3.1. EO Extraction
It would be expected that prolonging the duration of the extraction process would lead to a slight increase in the cumulative amount of EO [
22], especially considering the results obtained with the fractionated distillation process, which show significant yields up to 24 h (see
Section 2.1.2). However, the results obtained in this study do not strictly follow this pattern, as the expected amount of EO may decrease with time extension. This aspect was not investigated further, as it could be due to several unknown reasons. The 24 h SD and HD extractions on fresh MS (first harvest) gave the lowest amounts of EOs, even more than three times less than the 6 h extractions. In the case of dried plant material, these exceptions are particularly noticeable in the 3-hour SD extractions. This is consistent with some literature data reporting a 25–40% decrease in yield for some
Cymbopogon EOs with increasing time [
23]. Numerous studies have confirmed that EO yield decreases with time [
24,
25,
26] or at least reaches equilibrium at some point without further increase [
27,
28,
29]. There are also data from the literature suggesting different effects of drying on EO content [
30], as some studies noted a significant decrease in yield [
31], while others reported an increase [
26]. These differences could be due to the drying time as well as the temperature used [
30].
From the results here, it can be observed that the second harvest yielded more EOs in both SD and HD processes, with only one exception in the 6 h SD from dried material (
Supplementary Material, Figures S1 and S2). The results also show that drying MS significantly increases the EO content, similar to that reported for fractionated SD and HD (
Supplementary Material, Figures S3 and S4). Comparing the two distillation methods applied, it seems that HD is more efficient since it gives a higher yield, especially in the case of the second harvest (
Table 2 and
Supplementary Material, Figures S5 and S6). As the two harvests were performed with a 10-day time difference and considering the high weather variability during September (harvest month), variability in both yield and composition could be expected.
The additional fractions from the continuous part up to 24 h and the cumulative yields (up to 24 h) were also examined (
Table 2). Continuous 1-, 2-, 3-, and 6-hour distillations yielded from 14.42% to 49.39% of the total (24 h) yield (
Supplementary Material, Table S1). With a few exceptions, it is also noticeable that MS drying increased these percentages, which is particularly evident in the case of HD extractions (up to two-fold). However, these data are not consistent with those obtained for fractionated distillations (see
Section 2.1.2), where 6 h of extraction (first four fractions) were allowed to extract from 68.11% to 98.77% of the total EO amounts (
Supplementary Material, Table S2).
Comparison of the same extraction types from two harvests revealed very small differences between the yield trends (
Supplementary Material, Figures S7 and S8). In the case of fresh material subjected to SD, two main yield peaks were observed: the first between the first and second extraction hours, and the second between the third and sixth extraction hours. On the other hand, SD of the dried material yielded the highest amount of EO in the first hour of the extraction process, but with a remarkable addition in the last 18-hour fraction. In the case of HD, from 63.8% to 88.9% of the total amount of EO (extracted in 24 h) was isolated in the first 2 h (
Supplementary Material, Table S2), which is common for many Lamiaceae species. However, there are large differences in yields between fresh and dried materials, as drying increased the total yield up to seven-fold in the case of SD and even 15-fold in the case of HD (
Table 3 and
Supplementary Material, Figures S9 and S10).
Comparing the two distillation methods, HD yielded a higher percentage of total EO extracted in the first 3 h of extraction (up to 91.9%), which is the most common extraction time (
Supplementary Material, Table S2, and Figures S11 and S12 This is consistent with a previous report suggesting a 3-hour HD as the optimal duration for MSEO extraction [
14].
