Spent Material Extractives from Hemp Hydrodistillation as an Underexplored Source of Antimicrobial Cannabinoids

Hemp (Cannabis sativa L.) has been used for millennia as a rich source of food and fibers, whereas hemp flowers have only recently gained an increased market interest due to the presence of cannabinoids and volatile terpenes. Currently, the hemp flower processing industry predominantly focuses on either cannabinoid or terpene extraction. In an attempt to maximize the valorization of hemp flowers, the current study aimed to evaluate the phytochemical composition and antimicrobial properties of several extracts obtained from post-distillation by-products (e.g., spent material, residual distillation water) in comparison to the essential oil and total extract obtained from unprocessed hemp flowers. A terpene analysis of the essential oil revealed 14 monoterpenes and 35 sesquiterpenes. The cannabinoid profiling of extracts showed seven acidic precursors and 14 neutral derivatives, with cannabidiol (CBD) reaching the highest concentration (up to 16 wt.%) in the spent material extract. The antimicrobial assessment of hemp EO, cannabinoid-containing extracts, and single compounds (i.e., CBD, cannabigerol, cannabinol, and cannabichromene) against a panel of 20 microbial strains demonstrated significant inhibitory activities against Gram-positive bacteria, Helicobacter pylori, and Trichophyton species. In conclusion, this work suggests promising opportunities to use cannabinoid-rich materials from hemp flower processing in functional foods, cosmetics, and pharmaceuticals with antimicrobial properties.


Introduction
Hemp (Cannabis sativa L., Cannabaceae) has served for over 6000 years as a versatile resource for food, fibers, oils, medicines, and recreational or religious activities [1].Throughout the Middle Ages, hemp became a vital fiber crop for the production of textiles and ropes [2].However, following the discovery of the central nervous system (CNS)intoxicating effects of ∆ 9 -tetrahydrocannabinol (∆ 9 -THC), an important constituent in its flowers, hemp cultivation has decreased over the 20th century [3].Nevertheless, at the beginning of the 2000s, controlled genotypes (containing less than 0.2-0.3%∆ 9 -THC) were re-authorized to be marketed for agricultural purposes in the European Union [2].Consequently, more than 60 cultivars have been registered since [4].Thanks to the wide application of its derivatives, hemp cultivation has ramped up in the last few decades; for

Results and Discussion
In this study, the essential oil from hemp flowers obtained by hydrodistillation (HEO) was analyzed by gas chromatography coupled with mass spectrometry (GC-MS).In parallel, the residual water used for hydrodistillation was concentrated (HWE), whereas the processed hemp flowers (spent material after hydrodistillation) were extracted with hexane (HSE).For comparison purposes, the unprocessed hemp flowers were also extracted under the same conditions (HTE) (Table 1).The fatty acid composition of the solvent extracts was analyzed by GC-MS, whereas the cannabinoid profile was assessed qualitatively by liquid chromatography hyphenated with high-resolution tandem mass spectrometry (LC-HRMS/MS) and quantitatively by liquid chromatography coupled with diode array detection (LC-DAD).Subsequently, the antimicrobial activity of hemp essential oil, cannabinoid-containing extracts, and individual cannabinoids (i.e., CBD, CBG, CBN, and CBC) was tested against a panel of 20 microbial strains.Data are reported as mean ± SD of three experiments; different letters within columns indicate significant differences (p < 0.05).

Cannabinoid Profile of Hemp Extracts
According to the extraction yields (Table 1), the processed plant material gave a yield of extraction higher than the unprocessed one.This could be related to the fact that the water used for the hydrodistillation extracted many polar components from the initial hemp flowers.Therefore, the same dried mass of processed hemp flowers contains more hydrophobic components (easily extracted with hexane) than the same mass of the initial hemp flowers [17].The cannabinoid profile of HWE, HSE, and HTE was assessed by LC-HRMS/MS.Due to the lack of standards, the identity of the cannabinoids was tentatively annotated according to previously developed methodologies [29][30][31][32].For instance, the discrimination between CBD-, THC-, and CBC-type acidic cannabinoids was already presented by Piccolella et al. [32] for 20 different compounds identified in hemp pollen samples from Italy.Borille et al. [30] proposed the annotation of around 70 different cannabinoids from 68 Brazilian samples of Cannabis spp.(leaves, stems, and flowers).In addition, Berman et al. [29] performed the comprehensive metabolic profiling of 36 samples of Cannabis plants from Israel, identifying 94 cannabinoids belonging to 10 distinct subclasses.Barhdadi et al. [31] identified and quantified 17 cannabinoids in 20 CBD-based e-cigarette liquids from the Belgium market.

