activity for the hemp–cardamom mixture than for the separate materials. Repetition of measurements would be necessary for statistical confirmation of the synergistic effect of these two plants separately. A synergistic effect was also observed with the supercritical extraction of the mixture of industrial hemp and turmeric, which achieved a higher percentage of inhibition than the individual materials. Comparison of extraction methods showed that UAE gave higher antioxidant yields than SCE for most materials, with the exception of ginger and cardamom, where the opposite was true. Figure 1 shows broader interquartile ranges for ginger, turmeric, and hemp–turmeric, which are attributed to the differences in extraction methods. The median antioxidant activity of turmeric was 71.51 (±6.08)% for UAE extraction, while SCE achieved only 6.56 (±0.49)%. Similarly, for the hemp–turmeric mixture, the UAE method was again more efficient, with a median antioxidant capacity of 54.23 (±4.56)%, while the median value of SCE was 8.23 (±0.32)%. In contrast, the median antioxidant activity of ginger was higher with SCE, with the highest median antioxidant activity of 79.17 (±4.67)%, while the UAE extraction had a median antioxidant activity of 43.2 (±3.02)%. Thus, the effectiveness of the extraction method for antioxidant activity is highly dependent on the plant material. Since more polar components are extracted by UAE and more nonpolar components are extracted by SCE, we can claim that many polar and nonpolar components are present in the analyzed samples.

In Figure 1B, total phenolic content in mg GA per g extract is shown for plant materials and their mixtures. The corresponding data are also listed in the Table A1 in Appendix A. The values follow a similar distribution to the antioxidant activity, while significant differences between natural materials and their mixtures were also confirmed with ANOVA (F(6) = 8.028; p < 0.001). Like the antioxidant activity, ginger had the highest average total phenolic content of 237.96 (±13.2) mg GA/g extract, followed by the mixture of hemp–turmeric with 191.88 (±43.07) mg GA/g extract, the mixture of hemp–ginger with 189.20 (±50.00) mg GA/g extract, and turmeric with 179.14 (±61.32) mg GA/g extract. The value of total phenolic content was lowest in cardamom, with 52.93 (±8.35) mg GA/g extract, while hemp showed a comparatively higher extraction efficiency of total phenolic components of 159.84 (±3.74) mg GA/g extract. Synergistic effects were obtained via UAE for industrial hemp–turmeric mixtures and industrial hemp–ginger mixtures, as the values of total phenolics obtained in the above extractions were higher than in the extracts of the individual materials. UAE gave the highest mean of total phenolic content for all hemp and Zingiberaceae mixtures. Smaller interquartile ranges can be seen on Figure 1B compared to Figure 1A, but again, turmeric and hemp–turmeric show the greatest differences in extraction methods. The UAE for turmeric had nearly twice the values of SCE, with a mean of 210.67 (±3.58) mg GA/g extract compared to 108.47 (±13.02) mg GA/g extract.

With natural extracts, variations in observed content of bioactive molecules is expected and a consequence of differences in material preservation, harvest period, climatic, geomorphological conditions, etc. [20,21,22]. Nevertheless, the measured results of antioxidant activity and total phenolic content concur well with the literature, and a broad spectrum of bioactive components with antioxidant properties is characteristic for these plants, as well as high values of phenolic components [9,10,11,12,13]. The cardamom total phenolics, which were the lowest among the materials measured, aligns well with results that Oueslati et al. obtained from ethanolic extracts, 51.7 mg GA/g extract [23], and are even higher than results from ethanolic extracts that Tarfaoui et al. obtained, 33.45 (±0.35) mg GA/g extract [12]. Furthermore, the values for ginger and turmeric, with the highest TPC difference between materials, are similar to those reported by Sepahpour et al. for turmeric ethanolic extracts, 172.1 (±1.4) mg GA/g extract [24], and by El-Sayed for ginger methanolic extract, 266.7 mg GA/g extract [25].

2.2. Cannabinoid Content

In this study, seven cannabinoids were measured, namely Cannabidiolic acid (CBDA), Cannabidiol (CBD), Cannabigerolic acid (CBGA), Δ9–tetrahydrocannabinol (THC), Δ9-tetrahydrocannabinolic acid (THCA), Cannabichromene (CBC), and Cannabinol (CBN), of which the most abundant were CBDA, followed by CBD, as shown in Figure 2 and Table A2 in the Appendix A. The extraction of bioactive components of industrial hemp in the case of CBDA had highest values for industrial hemp as a single material, since in this case the majority was extracted, i.e., 160.07 (±18.3) mg CBDA/g extract. The most effective mixture was obtained in the interaction with cardamom, with 147.9 (±21.8) mg/g extract. Both extraction methods had similar CBDA values.