3.2. EO Chemical Composition
As previously reported, MSEO from Tarquinia belongs to the PO-rich chemotype [
14,
21,
32,
33,
34,
35]. PO can certainly be highlighted as the main component, and it was found in all samples with relative percentages ranging from 1.2% (EO087) to 71.7% (EO068), and it is present in 25% of all EO samples with a percentage higher than 55.6% (
Table 4 and
Table 5). The results show that this monoterpenoid epoxyketone is usually extracted in the first 6 h of the extraction process, in agreement with a previous study [
14]. This is particularly evident in the fractionated distillations: PO was the most abundant compound in the first four fractions, while its content was significantly reduced in the last 18 h of the extraction process. Examples include both SD and HD extractions, e.g., EO082, EO087, and EO107 (
Table 4). PO was mainly extracted during the first 3 or 6 h of the extraction process, as shown by the results of the continued distillations. Consequently, the extractions performed to complete the 24-hour distillation processes provided EO samples with the lowest PO content. For example, sample EO049 obtained with a 6-hour continued HD extraction (2_fc-HD_0-6) contained 71.3% PO, while the corresponding additional 18-hour extraction (2_fc-HD_6-24) yielded only 3.2% PO (EO054).
It is also observed that the HD method gave a better yield of PO than the SD method, but only in the case of fractionated distillations. Moreover, it seems that the drying of the plant material did not have a significant effect on the PO content, although there were samples where the PO content even doubled, from EO079 with 27.3% in the second fraction (2_ff-SD_1-2) isolated from the fresh material by fractionated SD to EO089 with a PO content of 60.2% obtained from the dried plant with the same extraction method and the same fraction (2_df-SD_1-2).
Regardless of the fact that PO was the compound driving the MS chemotype, among all the EO samples, NPL showed the highest values, with a percentage up to 79.8%, as found in EO112, obtained by HD on dried plant in the last fraction of a fractionated distillation (2_df-HD_6-24). In fact, NPL was always abundant in all the EOs obtained in the longer distillation, with percentages starting at 4.30%. Correspondingly, in the first fractions, NPL was always present at low percentages or even absent in a few samples (EO019, EO020, EO057, EO078, EO079, and EO093). Nepetalactones are atypical monoterpenes of the iridoid group produced by some plants as defense compounds. The term itself includes several analogous stereoisomers. NPL has been found among the main constituents of these EOs. Its significant amount was usually reported after the first 2 h of extraction, reaching a maximum in the last 18-hour fraction: 16.4–79.8% and 4.9–27.7% in HD and SD, respectively. This is also evident in the EOs obtained by the continuous distillation processes: the NPL content increased with the prolongation of the extraction, thus significantly enriching the chemical composition of the samples obtained by long distillations. For example, 33.6% of NPL was reported in EO053 obtained by the 3–24 h HD extraction and 32.8% in sample EO017 obtained by the corresponding SD. There was no difference in yield between HD and SD. In addition, the drying of the plant material did not seem to have a significant effect on the accumulation of NPL. However, it is interesting to note that the EOs extracted from the plant material of the second harvest were significantly more abundant in NPL.
Along with glycosylated iridoids, nepetalactones are found in many Lamiaceae species, the most common source being
Nepeta cataria L. (catnip), which is known for its feline-attracting properties [
36]. These lactones are produced via a monoterpenoid pathway involving several oxidation processes on geraniol and a two-step enzyme-controlled cyclization leading to the formation of specific stereoisomers [
36,
37,
38,
39].
Mentha species are not considered to be an important source of nepetalactones; they are usually present in traces or low amounts in their EOs [
40]. However, there are some reports of their significant content in
M. longifolia (L.) Huds. [
41,
42] and even in MS [
43,
44]. On the other hand, several studies have shown a possible biosynthesis of nepetalactones from some typical mint monoterpene constituents, such as pulegone (PUL) [
45,
46], CPO [
45], and LIM [
47,
48]. Considering the results presented here, it is possible that some transformations may occur due to the prolonged exposure to heat during the extended distillation process. In fact, LIM, for example, was mostly isolated during the first hours of extraction, while the NPL content became significant after this period.
