Green Extraction Approach for Isolation of Bioactive Compounds in Wild Thyme (Thymus serpyllum L.) Herbal Dust—Chemical Profile, Antioxidant and Antimicrobial Activity and Comparison with Conventional Techniques

The aim of this study was to provide a chemical profile and determine the antioxidant and antimicrobial activity of the essential oil (EO) and lipid extracts of Thymus serpyllum L. herbal dust obtained via conventional (hydrodistillation (HD) and Soxhlet extraction (SOX)) and novel extraction techniques (supercritical fluid extraction (SFE)). In addition, a comparative analysis of the chemical profiles of the obtained EO and extracts was carried out, as well as the determination of antioxidant, antibacterial and antifungal activity of the lipid extracts. According to the aforementioned antioxidant and antimicrobial activities and the monoterpene yield and selectivity, SFE provided significant advantages compared to the traditional techniques. In addition, SFE extracts could be considered to have great potential in terms of their utilization in the pharmaceutical and cosmetic industries, as well as appropriate replacements for synthetic additives in the food industry.


Introduction
One of the most taxonomically complex genera of family Lamiaceae, which includes 250-350 species and varieties of wild growing species of herbaceous sub-and perennial shrubs, is genus Thymus [1].It is widely spread across Europe, North Africa and Asia and represents a great potential in terms of its medicinal use through the wide spectrum of its pharmacological properties.Thymus species, being considered important medicinal plants, have been used as healing agents in traditional medicine for thousands of years [2].Different species of Thymus possess different types of bioactive compounds that directly affect their chemical compositions and pharmacological activities.In general, the most common compounds identified are thymol and carvacrol, as well as flavonoids and phenolic compounds [3].One of the well-known medicinal plants and a member of the genus Thymus is Thymus vulgaris L., which contains more essential oil (EO) than other species and has a great potential as an antioxidant, antimicrobial, anti-inflammatory, antiviral and insecticidal agent [4].Beside aforementioned, Thymus serpyllum L. represents an aromatic and medicinal plant with a highly potent source of bioactive compounds with antioxidant, antimicrobial, antitumor and cytotoxic properties.Additionally, wild thyme has been used Plants 2024, 13, 897 2 of 30 for centuries in traditional medicine, where its most common application was to treat health problems related to respiratory and gastrointestinal systems [5].It may be used as a substitution for synthetic antioxidants to treat pathological conditions caused by free radicals but also as a substitution for synthetic additives in the food industry for the purpose of reducing food deterioration and extending the product's shelf-life [6,7].In order to isolate bioactive compounds from the herb, several conventional and novel extraction techniques have been used.In the first place, conventional hydrodistillation (HD) is considered one of the most convenient for essential oil isolation.Another interesting and widely used technique, which represents a gold standard for the isolation of lipophilic compounds, is Soxhlet extraction (SOX).Although this method is not the most suitable technique in terms of moving production from lab to industrial levels, considering the obtained total extraction yields (Y), it could be used as a great comparative technique.On the other hand, what has been long sought after are novel extraction techniques, advanced in reducing the costs of time, energy, manpower and solvents, as well as increasing the qualitative and quantitative composition of the final product.Last but not least is ensuring a simpler flow of downstream processes.Precisely in that manner, the novel extraction techniques can respond to the challenges of the modern age to a greater or lesser extent.To extract polyphenolic compounds from T. serpyllum L., besides conventional solid-liquid extraction [8,9], also used were extractions assisted with ultrasounds [8,9] and microwaves [10], pulsed electric fields extraction [11], extractions with subcritical fluids [12], as well as extractions with natural deep eutectic solvents (NADESs) [13].
When focusing on the isolation of lipophilic compounds, supercritical fluid extraction (SFE) has to be taken into account as one of the most promising techniques.Although, at the moment, it cannot be seen as economically profitable, its exceptional advantage in obtaining an extract rich in bioactive compounds without traces of solvents places this technique in a very high position in relation to the mentioned techniques.One successful example of herbal dust utilization using SFE was described by Mrkonjić et al. [14], who developed various mathematical models for fitting the T. serpyllum SFE process but without insight into chemical profiling and bioactivity.With this in mind, the SFE of T. serpyllum herbal dust has been chosen in this work as the technique that is, at the moment, the most suitable for comparison with conventional techniques in terms of the obtained Y, antioxidant and antimicrobial activity and the quantity and quality profile of extracted compounds.
The EOs and lipid extracts consist of a mixture of volatile terpenes, their oxygenated derivatives and also non-volatile compounds that make their separation and identification difficult [15].Gas chromatography coupled with time-of-flight mass spectrometry (GC-TOF/MS) that is characterized by a high resolution that enables the fast identification of a large number of compounds in EOs was used in order to perform the complete determination of chemical profile of EO and lipid extracts.
Several papers focused on the chemical composition of the EO of T. serpyllum, where the presence of secondary metabolites such as terpenes, terpenoids and polyphenolic compounds was determined, have been published [16][17][18][19].However, to the best of our knowledge, this is the first paper where the GC-TOF/MS of T. serpyllum herbal dust SFE extracts has been carried out.
The main objective of our study was to provide the chemical profile of extracts and EO obtained due to conventional and SFE techniques from T. serpyllum herbal dust.Additionally, the determination of their Y and antioxidant and antimicrobial activity was conducted in order to evaluate the advantages and disadvantages of hydrodistillation, SFE and Soxhlet extraction.

