3.2. Estimation of Total Phenolics Content, Rosmarinic Acid Content, and Their Influence on Antioxidant Activity
Based on the HPLC-DAD analysis, there was one major compound detected commonly at 280 nm across all plant species. The main peak was identified as the RA by its specific MS and MS
2 spectrum derived from its fragmentation pattern (
Figure 2) through UPLC-Tandem MS analysis. Based on the MS and MS
2 spectrum, the fragmentation pattern (
Figure 2C) showed a peak at
m/
z 359 as the precursor ion and MS
2 spectrum exhibited the ions at
m/
z 197, 179, 161, and 135, corresponding to [caffeic acid (C
9H
8O
4) − H − CO
2]
−, [caffeic acid (C
9H
8O
4) − H − H
2O]
−, [caffeic acid (C
9H
8O
4) − H]
−, and [quinic acid (C
7H
12O
6) − H]
−, respectively.
Through qualitative analysis of selected
Lamiaceae plants, RA was observed as the main compound, as described in
Figure 3. Previously, phenolics including RA have had a high association with antioxidant activity (AOA), and this can result in the presence of diverse pharmacological properties in plants [
45]. Therefore, spectroscopic measurement was conducted for TPC and HPLC analysis detected at 280 nm was performed for quantification of RA (
Figure 4). Additionally, DPPH radical scavenging activity (DPPH) and ABTS radical scavenging activity (ABTS) were evaluated to assess their influence on antioxidant activities (
Figure 5). The distribution of TPC and RAC in
Lamiaceae plants varied significantly depending on plant species and their growth progression, as shown in
Figure 4.
In the case of AR, RAC increased during the vegetative phases (S1, 18.3 ± 4.6 mg/g; S2, 35.0 ± 6.8 mg/g; S3, 71.4 ± 5.2 mg/g) and decreased during the flowering phase (S4, 65.1 ± 10.1 mg/g). However, there were no significant differences in TPC observed from 60 DAS (
Figure 4A,B). The antioxidant activity, which was approximately 12.4 ± 2.2% for DPPH and 17.2 ± 1.8% for ABTS at 30 DAS, steadily increased and showed the highest DPPH (28.8 ± 2.3%) and ABTS (30.5 ± 1.4%) at 80 DAS at a sample concentration of 100 µg/mL and 50 µg/mL, as shown in
Figure 5A. This low trend concordance was once again confirmed through correlation analysis (Figure 7A). A strong correlation coefficient (
r) of 0.7181 (
p < 0.01) was observed between TPC and RAC, and both were positively affecting AOA. In the case of RAC, the
r values between DPPH (
r = 0.8291,
p < 0.005) and ABTS (0.8148,
p < 0.005) were similar, while TPC had a more significant positive correlation with ABTS (
r = 0.9564,
p < 0.005) than with DPPH (
r = 0.7181,
p < 0.005). A similar pattern showing a different tendency between TPC and RAC was also observed from SO. There was no dramatic change in TPC. However, in the case of RAC, it was found to increase significantly (
p < 0.05) approximately 3.88 times during the transition from S1 (30 DAS) to S2-1 (60 DAS), and this content was maintained during the vegetative phases (S2, 70–80 DAS). Not only TPC results but these secondary metabolites are also considered to affect the AOA of SO. In terms of AOA, DPPH activity was highly determined at 70 and 80 DAS with activity values of 62.9 ± 0.9 and 58.8 ± 1.5% at a sample concentration of 100 µg/mL, as depicted in
Figure 5D. Meanwhile, a non-significant difference (
p < 0.05) was observed in ABTS. A higher correlation coefficient between RAC and DPPH (
r = 0.8840,
p < 0.005) was observed than between TPC and DPPH (
r = 0.6420,
p < 0.05) (Figure 7D). Therefore, it was explained that RAC is a more important factor than other phenolics when determining the AOA of SO. In the case of MO, both TPC and RAC showed a moderate–strong correlation coefficient with growth parameters (weight and height) during the growth progression, indicating a consistent increase (
Figure 4). TPC and RAC increased from 232.5 mg GAE/g and 29.4 mg/g in S1 (30 DAS) to 323.7 mg GAE/g and 120.3 mg/g in S2-3 (80 DAS), following a similar trend that led to a moderate correlation between TPC and RAC (
r = 0.6292,
p < 0.05) (
Figure 6B). Consequently, the positive influence of both TPC and RAC on AOA was depicted. TPC showed a higher correlation with DPPH (
r = 0.9011,
p < 0.005) and ABTS (
r = 0.9220,
p < 0.005) than RAC (DPPH:
r = 0.7843,
p < 0.005 and ABTS:
r = 0.7423,
p < 0.01) (
Figure 7B). Therefore, it has been proven that the contribution of TPC to AOA is higher than that of RA.