As with PO, PCY and PIP were also found in each sample at concentrations ranging from 0.2% (EO011) to 47% (EO063) and 0.1% (EO098) to 32.9% (EO015), respectively. PCY appears to be randomly distributed across the EO samples, with a slight tendency to accumulate in the later fractions. Instead, a slightly different profile was observed for PIP, with this monoterpene ketone being characteristic of the latter fractions as its percentage reached a maximum after the first 3 h of extraction. This was particularly emphasized by the continuous distillations, in which the PIP content increased its percentage in the last 6 h, thus representing a chemical marker for long-lasting MS distillations. Interestingly, by comparing HD and SD, it could be excluded that SD allows a better extraction of PIP than HD, especially in the case of continuous SD, where its content was the highest. Regarding LIM and CPO, although they have a frequency of 69% and 75% (as shown in
Table 5), their percentage reaches a maximum of 30.8% and 17.70%, with averages of 3.5% and 3.0%, respectively. Contrary to PCY and PIP, generally higher LIM concentrations were found in the first hour of distillation (30.8% as in 2_ff-SD_0-1), decreasing with increasing distillation time until disappearing in the last 18 h. A similar profile was also observed for CPO and its
trans-geometric isomer (TPO), which were usually found in the first hours of extraction, particularly in the case of fractionated HD and SD, reaching up to 9.3% and 6.3%, respectively. Furthermore, in some samples from the continuous extractions, CPO and TPO participated to a large extent. For example, CPO was an important component in samples EO002, EO019, EO021, EO037, EO038, and EO039 (10.1–17.7%), and TPO reached a maximum of 20.1% in sample EO001. All of these samples were obtained by extractions lasting a maximum of 3 h. Continuous distillation processes showed that the content of PO and CPO (as epoxy forms) decreased with the extension of the extraction process, whereas the content of PIP increased. Considering the monoterpene biosynthetic pathway in
Mentha species [
49] and the very close relationships between the mentioned compounds, it is most likely that the longer extraction processes allowed different types of degradations or/and transformations, thus significantly changing the chemical outfits of these EOs.
The appearance of other constituents is related to the extraction time, but some of them are often present in significant amounts. A group of other monocyclic monoterpenes with a menthane skeleton should be mentioned here. LIM was often found in significant amounts in the EOs obtained by fractionating HD and SD. It was mostly isolated at the beginning of the extraction and significantly influenced the chemical composition of the first three fractions (up to 30.8%). The results led to the conclusion that SD was more potent in yielding LIM and that drying of the plant material seemed to enhance LIM accumulation. Interestingly, these findings were not supported by the results obtained with continuous distillation processes: LIM was not among the major compounds, although it was present in many of these samples.
The longer HD and the presence of two unsaturated double bonds make the LIM structure suitable for transformation into its numerous derivatives [
50]. LIM could be the ancestor of the group that exemplifies various oxygenated forms, such as the aforementioned PIP, PO, CPO, and TPO [
51]. However, a group of terpinenes is also formed from the same LIM-transformed intermediates, including γ-terpinene, which is thought to be the precursor of the phenolic derivative thymol (TYM) [
52,
53]. Alternatively, the same carbocation can be quenched by a water attack, resulting in the formation of α-terpineol. The formation of a heterocyclic ring on this alcohol leads to its conversion to 1,8-cineole (
syn. eucalyptol, EUC) [
51]. Significant amounts of EUC and TYM were found in the samples obtained by fractionated distillations. EUC was particularly abundant in those obtained by SD (up to 14.8%), mainly related to the first hours of extraction. In contrast, higher amounts of TYM were reported for the EOs isolated by HD (up to 11.4%), usually appearing in the last hours of extraction. Significantly lower amounts of EUC and TYM were found in EOs obtained by continuous HD and SD methods: up to 6.3% and 6.5%, respectively.