What is interesting about CBD is that the hemp–cardamon mixture (109.26 ± 20.89 mg CBD/g extract) exceeded the value of industrial hemp alone (107.64 ± 9.06 mg CBD/g extract). Although the values between materials are not significantly different, a strong synergistic effect of ginger on CBD for all replicates can be observed. Furthermore, the mixture of industrial hemp with cardamom approaches the CBD value of industrial hemp. Such behavior is achieved only for SCE of the industrial hemp–ginger mixture, where more CBD components were extracted from the mixture than from industrial hemp itself. Only the mixture of hemp–turmeric did not show a positive effect in any case.

The third most abundant cannabinoid in the hemp extracts was CBGA. The two-way ANOVA confirmed the influence of the extraction method (F = 22.05, p > 0.001) on the CBGA content but not for material mixtures (F = 1.33, p = 0.31). The SCE extraction was the most efficient method for CBGA extraction, and for this method, all hemp mixtures achieved higher CBGA values than hemp alone, but this was not the case for UAE, where only the hemp–ginger mixture surpassed the sole material. The highest extraction efficiency of CBGA was achieved for SCE of the hemp–cardamom mixture, with the highest measurement of 14.13 mg CBGA/g extract, while hemp’s highest measurement was 12.76 mg CBGA/g extract.

For THC, the two-way ANOVA showed a significant influence of the extraction method (p < 0.001), but not of the material (p = 0.31). SCE was most efficient for THC extraction for all materials, while the values of THC for SCE of hemp–turmeric exceeded the values of hemp as a single component but not for UAE. The same pattern was observed for THCA, where extraction method significantly influenced the obtained values (p = 0.04); however, the combinations of materials used did not (p = 0.07). Contrary to THC, UAE was more effective for the extraction of THCA. UAE of hemp–ginger produced the highest content of THCA, with 3.93 (± 0.64) mg THCA/g extract, even higher that UAE extraction of hemp alone, with 3.72 (±0.35) mg/g extract, yet not significantly different.

CBC and CBN values were present in minor amounts, and for both components, SCE reached significantly higher values compared to UAE (p < 0.05). Hemp as a separate material had the highest component content, followed by the hemp–turmeric mixture. Nevertheless, the latter also had the biggest difference in extraction method efficiency, as UAE extraction produced lower quantities of CBC and CBD than any other mixture.

Additionally, the sum of all cannabinoids was calculated, where SCE of sole hemp proved to be the most effective method, reaching 303.43 mg/g extract. However, hemp–cardamom SCE reached almost the same levels, with 302.60 mg/g extract. It is important to state that plants of the Zingiberaceae family do not contain any cannabinoids, and as hemp mixtures contained 50% less hemp material than the individual material, the cannabinoid content without any synergistic effects should therefore be 50% lower than in the mixtures, which is clearly not the case. Instead, the cannabinoid content in mixtures is comparable and, in several cases, even exceeds the hemp values extracted separately. Comparing to the literature, optimized SCE methods reached up to 305.8 mg CBGA/g extract for cannabis inflorescences [26], which is roughly twice the levels achieved in our extraction of hemp inflorescences and leaves. Furthermore, extractions from cannabis residues were directly comparable to our results in CBDA levels, which were between 157.6 (±6.0) mg CBDA/g extract and 261.4 (±2.2) mg CBDA/g extract, with lower CBD values between 18.28 (±0.68) mg CBD/g extract and 64.18 ± 2.98 mg CBD/g extract [27]. Similar results were found for CBD in medicinal cannabis SCE plant material extracts, with values between 103.2 mg CBD/g extract and 191.7 mg CBD/g extract, while THC values were, in this case, considerably higher, reaching 761.1 mg THC/g extract [28].

2.3. Other Bioactive Components

The obtained results, shown in Figure 3A,B with detailed information in Table A3 in the Appendix A, show that the contents of both 6-gingerol and 6-shogaol are significantly higher in ginger alone than in its mixture with industrial hemp. The addition of industrial hemp does not lead to higher extraction efficiency. In the case of 6-gingerol, its value in the mixture is reduced by a factor of one compared to ginger alone, while in the case of 6-shogaol, this value increases to about three times. 6-gingerol was isolated to the greatest extent during SCE of ginger alone (137.38 ± 7.21 μg/mg extract), followed by SCE of hemp–ginger (69.98 ± 6.39 mg/g extract). A similar trend of results can also be observed with 6-shogaol, where SCE of ginger also proved to be the most effective (271.12 ± 59.93 mg/g extract), followed by UAE of sole ginger (172.25 ± 8.24 mg/g extract).