Total Extraction Yield (Y) and Chemical Composition
SFE, SOX and HD were applied to determine the Y of the EO and lipid extracts (Figure 1).The SFE extraction yield ranged from 0.47 to 2.93%, while the Y for SOX-Hex and SOX-MeCl was 1.83 and 2.60%, respectively.Conventional HD gave considerably the lowest yield of wild thyme EO (0.15%).Considering that EO consists of volatile compounds without non-volatile lipids, it could be expected that the EO yield is lower compared to other techniques.Furthermore, according to Y, SFE stood out as the best technique.The highest Y was obtained under the following extraction conditions: pressure of 350 bar, temperature of 50 °C and flow rate of 0.3 kg CO2/h.By comparing several different conditions under which the extraction was performed, it can be concluded that the highest yield was achieved by increasing the pressure.Varying the pressure from 100 to 225 bar leads to a sudden increase in Y. Furthermore, the increase in pressure from 225 to 350 bar, enables obtaining the highest Y. Observing the temperature ranged from 40 to 60 °C under the pressure of 350 bar, temperature of 50 °C stood out as the best, which leads to reduced possibility of the thermal degradation of the target compounds.However, a higher flow rate of 0.4 kg CO2/h (2.66%) did not enable a higher Y compared to a flow rate of 0.3 kg CO2/h (2.93%) at the same pressure and temperature.SOX gave a significantly lower Y of wild thyme EO using hexane as the solvent (1.83%).However, methylene chloride stood out as a better solvent in terms of Y (2.60%).A wide spectrum of compounds characterized by wild thyme EO and extracts were identified via HS-GC-MS (Table 1).The SFE extraction yield ranged from 0.47 to 2.93%, while the Y for SOX-Hex and SOX-MeCl was 1.83 and 2.60%, respectively.Conventional HD gave considerably the lowest yield of wild thyme EO (0.15%).Considering that EO consists of volatile compounds without non-volatile lipids, it could be expected that the EO yield is lower compared to other techniques.Furthermore, according to Y, SFE stood out as the best technique.The highest Y was obtained under the following extraction conditions: pressure of 350 bar, temperature of 50 • C and flow rate of 0.3 kg CO 2 /h.By comparing several different conditions under which the extraction was performed, it can be concluded that the highest yield was achieved by increasing the pressure.Varying the pressure from 100 to 225 bar leads to a sudden increase in Y. Furthermore, the increase in pressure from 225 to 350 bar, enables obtaining the highest Y. Observing the temperature ranged from 40 to 60 • C under the pressure of 350 bar, temperature of 50 • C stood out as the best, which leads to reduced possibility of the thermal degradation of the target compounds.However, a higher flow rate of 0.4 kg CO 2 /h (2.66%) did not enable a higher Y compared to a flow rate of 0.3 kg CO 2 /h (2.93%) at the same pressure and temperature.SOX gave a significantly lower Y of wild thyme EO using hexane as the solvent (1.83%).However, methylene chloride stood out as a better solvent in terms of Y (2.60%).A wide spectrum of compounds characterized by wild thyme EO and extracts were identified via HS-GC-MS (Table 1).(20.94 µg/mL), thymoquinone (15.37 µg/mL) and β-bisabolene (15.15 µg/mL) were identified in the highest amounts (Figure S1a).Although SFE extracts emerged as the best in terms of Y, they are not distinguished as the best in terms of terpene content.However, the SFE-10 extract (350 bar, 50 • C, 0.4 kg CO 2 /h) was the most terpene-rich extract.o-cymene (167.72 µg/mL), carvacrol (108.15µg/mL), β-bisabolene (50.32 µg/mL), thymol (49.36 µg/mL) and 5-methyl-2-(1-methylethenyl)-, (R)-4-hexen-1-ol (43.15 µg/mL), were identified in this extract in the highest amounts.Subtle differences in composition between SFE extracts under different conditions could be observed.Although the SFE-2 extract was found to be the best in terms of Y, it was not the richest in terms of terpenes content.The results indicated that extracts obtained at the higher flow, pressure and temperature are richer in terpene content.When SOX was analyzed, the SOX-Hex extract dominantly contained nonane (27.15 µg/mL), carvacrol (19.83 µg/mL), geraniol (13.26 µg/mL) and o-cymene (10.68 µg/mL), whose chromatogram could be found in Figure S1b.The SOX-Hex extract was richer in monoterpenes content compared to SOX-MeCl.Although a higher yield of lipophilic extracts was obtained with methylene chloride, it does not represent an adequate solvent in terms of terpene content.Even though they were obtained via different techniques, SFE and SOX extracts do not differ significantly in terms of the content of dominant compounds.The most significant results were obtained via HD.Wild thyme EO was characterized by a significantly higher content of m-cymene (832.26µg/mL), γ-terpinene (192.52 µg/mL), α-terpinyl acetate (177.69 µg/mL) and isothymol methyl ether (151.73 µg/mL).α-copaene, β-thujene, β-myrcene, β-cububene, 2-hexenal, benzaldehyde, cis-sabinenehydrate, dihydrocarvone, hexanal and α-terpinyl acetate were identified only in HD-EO (Figure S1c).Considerable differences can be observed by comparing HD extracts with SOX and SFE extracts, mostly due to the absence of non-volatile lipids in HD-EOs.Therefore, HD represents the most economically profitable technique for obtaining oils with a high content of volatile terpenes.However, in terms of the environmental aspects and extraction yield, the advantage is given to SFE.