Similar results indicating a higher association of TPC to AOA were observed for OBP. All factors were found to have a significantly strong correlation, exhibiting sufficient evidence (
p < 0.05). During the cultivation period, TPC and RAC increased from 153.7 mg GAE/g and 30.2 mg/g in S1 (30 DAS) to 369.0 mg GAE/g and 224.2 mg/g in S3 (80 DAS), and these compound changes strongly influenced AOA (
Figure 5C). DPPH and ABTS measured at 100 μg/mL and 50 μg/mL increased from 16.3 ± 0.0% and 39.4 ± 4.5% in S1 (30 DAS) to 61.1 ± 1.0% and 65.0 ± 0.7% in S3 (80 DAS).
Natural antioxidants can delay or inhibit lipid oxidation by suppressing the initiation or propagation of oxidative chain reactions [
46]. Phenolic compounds are widely recognized for their desirable antioxidant properties. They act as reducing agents, single oxygen quenchers, hydrogen donors, and chelating agents of metal ions [
47,
48]. The antioxidant activity of phenolics typically depends on the arrangement and substitution pattern of hydroxyl groups, making the presence and proportion of active compounds crucial [
46]. Although all plants in this study contained RA as the main component, their contributions to antioxidant activity were differently confirmed. According to the influence on the AOA of TPC and RAC, expressed as correlation coefficients, they could be divided into two major groups. The dominant group of the combination of phenolics included AR and SO, while the dominant group of RAC alone included OBP and MO. This was explained by the compound profiles and their respective antioxidant capacity in each plant species.
For secondary metabolites of AR, numerous phenolics such as 4-hydroxybenzoic acid, chlorogenic acid, caffeic acid, cinnamic acid, and flavonoids including quercetin, rutin, kaempferol, tilianin, and acacetin have been reported as well as RA [
49,
50]. In particular, the amount and proportion of major phytochemicals such as tilianin, acacetin derivatives, and RA present in AR were found to vary [
51,
52]. It was explained by the increase in RA and decrease in other components (quercetin, tilianin, acacetin, etc.) during the flower development process [
49,
52]. The higher phenolic levels in flowers produced stronger antioxidant activity than in the stems and leaves, resulting in an improvement of the biological activities of the reproductive stage [
53]. Therefore, the highest antioxidant activity observed at 70 DAS and the significant contribution of RA to this activity can be well explained by the entry into the reproductive stage. In the case of SO, the previous literature reported that carnosol [
54], quercetin derivative [
55], and campherol [
56] are also included in SO as secondary metabolites. Furthermore, abietane-type diterpenoids, such as carnosic acid and carnosol, along with RA, significantly contribute to the antioxidant activities of SO [
57]. However, during the transition from the vegetative stage to the reproductive stage, there is an opposite trend in the accumulation of active compounds, with an increase in phenolic diterpenes (such as carnosic acid and carnosol) and flavonoids (such as apigenin, hispidulin, cirsimaritin, and naringin) and a decrease in RA [
58]. Additionally, Lu and Yeap Foo [
59] explained that flavonoids exhibit comparatively weaker antioxidant activity (AOA), while RA derivatives display stronger AOA, than Trolox in SO. During the experiment, SO did not enter the flowering period, so it was considered as a vegetative state with RA being dominant. Therefore, it was explained that the effect of RA on antioxidant activity was greater.