Cannabinoid Composition of Hemp Extracts
The quantification of cannabinoids (also referred to as 'cannabinoid potency') is an important step in assessing the quality of a hemp flower-derived product.According to the results presented in Table 4, CBD was the major cannabinoid in all extracts (up to 15.93 ± 0.02 wt.%).Compared to HTE, HSE displayed CBD levels around two times higher, while the CBD concentration in the residual water extract HWE was ten times lower.This tendency could be correlated to the lipophilic character of the cannabinoids, conferring them a low extractability in water and high extractability in organic solvents.Interestingly, the acidic precursor CBDA was found in significant amounts in the unprocessed flower extracts HTE but in very low levels in the spent material extracts (Table 4).This could indicate a thermal conversion of CBDA to CBD during hydrodistillation due to exposing CBDA to a high temperature (~100 • C) for a long time (3 h).The remaining cannabinoids (e.g., CBDV, CBG, CBC, CBN, etc.) were found in concentrations between 0.01 to 0.55 wt.% (Table 4).

Fatty Acid Composition of Hemp Extracts
Even though hemp flower extracts are known to contain lipids, their fatty acid composition is usually neglected.In all samples (Table 5), linoleic acid (C18:2) and linolenic acid (C18:3) were found in the highest amounts (up to 1.75 ± 0.07 wt.%).Significant concentrations of palmitic (C16:0) and arachidic acid (C20:1) were also noticed.Concerning its high polarity, the residual water extract HWE contained 15-50 times lower quantities of fatty acids than the solvent extracts.Generally, HSE displayed slightly higher values of fatty acids than HTE.This can be related to the fact that, during hydrodistillation, lipids are neither entrained with the volatile terpenes nor transferred into the residual water.To our knowledge, there are no previous data concerning the fatty acid composition of hemp flower extracts.However, there are numerous similarities with the composition of hemp seeds.Kriese et al. [35] showed that linoleic acid (C18:2) and linolenic acid (C18:3) are the predominant fatty acids in the oils extracted from the seeds of 51 C. sativa genotypes.Their levels varied from 15.00-19.89wt.% for linoleic acid and 5.15-8.24wt.% for linolenic acid.Furthermore, numerous studies confirmed that linoleic acid (C18:2) is the major constituent of hempseed oils (up to 60% of the total fatty acids) [36][37][38][39].