Sfaei-Ghomi et al. [20] proved that there are small differences in the composition of EOs obtained via HD between four different Thymus species (T.persicus, T. eriocalyx, T. daenensis subsp.daenensis and T. serpyllum L.).Carvacrol (14.94%), α-pinene (12.2%) and thymol (7.39%) were the major terpenes in T. serpyllum, with an EO yield of 1.2 ± 0.8%, which was in accordance with the present data.Furthermore, similar results were reported from T. serpyllum EO from the Mascara region [6].The obtained EO yield rich in carvacrol (66%), γ-terpinene (11.5%), thymol (7.5%) and p-cymene (3.9%) was 5.66%, which is higher than the Y obtained in this work.Bendif et al. [21] performed SFE, pressurized liquid extraction (PLE) and HD to obtain the extraction yield, chemical composition and antioxidant activity of the two Thymus munbyanus subspecies [21].SFE extracts (45 MPa, 70 • C and 2 L CO 2 /min) achieved a Y of 0.35 to 0.43%, which is significantly less than the results obtained by the same technique in this work.The main reasons for this are the extraction parameters, which could have led to the precipitation of waxes in the separator.However, the yield of EO obtained via HD (0.11 and 0.09%) was similar to the Y obtained in this work (0.15%).Comparing Thymus species collected from different regions of Iran, it was detected that thymol (12.4-79.74%),carvacrol (4.37-42.14%),geraniol (0.3-22.44%) and p-cymene (0.8-12.86%) were the most dominant [22].T. migricus provides a higher Y (3.87%) in comparison with T. fedtschenkoi-2 (0.29%).These yields are higher than the Y obtained in this work, so it can be concluded that differences in species are noticeable.Pavlić et al. [23] used sage herbal dust for discovering the adequate technique in terms of Y among HD, SOX and SFE.SOX stood out with both methylene chloride and hexane (14.68 and 10.84%, respectively), which is not in accordance with the results obtained in this work.Also, wild thyme is characterized by a significantly lower yield compared to sage.The differences in the chemical composition of the EO and extracts between these two plants are notable, even though they belong to the same plant family.Camphor, α-thujone, eucalyptol, viridiflorol and epirosmanol were the most abundant compounds in sage herbal dust, opposite to the wild thyme, where thymol, carvacrol and o-cymene were the most dominant.
Also, Pavlić et al. [24] applied HD, SOX, microwave-assisted extraction (MAE), ultrasound-assisted extraction (UAE) and SFE in order to obtain peppermint EO and lipophilic extracts.According to the GC-MS results, the most dominant compounds in peppermint extracts were monoterpenes.Among these monoterpenes, the most abundant compounds were menthol, menthone, isomenthol, isomenthone and eucalyptol.They concluded that the different techniques applied have a huge impact on the extracts' chemical profile because of their different selectivity.Kulisic et al. [25] proved that there is no particular difference in the qualitative composition between thyme and wild thyme EOs obtained via HD.GC-MS analysis showed that the predominant compounds in both thyme and wild thyme were γ-terpinene, p-cymene, thymol and carvacrol, which is in correlation with the results in this study.Another study from Aćimović el al. [17] examined the chemical composition of Mediterranean plants such as S. kitaibelii, T. serpyllum and O. vulgare, which belong to the Lamiaceae family.The obtained results showed that the most abundant compounds in T. serpyllum EO were geraniol (63.4%) and nerol (18.9%), which is not in accordance with the results obtained in this work.In addition, S. kitaibelii EO contained p-cymene, limonene and linalool, while O. vulgare EO contained germacrene D, 1,8-cineole, sabinene and trans-caryophyllene.Differences in the composition of terpenes between the same species, but also between species which belong to the same family, can be caused by growing conditions, geographical location and environmental factors.An investigation by Goyal et al. [19] confirmed variation in the dominant compounds (thymol, α-terpineol, p-cymene, camphor and γ-terpinene) in T. serpyllum.They observed that the content of thymol was different in three locations, Haldwani (84.63%),Auli (50.80%) and Pithoragarh (41.15%).In addition to thymol, which was dominant, camphor was the second most abundant terpene (36.34%).According to Verma et al. [26], T. serpyllum showed an extraction yield of 0.22%, whereas, using GC-MS, it was determined that the most abundant compounds were thymol, p-cymene, thymol methyl ether, borneol, sabinene, γ-terpinene and carvacrol methyl ether, which is in accordance with EO composition obtained in this work.In addition, Topal et al. [27] determined the chemical composition of nine EOs from different Turkish plants.Among them, T. serpyllum was used to obtain steam distillation (SD) EO and SFE extracts.GC-MS analysis showed that the dominant terpenes in SFE extracts were p-cymene (5.41%), carvacrol (47.79%), 2,4,6-trimethylanisole (24.95%) and βbisabolene (3.67%).The content of thymol in SFE extracts and in SD EO was 1.53 and 1.41%, respectively.The results obtained in these studies are in accordance with the presented results in this paper, and both show that thymol was one of the most dominant compounds found in the extracts.