It is important to note that hemp as an individual material does not contain 6-shogaol and 6-gingerol; therefore, without any synergistic effects during extraction, half of the amounts of both components are expected, which is the case for 6-gingerol. For 6-shogaol, the values in material mixtures were even lower, indicating that a negative synergistic effect has occurred. For ginger SCE extracts, Kamaruddin et al. reported 6-gingerol values between 72.19 mg/g extract and 167.78 mg/g extract [29], while for UAE ethanolic extracts, lower values were reported by Ok and Jeong (25 mg/g extracts of 6-gingerol and 57 mg/g of 6-shogaol) [30].

Figure 3C shows 1,8-cineole, which is present in cardamom. The corresponding data are presented on Table A4 in Appendix A. Using UAE, it was extracted in the highest amounts, but hemp mixtures far exceeded the half values and, in case of SCE, far exceeded the individual material, which indicates a potent interaction between materials. However, this cannot be said for alpha-terpinyl acetate and linalool, where the mentioned components are extracted almost twice as much from cardamom as from the hemp mixture.

Considering the influence of the extraction method on the secretion of 1,8-cineole, the largest quantity of it was obtained via UAE of cardamom (9.74 ± 0.33 mg/g extract), and slightly less through SCE of hemp–cardamom (8.89 ± 0.86 mg/g extract), while SCE of cardamom turned out to be the least efficient method (2.65 ± 0.06 mg/g extract). These results are supported by the solubility of biologically active components, which in our case means that 1,8-cineole is a more polar component and is extracted via UAE. In the case of alpha- terpinyl acetate, shown in Figure 3D, as a non-polar component, the situation is exactly the opposite. Most of the mentioned component was extracted precisely during the SCE of cardamom (485.54 ± 71.82 mg/g extract). As linalool is highly soluble in ethanol, the UAE of pure cardamom was the most efficient (12.23 ± 0.78 mg/g extract), as shown in Figure 3E. The hemp mixture reached higher values than half for linalool (7.18 ± 0.4 mg/g extract). The SCE values were lower but comparable for individual material (10.44 ± 0.4 mg/g extract) and mixtures (6.58 ± 0.28 mg/g extract).

Paul and Bhattacharjee reported that SCE extract contained between 11.93 mg of 1,8-cineole per g dry seeds and 32.18 mg/g dry seeds [31], which is comparatively higher than the results reported in this study. Furthermore, Morsy reported 1,8-cineole values in essential oils after different treatments that were between 10.93 mg/mL and 13.31 mg/mL, as well as alpha-terpinyl acetate levels between 8.20 mg/mL and 14.50 mg/mL [32]. To our knowledge, no studies have been published reporting the mass of 1,8-cineole, alpha-terpinyl acetate, and linalool in cardamom per mass of extract.

Figure 3F shows the content of aR-turmerone in extracts, while the corresponding data are also listed in the Table A5 in Appendix A. The results indicate a higher efficiency of turmerone secretion in turmeric SCE extract (147.93 ± 8.15 mg/g extract). This value decreases to 50.73 ± 0.52 mg/g extract in the mixture of hemp–turmeric. Using UAE, the values are even lower. Like ginger, turmeric works best as a stand-alone ingredient. The extraction methods used showed the highest efficiency of turmerone extraction in SCE of turmeric (147.93 ± 8.15 mg/g extract). UAE followed with the second highest values (66.01 ± 2.09 mg/g extract), where turmeric was extracted without the addition of industrial hemp, and the mixtures had lower values for both extraction methods. Kongpol et al. report 2.80 (±0.29) wt%/dry basis aR-turmerone for turmeric UAE ethanolic extracts [33], while Carvalho et al. reported a yield of aR-turmerone of up to 1.14% per raw material for SCE turmeric extracts [34]. However, we have again found no data on the aR-turmerone mass per mass of extract.

3.1. Ultrasonic Assisted Extraction with Ethanol

Ultrasonic-assisted extraction (UAE) is a technique for isolating bioactive compounds from a plant, which has applications in the food and pharmaceutical industries, among others. The energy of ultrasonic waves in the solvent causes cavitation, accelerating dissolution, diffusion of solutes, and heat transfer, positively affecting extraction efficiency. Ultrasonic treatment of each material was carried out for 30 min at a room temperature of 25 °C. The obtained mixture was filtered, and the solvent was removed under a vacuum at 40 °C. The procedure for obtaining the extract was repeated three times, and the obtained extracts were stored at −20 °C until analysis.

3.2. Supercritical Fluid Extraction with Carbon Dioxide

Supercritical fluid extractions (SCE) of natural products were performed according to the procedure as described by Žitek et al. [3]. The system was preheated and then maintained at 50 °C. The autoclave was filled with 25 g of natural ground material or a 1:1 mixture of materials. During the extraction process, the supercritical carbon dioxide (solvent to feet ratio 8.3) was uniformly introduced into the autoclave at a pressure of 250 bar. The supercritical fluid flowed through the material and dissolved the soluble components, which were then collected in a pre-weighed test tube. The procedure for obtaining the extract was repeated three times, and the obtained extracts were stored at −20 °C until analysis.