Meanwhile, MO and OBP showed other antioxidant active compounds’ effects on AOA. MO is known for containing RA as its main compound, along with other phenolic compounds including ferulic acid, gallic acid, chlorogenic acid, syringic acid, p-coumaric acid, and caffeic acid, which are present in high proportions [
60]. Caffeic acid, in particular, is found as another prominent phenolic in MO [
61,
62] and is more effective in inhibiting lipid oxidation and oil-in-water emulsion oxidation activities than RA [
63]. Phenolic compounds as well as flavonoids are important active compounds of MO, although their profiles vary depending on the variety. Abdellatif et al. [
64] explained that several flavonoids, including quercetin, luteolin, and kaempferol, are dominant active compounds, with relative contents exceeding 1%; among these, quercetin has excellent DPPH scavenging ability comparable to RA. Therefore, the relatively high amount of flavonoids in MO, showing AOA, was the reason for the lower correlation between RAC and AOA than between TPC and AOA [
65]. This similar trend is also observed in OBP, which is reported to contain other phenolics such as caffeic acid, caftaric acid, chicoric acid, etc. [
66]. Their distribution is differently reported following the varieties [
67] but, in the case of OBP used in this experiment, RA dominated the composition, with no other major components detected at a wavelength of 280 nm, which is generally used for detecting phenolics (
Figure 3C). While the color of OBP is typically attributed to anthocyanins, their content and composition could be responsible for the red and blue pigmentation in plants [
66]. However, there have been reports that explain the low contribution of anthocyanins to the AOA of basil. Therefore, focusing on the presence of a high amount of RA was sufficient to explain the high correlation between RAC-AOA and TPC-AOA [
66,
68,
69].
3.3. Profiling of VOCs and Their Distribution
Lamiaceae plants have a small number of primary ingredients that contribute significantly to their characteristic fragrance by comprising more than 40% of the VOCs’ content [
70]. These VOCs have gained recognition as promising metabolites, known for their safety and applications in antioxidant, antibacterial, and antimutagenic activities with non-toxicity. Ensuring the uniformity of VOCs is essential for maintaining quality due to their high variability. For the VOCs, not only the genetic variability, but also plant parts, different developmental stages, and environmental conditions regulate the VOCs content [
21]. The profile of VOCs is strongly influenced by plant species such as thyme (
Thymus vulgaris) of thymol [
71], rosemary (
Salvia rosmarinus) of α-pinene [
72], and peppermint (
Mentha piperita) of carvone [
73,
74]. The essential oil content was influenced by the growth stage in thyme [
26,
27], oregano (
Origanum vulgare) [
28], and rosemary [
23,
24], and these changes have also been confirmed to influence the antimicrobial [
30] and AOA [
31] of the
Lamiaceae plants. To profile the VOCs, an SPME analysis was conducted, and the results varied among the different plant species (AR:
Table 3; MO:
Table 4; OBP:
Table 5; SO:
Table 6). The relative abundance of identified peaks detected by GC-MS was calculated as a percentage of the total area of identified peaks, with peak areas normalized using the internal standard, 3-pentanol. Significant fluctuations in the VOCs of the selected plants based on sampling dates were observed.
One commonality among all plant species was the presence of (
E)-2-hexanal, a naturally occurring volatile C6 aliphatic aldehyde compound [
75]. It acts as a germination inhibitor [
76] and accumulates as a product of lipid peroxidation during the germination process [
77]. These aldehydes result from the degradation of hydroperoxides, which are produced from the conversion of unsaturated fatty acids compositing the main membrane of lipids susceptible to peroxidation into free radicals and C6/C9 hydroperoxides [
78]. The accumulation of (
E)-2-hexanal at 30 DAS indicated the germination process, while its decrease after 60 DAS signified the transition to the growth stages. Additionally, for the dominant volatile compounds, all the plant species showed an increased tendency in the initial vegetative stages, followed by a decrease in total VOCs content over time, coinciding with a decrease in the main component.
In the case of AR (
Table 3), a total of 16 compounds were identified and these consist of aldehydes (2), fatty alcohol (1), monoterpenes (4), sesquiterpenes (3), terpene alcohols (2), and phenylpropenes (4). A significant effect of the growth stage on total VOCs was observed by varied total VOCs content. It varied from 71.7 ± 7.0 ng/mg at 70 DAS to 243.4 ± 21.3 ng/mg at 60 DAS. As depicted in
Figure S3, the phenylpropene group constituted a substantial proportion, ranging from 96.7% to 97.8% of the total VOCs. The dominance was primarily attributed to the presence of estragole, the most abundant compound. Changes in estragole content influenced the overall total content, with a notable increase up to 60 DAS (S2), followed by a sharp decrease. AR proved to be a rich source of estragole, as indicated in
Table 3. Estragole is a well-known aroma ingredient widely used in food products as a flavoring agent [
79], and it accounted for a significant proportion, ranging from 46.7% to 94.6% in AR [
80,
81].