In Vitro Antioxidant Activity
The ability to neutralize DPPH and ABTS + radicals was determined in samples obtained via SFE, SOX and HD.The results of the antioxidant capacity of wild thyme extracts determined by DPPH and ABTS methods are presented in Figure 2.
The DPPH test values ranged from 8.18 to 58.32 µM TE/g, where the highest antiradical scavenging effect was shown in extracts obtained via SFE at the temperatures of 50 and 60 • C and at a fixed pressure (350 bar) and CO 2 flow rate (0.3 kg CO 2 /h) (58.32 µM TE/g and 52.66 µM TE/g, respectively) (Figure 2a).Regarding the DPPH assay, the sample SOX-Hex showed the lowest antioxidant activity (8.18 µM TE/g) (Figure 2e).In addition, at the pressure of 225 bar and with a CO 2 flow rate of 0.3 kg CO 2 /h, using different temperatures (40, 50 and 60 • C), it was observed that there were no big changes in antioxidant activity (Figure 2b).However, in the case of the ABTS assay, by increasing the temperature, the possibility of neutralizing ABTS + radicals increased as well (Figure 2b).In order to visualize the impact of the CO 2 flow rate on the antioxidant activity of SFE extracts, a pressure of 350 bar and temperature of 50 • C were used (Figure 2d).It could be concluded that in the case of DPPH, there is no notable effect of the CO 2 flow rate, in comparison with ABTS, where antioxidant activity increased with the elevated CO 2 flow rate.According to the DPPH values of samples obtained via conventional techniques, the sample SOX-MeCl showed the highest antioxidant activity (28.08 µM TE/g), followed by HD-EO (25.32 µM TE/g), while SOX-Hex showed the lowest antioxidant activity (8.18 µM TE/g) (Figure 2e).In the case of ABTS, the sample HD-EO (2402.95µM TE/g) showed four to six times higher antioxidant activity compared to the samples obtained via SOX (Figure 2f).It could be concluded that the pressure had the greatest influence on antioxidant activity in the case of the DPPH test, where by increasing the pressure, antioxidant activity increases as well.An increase in the pressure could cause the increase in CO 2 density, which directly affects the increase in its solvation power.Considering a close relationship between the solubility of target compounds and the solvent density, greater antioxidant activity and also the low selectivity of the extraction process toward the targeted compounds and the low level of purity of the obtained extracts could have occurred [33].The DPPH test values ranged from 8.18 to 58.32 µM TE/g, where the highest antiradical scavenging effect was shown in extracts obtained via SFE at the temperatures of 50 and 60 °C and at a fixed pressure (350 bar) and CO2 flow rate (0.3 kg CO2/h) (58.32 µM TE/g and 52.66 µM TE/g, respectively) (Figure 2a).Regarding the DPPH assay, the sample SOX-Hex showed the lowest antioxidant activity (8.18 µM TE/g) (Figure 2e).In addition, at the pressure of 225 bar and with a CO2 flow rate of 0.3 kg CO2/h, using different temperatures (40, 50 and 60 °C), it was observed that there were no big changes in antioxidant activity (Figure 2b).However, in the case of the ABTS assay, by increasing the temperature, the possibility of neutralizing ABTS + radicals increased as well (Figure 2b).In order to visualize the impact of the CO2 flow rate on the antioxidant activity of SFE extracts, a pressure of 350 bar and temperature of 50 °C were used (Figure 2d).It could be concluded that in the case of DPPH, there is no notable effect of the CO2 flow rate, in comparison with ABTS, where antioxidant activity increased with the elevated CO2 flow rate.According to the DPPH values of samples obtained via conventional techniques, the sample SOX-MeCl showed the highest antioxidant activity (28.08 µM TE/g), followed by HD-EO (25.32 µM TE/g), while SOX-Hex showed the lowest antioxidant activity (8.18 µM TE/g) (Figure 2e).In the case of ABTS, the sample HD-EO (2402.95µM TE/g) showed four to six times higher antioxidant activity compared to the samples obtained via SOX (Fig- Pavlić et al. [23] determined in vitro antioxidant activity from sage herbal dust extracts obtained using SFE, HD and SOX.A high DPPH value of SFE extracts was obtained at a pressure of 300 bar, temperature of 50 • C and CO 2 flow rate of 0.3 kg CO 2 /h (987.60 mM TE/g).These results were obtained under nearly identical conditions as in this paper.However, the sage herbal dust extract showed higher antioxidant effect than the wild thyme extract.Kulisic et al. [25] determined the antioxidant activity of thyme and wild thyme EOs that were obtained via HD.According to the DPPH assay, EO thyme had a slightly better antioxidant activity compared to wild thyme.Babovic et al. [34] isolated an antioxidant fraction from thyme via SFE at a pressure of 35 MPa, temperature of 100 • C and CO 2 flow rate of 0.3 kg CO 2 /h.The results of the DPPH test were presented as the IC 50 value, where the radical scavenging of the thyme extract was 0.08 mg/mL.This result was compared with synthetic antioxidant butylated hydroxytoluene (BHT), whose value of IC 50 was slightly better (0.03 mg/mL).Topal et al. [27] analyzed samples obtained via SD and SFE.They used nine Turkish plants from different families, including T. serpyllum from the Lamiaceae family.In this study, the EO of T. serpyllum showed effective DPPH radical scavenging potential when compared with BHT.SFE was performed at a pressure of 20-30 MPa and a temperature of 40-60 • C. The samples obtained via SFE had marginally higher free radical scavenging activity percentages than the EO samples obtained via SD.Similar research was executed by Petrović et al. [35], where bioactive compounds were isolated from Thymus praecox using the SFE method under the following conditions: at 100 bar and 40 • C and at 300 bar and 40 • C. At a lower pressure, IC 50 was 446.0 mg/mL, and with increasing pressure, IC 50 decreased (404.5 mg/mL), which means that the pressure had a positive impact on the antioxidant activity of the extracts, which corresponds to the results in this work.
The potential to scavenge ABTS + radicals is significantly higher compared to DPPH.ABTS test values ranged from 360.08 to 2402.95 µM TE/g.According to the ABTS test, the highest antioxidant activity was shown by HD-EO, while the SOX-Hex and SOX-MeCl showed a notable lower radical scavenging value (360.08 and 546.31 µM TE/g, respectively).Extracts obtained by SFE at a pressure of 100 bar, at a temperature of 50 and 60 • C and at a CO 2 flow rate of 0.3 kg CO 2 /h (Figure 2c) showed the highest antioxidant activity (834.81 and 932.08 µM TE/g, respectively).The lowest ABTS radical scavenger was shown by the extract obtained by SFE under the following conditions: pressure 225 bar, temperature 40 • C and CO 2 flow rate of 0.3 kg CO 2 /h (470.65 µM TE/g).When the temperature increased (Figure 2a), at constant pressure (350 bar) and at constant CO 2 flow rate (0.3 kg CO 2 /h), there was a decrease in antioxidant activity.This can be explained as a consequence of the degradation of antioxidants.
Pavlić et al. [24] evaluated the antioxidant activity of peppermint.According to the DPPH test, sample SOX-MeCl had the highest antioxidant activity (98.43 ± 2.39 mM TE/g).Meanwhile, the ABTS test showed the highest antioxidant activity in the sample obtained via MAE (138.62 ± 4.33 mM TE/g).In the case of SFE, 400 bar, 40 • C and a CO 2 flow rate were shown to be the best conditions for the obtained extracts with the highest DPPH and ABTS radical scavenging (33.27 ± 3.13 and 47.50 ± 4.02 mM TE/g, respectively).Compared to the results obtained in this work, it could be concluded that peppermint has a greater ability to reduce DPPH and ABTS + radicals than wild thyme.Similar results are reported in the review paper by Šojić et al. [7], who have found a practical application of T. serpyllum SFE extracts obtained at the following conditions: pressure of 100 bar, temperature of 40 • C for the first extract, and pressure of 350 bar and temperature of 50 • C for the second extract.In both extractions, the flow rate was 0.3 kg CO 2 /h.Leon-Méndez et al. [36] also reported that T. vulgaris EO, obtained via HD, showed good antioxidant activity obtained by the DPPH assay (IC 50 = 165.5 ± 1.05 µg/mL), while the IC 50 value obtained by the ABTS assay was 29.07 ± 0.07 µg/mL.In addition, the antioxidant activity of T. vulgaris EO obtained via HD was reported by Gladikostić et al. [37] as well.The value of the DPPH assay was 29.78 µM TE/g, which is very similar to the results in this work.In the case of the ABTS assay, the antioxidant activity value was 757.19 µM TE/g, which is significantly lower compared to the results in this paper.