Antimicrobial Activity of Hemp Essential Oil and Extracts
All obtained extracts were tested against a panel of human pathogens comprising nine Gram-positive bacteria, six Gram-negative bacteria, and five fungi (Table 6).Except for H. pylori (minimum inhibitory concentration, MIC ≤ 62.5 mg/L), the samples were inactive against the other Gram-negative bacteria.Nevertheless, most Gram-positive strains were sensitive to the treatment with the hemp extracts.The inactivity of hemp against most Gram-negative bacteria has already been proven by other authors [34].According to the criteria proposed by Kuete and Efferth [40], a sample is considered to have a significant antimicrobial activity if the MIC value is below 100 mg/L.Thus, HEO displayed the most potent antibacterial activity against S. aureus (MIC = 62.5 mg/L) and M. luteus (MIC = 15.6 mg/L); for the other strains, the MIC values were ≥125 mg/L.In general, the previous literature data also showed a modest activity of hemp EO.For instance, the MIC values ranged from 1200-4700 mg/L for a panel of bacteria comprising S. aureus, B. subtilis, M. luteus, E. coli, P. aeruginosa, and K. pneumoniae [41].Similar values (MIC = 8000 mg/L) were also reported for hemp EO against different strains of S. aureus [42].HTE showed the highest activity against S. aureus (MIC = 0.98 mg/L).For S. aureus MRSA, S. epidermidis, M. luteus, and B. cereus, the MIC values displayed by HTE were identical (3.9 mg/L).Candida species were not significantly inhibited; however, a strong anti-Trichophyton activity (MIC = 31.3mg/L) was noticed for HTE.Similarly, HSE acted as a potent inhibitor against the growth of S. aureus (MIC = 0.98 mg/L), M. luteus (MIC = 1.95 mg/L), E. faecalis (MIC = 7.8 mg/L), and B. cereus (MIC = 1.95 mg/L), and T. mentagrophytes (MIC = 31.3mg/L).The antimicrobial potential of HWE was better than HEO but considerably lower than that exhibited by HTE and HSE (Table 6).This behavior could be linked to the low cannabinoid concentration noticed in HWE compared to HTE and HSE (Table 4).
Our findings are in line with those observed by other authors.For instance, various hemp flower extracts displayed MIC values between 10 and 66 mg/L against E. coli, P. aeruginosa, B. subtilis, and S. aureus [43].Muscara et al. [34] reported MIC values of 39 mg/L against S. aureus.On the other hand, Serventi et al. [4] documented for various hemp extracts a potent inhibitory activity against B. subtilis (MIC = 1.5-25 mg/L), S. aureus (MIC = 12.5-100 mg/L), P. aeruginosa (MIC = 25-100 mg/L), and E. coli (MIC = 6.2-12.5 mg/L); however, the inhibition of B. cereus and S. typhy was negligible (MIC > 200 mg/L).Furthermore, the same group of authors also presented the activity of hemp extracts against several dermatophyte strains, with MIC values ranging from 25 to 100 mg/L against T. mentagrophytes, T. tonsurans, and T. rubrum [4].
In an attempt to link the antimicrobial activity of hemp flower extracts to the presence of cannabinoids, the effects of four previously isolated cannabinoids (i.e., CBD, CBG, CBN, and CBC) [44] were subsequently evaluated against the same panel of microbial strains (Table 7).All tested cannabinoids were inactive against Gram-negative bacteria except for H. pylori (MIC ≤ 0.98 mg/L for CBG, CBD, and CBN).Furthermore, the four compounds showed no significant inhibitory activity against Candida spp.and Trichophyton spp.However, the MIC values against the Gram-positive strains indicated promising antibacterial activity, in particular against S. aureus, methicillin-resistant S. aureus (MRSA), S. epidermidis, M. luteus, E. faecalis, and B. cereus (MIC between 0.49 and 15.6 mg/mL).Overall, the following decreasing order of the inhibitory activity against Gram-positive bacteria could be proposed: CBD > CBN > CBG > CBC.The evidence of antimicrobial activity for CBD has already been proven.For instance, MIC values of 1-4 mg/L were noticed against a diverse array of Gram-positive bacteria, including MRSA, MDR S. pneumoniae, E. faecalis, and anaerobic Clostridium difficile and Cutibacterium acnes [45].CBD also showed different levels of antibacterial activity against Grampositive bacteria, including susceptible and MDR strains (MIC = 2-4 mg/L for E. faecium, Enterococcus spp., Staphylococcus spp., M. luteus, and Rhodococcus equi.Furthermore, CBD displayed MIC values of 4-8 mg/L against MRSA, E. faecalis, and L. monocytogenes [46]. The antimicrobial activity of other minor cannabinoids has been scarcely investigated.For instance, Appendino et al. [5] showed that CBC, ∆ 9 -THC, CBN, and CBG could exert antibacterial activity against various MDR S. aureus strains, with MIC values between 0.5 and 2 mg/L.In addition, the same cannabinoids also inhibited MRSA, Streptococcus mutans, Streptococcus sanguis, Streptococcus sobrinus, and Streptococcus salivarius with MIC below 5 mg/L [47].Furthermore, S. aureus, S. epidermidis, and S. pyogenes were also shown to be impacted by CBG treatment (MIC = 10-75 µM) [48,49].Nevertheless, the inhibitory effects of CBG, CBN, and CBC against other human pathogens are reported herein for the first time.
Concerning the possible mechanisms of antimicrobial activity of cannabinoids, it was previously shown that CBD, in combination with bacitracin, caused defects in cell division and irregularities in the cell envelope [46].The treatment with CBG led to intracellular accumulation of membrane structures, induced membrane hyperpolarization, and decreased membrane fluidity of various bacterial strains [48].The antibacterial activity of cannabinoids against MRSA was shown to be mediated through the inhibition of biofilm formation, the eradication of pre-formed biofilms and stationary phase cells persistent to antibiotics [47].Furthermore, CBD proved a potent inhibitor of membrane vesicle release from E. coli, altering cell communication [50].Nevertheless, the antimicrobial mechanisms of CBD and minor cannabinoids remain elusive, requiring subsequent investigations.