Oxidative processes of p-cymene, a by-product of TYM biosynthesis, lead to the formation of p-cymenene (CYN) and p-cymen-8-ol (PCY), both of which were found to be important constituents of the EOs analyzed. CYN was extracted much better with the fractionated distillations, especially with SD. In some of these EO samples, its content reached its maximum, e.g., 12.3% (EO086) and 44% (EO087). PCY was present in each sample, and it is interesting that those EOs extracted from the plant material of the first harvest were much more abundant in this compound. Furthermore, drying of the material increased its content up to more than 10 times: e.g., 1.6% in the third fraction (EO075) isolated from the fresh material by fractionated SD, and 22.9% in the same fraction (EO085) but obtained from the dried material. This aromatic monoterpenoid was mainly extracted between the second and sixth hours of extraction, although some of the long-continued extractions gave samples very rich in it, e.g., 18th hour EO063 and 22nd hour EO061 with 47.8% and 38.9% of PCY, respectively.
Some compounds from the sesquiterpene group are worth mentioning since they appeared in a large number of EO samples: e.g., α-cadinol (up to 4.1% and 2.8% in the EOs obtained by continuous and fractionated extractions, respectively), trans-caryophyllene (up to 6.1% and 5.9% in the EOs obtained by continuous and fractionated extractions, respectively), and its oxide (up to 5.5% and 2.7% in the EOs obtained by continuous and fractionated extractions, respectively).
Regarding other EO components, some were found in significant concentrations only in certain samples. For example, shisofuran in sample EO036 (5.2%), phytol in sample EO018 (7.6%), trans-phytol acetate (6.8% and 7.3% in EO072 and EO043, respectively), and 6-hydroxy carvotanacetone (6.6% and 5.1% in EO027 and EO009, respectively). Interestingly, a further 2-hour SD of dried MS material from the first harvest gave sample EO020 an abundance of compounds not significantly present in other EO samples: cis-mercapto-p-menthan-3-one (16.5%), citronellyl propanoate (10.5%), camphene hydrate (7.2%), and cis-carvyl acetate (5.5%).
It is important to mention that various structural changes of thermally labile terpenes can occur during extraction. Moreover, this is an expected scenario considering that the whole distillation process was often prolonged, so hydrolytic and/or oxidative degradations and/or transformations may have occurred [
54]. For example, six-membered rings containing one or two double bonds may undergo dehydrogenation to form an aromatic system; this transformation has been reported for LIM and α-terpinene [
55]. As a result, p-cymene, CYN, or TYM may be formed. Another possible modification is the formation of epoxides; e.g., it has been reported that α-terpinene can be converted to EUC. In fact, the results presented here showed the opposite evolution of LIM and some other compounds that could be considered as its degradation products, such as the TYM of CYN: the former decreased and the latter increased with the extraction time. Furthermore, the lactone structures are expected to degrade easily under prolonged distillation conditions. As bicyclic monounsaturated terpenoids, nepetalactones undergo primary oxidation to an unstable product and are then converted to stable secondary oxidation products such as alcohols, aldehydes, ketones, epoxides, or acids. Heat and light have been found to promote the cleavage of the unique double bond in NPL by epoxidation or allylic oxidation to alcohols, ketones, and aldehydes [
56]. All of these possible changes should be considered, as the EOs analyzed showed great chemical variability depending on both the type and duration of extraction.
3.3. Anti-Candida Activity
The antimicrobial potential of MSEO has been thoroughly analyzed; there is a large amount of data in the literature, summarized in a previously published review [
20]. Considering the EO chemistry of this plant depending on its origin, a large variability in its antimicrobial efficacy has been reported. For example, strong activity was observed for PUL-rich MSEO, while those rich in PO and/or piperitone oxide showed much weaker activity [
57]. The importance of the chemical structure of several monoterpenoids for the expression of biocidal activity was investigated, most of which are common components of MSEO. The results obtained showed that LIM, carvone, and menthone were significantly less potent than PO, piperitone oxide, and PUL [
57]. The authors concluded that the presence of an additional double bond in the molecules of PUL and PO (compared to the others) may be responsible for the higher potency.