Antimicrobial Activity
The results of antimicrobial activity assay for extracts prepared from T. serpyllum by-products are presented in Table 3. Data are expressed as mean ± standard deviation (n = 3). 1 CHL data shown in the table column refer to antibacterial activity of chloramphenicol on the bacterial strains, while NYS refers to antifungal activity of nystatin on C. albicans. 2Means marked by different letters in the same row are significantly different at α = 0.05 (Tukey's HSD). 3 ND not detected (with the highest tested concentration of samples (20 mg/mL), or antibiotic (1000 µg/L) antimicrobial activity was not detected).
Two Gram-positive bacterial strains, S. aureus ATCC 25923 and methicillin-resistant S. aureus (MRSA), were the most sensitive to all three samples, with MIC values below 0.02 mg/mL.EO and SFE-2 expressed the same inhibitory effect against B. spizizeni (MIC value of 0.31 mg/mL) and, compared to SFE-7 (MIC value of 0.83 mg/mL), were over 2.5 times more effective.Both E. faecalis strains and L. monocytogenes were the most susceptible to HD-EO samples (Table 3).HD-EO and SFE-2 showed microbicidal effects on all tested Gram-positive bacteria.On the other hand, the MBC of SFE-7 for both E. faecalis strains was not determined.B. spizizeni (MBC value 0.31 mg/mL for HD-EO and SFE-2) and S. aureus ATCC 25923 (MBC value 0.31 mg/mL, SFE-2) stood out as the most sensitive.All three samples had an microbicidal effect on methicillin-resistant S. aureus (MBC value of 0.62 mg/mL HD-EO; MBC value of 2.5 mg/mL for SFE-2 and SFE-7).S. aureus is a producer of staphylococcal enterotoxin in foods which causes poisoning after ingestion and is a causer of skin infections as well as other systemic infections [38].Furthermore, the S. aureus (MRSA) strain is recognized as a top priority of worldwide public health systems as its prevalence is between 25 and 50%, and in some areas, it reaches over 60% of all S. aureus isolates [39].The outstanding activity of all T. serpyllum SFE extracts against S. aureus ATCC 25923 and S. aureus MRSA fulfills the enormous importance of tackling these problems.
Regarding the Gram-negative bacteria, observed differences between the samples were more pronounced.Namely, sample HD-EO expressed the highest efficiency, with MICs and MBCs established for all bacteria (MIC was between 0.10 and 2.5 mg/mL, and MBC ranged between 0.62 and 2.5 mg/mL).The most sensitive were P. hauseri and Y. enterocolitica (MIC 0.16 and 0.10 mg/mL, respectively).Additionally, HD-EO was the only sample that had both microbistatic and microbicidal effects on E. coli strains (Table 3).Regarding SFE-2 and SFE-7 extracts, differences in antibacterial activity were also observed.Namely, better activity was found for the SFE-2 extract which acted in an inhibitory way toward seven strains in comparison with the SFE-7 sample that inhibited the growth of four Gram-negative strains.P. hauseri and Y. enterocolitica were the most sensitive strains on both samples, but the MIC values were lower for SFE-2 (0.83 mg/mL) than for SFE-7 (2.5 mg/mL).Comparing MBCs, differences were even more pronounced since SFE-7 had an microbicidal effect only on three bacterial species, i.e., P. mirabilis, P. hauseri and P. aeruginosa, expressing MBC activity that was significantly lower compared with the SFE-2 extract.
Additionally, when comparing SFE-2 and SFE-7, a microbicidal effect on S. Typhimurium, S. Enteritidis and Y. enterocolitica bacterial species was found only for SFE-2.These findings are in accordance with the previous research when the SFE-2 extract, applied in ground pork meat, more efficiently reduced the total Enterobacteriaceae number and total number of microorganisms, showing better activity on its microbiological profile than the other extracts.Such an effect was the consequence of the extraction parameters that were set to obtain the total lipid amount and high polyphenolic terpenoids in the SFE-2 extract.The identification of sixteen compounds in SFE-2 that were not found in SFE-7 proved the more complex composition of SFE-2 [7].Differences between the samples were observed regarding antifungal activity as well.The SFE-7 sample did not express activity on C. albicans pathogenic yeast growth.HD-EO and SFE-2 exhibited both fungistatic and fungicidal effects.It can be seen that HD-EO had a fungistatic effect on C. albicans yeast at a lower concentration (1.25 mg/mL) compared with SFE-2 (2.5 mg/mL).A similar pattern was observed regarding fungicidal activity, with the MFC eight times lower for HD-EO (1.25 mg/mL) compared to SFE-2 (10 mg/mL).In the research of Jovanović et al. [40], lyophilized T. serpyllum extracts obtained after maceration, heat or ultrasound-assisted extractions did not express fungicidal activity against C. albicans, while the MIC value of SFE-2 (2.5 mg/mL) was lower.In addition, SFE-2 expressed lower MICs against S. aureus and Y. enterocolitica and a lower MBC on S. aureus, but higher concentrations of SFE-2 were necessary to inhibit the growth of L. monocytogenes, E. coli and Salmonella sp.compared to the lyophilized extracts [40].Activity absence toward C. albicans, Salmonella spp., E. faecalis, Ps. aeruginosa, Shigella flexneri and S. aureus was determined for hydroalcoholic extracts of T. serpyllum [41].Monoterpene alcohols are usually recognized as components with high antimicrobial activity [38,42].However, in this research, thymol and carvacol concentrations were lower in the SFE-2 sample (Table 1) that exhibited stronger antibacterial and antifungal activity than the SFE-7 extract.The highest concentration of linalool and terpinen-4-ol were found in HD-EO, followed by SFE-2 and SFE-7, which is in accordance with the detected antimicrobial activity of the samples (Table 3).Similarly, monoterpene hydrocarbons, i.e., camphene, γ-terpinene, m-cymene, α-pinene and β-pinene, were found in the highest concentration in HD-EO than in SFE-2 and were not detected in SFE-7 or were present in lower concentrations.These compounds were previously shown to express anticandidal and antibacterial activity [38].It was previously found that minor components might also contribute to better antibacterial activity via synergistic effects with major components [42,43].Even though the extracts obtained from T. serpyllum by-products acted at higher concentrations in comparison to applied antibiotics (Table 3), natural extracts are attracting high interest for application in food systems to prevent the occurrence of pathogenic microorganisms and to reach a prolonged product shelf-life.The complex chemical composition of phytopreparations that contain several active compounds allows the treatment of microbial infections due to the reduced possibility of resistance development in microorganisms.Additionally, it is possible to reduce the applied doses of antibiotics via the synergistic effects of active compounds from plant extracts with antibiotics [44].In that manner, the better activity of SFE-2 recommends it for applications in food, cosmeceutical or pharmaceutical products with antimicrobial properties.