3.3. Spectrophotometric Analyses

Spectrophotometric measurements were performed for examination of the extracts on their antioxidant activity and total phenolic content.

3.4. Determination of Antioxidant Activity

Extract antioxidant activity (AA) was determined using the 1,1-diphenyl-2-picrylhydrazyl (DPPH) reagent as described by Huang et al. [42]. Briefly, 10 mg of extract was dissolved in 10 mL of methanol. The solution was dissolved using an ultrasonic bath. A total of 3 mL of the DPPH solution and 77 μL of the extract solution were mixed and thermostated for 15 min at room temperature before the absorbance at 515 nm was measured.

3.5. Determination of Total Phenolic Content

The total phenolic content (TPC) was measured using the Folin–Ciocateu (FC) reagent, as described by Škerget et al. [43]. Briefly, 40 mg of extract was diluted in 20 mL of distilled water and added to 2.5 mL FC reagent diluted 10-fold with distilled water, and 2 mL Na₂CO₃ (75 g/L) was added. Distilled water was used as the control. Using gallic acid (GA), the calibration curve was prepared. The glass vials were thermostated at 50 °C for 5 min, cooled, and the absorbance was measured at 760 nm. The total phenolic content was expressed as mg GA per g of extract.

3.6. Chromatographic Analyses

Liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) and gas chromatography with mass spectrometry (GC/MS) were used to measure biologically active components in plant extracts. Representative chromatograms are shown in the Supplementary Materials.

3.7. Liquid Chromatography with Mass Spectrometry (LC-MS/MS)

The content of cannabinoids and ginger compounds was determined on Agilent 1200 HPLC apparatus coupled with an Agilent 6460 JetStream triple quadrupole mass detector (Agilent Technologies, Santa Clara, CA, USA).

For separation of cannabinoids, an Agilent Zorbax Eclipse Plus C18 (100 × 2.1 mm, 2.7 μm) analytical column at 35 °C was used using a mobile phase consisting of water containing 0.1% formic acid (A) and acetonitrile with the addition of 0.1% formic acid (B) with the following elution gradients: 0 min 50% B, 5 min 85% B, 7 min 85% B, 8 min 50% B till 10 min. The flow rate was set at 0.3 mL/min and the injection volume was 5 μL. The mass spectrometer was operated in negative ionization mode with parameters optimized for nitrogen as carrier gas at 300 °C and a flow rate of 6 L/min, the sheath gas temperature was 250 °C, the sheath gas flow was 11 L/min, and the capillary and nozzle voltages were at 3500 kV and 500 kV, respectively.

For analysis of 6-gingerol and 6-shogaol, a Phenomenex Kinetex C18 (150 × 2.1mm, 2.6 μm) analytical column was used. The mobile phase was composed of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). The following optimized mobile phase gradient program was used: 0 min 50% B, 3 min 90% B, 6 min 90% B, 7 min 50% B and maintained for 5 min. The flow of the mobile phase was 0.5 mL/min at 30 °C, and 5 µL of sample was injected onto column. Optimized mass spectrometer parameters were gas temperature of 150 °C, gas flow of 5 L/min, nebulizer voltage of 30 V, sheath gas temperature of 300 °C at a flow of 11 l/min, and the capillary and nozzle voltages were 3500 V and 1000 V, respectively.

The multiple reaction monitoring (MRM) mode was applied to detect and quantify ion transitions (Supplementary Materials: Table S1). The quantification of individual components was performed based on the calibration curve methodology.

3.8. Gas Chromatography with Mass Spectrometry (GC/MS)

The identification of volatile compounds in the extracts was performed on a Shimadzu GCMS-QP2010 Ultra GC-MS chromatograph (Shimadzu, Duisburg, Germany), using the NIST library, while quantification was performed using the external standard for each individual component. Volatile components from natural extracts were separated on a Phenomenex Zebron ZB5MS chromatographic column, with dimensions of 30 m × 0.25 mm, of 0.25-micron particles with a helium flow rate of 1 mL/min. Separation of aR-turmerone, 1,8-cineole, linalool, and alpha-terpinyl acetate was carried out under the following conditions: temperature gradient: 0–2 min 50 °C, raised to 60 °C at 2 °C/min, after 5 min at 60 °C raised at 10 °C/min to 150 °C, held for 2 min, and finally heated to 240 °C at a rate of 10 °C/min and held for 3 min; the detector and injector were set to a temperature of 250 °C, and the sample injection volume was 1 μL in SPLIT mode at the ratio of 1:10.

We thank Željko Knez for financial management of the project. We thank Simona Sedonja (University of Maribor) for her help in the extract preparation and spectrophotometric measurements.