Preparation of Essential Oil and Solvent Extracts
The powdered hemp flowers (50 g) were subjected to hydrodistillation on a Clevengertype apparatus with 500 mL water for 3 h.At the end of the hydrodistillation process, the amount of essential oil (HEO) was measured using the apparatus scale (in mL), collected, and dried over anhydrous sulfate.The water in the flask was filtered, and 250 mL were freeze-dried to afford the residual water extract (HWE).The solid plant material residue (spent hemp flowers) was dried in an oven at 40 • C for 48 h, and 7.5 g were extracted with hexane (75 mL) in an ultrasound bath for 30 min, for 3 repeated cycles, each time with the same volume of fresh solvent.After filtration, the solvent was evaporated under reduced pressure, yielding the spent extract (HSE).For comparison purposes, the unprocessed powdered hemp flowers were extracted with the same solvent under the same conditions, affording the total (unprocessed material) extract (HTE).For HEO, the yield was calculated with the following formula: %yield = volume of the obtained oil(mL) mass of the hemp flowers(g) × 100 For the spent extract (HSE), the yield was determined as follows: %yield = mass of the obtained extract after drying(g) mass of the spent hemp flowers(g) × 100 For the total extract (HTE), the yield was calculated with the formula: %yield = mass of the obtained extract after drying(g) mass of the hemp flowers(g) × 100 For HWE, the yield was determined using the formula: where m 1 is the mass of the obtained water extract after freeze-drying, m 2 is the mass of the hemp flowers, V 1 is the volume of the water introduced in the hydrodistillation, and V 2 is the volume of the freeze-dried water.All extractions were performed in triplicate, with the extraction yields provided in Table 1.

Terpene Profile of Hemp Essential Oil (GC-MS)
The terpene profile of HEO was assessed by GC-MS performed on a TRACE GC Ultra instrument (Thermo Fischer, Waltham, MA, USA).The chromatographic separations were conducted on a Zebron ZB-5MS (30 m × 0.25 mm, 0.25 µm) column (Phenomenex, Torrance, CA, USA), with helium as the carrier gas at a flow rate of 1.43 mL/min.The injection temperature was 250 • C; 1 µL was injected with a split ratio of 50:1.The column temperature was initially held at 60 • C for 4 min, then increased to 280 • C at a rate of 10 • C/min, and maintained at 280 • C for 5 min.The MS parameters included: transfer line temperature, 320 • C; source temperature, 230 • C; ionization energy, 70 eV.The linear retention indices (LRI) were determined for all compounds in the chromatograms using a standard mixture of alkanes ranging from 8 to 20 carbon atoms.Peak identification was achieved by referencing the NIST 11 Mass Spectra Library and comparing the calculated LRI with those found in the relevant literature.All analyses were performed in triplicate.
The detection was carried out in positive electrospray ionization mode, with the spectra recorded in the m/z 100-1000 Da range.The ion source parameters were as follows: drying gas (nitrogen) flow rate, 12 L/min; heated capillary temperature, 300 • C; nebulizer pressure, 35 psi; sheath gas temperature, 275 • C; sheath gas flow rate, 12 L/min; capillary voltage, 4000 V; nozzle voltage, 2000 V; fragmentor, 110 V; skimmer, 65 V; octupole radiofrequency peak voltage, 750 V.The MS/MS spectra were generated by automated fragmentation at a fixed collision energy of 30 V.
The cannabinoid concentration of solvent extracts (HTE, HSE, and HWE) was assessed by LC-DAD on a Shimadzu HPLC (Tokyo, Japan) containing a binary pump (LC-20AD), autosampler (SIL-20A), degasser (DGU-20A), and UV/VIS detector (SPD-M20A).The chromatographic separations were conducted on a Zorbax XDB-C18 (150 mm × 4.6 mm, 3.5 µm) column (Agilent Technologies, Palo Alto, CA, USA) at 30 • C. The mobile phase comprised (A) water and (B) acetonitrile, both containing 0.1% formic acid.The phases were delivered at a flow rate of 1.5 mL/min in the following gradient: 72% B for 4 min, 75-80% B from 4.01 to 11 min, and 90% B from 11.01 to 12 min.The sample injection volume was 10 µL, with the chromatograms recorded at 228 nm.The concentrations of cannabinoids were assessed using the calibration curves of the corresponding standards.All analyses were performed in triplicate.