Sample
The sample of wild thyme herbal dust was kindly donated by Macval Tea D.O.O.(Novi Sad, Serbia), and it constitutes industrial waste generated during filter tea production.By cutting, grinding and fractionating the raw plant material, a certain amount is separated and considered a by-product because of its mean particle size (≤0.315mm) and the impracticality of it being packed in the filter tea bags.The T. serpyllum herbal dust was stored in paper bags at room temperature prior to the extractions and distillations.

Supercritical Fluid Extraction (SFE)
SFE was performed using a high-pressure extraction system (HPEP, NOVA, Swiss, Efferikon, Switzerland), which consists of a gas cylinder with CO 2 , a diaphragm-type compressor (pressure range up to 1000 bar), an extractor with an internal volume of 200 mL (maximal operating pressure of 700 bar), a separator (maximal operating pressure of 250 bar), a temperature regulation system and pressure control and regulation valves.The sample (35.00 g) was placed in the extractor vessel, and the extraction process was carried out by varying the pressure, temperature and CO 2 flow rate.The separator conditions were fixed at 15 bar and 25 • C, and the extraction time was 180 min.The first 9 extracts were obtained by varying the pressure (100, 225 and 350 bar) and temperature (40, 50 and 60 • C)), while the other SFE extracts were obtained by varying the CO 2 flow rate (0.2, 0.3 and 0.4 kg CO 2 /h) under a fixed pressure of 350 bar and at a temperature of 50 • C. The Y was measured after a 180 min extraction time and presented as a mass of total extractable solids per 100 g of dry plant material (w/w).The obtained extracts were collected into plastic vials and stored at 4 • C prior to analysis.