In statistical analysis, estragole exhibited a very strong positive correlation with the total VOCs content, with a correlation coefficient (
r) of 0.9999. The data presented indicated that the total VOCs content (S2: 242.4 ± 21.3 ng/mg; S1: 207.3 ± 14.3 ng/mg; S4: 94.2 ± 11.7 ng/mg; S3: 71.7 ± 7.0 ng/mg) tended to decrease over time in tandem with the diminishing estragole content (
Figure 6A). As a result, the maximized stage (S2) was considered the most efficient for obtaining estragole. EOMT (eugenol-O-methyltransferase) activity and eugenol accumulation have a significant correlation and the decrease in the transcript expression level of EOMT with leaf age [
82] was expected to be responsible for the decrease in the total amount of VOCs.
In MO (
Table 4), 18 VOCs were identified, including aldehydes (3), monoterpenes (4), sesquiterpene (1), terpene alcohols (3), phenylpropene (1), ketones (2), furan (1), ester (1), lactone (1), and fatty alcohol (1). Monoterpenes composed of isoneral, (
Z)-neral (citral B), geranial (citral A), and methyl geraniate accounted for approximately 87% of the VOCs (
Figure 6B) and geranial was identified as the major component of MO (
Table 4). Specifically, the primary composition of VOCs in S2 stage MO, which exhibited the highest VOCs content, consisted of geranial (66.5%) and (
Z)-neral (30.1%). The combined total of these two main components represented a significant percentage in all MO samples, regardless of the stages, ranging between 87.0% and 96.8% (
Figure 6B). Consequently, the variation in total VOCs content across the stages was primarily attributed to changes in geranial and (
Z)-neral content (as shown in
Table 4), and the order from highest to lowest content was as follows: S2-1 (60 DAS; 215.6 ± 34.1 ng/mg) > S1 (30 DAS; 166.0 ± 4.2 ng/mg) > S2-2 (70 DAS; 35.2 ± 12.6 ng/mg) > S2-3 (80 DAS; 20.6 ± 11.1 ng/mg). In the case of MO, some previous reports have referred to the major components of VOCs of MO as the geranial, (
Z)-neral, citronellal, (
E)-caryophyllene, caryophyllene oxide, geraniol, etc. [
83]. Among these diverse monoterpenes, monoterpene aldehydes (geranial and neral) were identified as the main component of MO. Although MO exhibited relatively higher levels of geranial compared to (
Z)-neral, both compounds made significant contributions. These two stereoisomers of citral (mixture of geranial and neral) are also found in citrus fruits, lemongrass, and gingers to determine quality [
84], and were highly quantified in the S2-1 stage and then declined. Total VOCs content was statistically concerned with geranial and (Z)-neral with correlation coefficients of 0.9866 (
p < 0.005) and 0.9980 (
p < 0.005), respectively. Consequently, to obtain MO with a rich VOC content, the optimal stages were S1 (30 DAS) and S2-1 (60 DAS).
In contrast to the two previously discussed plant species, OBP and SO did not exhibit remarkable major VOCs. In OBP, 32 VOCs, which included aldehydes (4), monoterpenes (9), sesquiterpenes (6), terpene alcohols (8), phenylpropenes (4), and fatty alcohol (1), were found, as described in
Table 5. Their distribution varied across the different growth stages, with S2-1 (60 DAS; 259.0 ± 14.9 ng/mg) containing the highest VOCs, followed by S1 (30 DAS; 221.6 ± 17.2 ng/mg), S3 (80 DAS; 48.5 ± 7.3 ng/mg), and S2-2 (70 DAS; 11.1 ± 3.8 ng/mg), respectively. Notably, the terpene alcohols and phenylpropenes displayed opposite patterns as the days progressed. The proportion of phenylpropenes steadily decreased from 93.5% to 47.9%, while the terpene alcohols’ portion consistently increased from 3.1% to 37.1% over time (
Figure S3). The primary aromatic compounds in OPB were eugenol, methyl eugenol, and linalool (
Table 5). During the S1 (30 DAS) and S2-1 (60 DAS) stages, methyl eugenol was the major component, accounting for 58.2–74.9%, with lower amounts of eugenol (17.9%) and linalool (1.9%). However, during the S2-2 (70 DAS) stage, eugenol, and linalool content increased, reaching 35.6% and 31.9%, respectively, making them the most abundant compounds (see
Figure 6C). This composition of OPB consisted of reported paper [
85]. Methyl eugenol was the dominant constituent of OBP, followed by eugenol.