Fatty Acid Composition of Hemp Extracts (GC-MS)
The fatty acid composition of solvent extracts (HTE, HSE, and HWE) was assessed by GC-MS on an Agilent 6890 GC (Agilent Technologies, PaloAlto, CA, USA) coupled to a mass selective detector (MSD).The chromatographic separations were conducted on a Zebron Rtx-Wax (30 m × 0.25 mm, 0.25 µm) column (Restek, Centre County, PA, USA), with helium as the carrier gas at a flow rate of 1 mL/min.The injection temperature was 200 • C; 1 µL was injected with a split ratio of 7.5:1.The column temperature was initially held at 80 • C for 2 min, then increased to 180 • C at a rate of 7 • C/min, maintained at 180 • C for 10 min; then ramped up to 230 • C at a rate of 1 • C/min, and finally held at 230 • C for 10 min.The MS parameters included: transfer line temperature, 230 • C; source temperature, 230 • C; ionization energy, 70 eV.Before analyses, the samples (~20 mg) were dissolved in 1 mL hexane.Then, 0.2 mL of the obtained solutions were incubated at 90 • C for 1 h in the presence of 6 mL of methanol/hydrochloric acid (11:1, v/v) and 0.1 mL tridecanoic acid (1 mg/mL in hexane) as internal standard.Next, 1.7 mL hexane and 2 mL water were added; after shaking and phase separation, 1 mL of the upper phase was analyzed by GC-MS.The fatty acid content in the samples was expressed based on a calibration curve performed with the FAME mixture.All analyses were performed in triplicate.

Antimicrobial Assays
The antimicrobial assays were performed using the microdilution method, according to the European Committee on Antimicrobial Susceptibility Testing [51,52] ).Serial double dilutions of samples were prepared in MH broth or RPMI 1640 medium 2% glucose buffered with 0.165 M MOPS and supplemented with chloramphenicol 50 mg/L and cycloheximide 300 mg/L, for non-fastidious bacteria and fungi, respectively.The sterile 96-well flat-bottom polystyrene microtitrate plates (Nunc, Denmark) were prepared by dispensing 100 µL of the appropriate dilution of the tested extracts in broth medium per well by serial two-fold dilutions, to obtain the final concentrations of the tested extracts ranging from 2000 to 0.25 mg/L for bacteria and yeasts.The inocula were prepared with fresh microbial cultures in sterile 0.85% NaCl to match the turbidity of 0.5 McFarland standard, and were added to wells to obtain the final density of 5 × 10 5 colony forming units (CFU)/mL for bacteria, 5 × 10 4 CFU/mL for yeasts and 5 × 10 5 CFU/mL for dermatophytes.After incubation (non-fastidious bacteria and yeasts-35 • C for 24 h and dermatophytes-28 • C for 5 days), the growth of microorganisms was measured spectrophotometrically at 600 nm (BioTEK ELx808, BioTek Instruments, Inc., Winooski, VT, USA).The minimum inhibitory concentration (MIC) for H. pylori ATCC 43504 was determined using a two-fold microdilution method in MH broth with 7% of lysed horse blood at an extract concentration ranging from 2000 to 0.25 mg/L with a H. pylori suspension of 3 McFarland standard diluted 100 times (9 × 10 6 CFU/mL).After incubation at 35 • C for 72 h under microaerophilic conditions (5% O 2 , 15% CO 2 , and 80% N 2 ), the growth of H. pylori was visualized with the addition of 10 µL of 0.04% resazurin.The MIC endpoint was recorded after 4 h incubation as the lowest concentration of extract that completely inhibits bacterial growth.An appropriate DMSO control (at a final concentration of 10%), a positive control (containing inoculum without the tested extracts), and a negative control (containing the tested extracts without inoculum) were included on each microplate.The MIC was determined and reported for each sample and strain.Vancomycin, ciprofloxacin, ofloxacin, and nystatin/terbinafine were used as the standard reference drugs.All experiments were performed in triplicate.