Soxhlet Extraction (SOX)
SOX was performed using a Soxhlet apparatus where wild thyme herbal dust (10.00 g) was extracted with two organic solvents, hexane and methylene chloride (120 mL each), separately.After 6 h, the solvent was evaporated under vacuum, and the obtained extract was further dried at 40 • C for 24 h in a laboratory dryer (Sutjeska, Belgrade, Serbia).The obtained extracts (SOX-Hex and SOX-MeCl) were collected into glass vials and stored at 4 • C prior to analysis.

Hydrodistillation (HD)
HD was performed according to the official HD procedure [45].The Y was presented as percentage (%, v/w), and the obtained EO was collected into glass vials and stored at 4 • C prior to analysis.
3.6.Analysis of Chemical Composition via GC-MS Techniques 3.6.1.HS-GC-MS The quantification of identified compounds of T. serpyllum extracts and essential oil was achieved using an Agilent gas chromatograph-mass spectrometer equipped with a headspace sampler (HS-GC-MS).The system consists of an Agilent 7697A headspace sampler, a 6890 N gas chromatograph (Santa Clara, CA, USA) and a 5975C mass selective detector (Santa Clara, CA, USA).The chromatographic separation of compounds was performed on a DB-5MS (60 m × 0.25 mm ID, 0.25 m) capillary column.After achieving full equilibrium at 60 • C for 30 min, 1 mL of the headspace sample was injected into a capillary column in split mode (15:1) for 30 s.The transfer line was set at 150 • C. Helium with a constant flow rate at 1 mL/min was used as the carrier gas.Initially, the oven temperature was held at 50 • C for 2 min and then programmed to 300 • C at a rate of 10 • C/min and finally kept at 300 • C for 5 min.The mass spectrometer was operated in the electron ionization mode (70 eV) in the mass scan range of m/z 40-550 Da.The ion source temperature was set at 300 • C. The retention indices (RIs) of compounds were calculated using a series of n-alkanes (C 10 -C 24 ).

GC×GC-TOF/MS
The identification of volatile compounds of T. serpyllum extracts were performed using a LECO Pegasus 4D GC×GC-TOF/MS instrument (LECO Corporation, St. Joseph, MI, USA) equipped with an Agilent 7890B gas chromatograph.A set of non-polar and a mid-polar columns was used for GC×GC separation.The first dimension (1D) column was 30 m × 250 µm × 0.1 µm DB-5 MS (Agilent J&W GC Columns, USA), and the second dimension (2D) column was 2 m × 150 µm × 0.15 µm Rxi-17Sil MS Restek (Bellefonte, PA, USA).GC inlet and transfer line temperatures were set at 220 and 250 • C, respectively.Helium was used as the carrier gas at a constant flow of 1 mL min −1 .Cryogenic modulation was performed with a 4 s modulation period (PM).Sample injection volumes of 1 mL and split ratios of 10:1 were performed using a CombiPAL autosampler (CTC Analytics, Zwingen, Switzerland).The oven temperature for the first column was held at 40 • C for 1 min and then ramped up to 260 • C at a rate of 5 • C/min and held for 1 min.The second oven was operated at 10 • C higher than the first oven throughout the process.The modulation period was 4 s with a heat pulse of 1 s.A Pegasus ® IV time-of-flight mass spectrometer (LECO Corp.) was used as the detector.MS was operated in electron impact ionization mode (70 eV), and ions were collected in the mass range of 45-550 amu.The ion source temperature was set at 230 • C. The tentative identification of compounds was based on a similarity comparison of standard MS in NIST05 (National Institute of Standards and Technology, Gaithersburg, MD, USA) and Wiley (Wiley, New York, NY, USA) libraries.