The strong correlation between these two phenylpropenes (
r = 0.8458,
p < 0.005) suggests the possibility of conversion from eugenol to methyl eugenol through the catalysis of the EOMT gene [
86]. Both eugenol and methyl eugenol exhibited a very strong correlation with total VOCs content (eugenol:
r = 0.9451,
p < 0.005; methyl eugenol:
r = 0.9678,
p < 0.005, respectively). Meanwhile, the other major compound, linalool, which is one of the terpene alcohols, showed a weak correlation with total VOCs (
r = 0.2844,
p < 0.005). Therefore, eugenol and methyl eugenol were revealed as the principal factors to determine total VOCs emission in OBP. The highest content of phenylpropene compounds was observed during the initial vegetative stage (S2-1). This finding is consistent with the results of Renu et al. [
87], who reported an increase in total VOC content during the juvenile stage (S2-1), followed by a rapid reduction during the pre-flowering and flowering stages in Ocimum species.
In SO, 34 VOCs, including aldehydes (3), monoterpenes (17), sesquiterpenes (2), terpene alcohols (9), phenylpropenes (2), and fatty alcohol (1), were found, as described in
Table 6. Several compound groups constituted a large portion (more than 10%) of the total VOCs. Monoterpenes (63.4–79.2%), sesquiterpenes (10.1–15.6%), and terpene alcohols (9.1–18.2%) were associated with the aroma of SO (
Table 6 and
Figure S3). Unlike AR and MO, where several components determine the total amount, the sum of some of the main compounds, such as (
Z)-thujone (28.4–43.1%), camphor (15.5–29.4%), and humulene (8.8–13.7%), accounted for approximately only half (
Figure 6D). The total VOCs content varied across different growth stages, including S2-1 (60 DAS; 128.5 ± 8.5 ng/mg), S1 (30 DAS; 94.5 ± 7.5 ng/mg), S2-3 (80 DAS; 75.9 ± 12.8 ng/mg), and S2-2 (70 DAS; 61.5 ± 3.4 ng/mg) (
Table 6). Interestingly, there were no remarkable major VOCs in SO. Without overwhelming compounds, diverse volatile compounds were evenly distributed, and relatively predominant volatile compounds (making up more than 8% of the total VOCs content) included two monoterpenes, (
Z)-thujone and (−)-camphor, and one sesquiterpene, humulene. Among three high-ratio components, in correlation analysis, only (−)-camphor showed a very strong correlation coefficient (
r = 0.9638,
p < 0.005), while others ((
Z)-thujone and humulene) exhibited comparatively lower correlation coefficients (
r = 0.7560 and 0.7322, respectively). The aromatic properties of SO have been reported to be composed of thujone diastereomeric forms (α-thujone and β-thujone), 1,8-cineol, camphene, humulene, α-pinene, limonene, and borneol, among others, varying with the species. These compounds make SO a commonly used savory food flavoring in the form of dried leaves and essential oil [
88,
89]. Therefore, the worth of the S1 and S2-1 stages was evaluated to be higher due to higher total VOCs content.
For the
Lamiaceae plants, VOCs are important quality indicators. However relatively little is known about the effect of growth stages on VOCs distribution. Some studies have provided evidence that some VOCs stimulate seed germination and seedling stages [
90]. Thus, young stages show dramatic changes in VOCs emissions, and leaf ontogeny can greatly influence VOCs production [
91]. This study indicated that variations of the VOCs emitted by
Lamiaceae species significantly depend on the growth stage. Although the main components and their proportions were all different, the total VOCs of four
Lamiaceae plants showed the highest content while passing from the S1 to S2 stage. Consequently, it is suggested that the seedling stages of
Lamiaceae plants under specific periods (leaf development to stem elongation) can serve as rich sources of VOCs.
Figure 7.
Heatmap illustrating the correlation analysis of (A) AR, (B) MO, (C) OBP, and (D) SO generated by Pearson’s correlation coefficient (r) between phenolic relative factors, antioxidant activities, and VOCs. TPC, total phenolics content; RAC, rosmarinic acid content; DPPH, DPPH radical scavenging activity at 100 µg/mL; ABTS, ABTS radical scavenging activity at 50 µg/mL; VOCs-Etc, VOCs content excluding main components.
Figure 7.
Heatmap illustrating the correlation analysis of (A) AR, (B) MO, (C) OBP, and (D) SO generated by Pearson’s correlation coefficient (r) between phenolic relative factors, antioxidant activities, and VOCs. TPC, total phenolics content; RAC, rosmarinic acid content; DPPH, DPPH radical scavenging activity at 100 µg/mL; ABTS, ABTS radical scavenging activity at 50 µg/mL; VOCs-Etc, VOCs content excluding main components.