Statistical Analysis
Data are provided as mean ± standard deviation of three repeated experiments; ANOVA with Tukey's post hoc test was conducted; p < 0.05 was considered statistically significant.

Conclusions
The seminal findings provided by the phytochemical analysis are as follows: (i) 15 monoterpenes and 36 sesquiterpenes were identified in the EO, with sesquiterpenes accounting for ~85% of total peak area; (ii) a total of 7 acidic cannabinoids and 14 neutral derivatives were annotated in the post-distillation by-products, with CBD as the dominant compound (up to 16 wt.%);(iii) linoleic and linolenic acid were the representative fatty acids in the solvent extracts (up to 2 wt.%); (iv) the spent extracts displayed cannabinoid levels around 2-3 times higher than the unprocessed flower solvent extracts.Concerning the biological study, the hemp EO and extracts demonstrated potent antimicrobial activity (MIC < 62.5 mg/L) against Gram-positive bacteria (e.g., S. aureus, S. epidermidis, M. luteus, E. faecalis, B. cereus, S. pneumoniae), H. pylori, and Trichophyton spp.In addition, when CBD, CBG, CBN, and CBC were individually tested against the same panel of microorganisms, MIC values ranging from 0.49 and 15.6 mg/mL against Gram-positive bacteria were retrieved.The inhibitory activity generally decreased in the following order: CBD > CBN > CBG > CBC.For some cannabinoids, the antimicrobial properties against certain microbial strains were proven for the first time in the current study.
Considering that the hemp flower essential oil industry generates significant amounts of unused biomass rich in cannabinoids, the strategy implemented in the current work could afford high-added-value by-products within the hemp production chain, contributing to the principles of the circular economy and sustainability.Altogether, this work can open promising avenues for utilizing cannabinoid-rich materials obtained during hemp flower processing in functional foods or cosmeceutical and pharmaceutical products with antimicrobial properties.

Figure 1 .
Figure 1.Structures of main terpenes identified in the hemp essential oil.

Figure 1 .
Figure 1.Structures of main terpenes identified in the hemp essential oil.

Figure 2 .
Figure 2. Chemical structures of cannabinoids identified in the hemp extracts.

Figure 2 .
Figure 2. Chemical structures of cannabinoids identified in the hemp extracts.

Table 1 .
Extraction yields of hemp essential oil and extracts.

Table 2 .
Terpene profile of hemp essential oil.

Table 3 .
Cannabinoid profile of hemp extracts.

Table 3 .
Cannabinoid profile of hemp extracts.

Table 4 .
Cannabinoid concentration of the hemp flower extracts.
Data are expressed as mean ± SD of three repeated analyses; sample codes as in Table1; different letters within rows indicate significant differences (p < 0.05).

Table 5 .
Quantification of fatty acid derivatives in the hemp flower extracts.Data are expressed as mean ± SD of three repeated analyses; sample codes as in Table1; different letters within rows indicate significant differences (p < 0.05).

Table 6 .
Antimicrobial activity of hemp essential oil and extracts.

Table 7 .
Antimicrobial activity of selected cannabinoids.