Numerous studies have reported good or excellent antifungal activity of MSEO [
33,
35,
58,
59,
60]. A previous analysis of the EOs extracted from the material collected in Tarquinia (Italy) showed strong anti-
Candida activity for the samples rich in PO [
14]. However, excellent potency was also reported for the sample containing 5.61% PO, but it was abundant in other compounds such as CYN (26.64%) and cinerolone (18.96%). Therefore, this study confirmed the phytocomplex hypothesis reported in many other experimental observations, according to which the overall expression of activity may result from synergistic and/or antagonistic mechanisms and cooperative interactions between different constituents [
61]. Following this study, a further 82 EOs were tested, as presented here. The results obtained confirmed that PO can be considered the main active ingredient since the most effective EOs contained large amounts of it, e.g., EO032, EO037, EO064, EO065, EO108, and EO109 (51.3–71.1% of PO). However, the role of PO was found to be controversial considering that low antifungal potency was observed with pure PO. The results presented here are in good agreement with a previous study that demonstrated potent candidastatic and candidacidal activities of MSEO rich in PO in an in vitro experimental system [
33]. Some other authors also attributed the strong antifungal activity of various plant EOs to the high content of PO [
21,
62,
63,
64].
In addition, samples EO035, EO070, and EO072 with average PO content (11.9–38.6%) also showed good activities. These were abundant in NPL (18.1–27.2%), which can be considered to enhance the efficacy of PO. In support of this, some other EO samples characterized by low PO content but rich in NPL lacked any significant activity (e.g., EO077, EO083, EO092, and EO102). As mentioned above, nepetalactones are the most dominant constituents in the EOs of different
Nepeta species, often accounting for more than 80%. Several studies have demonstrated the high potency of these EOs against
Candida strains [
65,
66,
67,
68,
69]. However, the antimicrobial performance of these EOs and/or nepetalactones, known for their stereoisomeric diversity and associated variability in the expression of biological activities, may be difficult to predict.
Another important finding is the discovery of a cooperative interaction between the main components, which is also evident in some samples containing significant amounts of PIP, LIM, EUC, and PCY. For example, samples EO023 and EO024 showed good activity, most likely due to the large amounts of PO together with PIP (10.1% and 15.1%, respectively) and PCY (12.8% and 16.2%, respectively). The same was reported for samples EO073 and EO098, which were rich in LIM (15.3% and 10.9%, respectively) and EUC (12.8% and 14.7%, respectively) in addition to PO. Some other
Mentha species are also rich in PIP, and authors have often speculated about the importance of this monoterpenoid in exerting the potent anti-
Candida activity reported for their EOs [
62,
70,
71]. Although usually not abundant, some EOs rich in PCY showed good antifungal potential [
72,
73]. Nevertheless, other authors reported a lack of anti-
Candida activity [
10,
74]. On the other hand, as a common EO ingredient, LIM has been the subject of numerous studies, including those on its efficacy against
Candida [
75,
76]. Studies report that LIM inhibits
C. albicans growth by disrupting the cell membrane. It induces oxidative stress, leads to DNA damage, and results in cell cycle modulation and the induction of apoptosis [
77]. Furthermore, some authors have reported the enormous potential of LIM in the treatment of invasive candidiasis due to its ability to inhibit adhesion and biofilm formation [
78]. A review of the literature has revealed a wealth of data regarding the antifungal activity of EUC, including that against various
Candida strains. Low to moderate efficacy is usually reported [
79,
80,
81,
82,
83], although many authors suggest its synergistic activity with other EO components such as camphor [
84,
85]. Finally, the possibility of antagonistic effects should not be underestimated, as there are examples where large amounts of the mentioned monoterpenes did not induce any significant antifungal activity. For example, samples EO014 and EO015 were rich in PO (33.1% and 24.8%, respectively) and PIP (21.4% and 32.9%, respectively), whereas samples EO074, EO075, EO078, and EO079 were dominated by PO (23.8–29.1%) and LIM (25.5–30.8%). The authors believe that these data are important for further studies on the design of specific EO mixtures.