Antioxidant Activity
The antioxidant activity of the samples was determined via spectrophotometric methods for the scavenging of DPPH [46] and ABTS + radicals [47].The DPPH assay was performed by mixing 100 µL of the extract with 2900 µL of DPPH solution, which was previously prepared in a concentration of 26 mg/L of methanol and adjusted with the aim of reaching an absorbance of 0.70 ± 0.02 at a wavelength of 517 nm.After the extracts were left for 1 h in the dark and at room temperature, the absorbances were recorded at the same wavelength.
In order to perform the ABTS assay, the ABTS stock was first prepared.Moreover, 7 mM ABTS and 2.45 mM potassium peroxydisulfate were mixed (1:1, v/v) and stored in the dark at room temperature for 16 h.After 16 h, in order to prepare the ABTS reagent, the ABTS stock was mixed with acetate buffer pH 3.6 (1:40, v/v).In order to complete the preparation of the ABTS reagent, it was adjusted in order to reach an absorbance of 0.70 ± 0.02 at a wavelength of 734 nm.Subsequently, the ABTS assay could be performed by mixing 100 µL of the extract with 2900 µL of the ABTS reagent.After 5 h in the dark at room temperature, the absorbances were recorded at a wavelength of 734 nm.
All experiments were performed in triplicate, and the mean values were presented as mM of the Trolox equivalent (TE) per gram of the sample dry weight (mM TE/g).Enteritidis ATCC 13076, Salmonella ser.Typhimurium ATCC 14028, Shigella sonnei ATCC 29930, Yersinia enterocolitica ATCC 27729) were used.Antifungal activity of the samples was determined against one pathogenic yeast, Candida albicans ATCC 1231.For culturing E. coli H7:O157 and L. monocytogenes bacteria, tryptic soy agar/broths (HiMedia, Lab., LLC, Mumbai, India) were used, while the other bacteria were cultured in Müller Hinton agar/broth (HiMedia).For the antifungal activity assay, C. albicans yeast growth was performed using Malt agar/broth (HiMedia).From the appropriate agar plate, a 24 h old colony was sub-cultured to an adequate broth (5 mL), which was incubated at 37 • C for 18-24 h.The final concentration of microorganisms was adjusted to ~10 5 colony forming units per mL (CFU/mL) using the DEN-1 McFarland densitometer (Biosan, Riga, Latvia).

Broth Microdilution Method
To determinate the minimal inhibitory (MIC, mg/mL), minimal bactericidal (MBC, mg/mL) and minimal fungicidal concentration (MFC, mg/mL) of the three T. serpyllum extracts, the broth microdilution method was used [40,48].Prior to the testing, the samples were dissolved in 5% DMSO/water solution followed by vigorous stirring.Three-fold sample dilutions in the concentration range between 0.02 mg/mL and 20 mg/mL were prepared in a 96-well microtiter plate (Sarstedt, Nümbrecht, Germany).Bacterial and yeast suspensions were added to each well that contained a sample of the different dilution so that the final volume in each well was 100 µL.After seeding, microtiter plates were incubated at 37 • C for 18-24 h.As a positive control, only bacterial suspensions without the sample were used, while as a negative control, 5% DMSO/water solution was applied.Additionally, to assess the inhibitory activity of the tested samples, chloramphenicol and nystatin were used for bacteria and yeast, respectively.As an indicator of microbial cell growth, resazurin sodium salt was added, and its color change from blue to pink or colorless indicated microbial growth.The lowest concentration of a sample where no visible microbial growth was observed (i.e., there was no change in the indicator color) was designated as the MIC (mg/mL).To determinate the MBC and MFC, sample dilutions with an already established MIC were sub-cultured to the appropriate agar base, and the lowest tested concentration with no visible growth after incubation was designated as the MBC or MFC.The antimicrobial activity assay was performed in triplicate for each sample/microorganism, and the results were represented as the mean value ± standard deviation.The obtained results were analyzed using a single-factor analysis of variance (ANOVA) by means of the statistical program Origin Pro 9.0.The significance of the differences between the samples that were determined were tested using Tukey's HSD test at a significance level of p ≤ 0.05.

1 .
Total Extraction Yield (Y) and Chemical Composition SFE, SOX and HD were applied to determine the Y of the EO and lipid extracts (Figure1).

Figure 1 .
Figure 1.Y of EO, SOX and SFE extracts Antioxidant activity of EO, SOX and SFE extracts determined via DPPH and ABTS methods.

Figure 1 .
Figure 1.Y of EO, SOX and SFE extracts Antioxidant activity of EO, SOX and SFE extracts determined via DPPH and ABTS methods.

Figure 2 .
Figure 2. Antioxidant activity of (a) SFE extracts obtained at 350 bar and 0.3 kg CO2/h by varying the temperature, (b) SFE extracts obtained at 225 bar and 0.3 kg CO2/h by varying the temperature, (c) SFE extracts obtained at 100 bar and 0.3 kg CO2/h by varying the temperature, (d) SFE extracts obtained at 350 bar and 50 °C by varying the CO2 flow rate, (e) SOX extracts and HD-EO determined via DPPH and (f) SOX extracts and HD-EO determined via ABTS method.Means values obtained via DPPH and ABTS assays marked by different letters are significantly different at α = 0.05 (Tukey's HSD).

Figure 2 .
Figure 2. Antioxidant activity of (a) SFE extracts obtained at 350 bar and 0.3 kg CO 2 /h by varying the temperature, (b) SFE extracts obtained at 225 bar and 0.3 kg CO 2 /h by varying the temperature, (c) SFE extracts obtained at 100 bar and 0.3 kg CO 2 /h by varying the temperature, (d) SFE extracts obtained at 350 bar and 50 • C by varying the CO 2 flow rate, (e) SOX extracts and HD-EO determined via DPPH and (f) SOX extracts and HD-EO determined via ABTS method.Means values obtained via DPPH and ABTS assays marked by different letters are significantly different at α = 0.05 (Tukey's HSD).

Table 1 .
The quantification of volatile compounds detected in wild thyme EO and SFE and SOX extracts determined via HS-GC-MS.

Table 2 .
The chemical composition of SFE extracts and EO determined via GC-TOF/MS.

Table 3 .
Antimicrobial activity of wild thyme EO and SFE extracts.